scamp usb manual

Default Page

SCAMP Manual

Introduction

Software Installation

Cautions

Frequently Asked Questions

SCAMP USB MANUAL

PROFILING WITH SCAMP

--Profiling with SCAMP

--Establishing Communication

--Checking the Batteries

--Assembling the Stand

--Upwards Profiling

--Downwards Profiling

--Aborting a Profile

--Assembling the Fluorometer

--Data Upload & Storage

--Velocity & Gradient Gain

Adjustment

--Disassembly & Storage

SCAMP CONTROL DIALOG

--Overview

--Data Upload

--Mission Tab

--System Tab

--Test Tab

--Channel Tab

--Setup Tab

HARDWARE

--Overview

--Fast Temperature Sensor

--Conductivity Sensor

--Fast CT Sensor

--Accurate CT Sensor

--Turbidity Sensor

--Laser Turbidity Sensor

--Pressure Sensor

--Gradient Channels

--Fluorometer Sensor

--PAR Sensor

--Offset & Gain Blocks

--Additional Sensors

--A/D Converter

http://www.pme.com/USB_smanual/default.htm (1 of 2) [10/5/2005 5:09:49 PM]

Default Page

--CPU

--Memory

--References

CALIBRATION

--Conductivity

--Depth

--Fluorometer

--Laser Turbidity Sensor

--References

MAINTENANCE

--Sensor Cover

--Opening SCAMP

--Inspecting O-rings

--Electronics Covers

--Closing SCAMP

--Testing for Leaks

--Replacing a Sensor

--Replacing the Batteries

--Replacing the Fuses

--Conductivity Sensors

--Pressure Transducer

--I/O Connector

--Installing a Dummy Sensor

--Spare Seal Kit

TIPS

--Tips & Techniques

APPENDICES

--Specifications

--Wiring Chart

--SCAMP Seals

--Sensor Connection Drawings

--A/D Input Connections

--WINKLER.PAS

--Calculation of Engineering

Units

--

RS232 Communication

--

Download a New OS

--

L.E.D. Indications

MatLab Software

SCAMP Library

Warranty

Revision History

http://www.pme.com/USB_smanual/default.htm (2 of 2) [10/5/2005 5:09:49 PM]

SCAMP introduction

Precision Measurement Engineering,

Inc.

Introduction

The Self Contained Autonomous MicroProfiler (SCAMP) is a portable instrument designed to measure, record, and display extremely small scale (order 1mm) absolute values and fluctuations of electrical conductivity, temperature, oxygen concentrations (under development), and other parameters, in lakes, reservoirs, estuaries, and the oceans. These data can be used to infer the levels of dissipation of turbulent kinetic energy, in-situ fluxes of heat, salt, and oxygen, and the microstructure behavior of these parameters

The SCAMP will make measurements while either descending or ascending through a column of water. This instrument is provided with a flotation and drag system designed to move it through the water column in either direction at roughly 10 centimeters/second. These two deployment modes are provided to allow measurement either of the water column completely up to the surface, or the water column down to a few centimeters above the bottom. The SCAMP must be tethered to the operator on the surface in some way although this tether need not be an electrically conducting cable. The SCAMP is a small, lightweight unit that may be deployed from even a small dinghy.

When used in ascending mode, an expendable weight is connected to the SCAMP and the probe unit is released into the water. The SCAMP is designed to sink at a 45 degree angle, sinking under the water it will later measure. When the SCAMP sinks to the proper depth, then the weight is released and the SCAMP floats vertically upward measuring and recording the various parameters. The measurements are transferred to the HOST program operating on the laptop computer where they may be viewed and stored on the laptop disk drive.

When used in descending mode, the SCAMP is permanently weighted and released directly downward into the water. Upon reaching the bottom, the SCAMP is hauled back and the measurements are transferred, viewed, and stored.

The SCAMP is a descendant of a previous microprofiler designed and built at the Centre for

Water Research within the University of Western Australia located in Perth, Australia. The

SCAMP is produced by PME under license from the Centre for Water Research.

http://www.pme.com/USB_smanual/intro.htm [10/5/2005 5:10:05 PM]

Installing SCAMP Software

Precision Measurement Engineering, Inc.

Installing the SCAMP Software

SCAMP is supplied with software for SCAMP control and for data analysis.

SCAMP control is only implemented under Windows 2000 or Windows 98 SE (not recommended!!).

Communication with SCAMP is implemented via USB 1.0. No other operating systems or communication modes are presently supported.

Data analysis is supported in conjunction with the Mathworks program Matlab, version 5.1 and higher.

Matlab is not included in the SCAMP package and must be supplied by the user. PME’s Matlab software is a combination of *.m and mex *.dll files. PME supports Matlab on PC platforms, but the mex files will not operate on other platforms. PME will, at our option, provide source for the mex files so that these can be re-compiled on other platforms. Successful compilations for other platforms have been accomplished by various SCAMP users and may be available via the SCAMP user’s group.

DISK FILES

SCAMP software may be supplied on various media including zipped files. The file list shown below shows the file arrangement that the user must create from the supplied media. Various activities such as unzipping and copying may be required to create this arrangement. This arrangement must be created prior to performing the installation steps below.

C:\SCAMP directory

Control

WCT32DR3

ReadMe

Users

USBDriver exe

DLL

DOC

DOC

MATLAB

EngUnit

Flash

Minimon

USB_smanual

SN00xx

Windows dialog for SCAMP control

Graphics plotting library

Description of files in this directory

Contact information for SCAMP users

Folder containing USB drivers for Win98, Win2000, WinXP

Folder containing Matlab *.m, and *.dll files

Software to convert SCAMP profiles to ASCII files of engineering units

SCAMP's internal operating system

Program used to install SCAMP's operating systerm

This operator's manual in HTML format

Information specific to SCAMP SN00xx (calibrations, notes.doc)

INSTALLATION OF THE SCAMP CONTROL DIALOG

The SCAMP control dialog must be installed onto a PC compatible computer having at least:

500 MHz internal clock http://www.pme.com/USB_smanual/software.htm (1 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

USB 1.0 port

256 Meg RAM

Large capacity HDD

The following instructions are given for Windows 2000, but are substantially the same for Windows 98

SE.

Installation begins by first copying files and folders shown above from the supplied media.

Next, the USB drivers are installed. Connect SCAMP to the host computer via the supplied USB connector. After several seconds, the Found New Hardware Wizard should begin operation.

Click "Next".

http://www.pme.com/USB_smanual/software.htm (2 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

Select "Search for a suitable driver for my device (recommended)". Click "Next".

http://www.pme.com/USB_smanual/software.htm (3 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

Select "Specify a location" from the check boxes shown. Click "Next".

http://www.pme.com/USB_smanual/software.htm (4 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

The most likely case is that the proper folder will not be shown. Click "Browse" and locate the proper folder. In the screen shots that follow the proper directory was located on CD in the D: drive.

However if the instructions above were followed the proper folder will appear under C:\SCAMP.

http://www.pme.com/USB_smanual/software.htm (5 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

Select the proper folder for the operating system you are using.

http://www.pme.com/USB_smanual/software.htm (6 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software http://www.pme.com/USB_smanual/software.htm (7 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

At this point you have located the proper SCAMP.inf file. Click "OK". the Wizard will complete its job and display a completion screen. Click "Finish" and you are finished installing SCAMP's USB driver.

You must register a dll for use by the SCAMP Control Dialog program. Select Start from the lower left screen. Select Run from the pop-up menu.

Enter "regsvr32 winrtusb.dll". Click OK.

This completes the installation of SCAMP’s control software. Test the installation by running the

Control.exe program in C:\SCAMP with SCAMP connected to the USB port. The program should display its control dialog, shown below.

http://www.pme.com/USB_smanual/software.htm (8 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

If the following screen appears, then the driver installation has failed. Contact PME. (Note that it takes SCAMP about 4 seconds to be recognized by the host computer's USB controller after SCAMP is connected. The screen below will appear if the SCAMP Control Dialog is run during this time. If this is the case, click OK and re-run the SCAMP Control Dialog.) http://www.pme.com/USB_smanual/software.htm (9 of 10) [10/5/2005 5:10:20 PM]

Installing SCAMP Software

INSTALLATION OF THE SCAMP DATA ANALYSIS SOFTWARE

C:\SCAMP\Matlab contains the data analysis software. This software is compatible with Matlab V5.2

or later, which must be supplied by the customer. No special installation of SCAMP data analysis software is required. There are two zipped files. SCAMPGUI contains the main processing software in Matlab script format. These routines can be read and modified by customers. SCAMPGUI should be unzipped into any convenient directory. SCAMPTOOL.ZIP contains Mex format sub-routines that implement specific processing steps. These can not be viewed or changed by the customer and must be run on a P.C. platform. These should be unzipped into any directory in Matlab's path. A recommended directory is C:\matlab11\toolbox\scamptool\ since this is added to Matlab's path in s_process.m.

Software may be tested on a PC by beginning Matlab operation, changing directories to the directory containing the unzipped SCAMPGUI.ZIP and running SCAMP.M. Use the LOAD button on the screen to load the ‘_1820.raw’ example data file. The following screen should appear after data processing. Click to enlarge.

INSTALLATION IN OTHER DIRECTORIES

The C:\SCAMP\ directory arrangement is not absolutely required. After the USB driver is installed, the

USB Driver directory may be deleted. The remaining files may be moved to any convenient directory.

Matlab files may be placed anywhere that Matlab can find them using Matlab’s path rules.

S_process.m contains a convenient path statement that may be used to implement a Matlab toolbox for SCAMP.

http://www.pme.com/USB_smanual/software.htm (10 of 10) [10/5/2005 5:10:20 PM]

Cautions

Precision Measurement Engineering,

Inc.

Cautions

SCAMP has points where customers sometimes have difficulties. Here is an incomplete list!

SCAMP'S SHIPPING BOX

SCAMP will survive only 10g accelerations. The box, foam pads, and popcorn are designed to protect SCAMP from a drop onto concrete from a 5 foot elevation. PME has shipped SCAMP in this box for 10 years with good results. Do not discard the box when you receive SCAMP.

Retain it and use it to store and ship SCAMP. Do not make your own protective case, especially do not use plastic cases with foam inserts.

SCAMP's RETRIEVAL LINE

http://www.pme.com/USB_smanual/cautions.htm (1 of 9) [10/5/2005 5:10:33 PM]

Cautions

SCAMP's retrieval line is constructed of Spectra, 500 pound test. You cannot break it under tension. However Spectra can be cut and can be melted by excess heat. When using SCAMP insure that the line will not come against any sharp points anywhere on the boat or other places.

Inspect the retrieval line from time to time for frays, especially at the point shown by the pencil in the picture above. If you make repairs to the line and use heat shrink tubing, be very careful not to melt the line when shrinking the tubing.

SCAMP's 9V BATTERIES

SCAMP's 9V batteries implement the + and - analog power. These batteries rest against http://www.pme.com/USB_smanual/cautions.htm (2 of 9) [10/5/2005 5:10:33 PM]

Cautions

SCAMP's digital card (the bottom of which is visible when the batteries are removed).

Be absolutely certain that you plug the batteries properly onto the connectors. Do not even momentarily touch them in reverse position to the connectors. The batteries will not plug on this way, but they will attempt to power the analog circuits with reverse polarity. No permanent damage, but you will have to replace the analog fuses located in the wiring to the connectors.

Fuses are the small yellow cylinders shown in the picture above.

Do not press the batteries deeply into the chassis. This places pressure on the digital card and can cause the connector from the digital card to the analog card to become partially unplugged.

No permanent damage, but SCAMP will fail the analog test. If this happens, then squeeze the analog and digital cards back together to re-connect the connector. Lithium 9V batteries are somewhat larger than alkaline and are especially prone to this problem.

Remove these batteries if SCAMP will not be used for long periods.

SCAMP's AA BATTERIES

SCAMP's AA batteries implement the digital power. Note that there are 6 batteries required and that they are placed into the pack with different orientations. There is a legend within the battery pack that shows the orientation. Be sure you use all 6 batteries and that they are placed into the battery pack properly.

These batteries are heavy and are held in position by spring contacts. If SCAMP is accelerated axially, then the batteries will press against the spring contacts and may break electrical contact.

Axial accelerations may cause SCAMP to behave erratically.

SCAMP consumes about 1 mA from this pack during periods of in-activity. This limits the endurance of this pack if SCAMP is shelved for days.

http://www.pme.com/USB_smanual/cautions.htm (3 of 9) [10/5/2005 5:10:33 PM]

Cautions

Remove these batteries if SCAMP will not be used for a long period.

Lithium AA batteries work fine in this battery pack.

SCAMP's BUBBLE TEST

SCAMP's bubble test adaptor (shown above on a SCAMP with a PAR I/O connector, but similar on standard SCAMPs) screws into 10-32 threads that are tapped into the plastic end cap. These threads are exposed when the bubble test port screw is removed. This screw is shown in the picture above. These plastic threads are easily stripped.

There are three points where the threads may be stripped. The first is when tightening the bubble test port screw. This screw should be tightened finger-tight. Tighten enough to completely compress the o-ring and bring the screw head into contact with the end cap. Then tighten an additional 1/8 turn.

The second point where the threads can be stripped is when screwing the bubble test adaptor into the end cap. If the squeeze bulb (not shown in picture above) and tube are connected to the adaptor, then it is very difficult to get the adaptor properly started in the threads. Disconnect the tube from the adaptor prior to starting it into the threads. Install the tube after the adaptor is fully screwed into place. The tube needs only be pressed gently and partially onto the adaptor.

The third point is when the adaptor is unscrewed from the end cap. If the squeeze bulb and tube are connected to the adaptor, then they tend to bend it sideways at the last threads and will tear them out. Remove the bulb and tube before unscrewing.

SCAMP's I/O CONNECTOR

http://www.pme.com/USB_smanual/cautions.htm (4 of 9) [10/5/2005 5:10:33 PM]

Cautions

SCAMP's I/O connector implements connection from SCAMP to the host computer USB port.

The picture above shows this connector for the PAR cable, but a similar bulkhead connector is used on a non-PAR SCAMP.

The contacts within this connector are powered and are not water resistant. Always cover this connector with the supplied rubber cover prior to placing SCAMP into the water. When connecting to your computer, carefully dry the connector after removing the cover and prior to plugging the connecting cable.

Never twist the connecting cable. Plug it straight on and pull it straight off.

If you loose the rubber cover PME has supplied a spare in the items shipped with SCAMP.

A few drops of silicone oil on the rubber from time to time will make plugging/unplugging go more easily.

SCAMP's PAR I/O CONNECTOR

http://www.pme.com/USB_smanual/cautions.htm (5 of 9) [10/5/2005 5:10:33 PM]

Cautions

SCAMP uses a special connector when a PAR sensor is installed. The connector is shown in the picture above with the cable coupled.

The circuits within this connection are very sensitive to water. Carefully dry this connection prior to disconnecting. Be sure this connector is connected either to the connecting cable or to a dummy connector when placing SCAMP into water, for example when performing the bubble test.

The connecting cable is very fragile at the point where it joins to the metal connector. There is a re-enforcement placed at this point. This is the larger diameter black cylinder just next to the titanium colored metal. Bending of this cylinder, or nearby this cylinder, will cause breaks in the internal wiring of the cable. SCAMP will become unable to connect to the host computer via

USB. Be sure to never bend the cable in this region. The most likely time that the cable will be bent is when SCAMP is fully assembled and rested upon its bottom end - where the cable connects to SCAMP. The following picture shows the connection.

http://www.pme.com/USB_smanual/cautions.htm (6 of 9) [10/5/2005 5:10:33 PM]

Cautions

PME provides a protective cylinder that slips over the cable and connector. It will keep the connection straight in the event that SCAMP is accidentally rested upon the lower end. This cylinder is secured with a small cable tie pointed by the pencil in the picture.

SCAMP's PAR I/O CABLE

SCAMP's PAR I/O cable connects from SCAMP to the PAR sensor and also to the USB I/O connector. There is a 'Y' near the middle of this cable.

The 'Y' is not very strong and does not resist bending very well. Do not bend the cable in this region. When installing this cable onto SCAMP's drag plate insure that this region is not bent. http://www.pme.com/USB_smanual/cautions.htm (7 of 9) [10/5/2005 5:10:33 PM]

Cautions

Proper installation can be seen in the previous section picture above.

SCAMP's FLUOROMETER ELBOW

SCAMP's fluorometer uses two elbows to route water through the fluorometer, one of which is shown above. Water is routed through a clear tube within the end cap. This tube is sealed at each end with an o-ring. There is no retainer provided to hold this tube in its proper location. If the tube is pressed inwards from either end, then it will release the o-ring seal and water will enter SCAMP. The fluorometer elbows each have a white sleeve the prevents the elbow from being screwed too far into the end cap and pressing upon the glass tube. However the elbow can be screwed in too far by excessive force. Only screw the elbow in to the point where the white sleeve touches the end cap. If the elbow is not in the proper orientation, then do not screw it in further. Instead, unscrew it until the proper orientation is obtained.

The glass tube end should be 0.250" in from each side of the fluorometer.

SCAMP's INTERNAL WIRING HARNESS

http://www.pme.com/USB_smanual/cautions.htm (8 of 9) [10/5/2005 5:10:33 PM]

Cautions

SCAMP's internal wiring harness (flat amber ribbon in picture above) is very fragile and is easily damaged. Most damage occurs after sharp bending of the harness nearby any of the connectors. If the harness must be disconnected from a connector, do not pull on the amber ribbon. Instead gently pry the connector apart with a sharp blade or pull the connector apart with small pliers by grasping the connector posts visable above the ribbon.

http://www.pme.com/USB_smanual/cautions.htm (9 of 9) [10/5/2005 5:10:33 PM]

Frequently Asked Questions

Precision Measurement Engineering,

Inc.

Frequently Asked Questions

Q: Should I leave SCAMP connected to USB when it is not uploading data or having its parameters changed?

No. SCAMP will not return to its sleep state if it is connected to the USB port of the host computer.

Q. I would like to mention that, when installing the Scamp software , we cannot get the "Found new hardware wizard". As the first execution of the Control.exe program indicated that it could not find the

WINRTUSB.dll file, we tried to copy it in the right directory. Then, running Control.exe gave the "No USB connection " message.

Control.exe will absolutely not operate until the USB driver for SCAMP has been installed.

What Microsoft operating system are you using?

We have tested your SCAMP with Windows 2000 and also Windows 98SE. If you are using any other operating system, try a different computer that has Windows 2000 on it. You don't have to install the SCAMP driver, just look to see if the "Found new hardware" screen appears. You can select Cancel on the Found New Hardware

Wizard screen. If Windows 2000 recognizes SCAMP you'll have to give up using whatever operating system you have that doesn't recognize SCAMP.

Instructions for installing the driver are contained within the operator's manual at the Software Installation section. These instructions depend upon the operating system recognizing SCAMP at the USB port. When you connect the SCAMP for the first time to your computer's USB port, Windows 2000 should recognize it as a un-installed device and display "Found new hardware" as shown in the manual. If this does not happen, try a different computer that has Windows 2000 on it. You don't have to install the SCAMP driver, just look to see if the "Found new hardware" screen appears. You can select Cancel on the Found New Hardware Wizard screen.

If a different computer will recognize SCAMP then you have some problem with the original computer.

If no computer recognizes SCAMP, then replace SCAMP's batteries as described in Maintenance \ Replacing the

Batteries section of the manual. Might as well get good at this since you'll do it often when operating SCAMP.

Try computers again.

If you continue to fail to see the Found New Hardware screens when you connect SCAMP via USB, then go to

Device Manager within your Windows 2000. Connect SCAMP via USB port and observe the Universal Serial

Bus controllers part of the device tree. It will take something less than 10 seconds for Windows 2000 to recognize a USB device and post this info to Device Manager.

Q.

We have been trying to get data in the downwards profiling mode, testing the Scamp functioning in the 5 gallon bucket full of water. We have defined all the conditions in the Mission Tab as indicated in the manual, we have awakened the Scamp with the Start magnet (but release motor operates once instead of

http://www.pme.com/USB_smanual/faq.htm (1 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions

twice as mentioned in the manual ?) . But we cannot get any data file. What could be wrong?

The magnet causes SCAMP to wake up. It triggers its release motor once immediately to indicate it has awakened.

During the next 2 or 3 seconds SCAMP checks internal resources (batteries and memory) to be sure that it can perform a profile. If there are sufficient resources, then it triggers its release motor twice in quick succession. It then begins to evaluate the profile start conditions you defined.

When SCAMP determines that a profile start condition is true, it begins acquiring profile data and triggers its release motor continuously for 2 seconds (to drop any weight if in upwards mode).

After starting a profile, SCAMP acquires profile data and evaluates stop conditions to determine when to stop acquiring the profile.

Profile data acquisition is halted if a stop condition becomes true or if you connect to SCAMP via USB.

Since your release motor only runs one time, SCAMP is determining that there are insufficient resources for obtaining a profile.

Q. Another question is about the Scamp Control Dialog, in the System Tab. We get nothing in the battery status (no progress bar display on the right, voltage = 0), despite the fact that they are new and that main other features are operating (connection, channel testing, rotation of the release motor, ...)

The batteries should display a reasonable value, something like 7 to 9 Volts. Since they do not, then I suspect that there is a battery problem. This would explain you problem in question 1 above.

Q. In the Test tab, when we test the Fast T0 channel for instance, it seems that the display does not give the right temperature values. The water we are using is about 18°C and the display shows us 13°C. What does this test mean exactly? Other channels are also giving dubious results.

If the batteries aren't working, then sensor values will not be correct.

Your objective is to get the battery indication on the System Tab to give reasonable values. Sensors and missions won't work until reasonable values are displayed.

SCAMP has three battery packs: two 9 V transistor batteries, and a 6 X 1.5 V AA battery group. The CPU operates from the 6 X 1.5 AA battery group. The sensors and A/D system operate from the two 9 V transistor batteries. Since SCAMP will communicate via USB I suspect that the 6 X 1.5 AA batteries are not dead. However if either 9 V battery is dead then the A/D will not work and SCAMP will not have a way to determine the voltage for any battery including the 6 X 1.5 AA. This is how it will communicate even though the battery levels it displays are not reasonable.

I suspect that you have problems with one or both of the 9 V transistor batteries or with the fuses connecting them to SCAMP.

Open SCAMP, then remove all batteries. Measure their voltage. If either battery gives an unreasonable value, then replace it. If you get reasonable results, then the 9 V battery fuses may be blown. This can happen if you momentarily touch the 9 V battery backwards onto the connector or from some other unknown cause.

See the Maintenance|Replacing the Fuses in the SCAMP manual. Prior to doing anything, remove the 9 V battery harness (red,black, white wires and battery snaps). You will probably need to take off one electronics cover (Maintenance|Electronics Covers). Note how the connector is plugged on before you remove it. You MUST plug the connector back on in exactly the same way. It is not polarized and can be plugged on in two rotations.

Measure the resistance of each fuse. You can see the fuse on the red and white wires. Red goes to a 9 V battery

(+) and White goes to the other 9 V battery (-) so you can guess which terminal in the battery snap to touch the

Ohm meter to. You should measure about 6 Ohms resistance in each fuse. If you measure greater than 10

Ohms, the fuse is blown. If either fuse is blown then find the replacement 9 V battery harness we shipped with http://www.pme.com/USB_smanual/faq.htm (2 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions

SCAMP. Measure its fuses. This is new and should give the correct 6 Ohm reading. If not then you've done the measurement wrong and must re-check things until you get the proper reading.

If you determine that the 9 V battery harness that you removed from SCAMP has a blown fuse and that the replacement harness is OK, install the replacement harness. Be sure you plug the connector on properly.

Re-install the batteries, 9 V first, then 6 X AA. Be sure you don't touch the 9 V batteries the wrong way to the connectors.

You don't have to put the electronics cover on or close SCAMP.

Connect SCAMP via USB and run the control dialog. Batteries should indicate reasonable voltages. If so, put cover back on, close SCAMP, continue with testing.

If the fuses in the battery harness you remove are ok, or if after you replace them SCAMP still doesn't give a good battery reading then there are other problems.

Please tell me what resistance you measure and whether you get SCAMP going as I've described here.

Q. Why won't the MSDOS Host program, or the new Windows 2000 Matlab software load profiles produced by the MSDOS Host program?

You sent two profiles, both recorded in August 2003. Each profile consists of three files: *.txt, *.raw, and *.cfg.

Both cfg files seem to be corrupted.

The problem is not with the new Matlab software only; the profiles do not load properly into the old Host program either. *.CFG files are very mechanical things and should have all information in exact positions within the file. I looked at historical *.cfg files from and also at *.cfg files from other SCAMP units. All historical files from all units that I reviewed have 6127 bytes. However the two *.cfg files that you sent have 6148 bytes.

As for the two files from Aug 2003 having 6148 bytes - either they were written wrong at the time they were collected or they have been corrupted during storage or handling since they were written. Since files prior and after 2003 seem correct, then I suspect storage or transport, especially if they have been transported through or stored on a Unix or Linux system.

Customer reply: Thanks for the reply. I figured out that all .cfg files are 6127 bytes in size on our MSDOS platform. It was the same even in Linux (which we prefer to use) and when I transferred the files from Linux to my

Windows laptop to use your software, it jumped up to 6148. This is really funny to me, never realized this could happen. Anyway, now I changed all my files with the correct file sizes and it seems to work pretty well.

Q. I have question about the .dll called in Matlab m files. How can a .dll file be edited (to see what it is exactly doing)?

*.dll files cannot be edited by you. You must have the source *.c file, which you edit and then compile to create the *.dll. I do not distribute the source files with SCAMP, but will provide individual files in response to a specific need. I've attached the source file for s_segmen.c so you can see what it does. Now that you have this file, you could edit it and then compile it. Type 'help mex' into the Matlab command space. I think this brings back some information about mex files. In addition to Matlab you'll need a C compiler. I use Microsoft Visual C++ 6.0.

You'll also find that the *.dll are complied specifically for use on a Windows operating system. They cannot be used with Linux or Unix or any other OS. However others have compiled them for Linux and maybe even Unix.

Q: What are SCAMP's internal power modes?

SCAMP has three principal power modes:

Awake and idle - SCAMP enters this mode when connected to USB

Awake and acquiring data - SCAMP enters this mode after activation with the start magnet

Sleep - SCAMP enters this mode when not connected to USB and when data collection is completed http://www.pme.com/USB_smanual/faq.htm (3 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions

In the Awake modes SCAMP consumes substantial battery power. Batteries will last only a few hours in this mode. In Sleep mode SCAMP consumes very little battery power. Batteries will last weeks to months in this mode.

Q: Can SCAMP be used in a moored situation?

SCAMP's main objective is to determine the turbulent rate of energy dissipation. This is done by matching observed temperature gradient spectra to theoretical Batchelor spectra for a passively transported scalar. Where matches can be found the dissipation is computed from kB.

This method works well where the water is non-thermally uniform, but depends upon knowledge of the spatial distribution of temperature since the Batchelor spectral form is a function of wave number. Unfortunately,

SCAMP can only measure temperature at fixed time intervals. The time-spaced interval is transformed to the distance-spaced interval required for Batchelor fitting by using SCAMP's velocity through the water. Since

SCAMP is a vertically profiling device, this velocity can be determined from the rate of change of hydrostatic pressure.

In a moored situation some other method of determining velocity must be provided. Yes, you can moor SCAMP but it will not, of its own measurements, be able to determine dissipation. You must provide some other way to measure water velocity past SCAMP's temperature sensor.

Assuming that you provide some independent velocity measurement there are yet more subtle problems.

SCAMP's temperature sensor cannot respond to temperature variations that occur too rapidly. This limits

SCAMP's use to roughly 10 cm/sec. This velocity is established when profiling by adjusting SCAMP's buoyancy.

In a moored situation you'll have to take what you get. If you expect 5 - 12 cm/sec SCAMP may work.

Q. I would like to use SCAMP deeper than 100m. Is this possible?

SCAMP is designed for 100 meters maximum depth. We have no experience with its use at deeper points.

SCAMP's pressure transducer is a 10 Bar device (145 PSI). At 200 meters it will be 2X range and may be permanently damaged. This can be swapped for a higher pressure sensor, but PME can only calibrate to 70 meters so the higher range would have to be extrapolated unless the customer does the depth cal.

Also , I can't promise that the system will be free of leaks or will resist implosion. My gut feeling (about 80% confidence) is that it will survive. I wouldn't use a Fast C sensor, but the other sensors (Acc CT and Fast T) should be OK at 200 meters. F-meter shouldn't have any problems either.

So I think SCAMP will work but you should plan to install a pressure sensor with the appropriate range.

I strongly suggest that if you intend to use SCAMP deeper than 100m, make the necessary modifications, and then thoroughly test it at 200 meters. Better to have it implode in a test chamber then at the end of a retrieval line.>

SCAMPs built with serial numbers ranging from SN0001 to SN0030 hold data internally, but have space for only

100 meters of 8 channel data. We are re-designing the SCAMP to contain up to 8000 meters of 8 channel data in multiple files. These SCAMPs can be retro-fitted with this new feature.

Q. This experiment I am doing in Australia will be in a fairly benign, calm, sheltered piece of water. Would reducing the free-fall rate (sy to 5 cm/s) be a useful thing for me to try? Could I get better vertical resolution in the dissipation?

Sorry, I don't have a yes or no answer for you. Yes, you'll increase the wave number that you can resolve with

FP07. Maybe not as much as you think since there are boundary layers around the thermistor that grow thicker as velocity decreases. There is the "frozen turbulence" Taylor hypothesis which offers less guidance as velocity http://www.pme.com/USB_smanual/faq.htm (4 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions decreases. SCAMP's travel rate for slower speeds may become less steady... No quantitative information however. I don't know of anyone who has experience or comparisons at 5 cm/sec so no help there. Suggest you simply try both, then look at the data. I wouldn't just take 5 cm/sec data only however. A careful investigation of fall rate with comparisons to other instruments would make a nice paper!

Q. Just a quick question about the fluorometer range. I noticed that on a couple of casts the fluorometer pegged at the top of the range (not too surprising, there's a lot of chlorophyll here). Is there something in the calibration constants that I can alter to reduce the sensitivity and increase the range if necessary. I'm not sure what to expect in the estuary I'll be working in, so want to be able to cope with anything.

Yes, there are two ways to change the fluorometer scale. Some background:

Fluorometer signal begins as light pulses received by a photodiode within SCAMP's end cap. Photo diode converts these to small current pulses. These are converted to voltage pulses by the first stage amp on the fluorometer circuit, then AC amplified by a second stage. Third stage is an adjustable gain (potentiometer on the

5500 board - only one inside SCAMP!) DC amplifier. After this the signal is rectified, filtered, and passed to

SCAMP's A/D card by two more stages. A/D card has digitally controllable offset and gain circuits that add another point where overall sensitivity can be changed. Ok, so there are two points where the signal can be adjusted: potentiometer and A/D card.

Changing the A/D card is easy. Just go to SETUP\CHANNELS\FLUOROMETER and increase the max/min range shown there. This simple change could solve the problem, but maybe not. It turns out that the actual voltages of the pulses within circuits on the 5500 board must not exceed processing ranges on the board. This can only be determined by looking at the pulses on an oscilloscope under the condition of max C. If your high C range causes the circuit voltage to exceed limits, nothing permanently bad happens but you'll see a slope in the otherwise straight calibration curve in the high C regions. Your max/min change will only expand the range before the A/D clips.

It is safe, however, to DECREASE the circuit gain from the present setting. This is done by turning the pot (a 20 turn device) screw a few turns as required CCW. Remember how many turns in case you want to turn it back up again later. If you decrease the circuit gain then you'll maintain linearity and at the same time expand the effective A/D clipping range.

Q. What is the recipe for Copro?

Mix 1000 micro-grams of copro powder in 100 ml of 6N HCl. Then add distilled water to make 2 liters.

I purchased my copro from Sigma Chemical Co 1-800-325-3010. Their part number is C-7157 for 1 mg of the substance. The proper name is Coproporphyrin III Tetramethyl Ester. 1 mg isn't much, so I simply put the entire bottle in 100 ml and shook until the powder dissolved, then removed the bottle. I doubt that the resulting 2 liter concentration of 500 ug per liter is very accurate, but you can tell if the f-meter changes its sensitivity if you use the same solution. Wrap aluminum foil around the storage bottle and store in a dark place. They say it lasts for a while.

As I remember I mixed 500 ug/l concentrations. I had to trust the supplier to weigh the sample correctly. I think I received 1000 ug in a little bottle. I measured two liters of water, then completely rinsed the bottle in them. I stirred then poured the solution into two 1-liter bottles which I sealed and wrapped with aluminum foil. I've had my solutions for several years now.

Q. Do you think SCAMP, or any of the individual sensors, will get upset about high concentrations of hydrogen sulfide?

uDO won't like this. H2S is known to give Clarke type oxygen sensors problems. I'm not really sure what happens to them however. No experience on this score.

http://www.pme.com/USB_smanual/faq.htm (5 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions

Q. I changed the Acc CT sensor, and also put on a new DO sensor. I tried a communicate with SCAMP,

and the serial connection worked. I then tested all the channels, and they all gave by total rubbish. Both

C (Acc and fast) gave something like 40 (absolutely constant over the 100 samples), and temperatures were both large and negative. I asked for the battery voltage, and got a circuit temperature of 154 degrees (it's warmer in Australia, but not that warm) and all three battery voltages were -33 V. I took out the batteries and replaced them; same results. I then disconnected the two new sensors that I installed, again same results. I've had a careful look around and I am sure that I have not dislodged anything. Any ideas?

My bet is that you've blown a fuse in the 9v battery connection cable. It's really easy to do. It sometimes happens if you it the transistor cans when installing the A/D side shield. Here's how you can tell:

Method 1: Take off the A/d side cover. This reveals the 5026 board. Look in your manual in the back for pictures of the boards. You'll find the 5026 shown there if you don't know which one it is. This is the A/D board.

Get SCAMP to awaken by communicating with it. Measure the voltage from each T0-5 transistor can to the chassis. These should be almost exactly equal to the battery voltage, about 9.2 volts for fresh batteries.

Method 2: Take off the A/d side cover. Unplug the 9V battery connection wire. Measure resistance from battery terminals to corresponding pin on 3 pin board connector. Two of the battery connector terminals will go to the center pin (black wire) A (-) connector will go to the white wire pin and will have the fuse resistance of about 7 ohms. A (+) connector will go to the red wire pin, again with 7 ohms resistance. Blown fuses do not entirely open, they have blown resistances in the 1000 ohm range.

It is also possible to blow the fuses by contacting any of the many power supply pins on the boards so maybe there. My money is on a blow out due to the covers somehow. I favor this since it seems you described the problem as occurring right after sensor replacement, with no normal operation interval.

Q. T1 was removed and I had switched off the channel and forgot to switch off the gradient channel for

T1. What are the effects of this?

When no sensor is installed the + terminal of the op-amp that services Fast T1 is simply open - not connected to anything. The circuit output is undefined in this case but I've not seen any problems. It is not impossible however that Fast T1 oscillated badly and maybe swamped out both gradient channels. All gradient gains are implemented by the same IC, I use a dual DAC, one for each gradient channel so channel-channel cross talk at higher frequencies is not impossible.

Q. How do I test the gradient channels?

Disconnect a Fast T sensor and replace it with the circuit shown. Set the oscillator for 10 hz. Record a short profile. You should see a sinusoidal variation of temperature on the Fast T channel, and its gradient on the Grad

Fast T. This should work for both T0 and T1 provided you connect to the proper sensor port. One thing to watch out for: It is possible to drive the gradient circuit into over-range internally and not see the clipping distortion in the recorded data. (The over-range point occurs before the anti-alias filter and the action of the filter removes over-range clipping features from the data.) Do the math on the Fast T channel and check that the peak values don't exceed the limits shown by HOST for the Grad Fast T channel at whatever gain you are using. You may have to turn the oscillator amplitude way down...

Another simple test might be to stratify a bucket of water, then profile SCAMP through it by hand, being careful not to crash the sensors. You could review the profile data, verify that the gradient is bad, then connect a dummy sensor and repeat. Maybe the gradient will clean up. I don't know for sure that the gradient problem is related to the lack of a Fast T1 sensor, but it seems most likely at this time.

Q. What about using the SCAMP on an oceanographic ship?

http://www.pme.com/USB_smanual/faq.htm (6 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions

(feedback from: Jonathan Sharples, University of Southampton) We used SCAMP from a small vessel

(freeboard about 2.5 metres), deploying over the windward side. With 200m of cable we could profile down to about 40 to 50 metres before the wind drift of the ship made the line too taut (and it was not very windy, about 5 -

10 knots). SCAMP did not like swell too much either, and in swell 1 - 2 metres tended to somersault in the upper few metres (novel use of the PAR sensor....). Good data was not recorded until about 10 metres depth due to a combination of swell somersaults and the ship's wake.

We used a modified spinnaker boom to deploy SCAMP, which kept it away from the ship's side.

Paying out the line from the hand-held roll was ok (but have a good strong lanyard connecting the cable roll to the ship!) Pulling SCAMP back to the surface against the ship's drift was a little scary initially, as the line gets rather taut. But, we did not get close to the breaking strain; you just have to overcome your fear.

Recovery onto the boat was not so easy, but waiting for the boat to roll toward SCAMP and hauling inboard worked ok (we were not using the conducting cable: hauling SCAMP out of the water using that may not be a good idea).

Note also, profiling to 40 or 50 metres means you don't get many profiles to combine for statistically reliable dissipation estimates. We had to use the longer 200m non-conducting cable, so data downloads took a lot of time. Ultimately, we were able to estimate tidal averages of dissipation with confidence, but could not reliably resolve intra-tidal signals.

(feedback from Barry Ruddick, Dalhousie University) - I've actually been afraid to use SCAMP from a large ship. The two main factors are:

1. The 10 cm/s fall rate means that the ship drifts away from the instrument faster than it falls. Profiles are very limited in depth range because of that.

2. If it's windy, the drag plate could cause the instrument to swing and bang against the side of the ship, breaking sensors. I think it might be possible to use SCAMP from a relatively small ship, anchored in protected waters. I would try using it from the forepeak, so the instrument can swing without hitting the ship, and to prevent the instrument from seeing the wake.

The certain way is to use SCAMP from a launch, so it can be set directly into the water. John Dower and I have done this without problems.

(feedback from John Dower, University of British Columbia) - Barry and Jonathon are right, deploying and recovering SCAMP from a ship is a nail-biting experience! I found that deploying from a launch or (better yet) an inflatable boat was much easier on my blood pressure. In any case, even if it were easy to deploy from a ship the amount of turbulent noise created by the ship makes it preferable to get away from the ship in a small boat.

Even from a small boat, however, the downwind drift of the boat can make it hard to get deep profiles. My best results came when I measured off 100m of the kevlar line, attached a small float to it and then tossed (well, lowered, actually) the whole thing (i.e. SCAMP, the kevlar line, and the float) over the side of the inflatable and then simply followed the float until SCAMP had reached the desired depth. Just make sure the float is sufficiently large!

The only other problem I encountered was condensation inside the housing when SCAMP was used in *really* cold waters. In coastal Newfoundland, where I've used SCAMP since 1997, the summer air temperature can be a muggy 25C, the surface temperature can be 12C, while at 50m depth the temperature falls to about -1C. When

I opened the housing after a few such casts (to change batteries) I found a fine layer of condensation all over the inside of the housing. After that, I always taped a few of those small desiccant packs inside....that seemed to solve the problem.

Short of that, the only thing that I find to be a bit annoying is the long time taken to download a deep cast from

SCAMP while siting in a tiny rubber boat in the wind, rain, and hail!

http://www.pme.com/USB_smanual/faq.htm (7 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions

Q. Can you just confirm that the engineering units were calibrated as micro Einstein's per meter squared per sec?

No. Calibration units are shown in SCAMP Control Dialog

Channel Tab under the PAR section. The PAR

(photosynthetically active radiation) sensor comes to me with calibration units of (umol/(s*m**2))/uA, a calibration of intensity to sensor current output. I calculate the conversion factor from uA to SCAMP's CR based on a math model of the circuit performance. The engineering units you ultimately see expressed by the data are

(umol/(s*m**2)). This seemed a weird intensity unit to me but I guess it is useful if you are a plant chemist. I guess it makes sense since the photodiode inside the PAR sensor doesn't see the full light spectrum but rather a shaped subset. Shaping is done by filters within the sensor.

PAR is usually measured (by aquatic biologists anyway) in units of u mol photons /(m**2 s) (i.e. a rate of supply of photons). An Einstein is simply a mol of photons, so the units are correct.

Q. I have a query about the circuit temperature. It seems to be reading too low. In the field today the air temp was about 16 C and the water 14 C, but the circuit T always said about 10 C.

CircT has hardware bugs and doesn't work very well. I wouldn't worry about CircT. Do not turn CircT channel on for recording purposes.

Q. How do I measure current consumption within SCAMP?

There are ways to measure the current consumption. Open SCAMP, remove the longer electronics cover and

connect the RS232 port to your computer. (This is described in the RS232 Connection

Appendix within this manual) With RS232 connected SCAMP's CPU is always powered, but the analog section is not. You can power the analog section by sending the BAT command. SCAMP replies by repeatedly printing the battery voltage. This holds all power on. The print halts in response to the RETURN keypress. Take out 9 volt batteries from cavity but don't disconnect them. Measure voltage from SCAMP chassis (volt meter black lead) to each terminal of each battery. You can probe this from the side with the red volt meter lead. Each battery will have one terminal that has 0 volts. This is connected to ground. Each will have one terminal that measures the battery output. One will be about +9 volts the other about -9 volts depending on the battery charge. Next measure voltage from SCAMP's chassis to the metal can of each large transistor on the analog board where the batteries connect. These will be about +/- 9 volts (there are two transistors). It turns out that that battery lead has fuses in it that have about 6 ohms of internal resistance. If you subtract the appropriate (+ with +, - with -) transistor case voltage from the corresponding battery voltage you'll get the voltage drop across each fuse.

SCAMP analog consumes about 32 mA so with 6 ohm fuses you'll see something like 0.190 volt difference.

The current consumption in the AA batteries is harder. Get a small piece of printed circuit board with copper on both sides. Solder a 1 ohm resistor from one side to the other. Insert the printed circuit board between the two top AA batteries. You've inserted a 1 ohm resistor in series with the batteries in this way. SCAMP digital consumes about 60 mA so expect about 60 mV when you measure voltage across the resistor.

Remember to sent the RETURN keypress to tell SCAMP it can turn off the analog section. Disconnect RS232 to cause SCAMP to sleep again.

Q. What water density is used to convert pressure (in psi) to depth for the SCAMP calibrations?

The conversion factor that I use is Depth = 0.7030696 * pressure (PSI)

This factor is actually located within the ParoScientific pressure transducer ROM. When calibrating a SCAMP I simply read depth (meters) from the Paros. The Paros applies the above conversion from measured pressure to depth. Essentially the Paros is telling me depth in fresh water. You can read more about depth calibration at the

Depth

chapter within the Calibration section of the operator's manual.

http://www.pme.com/USB_smanual/faq.htm (8 of 9) [10/5/2005 5:10:41 PM]

Frequently Asked Questions http://www.pme.com/USB_smanual/faq.htm (9 of 9) [10/5/2005 5:10:41 PM]

Profiling with SCAMP

Precision Measurement Engineering,

Inc.

Profiling with SCAMP

This section presents the information required to begin profiling with the SCAMP, organized roughly in time-sequence. This component assumes that the SCAMP has already been calibrated and otherwise prepared for field use, or that it has been newly received from PME.

This information is provided to get you up-and-going with as little reading as possible. The reader is encouraged to read this section completely, practice with the SCAMP, and then read the other sections.

The old saying "Practice makes perfect," applies especially to the SCAMP. The high resolution sensors used by the SCAMP are quite fragile. Practice setting up and dismantling the SCAMP in your lab before attempting to use it in a real situation. Choose only the best conditions for the first use of the SCAMP. It is a lot like fishing... there is a technique to be learned that can't be taught here in the manual.

PROFILING OVERVIEW

The SCAMP can be used to make profiles in the upward and downward direction as shown in

the SCAMP Profiling Modes diagram

. In downward mode, the floats are positioned against the drag unit and the nose weighted with lead ballast. The SCAMP is placed in the water with the sensors down and descends straight downwards. Data are recorded during the descent as shown in this sketch.

In upward mode, the floats are positioned against the SCAMP nose and an expendable weight

(rock) is connected to the SCAMP's release screw by a short wire. The SCAMP is placed in the water with the sensors up and initially sinks. See the photo below.

http://www.pme.com/USB_smanual/profile.htm (1 of 3) [10/5/2005 5:10:47 PM]

Profiling with SCAMP

The drag plate is hinged and allows the SCAMP to sink sideways as shown in the

SCAMP

Profiling Modes diagram

. When the SCAMP reaches the bottom or a pre-programmed depth, the release is activated and the weight drops off. The SCAMP then ascends vertically through undisturbed water as it records the data.

In each direction, the SCAMP must be connected to the surface using its retrieval line.

STEPS FOR PROFILING

Before you begin to profile you will need to:

● establish communication

,

● check the batteries

,

● define the mission,

● and assemble the SCAMP stand.

Next, the SCAMP must be set up for the appropriate mode and a profile taken (If the SCAMP

has a fluorometer

, then extra steps need to be followed regarding its assembly):

Upwards profiling

Downwards profiling

http://www.pme.com/USB_smanual/profile.htm (2 of 3) [10/5/2005 5:10:47 PM]

Profiling with SCAMP

The resulting profile(s) should be reviewed and saved:

Data upload & storage

.

This profile should be reviewed to determine if SCAMP is properly set for current environmental conditions:

Velocity or gradient gain adjustment .

When these steps are completed, then future profiles can be obtained and stored in the same manner. Gradient gains should be reviewed from time to time as environmental conditions change.

When profiling is completed:

Disassembling and storing the SCAMP

.

Profiling is now completed. http://www.pme.com/USB_smanual/profile.htm (3 of 3) [10/5/2005 5:10:47 PM]

Establishing Communication

Precision Measurement Engineering, Inc.

Establishing Communication

SCAMP communicates with the host computer by USB 1.0. SCAMP is connected to the host computer via the supplied I/O cable. This cable plugs onto SCAMP’s I/O connector and into any USB connector on the host computer. The host computer will automatically recognize SCAMP within 5 seconds after this connection is made. There will be no obvious screen display, but SCAMP will appear in Device Manager. The images below show USB I/O cable connection to the SCAMP.

Be careful not to bend or twist the rubber boot at the SCAMP connector end . Even slight bending will break the internal wires. Twisting may cause the SCAMP connector to unscrew.

A little silicone grease sparsely applied on the SCAMP connector outside diameter will help.

Begin SCAMP control dialog by browsing to the C:\SCAMP directory and selecting the Control.exe

program. After a few seconds the control dialog will begin operation. The following Windows dialog box will appear.

http://www.pme.com/USB_smanual/comm.htm (1 of 2) [10/5/2005 5:10:54 PM]

Establishing Communication

SCAMP communication is now initialized.

SCAMP’s internal clock is set to host computer time at this point.

HOST COMPUTER

The HOST computer must be prepared for profiling. You must insure that sufficient battery power exists within the computer to allow it to operate for the duration of the profile. It is not unusual for the

SCAMP batteries to outlast laptop computer batteries. One convenient way to provide extra battery power is to use a large lead-acid rechargeable battery, an inverter to generate normal AC service voltages, and then use the laptop computer's external AC power adapter.

The HOST computer often has an automatic power saver function that shuts it down during periods of inactivity. This may inconveniently occur during SCAMP data upload where several minutes pass with no keyboard entry. There is usually a software control supplied with the computer that will disable this feature. Disable it.

http://www.pme.com/USB_smanual/comm.htm (2 of 2) [10/5/2005 5:10:54 PM]

Checking the Batteries

Precision Measurement Engineering,

Inc.

Checking the Batteries

The batteries should be checked in the lab before using the SCAMP in the field. It is much easier to change batteries in a dry environment compared to a wet one.

Batteries may be checked by connecting SCAMP to the host computer's USB port and running

the SCAMP Control Dialog. Battery voltages and depletions are shown on the System tab . Note

that voltages and depletions are read from SCAMP only once upon USB connection and are not updated thereafter.

There are 3 battery packs within the SCAMP. These will last for approximately 4 hours of continuous operation. The SCAMP is designed for power conservation and turns itself off whenever possible. Profiles should not be initiated when any battery voltage is close to depletion as shown by the bar graphs. SCAMP itself checks for battery depletion after it is magnetically triggered and will not begin a profile should any battery reach the depleted state.

Be aware that deeper upward profiles require more torque from the release motor due to the increased axial pressure loading of the release screw. The digital battery minimum voltage requirement increases with increasing profile depth. Since the mechanical drag on the release screw, due to o-ring seals and other parameters varies from SCAMP to SCAMP, it may happen that SCAMP will be unable to operate its release motor prior to sensing digital battery depletion.

If the batteries need to be replaced, then read the

Battery Replacement

section.

http://www.pme.com/USB_smanual/battery.htm [10/5/2005 5:10:58 PM]

Assemble SCAMP stand

Precision Measurement Engineering,

Inc.

Assembling the SCAMP Stand

The SCAMP is supplied with a stand that will support it while it is not profiling. A 5-gallon plastic bucket, not supplied, is required in addition to this stand. This stand performs two functions: it provides a safe location to store the SCAMP between profiles and it provides a convenient way for keeping the micro-conductivity sensor wet. If this sensor, and to a lesser extent the accurate conductivity sensor dries out, then the salt crystals left behind will disturb the electroplated electrodes. This will not damage the sensor, but may result in small calibration shifts. If the

SCAMP is immediately placed in the stand with the sensors in the bucket, then drying will not be a problem.

This

sketch shows the separate parts of the SCAMP stand.

Assemble the SCAMP stand as seen in the images below. Click on an image to view it magnified. Make sure that all plastic screws are tightly screwed into the PVC pipe.

To use the SCAMP stand and bucket on the boat, locate the stand securely on the deck of your boat where it will be convenient to deploy/retrieve the SCAMP. Place a standard 5 gallon bucket within the SCAMP stand. Tighten the bucket retaining strap securely so that the empty bucket is lifted from the deck. Fill the bucket half full with water from the lake or ocean; do not used distilled water . The water will help anchor the stand.

http://www.pme.com/USB_smanual/stand.htm (1 of 2) [10/5/2005 5:11:03 PM]

Assemble SCAMP stand

Mount the laptop tent on the stand table and place the wire support inside. Place the laptop computer inside the tent and anchor it with the straps supplied. Insure that this computer is well-protected from water that may drip from the SCAMP or be splashed over the side of the boat.

http://www.pme.com/USB_smanual/stand.htm (2 of 2) [10/5/2005 5:11:03 PM]

Upwards Profiling

Precision Measurement Engineering, Inc.

Upwards Profiling

Upward direction profiling is used when measurement of the surface layer is desired. The host computer and SCAMP control dialog must be properly installed as described in the

Installing the

SCAMP Software

, Checking the Batteries

, and the

Assembling the SCAMP stand

sections.

Assembly of SCAMP

SCAMP is most conveniently shipped and stored dis-assembled. It must be assembled prior to profiling. Assembly is accomplished by the following steps:

Locate enough expendable weights

(rocks) for the profiles you intend to make. Make wire loops from the wire provided by threading the wire through the collar provided to make a loop, then adjusting the loop size using the tool provided. Crimp the collar in place with pliers. Connect the other ends of the wires to the expendable weights so that the weights will hang below the SCAMP by at least 12 inches (30cm).

Place the 5151 float retainer ring over the sensor cover and onto the

SCAMP.

Place the retrieval line loop over the sensor cover and onto the

SCAMP.

Place the float assembly over the sensor cover and onto the SCAMP.

http://www.pme.com/USB_smanual/profile_up.htm (1 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling

Carefully remove the sensor cover. Do this in a sitting position with the

SCAMP rotated so that the

Accurate CT sensor is either on the right or left side. Do this so that the relative slant of the sensor cover as it comes off the retaining threads is visible relative to the Accurate CT sensor. Since the

Accurate CT sensor is the farthest sensor from

SCAMP's center it is the most likely sensor to be damaged when removing the sensor cover.

Screw the sensor guard into place. Use a small cable tie or bit of dental floss to tie one of the guard's legs to one of the sensor seal nuts bases. This will prevent the unscrewing of the guard during a SCAMP deployment. Should the guard come unscrewed the retrieval cable can slip loose and the

SCAMP will be lost. The sensor guards screw on tightly, but this extra precaution is low-cost insurance. The picture displays an older version of this guard.

Move the float assembly back against the sensor guard base.

http://www.pme.com/USB_smanual/profile_up.htm (2 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling

Move the cable loop and the 5151 float retainer ring back against the float.

Tighten the thumbscrew on the float retainer ring.

Install the drag assembly and tighten the thumbscrews. Be sure that the thumbscrews find the groove in the end sleeve on the SCAMP. If the SCAMP slips from the drag assembly, then it can be lost.

Place the SCAMP in the SCAMP stand with the sensors in the water contained by the bucket. The tent for the computer is not shown in the photos.

http://www.pme.com/USB_smanual/profile_up.htm (3 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling http://www.pme.com/USB_smanual/profile_up.htm (4 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling

Insert the drag plate wing retainer pin located on the cable into the drag plate arms.

Insure that the I/O connector cover is securely in place over the I/O connector.

Water must not enter this connector!!

Mission Definition

You must program SCAMP to implement the desired profile. Once programmed, SCAMP will implement the profile over and over with data from each profile stored separately in SCAMP’s memory.

Missions are defined by the following steps:

Establish

communication with the SCAMP.

Select the

Mission tab , shown below.

http://www.pme.com/USB_smanual/profile_up.htm (5 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling

Select Upwards for the Operational Mode.

Define Starting Conditions for Data Recording. The conditions you define are ‘OR’d by SCAMP for the purposes of determining when to begin a profile. SCAMP will begin a profile when it determines that any one of the conditions you define are true. You must carefully consider the definitions you make here.

In an upwards profile, SCAMP must sink to the Begin Depth depth before it releases its weight and begins its ascent. You must provide Begin Delay time for SCAMP to be deployed and reach the desired depth. You can supply substantially more Begin Delay time, but then if SCAMP reaches the bottom before it reaches the Begin Depth you will have to wait this entire time or loose the profile by hauling SCAMP back. Also, SCAMP will not begin a new profile until either you connect it to the host computer or it ends the current profile so you will be faced with waiting out the Begin Delay time or connecting SCAMP.

Define Ending Conditions for Data Recording. At the time the profile begins scanning and operates its http://www.pme.com/USB_smanual/profile_up.htm (6 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling release motor, releasing the attached weight. Thereafter SCAMP ascends while collecting data. You should define Acquire Scans with sufficient scans to collect data over the length of the water column.

SCAMP scans at 100 scans per second and nominally travels at 10 cm/second giving a nominal 1000 scans per meter. Allow extra scans to cover uncertainties in depth and SCAMP travel rate. Disconnect

SCAMP from the host computer and place the I/O cover onto the I/O connector prior to profiling.

Profiling

SCAMP will acquire multiple profiles without uploading of data or re-definition of missions. Between profiles SCAMP enters a low power ‘sleep’ state to conserve batteries. Prior to each profile SCAMP must be ‘awakened’. After SCAMP is awake it can be placed into the water and perform its mission.

(Note that SCAMP will not acquire a profile if connected to the host computer!)

Profiles are implemented with the following steps:

Place a finger against the release screw. Hold the Start Magnet near the SCAMP, about half way between the ends. Magnet should be alligned with the Accurate CT sensor. When SCAMP senses the magnet it will immediately rotate the release screw for a brief period. The magnet should be removed at this time. SCAMP reviews its internal state to determine if sufficient battery and memory resources exist to acquire the impending profile. If insufficient resources are available SCAMP will return to sleep mode. if sufficient resources are available and after about 2 seconds, SCAMP will operate its release motor twice briefly. At this time SCAMP will define the current pressure as zero depth and begin evaluating starting conditions for data recording as described in the section above.

(Note that since SCAMP's pressure transducer is located at end opposite SCAMP's sensors, sensor zero depth will be different and must be adjusted by the customer's software.) Once it begins evaluating start conditions, SCAMP must complete a profile before it will return to the sleep state.

(See the Aborting a Profile section if a profile must be aborted.)

Connect a weight to the SCAMP by placing the wire loop over the release screw and post, with the collar inward along the post. Move the loop to the base of the lead screw. Pull the wire to test that the loop will not slip off. Place the weight on the underside (presently in the upwards position with the SCAMP on the stand) of the drag plate.

http://www.pme.com/USB_smanual/profile_up.htm (7 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling

Lift the SCAMP from the stand while supporting the expendable weight. Hold the

SCAMP with the sensors up and let the weight hang down free. The weight is not shown in the picture.

Place the SCAMP in the water with the hinged side of the drag plate away from the boat.

This is important since the SCAMP will sink at a 45 degree angle toward the hinged side of the drag plate. Release the SCAMP and pay out the cable as it sinks.

The SCAMP must be free of cable tension as it travels in the water column. You must pay out sufficient retrieval line to prevent cable tension as the SCAMP travels away from the boat. Also, in windy conditions the boat will be blown away from the SCAMP and the retrieval line must be paid out to compensate. You may attempt to hold station with the boat, but you risk cutting the retrieval line with the propeller. This operation takes some practice and people new to SCAMPing are advised to practice on calm days first.

http://www.pme.com/USB_smanual/profile_up.htm (8 of 9) [10/5/2005 5:11:22 PM]

Upwards Profiling

When you see the SCAMP, (the SCAMP may be hard to see on windy days), or when sufficient time has elapsed for the SCAMP to sink and return to the surface, recover the SCAMP by hauling it back with the cable. Wind it onto the reel in the reverse manner from the way it was unwound. First, give a gentle tug on the retrieval line to pull the pin holding the drag unit wings allowing these to fold back.

This will make hauling the SCAMP back much easier.

When the SCAMP is alongside the boat, first grasp the retrieval line within 1 meter of the

SCAMP. The retrieval line is reinforced in this area. Lift the SCAMP by this retrieval line or the sensor guard (do not touch the sensors) until you can grasp the SCAMP tube below the floats. Lift it into the boat. Once retrieved, then place the SCAMP again in the stand with the sensors in the bucket of water.

At this time you may again conduct a profile by repeating the steps in this Profiling section. Alternately,

you may upload SCAMP’s data

.

A final note: SCAMP profile files are named with the time that SCAMP determined a starting condition became true and began the profile.

http://www.pme.com/USB_smanual/profile_up.htm (9 of 9) [10/5/2005 5:11:22 PM]

Downwards Profiling

Precision Measurement Engineering, Inc.

Downwards Profiling

Downward direction profiling is used when measurement of the bottom boundary layer is desired. The

host computer and SCAMP control dialog must be properly installed as described in the Installing the

SCAMP Software

,

Checking the Batteries , and the

Assembling the SCAMP stand

sections.

Assembly of SCAMP

SCAMP is most conveniently shipped and stored dis-assembled. It must be assembled prior to profiling. Assembly is accomplished by the following steps:

Place the retrieval line loop over the sensor cover and onto the SCAMP.

Place the float assembly over the sensor cover and onto the SCAMP.

Place the 5151 float retainer ring over the sensor cover and onto the SCAMP.

Carefully remove the sensor cover. Do this in a sitting position with the

SCAMP rotated so that the

Accurate CT sensor is either on the right or left side. Do this so that the relative slant of the sensor cover as it comes off the retaining threads is visible relative to the Accurate CT sensor. Since the

Accurate CT sensor is the http://www.pme.com/USB_smanual/profile_down.htm (1 of 7) [10/5/2005 5:11:35 PM]

Downwards Profiling farthest sensor from

SCAMP's center it is the most likely sensor to be damaged when removing the sensor cover.

Screw the sensor guard into place. Use a small cable tie or bit of dental floss to tie one of the guard's legs to one of the sensor seal nuts bases. This will prevent the unscrewing of the guard during a SCAMP deployment. Should the guard come unscrewed the retrieval cable can slip loose and the

SCAMP will be lost. The sensor guards screw on tightly, but this extra precaution is low-cost insurance. The picture displays an older version of this guard.

Install the drag assembly and tighten the thumbscrews. Be sure that the thumbscrews find the groove in the end sleeve on the SCAMP. If the SCAMP slips from the drag assembly it can be lost.

http://www.pme.com/USB_smanual/profile_down.htm (2 of 7) [10/5/2005 5:11:35 PM]

Downwards Profiling

Wire the drag assembly in the 90 degree from SCAMP axis position using the holes in the arms provided and bits of weight wire.

Move the retrieval line loop and float assembly against the drag plate.

Move the 5151 float retainer ring against the float. Tighten the thumbscrew on the float retainer ring.

Place a sufficiently large ballast weight (sheet lead supplied with the SCAMP) around the sensor end on the SCAMP tube just aft of the sensor guard. Tape this in place. The SCAMP must be nose heavy. You may have to add floats and weight.

http://www.pme.com/USB_smanual/profile_down.htm (3 of 7) [10/5/2005 5:11:35 PM]

Downwards Profiling

Place the SCAMP on the SCAMP stand with the sensors in the water contained by the bucket.

Insert the drag plate wing retainer pin located on the retrieval line into the drag plate arms.

Insure that the I/O connector cover is securely in place over the I/O connector.

Water must not enter this connector!!

Mission Definition

You must program SCAMP to implement the desired profile. Once programmed, the SCAMP will implement the profile over and over with data from each profile stored separately in SCAMP’s memory.

Missions are defined by the following steps:

Establish

communication with the SCAMP

Select the

Mission tab , shown below

http://www.pme.com/USB_smanual/profile_down.htm (4 of 7) [10/5/2005 5:11:35 PM]

Downwards Profiling

Select Downwards for the Operational Mode.

Define Starting Conditions for Data Recording. The conditions you define are ‘OR’d by SCAMP for the purposes of determining when to begin a profile. SCAMP will begin a profile when it determines that any one of the conditions you define are true. You must carefully consider the definitions you make here.

In a downwards profile SCAMP must usually begin sampling as near to the surface as possible. The

Begin Depth should be defined to be 1 meter although it can be defined deeper should a deeper start be desired. You must provide sufficient Begin Delay time for SCAMP to be deployed and reach the desired depth. You can supply substantially more Begin Delay time and thereby depend upon a depth start, but normally a Begin Time start is used and scanning begun with SCAMP above the surface.

Define Ending Conditions for Data Recording. At the time the profile begins scanning and operates its release motor for two seconds. Thereafter SCAMP descends while collecting data. You should define

Acquire Scans with sufficient scans to collect data over the length of the water column. SCAMP scans http://www.pme.com/USB_smanual/profile_down.htm (5 of 7) [10/5/2005 5:11:35 PM]

Downwards Profiling at 100 scans per second and nominally travels at 10 cm/second giving a nominal 1000 scans per meter. Allow extra scans to cover uncertainties in depth and SCAMP travel rate. Disconnect SCAMP from the host computer and place the I/O cover onto the I/O connector prior to profiling.

Profiling

SCAMP will acquire multiple profiles without uploading of data or re-definition of missions. Between profiles SCAMP enters a low power ‘sleep’ state to conserve batteries. Prior to each profile SCAMP must be ‘awakened’. After SCAMP is awake it can be placed into the water and perform its mission.

(Note that SCAMP will not acquire a profile if connected to the host computer!)

Profiles are implemented with the following steps:

Place a finger against the release screw. Hold the Start Magnet near the SCAMP, about half way between the ends. Magnet should be alligned with the Accurate CT sensor. When SCAMP senses the magnet it will immediately rotate the release screw for a brief period. The magnet should be removed at this time. SCAMP reviews its internal state to determine if sufficient battery and memory resources exist to acquire the impending profile. If insufficient resources are available SCAMP will return to sleep mode. if sufficient resources are available and after about 2 seconds, SCAMP will operate its release motor twice briefly. At this time SCAMP will define the current pressure as zero depth and begin evaluating starting conditions for data recording as described in the section above.

(Note that since SCAMP's pressure transducer is located at end opposite SCAMP's sensors, sensor zero depth will be different and must be adjusted by the customer's software.) Once it begins evaluating start conditions, SCAMP must complete a profile before it will return to the sleep state.

(See the Aborting a Profile section if a profile must be aborted.)

Lift the SCAMP from the stand and place it in the water with the sensors downward.

Release it and pay out the retrieval line as it sinks.

The SCAMP must be free of retrieval line tension as it travels the water column. You must pay out sufficient retrieval line to prevent cable tension as the SCAMP travels away from the boat. Also, in windy conditions the boat will be blown away from the SCAMP and then the retrieval line must be paid http://www.pme.com/USB_smanual/profile_down.htm (6 of 7) [10/5/2005 5:11:35 PM]

Downwards Profiling out to compensate.

After more than (10 seconds/meter * depth) seconds have passed, then recover the SCAMP by hauling it back with the retrieval line.

When the SCAMP is alongside the boat, lift it by the retrieval line or the sensor guard (don't touch the sensors) until you can grasp it by the

SCAMP tube. Lift it into the boat. Once retrieved, place the SCAMP again in the stand with the sensors in the bucket.

At this time you may again conduct a profile by repeating the steps in this Profiling section. Alternately,

you may upload SCAMP’s data

.

A final note: SCAMP profile files are named with the time that SCAMP determined a starting condition became true and began the profile.

http://www.pme.com/USB_smanual/profile_down.htm (7 of 7) [10/5/2005 5:11:35 PM]

Aborting a Profile

Precision Measurement Engineering,

Inc.

Aborting a Profile

SCAMP acquires profile information according to conditions set at the

Mission tab

of the SCAMP

Control Dialog. Once activated by the magnetic switch, SCAMP will begin evaluating start conditions until one or more become true. Thereafter SCAMP will acquire the number of scans requested. SCAMP will return to sleep (power conservation) mode only after all scans are acquired.

Once triggered, SCAMP must complete its profile before it can be triggered again or enter sleep mode.

It may sometimes be necessary to abort a profile. This may occur if too many scans are selected, or if unresonable starting conditions are specified. The only method available to abort a SCAMP profile is to connect it to the USB port of the host computer. Once SCAMP recognizes this connection it will abort the profile presently in operation, save any collected data, and begin responding to USB commands. http://www.pme.com/USB_smanual/profile_abort.htm [10/5/2005 5:11:40 PM]

Fluorometer Assembly

Precision Measurement Engineering, Inc.

Assembling the Fluorometer

The SCAMP fluorometer must be carefully assembled prior to field use. This section describes how to perform the assembly.

Assembling a fluorometer version of SCAMP is very similar to assembling the standard SCAMP. The description below shows the procedure for upward mode. The other modes require only slight changes.

This photo shows the items to be assembled.

Begin the assembly by first sliding the 5151 float retainer ring over the sensor cover and down onto the SCAMP body.

Place the retrieval line loop over the sensor cover and onto the SCAMP body.

Place all floats over the sensor cover and onto the SCAMP body.

http://www.pme.com/USB_smanual/fluorometer.htm (1 of 4) [10/5/2005 5:11:54 PM]

Fluorometer Assembly

Next, carefully remove the sensor cover. Do this in a sitting position with the

SCAMP rotated so that the

Accurate CT sensor is either on the right or left side. Do this so that the relative slant of the sensor cover as it comes off the retaining threads is visible relative to the Accurate

CT sensor. Since the

Accurate CT sensor is the farthest sensor from SCAMP's center it is the most likely sensor to be damaged when removing the sensor cover.

The picture shows the proper hand position.

Next, install the sensor guard.

This is screwed onto the same threads that retain the sensor cover.

After the sensor guard is installed, next install the fluorometer inlet and exhaust elbows. Note that these must not be screwed deeply into the fluorometer ports. If screwed too deeply, then they will push the fluorometer sample tube towards the other side and may cause the o-ring seal to fail.

The elbows provided with

SCAMP have white plastic collars that will stop the elbow at a safe depth. Note that the elbows need not be tight in the ports. There may be a slight water leakage through the threaded region, but it will be very small compared to the normal flow through the elbow. The picture shows the elbows being installed.

http://www.pme.com/USB_smanual/fluorometer.htm (2 of 4) [10/5/2005 5:11:54 PM]

Fluorometer Assembly

Next, install the fluorometer exit tube. This long black tube is first placed through the holes drilled in the floats. Be careful when sliding it through the holes that it does not come into contact with the sensors.

Watch out for fingers at this time also.

Place the fluorometer exit tube over the exhaust elbow. Slide the floats upward until they contact the bottom edge of the sensor guard. Move the retrieval line loop up against the floats and secure them all in place with the float retainer ring.

Now attach the drag plate.

The SCAMP must be rotated so that the fluorometer exit tube passes through the relief in the drag plate and passes into the gray cylinder glued just below the drag plate. The picture below shows the drag plate installation. The red-capped sensor attached to the drag plate is the PAR sensor.

http://www.pme.com/USB_smanual/fluorometer.htm (3 of 4) [10/5/2005 5:11:54 PM]

Fluorometer Assembly

The pin on the retrieval line is passed through the hole in the drag plate arms to secure the drag plate in the open position. When SCAMP is retrieved after a profile, this pin will pull allowing the drag plate to fold up for easier retrieval.

Disassembly of SCAMP after use is simply the reverse of the above procedure. When disassembled, cover the fluorometer ports with electrical tape to prevent dust and particles from entering.

Clean the fluorometer tube occasionally with a q-tip and alcohol.

If SCAMP is used in downwards mode, then the float retainer ring is not installed first, but rather installed after the floats and is used to press the floats against the drag plate. Installation of the fluorometer parts remains in the same order. If the fluorometer tube is loose on the exhaust elbow, then it may be secured with a cable tie.

http://www.pme.com/USB_smanual/fluorometer.htm (4 of 4) [10/5/2005 5:11:54 PM]

Data Upload/Storage

Precision Measurement Engineering, Inc.

Data Upload & Storage

SCAMP can contain 1 or many more profiles within its memory. This information must be transferred

(‘uploaded’) from SCAMP to the host computer from time to time. Uploading occurs automatically whenever SCAMP is connected to the host computer. The steps for uploading are:

Establish communication with the SCAMP

Nothing else is required. When the SCAMP Control program begins operation it queries SCAMP about profiles. If any profiles are stored within SCAMP, then they are transferred to the host computer automatically. Thereafter, the SCAMP memory that contained the profiles is erased. There are screen displays that give the progress of this operation. USB transfers are very fast but owing to the volume of data that SCAMP can store, the transfer process may take several minutes to complete.

Uploaded profiles are stored on the host computer HDD using a file name constructed based upon the

SCAMP time that the profile was begun (when any Starting Conditions for Data Recording became true). The format is day-month-year hour-minute-second .raw An example file name is

15JAN2004 162403.raw

These files will appear in the current directory of the host computer.

After uploading is completed the SCAMP control dialog screen appears.

http://www.pme.com/USB_smanual/data.htm (1 of 2) [10/5/2005 5:11:59 PM]

Data Upload/Storage

If there are no activities required on this screen, then it may simply be closed.

However it is often valuable to perform

velocity and gradient gain adjustments .

http://www.pme.com/USB_smanual/data.htm (2 of 2) [10/5/2005 5:11:59 PM]

Velocity & Gradient Gain Adjustment

Precision Measurement Engineering, Inc.

Velocity & Gradient Gain Adjustment

The sensors and circuitry within the SCAMP are designed to operate at a velocity of 10 centimeters/second. The actual velocity depends upon the weights, floats, and the density of the water in the water column. When received, the SCAMP will not be properly ballasted. Ballasting is the customer's responsibility.

SCAMP contains special circuits that perform analog signal processing of selected channels. Normally there are two of these channels, each connected to a Fast T sensor. The analog processing consists of a gradient circuit (d/dt) followed by a programmable gain amplifier, followed by a 6-pole, 45 Hz, anti-alias filter. The outputs of these circuits are connected to SCAMP's "gradient" channels. The programmable gains must be adjusted by the customer in response to profile conditions.

Since ballasting and gradient gain depend upon actual environmental conditions they cannot be set in advance. Instead, preliminary profiles must be conducted, the results reviewed, and appropriate corrections made until the velocity and gradient gains are acceptable.

Preliminary profiles are conducted as described in the

Profiling with SCAMP section of this manual.

After data upload and storage

, select the System tab from the SCAMP Control Dialog. The screen shot below shows this tab.

http://www.pme.com/USB_smanual/adjust.htm (1 of 3) [10/5/2005 5:12:06 PM]

Velocity & Gradient Gain Adjustment

This screen allows the user to view gradient plots, set gradient gains, and to view velocity plots. (Note that the plots result from the last profile uploaded from SCAMP at the beginning of this SCAMP

Control Dialog. If the Dialog is begun and SCAMP has no profiles to upload, then the display buttons will be inactive.)

VELOCITY

Click the Display button in the velocity section. A plot of SCAMP’s velocity from the most recent profile will appear. SCAMP should ideally travel at 10 cm/sec. You may have to add or remove ballast weights to obtain this velocity. In

Upwards mode

, weights should be hung on the rods provided at the bottom of the drag plate. In

Downwards mode

, major weight adjustments should be made to the weights at the sensor end (see pictures below), but small adjustments may be made by hanging weights on the rods provided at the bottom of the drag plate.

http://www.pme.com/USB_smanual/adjust.htm (2 of 3) [10/5/2005 5:12:06 PM]

Velocity & Gradient Gain Adjustment

SCAMP’s velocity will not be constant throughout the water column since the density of the water is different at different points. Try to make an adjustment that gives 10 cm/sec in interesting regions.

GRADIENT GAIN

Click the Display button in the Gradient Channels section. A plot of the selected gradient channel from the most recent profile will appear. You must select the amount of programmable gain used in each gradient circuit. This gain cannot be set prior to field use since the gradients in the water column vary depending upon environmental conditions such as recent winds. The user must set the gain based on conditions actually observed in profiles, and should generally be aware that conditions can change hourly making gain settings obsolete.

The objective of setting gradient gains is to obtain signals from the gradient circuits that occupy a large fraction of the A/D input range. The plot presents the selected Gradient Channel plotted on a fixed range of +/- 1.0, where 1.0 represents full A/D count (32768). Try to select a gradient gain that gives 30% usage without any data points reaching the A/D clipping value of +/- 1.0. Set the gradient gains by entering them in the Gradient Gain box. Note that the gain you set here will only influence future profiles and will not change the present plot or any already uploaded profiles. Normally gradient gains for similar sensors (like Fast T0 and Fast T1) are set equal to each other. In special cases gains can be set differently to capture different sections of the water column. Note also that data obtained when the sensors are above the water surface may contain very large gradients. Ignore these when evaluating the plot.

MONITORING CONDITIONS

When gradient gains and velocity are set you may use SCAMP for profiling. Velocity will not normally require modification as profiles are obtained. However gradients in the water column can change quickly in response to surface wind conditions. Profile groups should be uploaded from time to time and gradients reviewed to insure that they continue to be set appropriately for current conditions.

http://www.pme.com/USB_smanual/adjust.htm (3 of 3) [10/5/2005 5:12:06 PM]

Disassembly & Storage

Precision Measurement Engineering,

Inc.

Disassembling & Storage of the SCAMP

When profiling is completed, disassemble the SCAMP by the following steps:

1.

Remove the SCAMP from its stand.

2. Remove the ballast weights if ballasted for downwards mode. Weights attached to the drag plate can optionally remain.

3. Disconnect the short I/O cable connector. Install the I/O connector cover.

4.

Loosen the thumb nuts and remove the drag unit.

5.

Loosen the thumb nut on the 5151 float retainer ring and move the floats away from the guard end if required.

6.

Completely rinse the entire SCAMP, drag unit, and cable in tap water.

7. Rinse the sensors, shafts, and seal nuts with a squeeze bottle of de-ionized water.

8.

Rinse the pressure transducer by directing a stream of de-ionized water from the squeeze bottle into the 1/4" hole in the release endcap. There is a small secondary hole drilled to allow this water stream to exit. Do not place the nose of the squeeze bottle in the hole since it may penetrate far enough to damage the fragile pressure transducer diaphragm, or cut the nose so that it cannot penetrate too deeply . Do this operation for at least 15 seconds.

9. Unscrew and remove the sensor guard.

10.

Install the sensor cover. If storing the SCAMP for a long period, or if the SCAMP has a fluorometer installed, leave the sensor cover empty and leave the plug out so that the interior will dry. Otherwise, partially fill the cover with tap (not distilled) water. Check for http://www.pme.com/USB_smanual/disassemble.htm (1 of 2) [10/5/2005 5:12:11 PM]

Disassembly & Storage any foreign materials.

11.

Remove the 5151 float retainer ring, floats, and cable or retrieval line loop. Pack these.

12.

If saltwater contamination of the I/O connector is suspected, rinse out both connectors with de-ionized water, blow out, and dry. Leave SCAMP I/O connector cover off so that the inside of the SCAMP I/O connector dries..

13.

Disassemble the sensor guard if it is to be packed for shipping.

STORAGE

If the SCAMP is to be stored for an extended period, then remove the batteries.

Flood the sensor cover with distilled water several times to clean the sensors of salt. Drain the sensor cover completely . Store with the plug removed so that the sensors dry completely. Before using the

SCAMP again, fill the sensor cover with water similar to the expected environment for several hours prior to use.

When disassembled, cover the fluorometer ports with electrical tape to prevent dust and particles from entering.

http://www.pme.com/USB_smanual/disassemble.htm (2 of 2) [10/5/2005 5:12:11 PM]

SCAMP Control Dialog - Overview

Precision Measurement Engineering, Inc.

SCAMP Control Dialog

Overview

The SCAMP Control Dialog (SCD) performs data upload, display, and control tasks. The Dialog operates on a P.C. computer under Windows 2000. The dialog communicates with SCAMP via the

host computer’s USB port. See the Software Installation

section of this manual for installation

instructions and the Establishing Communication

section for SCAMP connection instructions.

The SCAMP Control Dialog begins its operation by attempting to connect to an attached SCAMP. It will service only one SCAMP at a time. If no SCAMP is connected, then the SCD displays an error message and terminates. If a SCAMP is found, then the SCD attempts to load profile data. If profile data are available, then the SCD displays an upload notification and loads the data. After all data is uploaded, then the SCD erases SCAMP’s flash memory and continues. If no profile data are available, then the upload is skipped and the SCD continues by reading parameters from SCAMP. After parameters are read, SCAMP’s internal time clock is set to the host computer’s time and the following screen is displayed.

http://www.pme.com/USB_smanual/overview.htm (1 of 2) [10/5/2005 5:12:16 PM]

SCAMP Control Dialog - Overview

The various tabs allow the user to set SCAMP parameters, view data from the last uploaded profile (if uploaded during this SCD session), and perform tests of the SCAMP.

SCD operation can be ended by clicking the OK or Cancel buttons, or by clicking the X box in the upper right corner. If the OK button is clicked the SCD transmits the current parameters (which may have been modified by the user during the SCD session) back to SCAMP and then terminates. If the

Cancel button or the X box is clicked, then the SCD does not transmit parameters (and so any changes are lost) and simply terminates.

Note that SCAMP is updated only when 'OK' is selected, NOT when changes are made to the individual controls within the dialog.

http://www.pme.com/USB_smanual/overview.htm (2 of 2) [10/5/2005 5:12:16 PM]

SCAMP Control Dialog - Data Upload

Precision Measurement Engineering,

Inc.

SCAMP Control Dialog

Data Upload

When the SCAMP Control Dialog (SCD) is begun it automatically establishes communication with the SCAMP and uploads any profile data that may be available. Data are uploaded profile-by-profile and are stored on the disk at the current directory location. Profiles are named with the following format: ddmmmyyyy hhmmss.raw

ddmmmyyyy hhmmss.txt

Two files are recorded for each uploaded profile. The *.raw file contains SCAMP parameters and raw data in binary format. The *.txt file contains SCAMP parameters and other info in text format, but does not contain any measured data.

The file name is constructed from SCAMP’s internal time. The time used is the time that one of

SCAMP’s start conditions became true thereby starting the profile. For example if SCAMP was programmed with a 60 second delay and a start depth of 10 meters, and it either reached 10 meters or 60 seconds elapsed on the 15 th

of January 2004 at 16:24:03 SCAMP internal time, then the files recorded would be named:

15JAN2004 162403.raw

15JAN2004 162403.txt

These files would appear in the current directory, usually the directory that contains the SCAMP

Control Dialog. (Note that SCAMP internal time is set to the host computer’s time each time the

SCAMP Control Dialog begins operation.)

If SCAMP has acquired multiple profiles since the last data uploading, then all profiles are loaded in sequence and a pair of files are recorded for each.

The *.txt file is text format and can be viewed with any text reader such as Notepad. The *.raw

file is binary format and can be accessed only with special software supplied by PME. Two methods are supplied: one for

Matlab and a ENGUNIT.EXE program that converts raw data

into ASCII files of calibrated engineering units. C-language source for ENGUNIT is available from PME for users who want to create their own data interface.

http://www.pme.com/USB_smanual/dataupload.htm [10/5/2005 5:12:24 PM]

SCAMP Control Dialog - Mission Tab

Precision Measurement Engineering, Inc.

SCAMP Control Dialog

Mission Tab

The Mission tab of the SCAMP Control Dialog allows users to define SCAMP’s mission. The Mission tab is shown below.

http://www.pme.com/USB_smanual/missiontab.htm (1 of 3) [10/5/2005 5:12:33 PM]

SCAMP Control Dialog - Mission Tab

Operational Mode

The user may select the profiling direction. See the Upwards Profiling or

Downwards Profiling sections

elsewhere in this manual.

Starting Conditions for Data Recording

The user may define the starting conditions. The starting conditions are ‘OR’ed meaning that if any starting condition becomes true then SCAMP will start collecting data.

Begin Delay sets the amount of time (in seconds) that SCAMP will wait after SCAMP is awakened by the magnet before recording data. SCAMP will start recording data after this time unless another starting condition becomes true first.

Begin Depth sets the depth (in meters) that SCAMP will start recording data. Each time SCAMP is awakened by the magnet it adjusts its depth channel coefficients so that the current pressure results in a computed depth of 0 meters. After this adjustment it samples the current depth about once per second. If the current depth is greater than the Begin Depth entry data recording begins.

Ending Conditions for Data Recording

Acquire Scans sets the number of scans SCAMP will acquire before it stops recording data. SCAMP scans at 100 scans per second so if the number entered is divided by 100 the number of seconds of data recording is obtained.

Strategy

The conditions should be defined so that data are recorded in an optimum way and so that the user can synchronize SCAMP’s motion through the water to the data recording.

In downwards mode, if near-surface measurements are desired, the Begin Depth condition is not very useful. The first reason is that SCAMP is unable to measure depth until the release end (containing the pressure transducer) is under water. Secondly, SCAMP’s depth channel has inaccuracies that can amount to roughly 25 cm. The best selection is to set Begin Depth to 1 meter, thereby insuring a start, but set time to something like 30 seconds or however much is required to get the SCAMP into deployment position after magnetic awakening. SCAMP indicates by activating its release screw that a profile is beginning and can be released at this time.

In downwards mode where deeper water is to be measured the Begin Delay condition should be set to insure sufficient time is available to deploy SCAMP and have it sink to the starting depth. The Begin

Depth should be set to the starting depth desired. Do not set Begin Time to a huge value since it is important that SCAMP begin recording data in case it never reaches the desired starting depth. If

SCAMP does not begin (and end) recording data it cannot be magnetically awakened for a new profile.

See the

Aborting a Profile

section in this manual.

In upwards mode the Begin Delay condition should be set to give sufficient time for SCAMP to be deployed and reach the Begin Depth. Do not set Begin Time to a huge value for the same reason described in the paragraph above.

If a start from the bottom is desired, Begin Depth should be set to a value larger than the depth of the http://www.pme.com/USB_smanual/missiontab.htm (2 of 3) [10/5/2005 5:12:33 PM]

SCAMP Control Dialog - Mission Tab bottom, and Begin Delay set to a value that will allow SCAMP to reach the bottom. SCAMP will then descend to the bottom but not reach the Begin Depth, wait out the remaining delay, and release its weight for ascent.

In all cases data recording must be ended by setting the Acquire Scans parameter. When travelling at

10 cm/sec SCAMP acquires 1000 scans per meter. Velocity will vary so Acquire Scans should be set to give a few meters extra data recording beyond the length of data recording desired in the water column.

http://www.pme.com/USB_smanual/missiontab.htm (3 of 3) [10/5/2005 5:12:33 PM]

SCAMP Control Dialog - System Tab

Precision Measurement Engineering, Inc.

SCAMP Control Dialog

System Tab

The System tab of the SCAMP Control Dialog allows users to set SCAMP system parameters and to view results. The System tab is shown below.

http://www.pme.com/USB_smanual/systemtab.htm (1 of 2) [10/5/2005 5:12:40 PM]

SCAMP Control Dialog - System Tab

Gradient Channels

SCAMP’s gradient channel gains can be set by entering a value between 1 and 128 into the Gradient

Gain box. This is applied to the Gradient Channel shown in the drop down box. SCAMP inplements gradient gains by the formula

Gain = 256/NGRAD

Where NGRAD is an integer between 2 and 255. Not all possible gradient gains can be implemented by this scheme. When focus shifts from the Gradient Gain edit box the SCAMP Control Dialog computes the closest NGRAD and then re-computes and re-displays the Gradient Gain.

Gradient Gains must be set in response to environmental conditions (see Velocity and Gradient Gain

Adjustment section elsewhere in this manual). If at least one profile was uploaded from SCAMP at the

start of the SCAMP Control Dialog session then clicking the Display button will produce a plot of the selected gradient channel from the most recently uploaded profile. If no profiles were uploaded this session then the Display button is inactive.

Velocity

SCAMP’s velocity must be set by ballasting in response to environmental conditions (see

Velocity and

Gradient Gain Adjustment

section elsewhere in this manual). If at least one profile was uploaded from

SCAMP at the start of the SCAMP Control Dialog session then clicking the Display button will produce a plot of the selected gradient channel from the most recently uploaded profile. If no profiles were uploaded this session then the Display button is inactive.

Battery Status

When the SCAMP Control Dialog begins its operation it reads SCAMP’s internal parameters. These include the battery voltage at the time SCAMP is connected to the USB port. The voltages read are displayed on the left side with progress bars on the right that give the approximate remaining life of each battery. The user should insure that sufficient battery life remains for the next profile or group of profiles. SCAMP will operate about 4 hours on a set of Duracell batteries. SCAMP will obtain about

1500 meters of data in 4 hours, not counting time spent awaiting a Starting Condition. See the

Checking the Batteries

section for instructions on replacing batteries.

http://www.pme.com/USB_smanual/systemtab.htm (2 of 2) [10/5/2005 5:12:40 PM]

SCAMP Control Dialog - Test Tab

Precision Measurement Engineering, Inc.

SCAMP Control Dialog

Test Tab

The Test tab of the SCAMP Control Dialog allows users to perform tests of SCAMP and to view results. The Test tab is shown below.

http://www.pme.com/USB_smanual/testtab.htm (1 of 2) [10/5/2005 5:12:46 PM]

SCAMP Control Dialog - Test Tab

Analog Test

SCAMP’s electronic circuits contain a A/D converter and also an offset and gain amplifier. The analog test programs the gain = 1 and ramps the offset through its full range while measuring the analog voltage produced with the A/D. This effectively tests the A/D converter and the offset section of the offset and gain amplifier. Channel electronics and sensors are not tested.

Channel Test

The Channel Test causes SCAMP to acquire a short amount of selected channel readings that are uploaded and displayed by the SCAMP Control Dialog. This is useful to establish that an individual channel is operating correctly.

Release Motor Test

SCAMP activates its release motor for 2 seconds in response to this test.

http://www.pme.com/USB_smanual/testtab.htm (2 of 2) [10/5/2005 5:12:46 PM]

SCAMP Control Dialog - Channel Tab

Precision Measurement Engineering, Inc.

SCAMP Control Dialog

Channel Tab

The Channel tab of the SCAMP Control Dialog allows users to set SCAMP’s channel parameters. Users will not normally have to make changes on this tab if PME supplies sensor calibrations, except in the case where a user must install a spare sensor. The Channel tab is shown below.

http://www.pme.com/USB_smanual/channeltab.htm (1 of 2) [10/5/2005 5:12:51 PM]

SCAMP Control Dialog - Channel Tab

Name

This is the name defined for the channel information displayed elsewhere on this tab. Different channels can be selected from this drop-down box. The user may not change channel names on this tab.

Units

The user may change channel units by entering new units here. This information will appear in *.TXT

files and on Matlab plots but is informational only. It is not used in any computations. PME strongly advises that Units not be changed since some processing routines depend upon certain units for correct operation.

CH#

SCAMP has 32 channels numbered 0 to 31. This displays the channel number.

Status

ON channels appear in the data. OFF channels do not. This applies only to the data stream. Channels operate electrically and consume power without regard to their Status.

Sensor ID

This is an informational string and is not used in processing. It appears in *.TXT file outputs.

Sensor Position

These boxes give the sensor position (in mm) relative to a datum on the SCAMP. All sensors are measured relative to the same datum so sensor-sensor distances can be computed without knowing the actual datum used.

Calibration

SCAMP raw data are converted to engineering units by using the relation:

EU = C0 + C1 * CR + C2 * CR^2 + C3 * CR^3

Where EU is the engineering unit (deg C for a temperature channel for example) and CR is the channel ratio observed. C0..C3 are supplied by PME for sensors we calibrate. Users must supply their own

C0..C3 if they perform calibrations.

Analog Processing

These parameters show the scaling that SCAMP applies to channel outputs. Scaling is complex and modifications to these parameters should not be attempted by users.

http://www.pme.com/USB_smanual/channeltab.htm (2 of 2) [10/5/2005 5:12:51 PM]

SCAMP Control Dialog - Setup Tab

Precision Measurement Engineering, Inc.

SCAMP Control Dialog

Setup Tab

The Setup tab of the SCAMP Control Dialog allows PME to define SCAMP’s channels when setting up new SCAMPs. Users will not have access to this tab unless they enter the correct Authorization

Password number. PME may supply the Authorization Password number at its discretion. Contact

PME.

http://www.pme.com/USB_smanual/setuptab.htm (1 of 2) [10/5/2005 5:12:57 PM]

SCAMP Control Dialog - Setup Tab http://www.pme.com/USB_smanual/setuptab.htm (2 of 2) [10/5/2005 5:12:57 PM]

Hardware - Overview

Precision Measurement Engineering,

Inc.

Discussing the Hardware

OVERVIEW

The hardware section presents a description of the electrical operation of the SCAMP, organized by sensor or circuit card location, and roughly in the order that the measured information is processed by the circuits. It is provided to give you insight into how information about the physical environment is transformed into measurements, and to give you a basis for understanding the software morphology

The paramount design criteria in both the SCAMP electrical design and mechanical packaging is to obtain a measurement system designed for spectral analysis of measurements and having the lowest measurement noise possible. All other features are chosen with this criteria in mind.

Certain features that may seem odd at first, such as three different battery voltage supplies or the short battery life, make sense when measurement noise is considered.

Processing begins with the sensors shown at the upper left section of the general

block diagram

of the SCAMP system. The various sensors operate in different ways depending upon the parameter being sensed. In all cases, these together with the sensor electronics, produce an analog voltage that is in some way related to at least one and possibly more sensed parameters.

In some cases, the analog voltage from a sensor electronics is routed to a gradient, programmable gain, and filter processor sequence. These signals are marked 'TO d/dt' on the diagram and appear at the lower left. These analog circuits compute the time rate of change of the sensor electronics output voltage, then amplify and filter the result.

There are housekeeping circuits that provide voltages related to SCAMP internal features such as battery voltages or circuit/ temperature. These are shown in the four boxes at the lower left.

Like the environmental sensors, these circuits have voltage outputs.

Voltages from the sensor electronics, from the gradient and filter processors, and from the housekeeping circuits are all routed to the A/D converter through two 16:1 analog multiplexors and one 2:1 multiplexor. The A/D converter is able to produce a 16 bit digital output for each of the voltage inputs.

Multiplexor channels 0 to 15 arrive at the A/D converter after processing by the analog offset and gain circuits. These circuits allow the A/D input voltage range of +/- Vref (+/- 3.000 volts) to be mapped onto sub-ranges of channel 0 to 15 voltages. Multiplexor channels 16 to 31 are directly http://www.pme.com/USB_smanual/hardware1.htm (1 of 2) [10/5/2005 5:13:03 PM]

Hardware - Overview routed to the A/D.

The digital output of the A/D converter is received by the Central Processor Unit (CPU) where it is stored in the RAM memory. When a profile is ended, the data in the RAM memory is appended to SCAMP parameters and the result moved to the MultiMedia Card (MMC). When

SCAMP is connected to a host computer all profiles in MMC are automatically uploaded and stored on the host computer disk drive.

The CPU program is stored within the flash memory and utilizes some of the RAM memory. The

CPU provides USB I/O for normal SCAMP operations and RS232 I/O for housekeeping functions. The CPU is always powered by 3.3 Volts but the other voltage supplies are turned off when not required. During sleep time the CPU enters a low power mode, but can respond to magnetic switch generated interrupts.

http://www.pme.com/USB_smanual/hardware1.htm (2 of 2) [10/5/2005 5:13:03 PM]

Hardware - Temp sensor

Precision Measurement Engineering,

Inc.

Discussing the Hardware

FAST TEMPERATURE SENSOR

The fast temperature channel consists of the 5316 Fast Temperature sensor and associated circuitry. These circuits are located on the 5264 uDO & Dual T circuit board.

The 5316 Fast Temperature sensor consists of a fast thermistor and a compensating resistor.

The image below shows the sensor. The compressed seal fitting uses the nut shown to compress an internal Teflon ferrule.

Completed sensors are extensively thermal-cycled and pressurized before being electrically tested. Our electrical test will detect any cracks that occur anywhere in the assembly.

The sensor is designed to be mounted on the SCAMP and is bent such that it can be closely aligned with other sensors.

The fast thermistor is a Thermometrics FP07DB104N. For more information, view the

manufacturer's data sheet . This thermistor is specified as having a resistance at 25 deg C of

100K ohms, plus or minus 25K ohms. The resistance of this type sensor changes in response to the temperature of the sensor. The change is not a linear function of temperature and can be described by

R(T) = R(To) * exp [Beta * (1/T - 1/To))] where

T - sensor temperature, deg K,

To - reference temperature, 296.15 deg K (25 deg C), http://www.pme.com/USB_smanual/hardware2.htm (1 of 2) [10/5/2005 5:13:09 PM]

Hardware - Temp sensor

R(To) - sensor resistance at reference temperature,

R(T) - sensor resistance at temperature T,

Beta - thermal sensitivity of thermistor, 4015.

The thermal time constant is given by the manufacturer as 0.007 seconds giving an approximate response frequency of 22 Hz. At the SCAMP travel rate of 10 cm/sec, this response is 2.2

cycles/cm, 220 cycles/meter.

PME allows a range of +/- 25% variation of R(To) that makes this part much easier, and therefore less expensive for the manufacturer to build. Unfortunately, this tolerance is too wide to allow a single circuit to operate all possible thermistors. PME therefore measures every thermistor when received and determines a compensating resistance. This resistor is included within the 5316 sensor shaft. The

electrical wiring diagram shows the wire color definition.

See REFERENCES

for more information.

The electronic circuit applies Vref reference voltage to R1. This voltage is divided by R1 and the thermistor resistance to produce a voltage that is a quasi-linear function of temperature. The circuit amplifies this voltage, filters it, and transmits it to the multiplexor and also to the gradient circuit. Fast T - example of a

typical calibration of this channel. Since a compensating resistor is

selected for each thermistor, this calibration will vary only slightly from sensor to sensor.

http://www.pme.com/USB_smanual/hardware2.htm (2 of 2) [10/5/2005 5:13:09 PM]

Hardware - Fast CT

Precision Measurement Engineering,

Inc.

Discussing the Hardware

FAST CT SENSOR

The 5127 Fast Conductivity Temperature sensor consists of a Fast Temperature sensor (5316) and a Fast Conductivity sensor (5346) potted together into one stainless steel tube. The

mechanical drawing shows that the sensor is designed to be mounted on the SCAMP and is

bent such that it can be closely aligned with other sensors. The image below shows a picture of this sensor.

The Fast CT sensor is a combination of the sensors used in

the Fast Temperature sensor

(1) and Fast Conductivity sensor (2). The position of the Fast Conductivity sensor's electrodes is adjusted to be slightly ahead of the Fast

Temperature sensor and 1 mm away. Please refer to the sections above for descriptions of these two sensors and wiring diagrams.

The Fast CT sensor is supplied when accurate alignment of the Fast Conductivity and Fast Temperature sensors is required or when many sensors are mounted on the

SCAMP.

The Fast CT sensor provides the two fast sensors used by the fast temperature channel and fast conductivity channel.

Head, M. J., The Use of Miniature Four-Electrode Conductivity Probes for High Resolution

Measurement of Turbulent Density or Temperature Variations in Salt-Stratified Water Flows

,

Ph.D. Thesis, University of California, San Diego, 1983.

http://www.pme.com/USB_smanual/hardware4.htm [10/5/2005 5:13:15 PM]

Hardware - Acc CT

Precision Measurement Engineering,

Inc.

Discussing the Hardware

ACCURATE CT SENSOR

The 5199 Accurate Conductivity Temperature sensor consists of a conductivity sensor, a thermistor, and a compensating resistor. The ceramic substrate and thermistor housing are molded thus causing a more reliable seal. Once the assembly is inserted into the stainless steel shaft, shrink tubing is used to cover this interface.

The mechanical drawing

shows that this sensor is designed to be mounted on the SCAMP, but has a straight shaft and will not come into close alignment with the other sensors.

The thermistor is a Thermometrics T1201/B3A-B07KA104N. For more information, view the

manufacturer's data sheet.

This thermistor is specified as having a resistance at 25 deg C of

100K ohms, plus or minus 20K ohms. The resistance of this type sensor changes in response to the temperature of the sensor. The change is not a linear function of temperature and can be described by

R(T) = R(To) * exp [Beta * (1/T - 1/To))] where http://www.pme.com/USB_smanual/hardware5.htm (1 of 2) [10/5/2005 5:13:24 PM]

Hardware - Acc CT

T - sensor temperature, deg K,

To - reference temperature, 296.15 deg K (25 deg C),

R(To)- sensor resistance at reference temperature,

R(T) - sensor resistance at temperature T,

Beta - thermal sensitivity of thermistor, 4243.

The thermal time constant of this sensor is approximately 0.2 seconds.

PME allows a range of +/- 20% variation of R(To) that makes this part much easier, and therefore less expensive for the manufacturer to build. Unfortunately, this tolerance is too wide to allow a single circuit to operate all possible thermistors. PME measures every thermistor when received and determines a compensating resistance. This resistor is included with each sensor.

The conductivity sensor is manufactured by PME. The sensor consists of an alumina substrate on which 4 electrodes, two on each side, have been deposited by a high temperature thick film process. These electrodes are made of platinum and are platinized. The sensor functions by being a dimensionally stable contact between the salt water and the electronic circuit connected to the sensor. The electronic circuit makes a 4 terminal A.C. measurement of the conductance of

the salt water. The Accurate Conductivity diagram

shows the sensor's wiring and wire color definition.

The measurement made by the combination of the conductivity sensor and its electronic circuit can be shown to result from a volume weighted average of the conductivity of the salt water near the sensor and within the sensor cylinder. This sensor has a spatial response that is effectively a low pass filter in space. The roll off point is presently unknown but can be estimated as a running boxcar average of conductivity over the length of the cylinder, 1.5 cm. There is also a time related response, which is presently unknown.

The purpose of the Accurate CT sensor is to provide a convenient calibration reference for the faster sensors, especially the Fast Conductivity sensor. Because of its small size, the Fast

Conductivity sensor suffers from drift induced by fouling. The Accurate Conductivity sensor is much larger, has much larger electrodes, and resists fouling. Information from this sensor can be used to determine if the Fast Conductivity sensor is reporting correct values during a profile and to provide information to correct the Fast Conductivity sensor if it is not.

The accurate temperature channel circuitry is the same as that used for fast temperature. This circuitry is located on the 5132 Conductivity-Temperature circuit board. The accurate conductivity channel circuitry is the same as the fast conductivity circuitry with the exception of the scale selection resistor value and type and is located on the 5132 circuit board also.

http://www.pme.com/USB_smanual/hardware5.htm (2 of 2) [10/5/2005 5:13:24 PM]

Hardware - Turbidity sensor

Precision Measurement Engineering,

Inc.

Discussing the Hardware

TURBIDITY SENSOR

The turbidity sensor and electronic circuit are supplied to PME by D & A Instruments. The sensor operates by sending light into the water and measuring the amount reflected back by particles. The sensitivity for this sensor is adjustable. The most sensitive range is 0 - 100 NTU.

At the present time, this sensor cannot be used with the SCAMP's fluorometer. For more information, view the manufacturer's

manual .

http://www.pme.com/USB_smanual/hardware7.htm [10/5/2005 5:13:32 PM]

Hardware - Laser Turbidity Sensor

Precision Measurement Engineering,

Inc.

Discussing the Hardware

LASER TURBIDITY SENSOR

The Laser Turbidity sensor is designed to be used with SCAMP's

fluorometer . The sensor is

shown in the picture below. It is the black cylinder located at the center of the end cap, amoung the other sensor sealing nuts. http://www.pme.com/USB_smanual/hardware18.htm (1 of 3) [10/5/2005 5:13:43 PM]

Hardware - Laser Turbidity Sensor

The Laser Turbidity sensor operates in a similar way to the other SCAMP Turbidity sensor. The

sensor housing is shown in the following drawing. The sensor emits pulsed infrared light from a laser diode located in the cavity shown. These light pulses illuminate the water directly outside above the cavity. The photo diode detector within the cavity shown views a small section of the illuminated volume. If particles (turbidity) are present in the viewed volume then some of the light will be scattered into the photo diode and detected. The small electrical signal that results is amplified and recorded within SCAMP.

The turbidity sensor components and electronic circuit are supplied to PME by D&A

Instruments . The housing shown above is manufactured by PME and parts supplied by D&A are assembled into it.

The Laser Turbidity sensor differs from SCAMP's other turbidity sensor in two ways. First, it can be used concurrently with SCAMP's fluorometer. Second, it is more sensitive to turbidity since the laser emits somewhat more light and the viewing angle used in the Laser Turbidity sensor is better to receive scattered light. http://www.pme.com/USB_smanual/hardware18.htm (2 of 3) [10/5/2005 5:13:43 PM]

Hardware - Laser Turbidity Sensor

The sensor has two sensitivities, selectable via a jumper on the circuit board within SCAMP.

Highest range of sensitivity is 0 to 300 NTU.

http://www.pme.com/USB_smanual/hardware18.htm (3 of 3) [10/5/2005 5:13:43 PM]

Hardware - Pressure sensor

Precision Measurement Engineering,

Inc.

Discussing the Hardware

PRESSURE SENSOR

The pressure sensor is a PAA-10 supplied to PME by Keller PSI. For more information, view the

manufacturer's data sheet.

Maximum sensor pressure varies depending upon the depth

specified for the SCAMP; 10 bar is typical. The sensor consists of a stainless steel diaphragm that transmits pressure via a silicone oil filled cavity to a thin diaphragm chemically milled on a silicon substrate. Strain sensitive resistors are diffused into this diaphragm in a bridge configuration. Applied pressure causes the values of these resistors to shift in a linear manner.

The pressure channel circuitry is simply a constant current source supplying the sensor bridge and a gain of 100 D.C. differential amplifier connected across the sensor bridge. This circuit is located on the 5026 Main Analog circuit board.

The Depth Channel Calibration graphs shows a typical pressure calibration. Both the slope and

the zero intercept will vary from sensor to sensor.

http://www.pme.com/USB_smanual/hardware8.htm [10/5/2005 5:13:49 PM]

Hardware - Gradient channel

Precision Measurement Engineering,

Inc.

Discussing the Hardware

GRADIENT CHANNELS

There are several gradient channels within the SCAMP. The actual number depends upon the options purchased and can be as many as 8. All are identical. They are analog signal processors that first compute 0.2 times the time derivative of the voltage supplied to the circuit, then amplify by a programmable gain given by:

G = 256/NGAIN where

G - circuit gain block gain Volt/Volt

NGAIN - gain integer 1 to 255 then filter the output.

These two graphs show the frequency response and step response, while this

graph shows the

ramp response for a gradient channel with gain of 1. The circuits are located on the 5267 Dual

Gradient & Filter circuit boards, two complete circuits to a board.

http://www.pme.com/USB_smanual/hardware9.htm [10/5/2005 5:13:56 PM]

Fluorometer sensor

Precision Measurement Engineering,

Inc.

Discussing the Hardware

FLUOROMETER SENSOR

The fluorometer channel consists of a fluorometer endcap , blue excitation L.E.D., optical filters,

mirrors, lens, and associated circuitry. The circuits are located on the 5500 circuit board. The

other components are located within the fluorometer endcap. The fluorometer's block diagram

shows the electrical circuit.

The fluorometer operates by exciting the sample volume with a pulsing 455 NM blue light at

1000 Hz with a 50% on, 50% off duty cycle. Under this condition, chlorophyll emits a pulsing red

685 NM light that is received by a photo-diode. The photo-diode output is first amplified by a DC amplifier, then strongly amplified by an AC amplifier. These circuits together have the effect of rejecting constant ambient lighting and only passing the pulsing chlorophyll response. The output of the AC amplifier is rectified by a phase synchronous rectifier, filtered, and passed to the circuit output where it is connected to the SCAMP fluorometer channel.

The circuit has an adjustment that controls the amount of amplification, thus allowing overall sensitivity control. The system has a linear sensitivity to fluorescence, but has significant output with only distilled water in the sample volume due to the slight fluorescence of various materials used to implement the optical components. The photos below show the fluorometer addition.

http://www.pme.com/USB_smanual/hardware10.htm (1 of 2) [10/5/2005 5:14:02 PM]

Fluorometer sensor http://www.pme.com/USB_smanual/hardware10.htm (2 of 2) [10/5/2005 5:14:02 PM]

Hardware - PAR sensor

Precision Measurement Engineering,

Inc.

Discussing the Hardware

PAR SENSOR

The PAR (Photosynthetically Active Radiation) responds to ambient light in the photosynthetically active band. The sensor is essentially a photo diode that receives light through a bandpass optical filter. The sensor produces an electrical current that is directly proportional to the flux of photons into the photo diode. This current is amplified by circuits on

SCAMP's 5550 fluorometer circuit assembly, low pass filtered, and digitized by SCAMP's A/D.

The sensor is a Li-Cor LT-192SA. For more information about this sensor see the

LT-192SA manual .

The PAR sensor is directional. Presently it can be mounted so that it looks away from SCAMP from the weight release end as shown in the picture below. When connected this way it is sensitive to downwelling light when SCAMP profiles in downwards mode and to upwelling light for SCAMP's upwards mode profiling. It is possible to place the sensor on the other side of the drag plate to get downwelling light in upwards mode. This requires modifications. Contact PME for more information. The picture also shows two white floats that must be added to compensate the system for the extra weight of the PAR sensor. http://www.pme.com/USB_smanual/hardware11.htm (1 of 2) [10/5/2005 5:14:08 PM]

Hardware - PAR sensor

The PAR sensor is connected to circuits inside SCAMP by using a special I/O cable assembly, and a special 4 pin I/O connector. The I/O cable assembly connects to the SCAMP I/O connector, then 'Ys' to connect to the PAR sensor and also to provide the usual I/O connection for uploading SCAMP.

USB SCAMP has a slightly different I/O connector mounted on the SCAMP release end cap.

This connector has proven quite susceptable to damage due to bending the attached cable near the connector. PME provides a cable protector as shown in the previous picture. The protector is a white cylinder that slips over the connector and cable. The end with the small transverse hole goes against SCAMP's end cap. The cylinder is secured into position by threading a small cable tie throught the transverse hole in such manner that it does not interfere with the release motor or weight cable. The pencil in the above picture points to the protector and cable tie.

PME does not calibrate the PAR sensor. Instead PME takes the calibration provided by LI-COR and implements it taking into account how the SCAMP electronics responds to sensor current.

PAR sensors are usually ordered when SCAMP's fluorometer is installed. SCAMPs can be retrofitted to add PAR at any time.

http://www.pme.com/USB_smanual/hardware11.htm (2 of 2) [10/5/2005 5:14:08 PM]

Hardware - Programmable Offset

Precision Measurement Engineering,

Inc.

Discussing the Hardware

OFFSET AND GAIN BLOCKS

These two circuits and analog signal processors that implement

Vad = (256/NGAIN) * (NOFF/256 +/- Vin) where

Vad - circuit output voltage

Vi - circuit input voltage

NGAIN - gain integer, 0 to 255

NOFF - offset integer, 0 to 255

Sub-ranges of channel voltage ranges can be mapped to the full A/D input voltage range by proper selection of NOFF and NGAIN. SCAMP internal software sets NOFF and NGAIN for each channel prior to A/D conversion of the channel. Thus each of channels 0 to 15 can have its own scaling using the same circuitry. Channels 16 to 31 do not have this scaling ability.

http://www.pme.com/USB_smanual/hardware12.htm [10/5/2005 5:14:13 PM]

Hardware - Additional sensors

Precision Measurement Engineering,

Inc.

Discussing the Hardware

ADDITIONAL SENSORS

There are unused analog channel connections to the multiplexors in the SCAMP system. These are available for future sensors that may be added by PME or by the customer. Extra +/- 5 volt analog power is available as well as extra +5 volt digital power. The SCAMP chassis contains a small amount of extra space depending upon which sensors are installed, and can be custom-engineered to provide substantial extra room by simply making the SCAMP longer.

HOST software is designed to include future sensor channels easily.

http://www.pme.com/USB_smanual/hardware13.htm [10/5/2005 5:14:22 PM]

Hardware - A/D converter

Precision Measurement Engineering,

Inc.

Discussing the Hardware

A/D CONVERTER

The A/D converter is a 20K sample/second, 16 bit circuit block. It maps the analog voltage range of +/- 3.000 volts onto the integer range +32768 to -32767 according to the approximate equation

Nad = Vad/Vref * 32768 where

Nad - A/D digital output

Vad - A/D input analog voltage

Vref - reference voltage (+3.000 volts) http://www.pme.com/USB_smanual/hardware14.htm [10/5/2005 5:14:27 PM]

Hardware - CPU

Precision Measurement Engineering,

Inc.

Discussing the Hardware

CPU

The CPU block is mostly implemented within a single micro-controller. Basic clock speed is 36

MHz with a pipe-lined instruction execution that allows effectively single-cycle instruction execution. This micro-controller is quite versatile and is easily programmed via a sophisticated development environment.

http://www.pme.com/USB_smanual/hardware15.htm [10/5/2005 5:14:58 PM]

Hardware - Memory

Precision Measurement Engineering,

Inc.

Discussing the Hardware

MEMORY

SCAMP contains three types of memory:

512K X 8 low power RAM,

512K X 8 flash, and a

64M X 8 MultiMedia Card.

The RAM and flash are fixed, but the MultiMedia Card can be removed and replaced with larger or smaller versions. (Contact PME if MultiMedia Card replacement is required!)

During profiling data flow into the RAM. At the completion of a profile, the data are appended to

SCAMP parameters (calibrations etc.) and saved as a profile block in the MultiMedia Card

(MMC) memory. Profiles are stored until uploaded to the host computer. The MMC is a flash memory and will store profiles even if all batteries are removed from SCAMP. The storage format is not MSDOS compatible. The MMC cannot be read in a PC card reader.

The flash memory holds SCAMP’s internal program.

http://www.pme.com/USB_smanual/hardware16.htm [10/5/2005 5:15:05 PM]

Hardware - References

Precision Measurement Engineering,

Inc.

Discussing the Hardware

REFERENCES

Head, M. J. (1983) The Use of Miniature Four-Electrode Conductivity Probes for High Resolution

Measurement of Turbulent Density or Temperature Variations in Salt-Stratified Water Flows,

Ph.D. Thesis, University of California, San Diego.

Thermometrics (1995) , Thermistors Catalog, Thermometrics Inc., 808 U.S. Hwy 1, Edison,

New Jersey 08817, Phone 1(800) 246-7019. This catalog is free upon request from

Thermometrics. www.thermometrics.com

http://www.pme.com/USB_smanual/hardware17.htm [10/5/2005 5:15:13 PM]

Calibration - Conductivity

Precision Measurement Engineering,

Inc.

Discussing Calibration

CONDUCTIVITY

The range of sensitivity for both accurate and fast conductivity channels can be changed. In the case of accurate conductivity, one of three ranges can be selected. Many ranges are possible for fast conductivity since ranges are set by installation of a custom resistor. In all cases, the lower end of the range remains fixed at 0.05 S/m; only the upper limit is adjustable. SCAMP allows scaling of channels whereby parts of the range are mapped onto the full +/- 32768 integer range of the A/D.

Accurate conductivity is supplied with 3 ranges. These are selected by moving a jumper block located on the

accurate conductivity circuit.

This diagram shows this circuit and the jumper block

positions that correspond to the three ranges. Fast conductivity is supplied with only 1 range.

This is set by the value of the resistor connected in the 9 S/m position on the fast conductivity circuit. Other ranges are selected by removing this resistor and installing a replacement.

PME calibrates only the 0 to 9 S/m range for both accurate and fast conductivity channels. If a different range is selected, then the corresponding calibration coefficients must be modified or the SCAMP re-calibrated. The recommended approach is to select a new range resistor, then re-calibrate the SCAMP using an accurate independent reference. If this is not possible, then modify the calibration coefficients as described below.

The modification is as follows:

Definitions:

R

R'

CR

CR'

Resistor for which circuit is calibrated

New resistor

Channel output with resistor R

Channel output with resistor R'

C Conductivity in engineering units

C0..C3

Calibration equation coefficients with resistor R

C0'..C3' Calibration equation coefficients with resistor R' http://www.pme.com/USB_smanual/calibration1.htm (1 of 2) [10/5/2005 5:15:20 PM]

Calibration - Conductivity

When the SCAMP is calibrated, a relation between conductivity, in engineering units, and channel output is developed for R,

C = C0 + C1*CR + C2*CR^2 + C3*CR^3

The conductivity circuit operates in a way that causes CR to be very nearly a linear function of

R. At a given conductivity, the relation between circuit outputs and the resistors is approximately,

CR' /CR = R'/R

Solving this for CR and substituting into the calibration equation above gives

C = C0 + C1*(CR'*R/R') + C2*(CR'*R/R')^2 + C3*(CR'*R/R')^3

The new coefficients are identified from this form as,

C0' = C0

C1' = C1*(R/R')

C2' = C2*(R/R')^2

C3' = C3*(R/R')^3 giving,

C = C0' + C1'*(CR') + C2'*(CR')^2 + C3'*(CR')^3

The accurate conductivity channel is supplied with three scaling resistors. The normal values for these resistors are:

0.05 to 9 S/m range

0.05 to 0.5 S/m range

0.05 to 0.1 S/m range

75 ohms +/- 1%

1370 ohms +/- 1%

6810 ohms +/- 1%

In some cases, different resistors may be supplied for special ranges. The values are printed on the resistors next to the jumper. Please check these.

The fast conductivity channel is supplied with a single resistor that is chosen to match the actual conductivity cell in-use. You must measure this resistor and then calculate the correct resistor to use for the desired range.

The new range will not be as accurate as the calibrated range. The first major contributor is the

1% resistor tolerance. This can be avoided by removing the jumper and measuring the resistances involved directly with an accurate ohm meter. The second contribution results from the observation above that the CR, CR' relation is only approximately correct. Presently, the amount of error that this contributes is unknown, but is believed to be in the few percent range.

http://www.pme.com/USB_smanual/calibration1.htm (2 of 2) [10/5/2005 5:15:20 PM]

Calibration - Depth

Precision Measurement Engineering,

Inc.

Discussing Calibration

DEPTH

The SCAMP computes depth from measurements of ambient pressure made using the

SCAMP's pressure sensor. Depth is computed from pressure by knowing the density of the surrounding water. The SCAMP is supplied with calibration coefficients for fresh water. These are computed for the pressure - depth relation: depth (Meters) = 0.7030696 * pressure (PSI)

If depth is desired for different density, seawater for example, then an equation such as the one

above must be determined. Each of the C's (C0, C1, C2, C3) in the SCAMP Control Dialog

Channel Tab

the depth channel must be multiplied by the ratio of the coefficient in the new depth-pressure relation divided by 0.7030696.

For example, if the SCAMP is used in sea water of 35 ppt at 15 deg C, then density is shown in sigma-T tables: density (d) = acceleration of gravity (g) = conversion constant = pressure =

C0' =

C1' =

C2' =

C3' =

1.02599 gm/cm**3

980.621 cm/sec**2 (sea level)

1.45034E-5 PSI per dyne/cm**2 density * acceleration of gravity * depth

C0 / (d * g * 100 * 1.45034E-5 *

0.7030696)

C1 / (d * g * 100 * 1.45034E-5 *

0.7030696)

C2 / (d * g * 100 * 1.45034E-5 *

0.7030696)

C3 / (d * g * 100 * 1.45034E-5 *

0.7030696) http://www.pme.com/USB_smanual/calibration2.htm (1 of 2) [10/5/2005 5:15:27 PM]

Calibration - Depth

Note that the pressure sensor makes errors of a few tenths of a percent. Very slight density corrections such as the change in density of fresh water with temperature will not make significant contributions to the overall depth channel error.

http://www.pme.com/USB_smanual/calibration2.htm (2 of 2) [10/5/2005 5:15:27 PM]

Calibration - Fluorometer

Precision Measurement Engineering,

Inc.

Discussing Calibration

FLUOROMETER

The fluorometer is supplied uncalibrated. Calibration must be completed by the customer.

Quantitative calibration of the fluorometer is very difficult since the fluorescent response of features in the water column varies in response to environmental conditions that are difficult to predict or model in the laboratory. The best calibration method is to capture bottle samples at several depths, assay these chemically, then match SCAMP fluorometer response with the results. If no quantitative calibration is available, then the fluorometer will still provide a qualitative indication of the distribution of chlorophyll in the water column.

The fluorometer must have its full scale adjusted to match the maximum fluorescence expected.

It may be difficult to estimate this in advance. It is most likely that the fluorometer full scale will need to be adjusted on-site after viewing initial profiles. Alternatively, the elbows can be removed, one side of the fluorometer taped shut with black electrical tape, a sample poured in from the other side, and the second side taped shut. Tape is required to contain the sample and also to prevent ambient light from entering the measurement volume.

PME supplies the proper C0..C3 coefficients to record the output of the fluorometer in Volts.

Because of internal circuit conditions, the fluorometer only has linear output to 2 volts. It will produce voltages that exceed this value but these are not linearly related to the fluorescence intensity. The fluorometer should, therefore, be adjusted such that 2 volts maxim is observed when it is measuring the maximum fluorescence.

The fluorometer sensitivity adjustment is accomplished by adjusting the gain potentiometer (R13) located on the

5550 fluorometer circuit.

This potentiometer is accessed by opening the SCAMP ,

removing the long electronics cover , and locating the 5550 circuit card. Clockwise adjustment

increases the gain (sensitivity) while counter-clockwise adjustment decreases it. In addition,

PME supplies scaling resistors that, when installed on the proper circuit connector (located over

R8 on the 5550 circuit), multiply the fluorometer sensitivity by factors that are less than 1 (make the fluorometer less sensitive). Resistor connectors are marked with the appropriate factor.

(Important: Some SCAMPs have ribbon wiring harnesses that cover the scaling resistor connector. The resistor can be installed but the ribbon wiring harness must be temporarily

removed to do so. The wiring harness is fragile. Please see the Cautions section

for information on removing the harness. After installing the scaling resistor the harness must be http://www.pme.com/USB_smanual/calibration5.htm (1 of 2) [10/5/2005 5:15:38 PM]

Calibration - Fluorometer re-connected).

Note that a quick test to determine if the fluorometer is operational is to record a profile while inserting/removing a blade of grass from the fluorometer measurement tube. This can be compared to the effects from inserting/removing a bit of paper. Take care for preventing the entry of ambient light while measuring. Ambient light won't damage the fluorometer, but it ruins the measurement.

http://www.pme.com/USB_smanual/calibration5.htm (2 of 2) [10/5/2005 5:15:38 PM]

Calibration - Laser Turbidity Sensor

Precision Measurement Engineering,

Inc.

Discussing Calibration

LASER TURBIDITY SENSOR

The Laser Turbidity sensor may be calibrated in two ways. The most accruate way is to obtain a 5 gal black polyethelyene bucket, place calibration solutions in the bucket, and make measurements with

SCAMP having its sensor guard in the normal position. This most closely matches the actual deployment situation. The disadvantage is that large amounts of calibration standard solutions must be mixed.

An alternate method is to use small amounts of calibration solution that are poured into SCAMP's clear sensor cover. The disadvantage here is that there is a slight error due to the reflectivity of the sensor cover. The instructions below are for calibration using this alternate method.

Mix 1000 ml of 100 NTU solution from the 4000 NTU stock solution. This is accomplished by placing 25 ml of 4000 NTU stock into a 1000 ml graduated cylinder, then adding distilled water to make 1000 ml.

This will become the stock solution for further dilutions.

Set SCAMP Laser Turbidity channel calibration coefficients to C0=0.0, C1=0.0117188, C2 = 0.0, and C3

= 0.0. Set Units to "Volts". Select Noff = 0, Noffpolar = 1, NGain = 255.

Begin at the lowest turbidity of interest.

Carefully rinse SCAMP's sensors by pouring about 100 ml of distilled water into the sensor cover, rinsing the interior surfaces, then pouring the water out. There will be a small amount of water retained within the sensor cover. Shake as much of this out as possible.

Rinse the mixing container with distilled water.

Mix 100 ml of the desired turbidity standard solution by weight, assuming the density of both water and stock solution is 1 gm/ml.

Pour the standard into SCAMP's sensor cover. Verify there are no bubbles present on the turbidity sensor near the sensing volume. Hold SCAMP upright in a dark place and record 1000 or more scans. Note that the formazin particles will settle slightly over a few minutes time. Perform the measurement quickly.

Repeat the above steps, working up to the maximum turbidity.

The results of calibrating SCAMP SN0034 are shown in the table and figure below.

http://www.pme.com/USB_smanual/calibration6.htm (1 of 3) [10/5/2005 5:15:46 PM]

Calibration - Laser Turbidity Sensor

Weight of 100

NTU stock

0

10.44

20.82

31.47

41.12

Total Weight of Standard

100

101.05

100.77

102.12

100.58

50.57

60.61

71.16

80.03

90.54

100.85

100.04

100.04

100.62

100.09

100.32

100.85

NTU of

Standard

0

10.33

20.66

30.82

40.89

50.55

60.59

70.72

79.96

90.25

100.00

File Name

FILE1

FILE2

FILE3

FILE4

FILE5

FILE6

FILE7

FILE8

FILE9

FILE10

FILE11

Sensor Output

Voltage

0.0307

0.0984

0.1660

0.2280

0.2865

0.3469

0.4143

0.4755

0.5392

0.6175

0.6960

A linear least square fit of the data gives

Turbidity EU

(NTU) = -3.9075

+ 153.3055 *

Sensor Output (Volts)

The C1 calibration coefficient can not be entered directly into the C1 box on the SCAMP Control Dialog

Channel tab. It must be adjusted by multiplying by 0.01172. The values to be entered into the Laser http://www.pme.com/USB_smanual/calibration6.htm (2 of 3) [10/5/2005 5:15:46 PM]

Calibration - Laser Turbidity Sensor

Turbidity calibration coefficients are C0 = -3.9075, C1 = 1.797, C2 = 0.0, C3 = 0.0. Change the Units to

'NTU'. Add a comment giving the date of calibration or other information.

This completes the turbidity calibration. SCAMP should be tested by pouring a standard into the sensor cover, recording a data file, and examining the file to be sure that the correct NTU units are shown.

Range of Calibrated Response

The Laser Turbidity sensor electronics can produce up to approximately 2.5 Volts. The above calibration shows that about 379 NTU will cause the sensor to produce this maximum. The Span Limits may be changed as required to give additional sensitivity and reduce the range available. If this is done the calibration may optionally be repeated.

Calibration Standard Solution

4000 NTU Formazin solution, 500 ML (Cost USD $45.80) Avaliable from Hach Company

( www.hach.com

) their P/N 246149.

Dilution of Formazin solution:

Tstd = Tstock * (Vstock)/(Vstock + Vwater)

Tstd = turbidity of standard to be mixed up

Tstock = turbidity of stock solution (4000 NTU)

Vstock = volume of stock solution to which water is added

Vwater = volume of distilled water added to stock solution http://www.pme.com/USB_smanual/calibration6.htm (3 of 3) [10/5/2005 5:15:46 PM]

Calibration - References

Precision Measurement Engineering,

Inc.

Discussing Calibration

REFERENCES

Hitchman, Michael L., Measurement of Dissolved Oxygen, John Wiley & Sons pub., (1978).

Knapp, George P., et al., Dissolved Oxygen Measurements in Sea Water at the Woods Hole

Oceanographic Institution, Woods Hole Technical Report #WHOI-89-23 (1989).

Strickland, J.D.H. and Parson, T.R., A Manual of Sea Water Analysis, Bulletin No. 125

Published by the Fisheries Research Board of Canada (1960).

http://www.pme.com/USB_smanual/calibration4.htm [10/5/2005 5:15:52 PM]

Maintenance - Sensor Cover

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

HANDLING THE SENSOR COVER

1.

To remove the sensor cover, first inspect the water-filled cover to be certain that there are no large bits of foreign matter floating inside. These can snag or damage the microconductivity sensor. Next, remove the plug in the cover end and carefully drain the water from inside the cover. Angle the SCAMP so that the fast conductivity sensor is above the water flow. This helps keep the foreign matter away from this sensor. When all the water is drained, replace the plug so that it does not become lost.

2.

Next, carefully unscrew and remove the cover. Do this in a sitting position with the SCAMP rotated so that the Accurate CT sensor is either on the right or left side. Do this so that the relative slant of the sensor cover, as it comes off the retaining threads, is visible relative to the Accurate CT sensor. Since the Accurate CT sensor is the farthest sensor from

SCAMP's center, it is the most likely sensor to be damaged when removing the sensor cover. The pictures below show the proper hand position.

http://www.pme.com/USB_smanual/maintain1.htm (1 of 3) [10/5/2005 5:16:02 PM]

Maintenance - Sensor Cover

Do not shortcut this procedure!

If you try to separate the SCAMP from the cover with water in the cover, then you will find yourself holding the cover stationary and lifting the SCAMP away. The SCAMP is heavy and the slightest slip will break sensors.

If it is difficult to loosen the sensor cover, ( SCAMP turns within its outer tube), then don't tighten the SCAMP retaining ring at the release end. Over tightening the retaining ring can pull the endcaps loose and damage them . Instead, open the

SCAMP as though you intended to change the batteries and grip the metal support and covers while loosening the sensor cover. Only loosen the cover, don't unscrew it much. Close the SCAMP and then remove the cover. If the cover is removed with the

SCAMP open, then some of the residual water within the cover may drip into the exposed electronics.

http://www.pme.com/USB_smanual/maintain1.htm (2 of 3) [10/5/2005 5:16:02 PM]

Maintenance - Sensor Cover

Replace the cover by reversing the removal sequence.

http://www.pme.com/USB_smanual/maintain1.htm (3 of 3) [10/5/2005 5:16:02 PM]

Maintenance - Opening the SCAMP

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

OPENING THE SCAMP

1.

If the SCAMP is wet, remove the closure ring (located at the release end) completely. Dry the exposed end cap completely with a towel. Replace the closure ring. This will reduce the slight amount of water that the release end cap leaves on the inside of the pressure tube when the tube is removed. Skip this step if the SCAMP is dry.

2.

Loosen the SCAMP closure ring (located at the release end) roughly one turn. Do not remove this ring . Press the ring toward the SCAMP body until the inner electronics slip forward. Repeat this procedure until the o- rings release. After the o-rings release, remove the ring completely. Note that this incremental removal procedure, if properly done, will not provide shock forces to the sensor when the internal electronics slips forward. Since the retaining ring remains on for the whole procedure, the internal electronics (and the sensors) will not unexpectedly fall out!

3.

Pull the SCAMP electronics from within the outer tube. Be careful that no water drips into the electronics or the tube. Be careful not to scratch the o-ring seals as the SCAMP electronics is withdrawn.

http://www.pme.com/USB_smanual/maintain2.htm (1 of 2) [10/5/2005 5:16:10 PM]

Maintenance - Opening the SCAMP http://www.pme.com/USB_smanual/maintain2.htm (2 of 2) [10/5/2005 5:16:10 PM]

Maintenance - Inspecting the O-rings

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

INSPECTING THE O-RINGS

1.

Be sure to CAREFULLY inspect the o-ring seals at each end of the SCAMP before closing.

The velcro used for the battery pack tends to lose fuzz. This fuzz can find its way to the o-ring and break the seal allowing water to enter the SCAMP and damage it seriously. The inside of the SCAMP tube should be wiped occasionally by pressing a balled-up paper towel through it to catch any velcro fuzz that the release endcap might pick up.

2.

Wipe the o-ring carefully with a lint free towel.

3. Inspect the o-ring by gently rubbing it. You are attempting to feel any scratches or cuts.

After this, carefully inspect the o-ring surface and within the groove for foreign matter. If the o-ring seems old, then stretch it gently and visually inspect for cracks.

4.

Replace the o-ring if any cuts or cracks are found. If replacing the sensor end o-ring, then slide it off/on over the SCAMP electronics with the covers on. Be careful not to nick it on the cover retaining screws.

5.

Apply a thin coating of silicone oil, grease or other o-ring lubricant designed for butyl nitrate type o-rings exposed to salt water.

6.

Completely wipe the stainless steel sealing ring within each end of the SCAMP outer tube.

Press a wadded up paper towel though this tube to capture any foreign materials (such as velcro fuzz).

7. Lubricate each sealing ring surface with a thin coating of lubricant.

Normal o-ring maintenance is complete. There are other o-rings located around the pressure transducer, below the I/O connector and a pair around the release screw shaft. These are not normally exposed and do not require routine maintenance.

http://www.pme.com/USB_smanual/maintain3.htm [10/5/2005 5:16:18 PM]

Maintenance - Electronics Covers

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

REMOVING & REPLACING THE ELECTRONICS COVERS

Electronics covers are secured by small screws, or in later models, simply taped on. In some cases black electrical tape has been applied over these screws and running the length of the edge of the electronic cover. Remove the tape if required. Unscrew the small screws and place them in a location where they will not be lost. Covers simply lift off after all screws are removed.

When replacing either cover be sure that the proper end is forward. There is a serial number stamped on the chassis and also scratched into each cover. When the covers are installed properly, all three serial numbers (two covers and the chassis - see picture) line up. When replacing screws be sure not to use too much torque when tightening. Insure that the screw head is flush or nearly flush with the cover. You may have to press on the cover to obtain this.

Replace the electrical tape with fresh tape. (Electrical tape reduces the chance that the screw heads will scratch SCAMP's outer tube o-ring sealing rings as you slide SCAMP back together.) http://www.pme.com/USB_smanual/maintain4.htm (1 of 2) [10/5/2005 5:16:28 PM]

Maintenance - Electronics Covers http://www.pme.com/USB_smanual/maintain4.htm (2 of 2) [10/5/2005 5:16:28 PM]

Maintenance - Closing the SCAMP

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

CLOSING THE SCAMP

1.

Examine the o-rings as described in the

Inspecting the O-rings section.

2.

Insert the SCAMP electronics back into the tube until the o-rings engage.

3. Replace the closure ring and tighten it, pulling the SCAMP electronics completely into the tube. Only tighten this ring until the electronics are completely in the tube. Loosen the ring slightly so that it will be easy to unscrew at a later date.

http://www.pme.com/USB_smanual/maintain5.htm (1 of 2) [10/5/2005 5:16:34 PM]

Maintenance - Closing the SCAMP http://www.pme.com/USB_smanual/maintain5.htm (2 of 2) [10/5/2005 5:16:34 PM]

Maintenance - Testing for Leaks

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

TESTING FOR LEAKS

PME supplies a squeeze bulb and connecting hose for leak testing the SCAMP. Leak testing is recommended whenever the batteries are replaced and absolutely required every time a sensor is replaced.

1.

To leak test the SCAMP, unscrew the 10-32 stainless screw from the release endcap.

Remove the adaptor from the tubing connecting it to the squeeze bulb. It is much easier to align it with the treaded hole in this condition. Screw the adaptor into the exposed port.

Be careful not to cross-thread this connection.

Push the tubing back onto the adaptor.

2.

Squeeze the bulb several times until no more air can be forced into the SCAMP.

3. Hold the SCAMP with the sensors up and check for bubbling into the water within the sensor cover.

http://www.pme.com/USB_smanual/maintain6.htm (1 of 2) [10/5/2005 5:16:46 PM]

Maintenance - Testing for Leaks

4. Place each end of the SCAMP in a bucket and check for bubbles coming from the endcap seals. In the case of the release endcap check for bubbles coming from the release screw, the pressure transducer port or the I/O connector.

When testing is completed, dry the area around the adaptor, then unscrew and remove it.

Replace the stainless screw immediately. Don't forget!

The SCAMP will become flooded if this screw is not replaced. Bubble testing is complete.

http://www.pme.com/USB_smanual/maintain6.htm (2 of 2) [10/5/2005 5:16:46 PM]

Maintenance - Replacing a Sensor

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

REPLACING A SENSOR

1.

Drain and

remove the sensor cover .

2.

Open the SCAMP. See

Opening the SCAMP

section for more detailed information. This is a delicate operation since the sensors are exposed at this time. Insure no residual water droplets run down into the electronics region.

3.

Clamp the SCAMP electronics horizontally in two vises by clamping the chassis, not the electronics covers. The sensor being replaced should be up. The following pictures show the Fast Temperature sensor (5316) with its associated resistor. Use two vises. If one vise is used, and the SCAMP clamped in the middle, it will unexpectedly slip and rotate the sensors into the table top.

4.

Remove the electronics cover. See Removing and Replacing the Electronics Covers

section for more detailed information.

http://www.pme.com/USB_smanual/maintain7.htm (1 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor

5. Cut the wires from the sensor being replaced so that 1/4" of colored insulator remain attached to the post. This will indicate the proper wiring of the new sensor. Click on the picture for a more detailed image.

6.

Use the pliers to gently unscrew (counter clockwise seen from the sensor side) the sensor sealing nut. Slide this nut up the sensor shaft.

7.

Loosen the set screw in the chassis end plate that mechanically retains the sensor.

http://www.pme.com/USB_smanual/maintain7.htm (2 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor

8.

Rotate the sensor so that it is not near the other sensors.

9.

Gently pull the sensor out of the endcap being careful that the wires don't crash the other sensors.. Place a protective cover on it.

10. Installing the new sensor. Inspect the wires-end of the shaft to make sure that there are no burrs, glue, other foreign materials, or scratches in the area where the seal is made.

Remove the plastic thermistor cover. This is a very delicate operation!! Place the retaining nut on the shaft. The image below shows a Fast Temperature sensor.

http://www.pme.com/USB_smanual/maintain7.htm (3 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor

11.

Gently slide the sensor into the endcap.

12. Position the sensor properly. You may be required to bend the sensor shaft somewhat if close alignment is desired. Tighten the set screw when this is accomplished.

http://www.pme.com/USB_smanual/maintain7.htm (4 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor

13.

Tighten the sealing nut firmly with pliers. CAREFUL! THE RETAINING NUT MAY SHIFT

THE SENSOR POSITION SLIGHTLY OR TWIST THE SENSOR IF THE RETAINING

SCREW IS NOT TIGHT CAUSING THE SENSOR TO CRASH INTO THE OTHERS .

14.

Cut the sensor wires to the appropriate length (leave a long loop) and strip the last 1".

New sensors should already have the wires stripped at the correct length. If a wire must be stripped and the proper 30 gauge stripping tool is unavailable the insulation can be removed by melting through it with a soldering iron while pulling on the wire. One at a time unwind the old sensor wire and replace it with the same color new sensor wire. Remember when replacing the Fast Temperature sensor that there is a resistor to replace as well.

Use the wire wrapping tool supplied by PME to wire wrap the new wires. DO NOT

SOLDER!

15.

Replace the electronics cover. See Removing and Replacing the Electronics Covers

section for more detailed information.

http://www.pme.com/USB_smanual/maintain7.htm (5 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor

16.

Examine the o-rings as described in the

Inspecting the O-rings section.

17.

Insert the SCAMP electronics back into the tube until the o-rings engage.

18. Replace the closure ring and tighten it, pulling the SCAMP electronics completely into the tube. Only tighten this ring until the electronics are completely in the tube. Loosen the ring slightly so that it will be easy to unscrew at a later date.

http://www.pme.com/USB_smanual/maintain7.htm (6 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor

19.

Replace the sensor cover

. Pressurize the interior as described in the

leak test , then

partially fill the sensor cover with water. Carefully check for bubbles exiting between the sealing nut and sensor shaft.. If bubbles appear, then drain and remove the cover.

Re-tighten the sealing nut. This nut is only plastic so don't tighten it too much. Repeat the leak test.

20.

If bubbles persist, then follow the above steps to remove the sensor. Inspect the sensor shaft. Remove the teflon ferrule within the endcap hole and replace it. Repeat the procedure.

21. Now that the sensor is physically installed on the SCAMP, its calibration must be installed

into the software. This is done by using the SCAMP Control Dialog’s Channel Tab to place

the correct values into the C0…C3 and NOFF, NOFFPOLAR, and NGAIN boxes.

Calibrated sensors supplied by PME will have documentation that give these values. User http://www.pme.com/USB_smanual/maintain7.htm (7 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing a Sensor supplied sensors must be calibrated and the results placed here.

http://www.pme.com/USB_smanual/maintain7.htm (8 of 8) [10/5/2005 5:17:01 PM]

Maintenance - Replacing the Batteries

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

REPLACING THE BATTERIES

Replace the batteries by following these steps:

1.

Do not attempt to replace batteries in a location where the SCAMP is subject to becoming wet while the pressure case is open.

2.

Remove the I/O cable connection if required.

3.

Open the SCAMP as described in the Opening the SCAMP section.

4.

Pull the cable loose from between the batteries and the chassis wall. Next, pull the 9v batteries out. Disconnect these.

5.

Open the velcro flaps and remove the AA batteries.

http://www.pme.com/USB_smanual/maintain8.htm (1 of 3) [10/5/2005 5:17:10 PM]

Maintenance - Replacing the Batteries

6.

Install fresh AA batteries and close the velcro straps. When installing insure that the polarity is proper. There is a diagram located in the AA battery cavity that shows proper

AA battery installation.

7.

Install fresh 9V batteries. BE CAREFUL NOT TO REVERSE THE POLARITY. IF YOU

TOUCH A BATTERY TO THE CONNECTOR IN THE REVERSE DIRECTION, EVEN FOR

AN INSTANT, THEN THE FUSE LOCATED IN THE 9V CONNECTOR WIRES WILL BE

BLOWN. THIS WILL NOT DAMAGE THE SCAMP, BUT A NEW FUSE MUST BE

INSTALLED. See

Replacing the Fuses if this happens.

8.

Replace the 9V batteries. Press the cable between the batteries and the chassis.

9.

Close the SCAMP as described in Closing the SCAMP

section.

10.

Leak test as described in Testing for Leaks

section.

11.

Replace the I/O cable if previously connected.

12.

SCAMP forgets its time and date when digital batteries are replaced. This must be re-initialized by connecting SCAMP to the host computer and running the SCAMP Control

Dialog. Once the Dialog screen appears the time and date have been automatically http://www.pme.com/USB_smanual/maintain8.htm (2 of 3) [10/5/2005 5:17:10 PM]

Maintenance - Replacing the Batteries initialized from the host computer’s time and date. No further action is necessary.

http://www.pme.com/USB_smanual/maintain8.htm (3 of 3) [10/5/2005 5:17:10 PM]

Maintenance - Replacing the Fuses

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

REPLACING THE FUSES

There are two fuses located within the 9v battery connector wiring. PME supplies two additional spare fuses with each system along with a 9v battery connector assembly. This assembly can be installed and used while the faulty connector is being repaired. Determine that the fuse is defective by measuring resistance. Good fuses measure about 6.5 ohms. Blown ones are much higher.

1. To replace a fuse, remove the batteries.

2.

Unplug the connector wiring and cut the shrink tubing.

3.

De-solder the bad fuse and install the new shrink tubing.

4. Solder the replacement fuse in place and shrink the tubing. Fuses blow for a reason. Try to determine the reason before re-connecting the batteries.

http://www.pme.com/USB_smanual/maintain9.htm (1 of 2) [10/5/2005 5:17:18 PM]

Maintenance - Replacing the Fuses http://www.pme.com/USB_smanual/maintain9.htm (2 of 2) [10/5/2005 5:17:18 PM]

Maintenance - Conductivity Sensors

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

CONDUCTIVITY SENSORS

To obtain maximum accuracy, the conductivity sensors should be soaked in salt water for several hours before use. To maintain maximum accuracy do not allow salt water to dry on the sensor. The resulting crystals can distort the electrode electroplating and change the calibration.

While actively using the SCAMP, store it in its stand with the sensors in the bucket half full of water. Lake water or whatever water is at hand ( not distilled ) may be used for this purpose.

Minimize the time that the conductivity sensors are in the air.

While the SCAMP is being stored temporarily, keep water in the sensor cover. Do not use distilled water for this purpose . Potable, clean, tap water having conductivity of greater than 0.5

mS/CM is best for this purpose. Some small amount of conductivity is necessary for the electronic circuit operation.

If the SCAMP is to be stored for a long period, flush the sensors with distilled water by repeatedly filling the sensor cover with distilled water. Drain all water. Replace the sensor cover plug.

http://www.pme.com/USB_smanual/maintain10.htm [10/5/2005 5:17:26 PM]

Maintenance - Pressure Transducer

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

PRESSURE TRANSDUCER

The pressure transducer is located within the release endcap and receives external pressure through the 1/4" diameter hole drilled approximately in the center of this cap. The pressure transducer should be carefully flushed after deployment in salt water.

1.

Cut the nozzle of a squeeze bottle so that it will only slightly penetrate the pressure transducer port. Be sure that the nozzle will not penetrate so deeply that it touches the pressure transducer diaphragm. This can damage the transducer.

2.

Gently squeeze distilled water into the pressure transducer port. This water will be expelled from the pressure transducer cavity through the small hole provided for this purpose drilled next to the port. Flush thoroughly. Click on the picture to get a closer view.

http://www.pme.com/USB_smanual/maintain11.htm [10/5/2005 5:17:37 PM]

Maintenance - I/O Connector

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

I/O CONNECTOR

Installation/removal of mating connectors or protective covers is greatly eased if a small amount of silicone grease is applied on the inside of the rubber shells of the mating connector.

If salt water gets into the connector by some accident, it won't cause permanent damage to the

SCAMP. Use a squirt bottle of distilled water to thoroughly wash the contacts, then blow dry using the leak test bulb. This same procedure applies to mating connectors.

Don't twist the connector . The plastic used in these connectors is very brittle and breaks easily in tension. The connector is tightened to 15 inch-pounds. This is surprisingly little torque.

http://www.pme.com/USB_smanual/maintain12.htm [10/5/2005 5:17:46 PM]

Maintenance - Installing a Dummy Sensor

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

INSTALLING A DUMMY SENSOR

Dummy sensors should be installed whenever a sensor is damaged or removed. All of

SCAMP's channels, except uDO, require that a sensor or a dummy sensor be connected. It is obvious that if no sensor or dummy sensor is connected to a channel then the reading on that channel will be incorrect. It is not so obvious, but true, that the readings on the other channels will also have serious errors even if sensors or dummy sensors are properly connected.

SCAMP is supplied with a dummy sensor kit, PME part number 5637. Two T sensor dummies and two C dummies are supplied. These can be installed on either Fast or Accurate circuits.

The dummy wire colors correspond to the same colors on the actual sensor. Installation is obvious: remove the existing sensor and install the dummy sensor wires to the same pins according to the color connection for the existing sensor. If the existing sensor is already removed and the color connection is lost, then see the fast temperature

sensor connection diagram , accurate temperature

sensor connection diagram or the conductivity sensor connection diagram . Wire-wrap wires to circuit posts using the tool provided. Do not solder!

http://www.pme.com/USB_smanual/maintain13.htm (1 of 3) [10/5/2005 5:17:54 PM]

Maintenance - Installing a Dummy Sensor

Broken sensors can be left physically connected to the SCAMP or removed and replaced with a

plug. Two sensor plugs are supplied with the Spare Seal kit

. Remove the sensor and replace it with the plug. The sensor plug is installed in the same way that a sensor is installed. See

Replacing a Sensor section for more information. Normally you will not have to replace the

plastic ferrule that provides the seal around the sensor shaft. One important note: Be sure to insert the plug fully into the SCAMP end cap so that it extends very slightly within SCAMP's electronics case, then TIGHTEN THE SENSOR RETAINING SCREW (a stainless set screw in

SCAMP's chassis). The picture below shows the location for this screw although it shows a real sensor and not a dummy plug.

http://www.pme.com/USB_smanual/maintain13.htm (2 of 3) [10/5/2005 5:17:54 PM]

Maintenance - Installing a Dummy Sensor

When SCAMP is submerged there is an external hydrostatic force that attempts to press the plug into the SCAMP. If the retaining screw is not tight, then the plug can slide into the SCAMP to the point where the ferrule seal releases allowing the SCAMP to become flooded.

Special note for SCAMPs having a Fluorometer installed. The dummy plug for a Fluorometer is longer than those supplied for other SCAMPs. Please insure that you have the right length.

Sensor plugs should be long enough to penetrate slightly inside the SCAMP's electronic housing while still extending out through the sensor seal nut.

Tighten the ferrule compression nut as done with a normal sensor.

Bubble test

the SCAMP to insure that you've tightened it properly.

The software should be informed that a dummy sensor has been installed. This is not critical, but

it may save confusion when the data are viewed subsequently. Use the SCAMP Control Dialog’s

Channel Tab

to change the Sensor ID to ‘Dummy’. A note can be made in the Comments box as well. These will appear in subsequent *.txt profile files.

http://www.pme.com/USB_smanual/maintain13.htm (3 of 3) [10/5/2005 5:17:54 PM]

Maintenance - Spare Seal kit

Precision Measurement Engineering,

Inc.

Discussing SCAMP Maintenance

SPARE SEAL KIT

SCAMP is supplied with a spare seals kit, PME part number 5107. The picture shows the kit's contents. The large o-rings are used to seal the outer tube. The next smaller, thin, o-ring seals the pressure transducer. Next smaller size o-ring seals the weight release shaft. The smallest o-ring seals the pressure test port screw. The white ferrules seal sensor shafts. The clear rods are sensor plugs for sealing SCAMP when sensors are removed. http://www.pme.com/USB_smanual/maintain14.htm [10/5/2005 5:18:01 PM]

Tips & Techniques

Precision Measurement Engineering,

Inc.

Discussing Tips & Techniques

PERSONAL DANGER

FLOOD

SHIPPING

SENSOR GUARD

CLOTHING

ROCKS

RELEASE CABLE

DAMAGED SENSORS

TESTS BEFORE A FIELD TRIP

USEFUL ITEMS (NOT SUPPLIED BY PME)

PERSONAL DANGER

If the SCAMP becomes substantially filled with water at depth, then it can return to the surface containing high pressure. If you have reason to believe that the SCAMP has flooded, such as failure to communicate or bubbling at any seal, then treat the SCAMP with extreme caution.

Internal pressure may be released by carefully loosening the pressure test screw and allowing several minutes to pass. This can be done from the side using a screwdriver. Do not allow any part of your body to come into alignment with either endcap since these may unexpectedly be explosively expelled by internal pressure.

FLOOD

Should a SCAMP seal fail and allow the SCAMP to become flooded with water, open it ,

immediately drain the water, and remove all batteries .

Remove the electronics covers and

completely flush the covers and all circuits with large quantities of distilled water. Leave the

SCAMP open so that it will dry. Allow it to dry completely, then

close it

and return it to PME for inspection.

SHIPPING

Save the cardboard shipping box (12" x 12" x 36") that contains the SCAMP. This may be

re-used to ship SCAMP from point to point or to return the SCAMP to PME. Do not ship SCAMP in any other container. If the cardboard box becomes unusable, then obtain another either from

PME or some other convenient source.

http://www.pme.com/USB_smanual/tips.htm (1 of 5) [10/5/2005 5:18:09 PM]

Tips & Techniques

When re-packing the SCAMP in the shipping box be sure to use all four pieces of foam. The two pieces with the hole in the middle of the SCAMP and the other two at either end of the box. This will help keep the SCAMP centered in the box. Fill box with popcorn.

The grey flip pack will contain all the SCAMP elements except the SCAMP itself. It takes a little arranging, but everything will fit. The tent stand PVC pipes lay on the bottom at one end. Place the cable assembly upright at the other end. Next, the drag plate unit fits well between the PVC pipes and the cable assembly. The guard can fit in this area too, but might need to be nested somewhat with the PVC pipes inserted into its interior. Make sure that it is positioned so that the supports do not become bent. The tent folds flat and can be placed on top.

SENSOR GUARD

When storing or shipping be sure that the sensor guard is securely packaged so that it cannot be bent. No disassembly is required.

After screwing the sensor guard onto the SCAMP, use a small cable tie or bit of dental floss to tie one of the guard's legs to one of the sensor seal nuts bases. This will prevent the unscrewing of the guard during a SCAMP deployment. Should the guard come unscrewed the retrieval cable can slip loose and the SCAMP will be lost. The sensor guards screw on tightly, but this extra precaution is low-cost insurance.

CLOTHING

You will find it difficult at best to see the laptop display in direct sunlight. Experience has shown that much of the glare on the laptop screen can come from light reflected from the operator's clothing. Wear something dark.

ROCKS

Tying the release cable to the rock is often a problem. Some suggestions are: cable tie the rock first, then tie to the cable tie, use a plastic zip lock bag to contain the rock, or use several rubber bands. Additional suggestions will be greatly appreciated.

RELEASE CABLE

Make these in advance of a field trip. They are only used for upwards profiling. There is a tool provided for creating the proper size loops to fit on the release screw.

Locate enough expendable weights (rocks) for the profiles you intend to make. Make wire loops from the wire provided by threading the wire through the crimp to make a loop, then adjusting the loop size using the tool provided. Crimp the collar in place with pliers. Connect the other ends of the wires to the expendable weights so that the weights will hang below the SCAMP by at least

12 inches (30cm).

http://www.pme.com/USB_smanual/tips.htm (2 of 5) [10/5/2005 5:18:09 PM]

Tips & Techniques

When you run out of release cable and crimps you can re-order these from PME, purchase similar materials at your local fishing shop, or make release cables from 50# test fishing line.

(For fishing line tie a slip-knot so that it tightens due to the weight of the rock. These aren't optimal since they do tend to slip off if the rock tension is lost.)

PME part numbers for additional crimps and release cable are:

4580 - crimp, each (please order in groups of 100 crimps)

4581 - release cable, roll of 100 feet

DAMAGED SENSORS

Damaged conductivity sensors should be replaced, or at least have dummy sensor plugs installed in their place. Damage to Fast Conductivity sensors can admit water into the SCAMP by allowing it to travel down the Fast C glass tubes and along the connecting wires. The conductivity circuits operate the sensor using an AC signal. In some cases, if the sensor is damaged or not installed, the circuit can enter an oscillating mode. This doesn't damage the circuit, but can induce a large amount of electrical noise in the other sensor channels in the

SCAMP. Dummy sensors must be electrically connected in place of damaged conductivity sensors.

Damaged temperature sensors need not be replaced since there are no likely leak paths.

Temperature circuits will not oscillate, but if the sensor is seriously damaged they can put out large voltages that will affect the other channels. Damaged temperature sensors must be at least electrically disconnected and dummy sensors connected in their place.

Damaged oxygen sensors should be replaced because of leak possibilities along the wires within the sensor. There is no dummy sensor required. Simply disconnect the oxygen sensor.

See Replacing a Sensor

section for more detailed information concerning sensor replacement

and Installing a Dummy Sensor for dummy sensor information..

http://www.pme.com/USB_smanual/tips.htm (3 of 5) [10/5/2005 5:18:09 PM]

Tips & Techniques

TESTING BEFORE A FIELD TRIP

Many features of the SCAMP should be tested prior to using it for data collection. These tests should be accomplished several weeks prior to the expected use to allow time for repairs and re-calibrations should these become necessary. Some of the tests should also be done on a daily basis or at other intervals. Many of these assume that the SCAMP is connected to the host computer and that the SCAMP Control Dialog is running.

Analog test. This test should be done in advance of a field trip. Click the Display button in the

Analog Test section of the Test Tab

. See the Analog Test section for the proper test result.

Battery test. This test should be done in advance of a field trip and frequently during use.

Simply review the

System Tab

battery indications.

Temperature sensor test. This test should be done in advance of a field trip and on a daily basis. If not already done, fill the sensor cover with tap water at ambient temperature or at the highest limit of SCAMP temperature channel calibration. Rock the SCAMP back and forth to gently mix the water in the sensor cover. Make sure that it is completely mixed. Stop rocking the

SCAMP and wait a moment. Record a brief profile with SCAMP. Pour out the water in the sensor cover and replace it with ice water, or water cooled to the lowest limit of SCAMP temperature channel calibration. Be sure that no ice enters the cover as it can damage the fast sensors. Record another brief profile. Upload both profiles and review these using the Matlab software. Check that all temperature sensors agree within 0.050 deg C. If larger differences are observed at either temperature, then contact PME.

Conductivity sensor test. This test should be done in advance of a field trip and on a daily basis. Fill the sensor cover with salty water of conductivity near the upper limit of SCAMP conductivity channel calibration. Rock the SCAMP back and forth to gently mix the water in the sensor cover. Make sure that it is completely mixed. Stop rocking the SCAMP and wait a moment. Record a brief profile with SCAMP. Upload this and review using the Matlab software.

The difference between Accurate and Fast Conductivity should be less than a few percent. If it is not, then carefully inspect the Fast Conductivity with a magnifier for fouling of the electrodes or cracks in the glass near the base or near the electrodes. Cracked sensors must be replaced.

Fouled sensors can be cleaned (process being developed now). Re-calibrate the Fast

Conductivity sensor using the Accurate Conductivity as a reference.

Fluorometer test (for SCAMPs with fluorometers installed). Record a brief profile with

SCAMP while inserting and removing a blade of grass into the fluorometer measurement tube.

Upload the profile and review using the Matlab software. The fluorometer should show a low voltage in the 0.1 to 0.5 Volt range with no grass in the tube and show a high voltage with grass inserted. The actual amplitudes will depend upon the sensitivity adjustment of the fluorometer.

Seal leakage test. This test should be done in advance of a field trip and after changing SCAMP batteries. For more information see the

Testing for Leaks

section.

If the above tests are all completed successfully, then nearly all of SCAMP's features will be verified.

http://www.pme.com/USB_smanual/tips.htm (4 of 5) [10/5/2005 5:18:09 PM]

Tips & Techniques

USEFUL ITEMS (NOT SUPPLIED BY PME)

scissors duct tape screwdrivers heavy dark cloth pliers voltmeter cable ties laptop computer power spare batteries laptop serial port squirt bottle float o-ring grease spare sensors panavise for cutting lead ballast weights for connecting the weights to the SCAMP small standard and phillips used to view the laptop computer; cover the computer tent and yourself to block the sun for squeezing the weight release line collars and operating the sensor seal nuts for checking battery voltages for securing the guard against unscrewing use a large capacity battery with an AC invertor so that the laptop can be powered by using its AC plug-in power supply for use in the SCAMP and especially the laptop

An extender cable is required for the laptop serial port since the SCAMP RS485-RS232 adaptor is too long to plug directly into the port. Allows the computer to easily fit into the tent.

fill with DI water to clean the sensor shafts and pressure transducer

If the SCAMP becomes tangled on the bottom, then it may be necessary to leave it to obtain diving gear. Attach the float to the retrieval line to act as a marker.

for lubricating the various o-rings for replacing damaged sensors use two for holding the SCAMP to replace sensors or service the instrument http://www.pme.com/USB_smanual/tips.htm (5 of 5) [10/5/2005 5:18:09 PM]

Specifications

Precision Measurement Engineering,

Inc.

Specifications

SCAMP

Travel

Rate

10 cm/sec

Overall

Length

Weight

76 cm

6 kilos (13 lbs.)

Maximum

Depth

100 meters

Sample

Rate

Batteries

Required

Battery

Life

100 scans/sec

6 alkaline AA; 2 alkaline 9V

4 hrs continuous operation minimum

FAST CONDUCTIVITY CHANNEL

Sensor

Range

PME 4-electrode microsensor

(being revised)

Ranged by scaling resistor(s):

5E-2 to 0.1 siemens/meter

min.

5E-2 to 9.0 siemens/meter

max.

http://www.pme.com/USB_smanual/specifications.htm (1 of 5) [10/5/2005 5:18:19 PM]

Specifications

Accuracy

Stability

+/- 5% full scale depending on calibration

+/- 5% full scale/4 hours

Bandwidth

Approximately 400 cycles/meter

Filter 4 pole Butterworth @ 50 Hz

Calibration Performed by PME

CONDUCTIVITY CHANNEL

Sensor

Range

PME 4-electrode ceramic sensor

Jumper selected ranges:

5E-2 to 0.1 siemens/meter

5E-2 to 0.5 siemens/meter

Accuracy

Stability

Spatial

Response

Filter

5E-2 to 9.0 siemens/meter

+/- 0.2% full scale (9 siemens/ meter range)

+/- 0.2% full scale/month

Approximately 33 cycles/meter

4 pole Butterworth @ 50 Hz

Calibration Performed by PME

FAST TEMPERATURE CHANNEL

(2 available)

Sensor

Thermometrics FP07 thermistor

Range 0 to 30 degrees C

Accuracy +/- 0.050 degrees C

Stability +/- 0.050 degrees C/month http://www.pme.com/USB_smanual/specifications.htm (2 of 5) [10/5/2005 5:18:19 PM]

Specifications

Bandwidth Approximately 20 Hz

Filter 2 pole Butterworth @ 50 Hz

Calibration Performed by PME

TEMPERATURE CHANNEL

Sensor

Thermometrics 1200 style thermistor

Range 0 to 30 degrees C

Accuracy +/- 0.020 degrees C

Stability +/- 0.020 degrees C/month

Bandwidth Approximately 0.3 Hz

Filter 2 pole Butterworth @ 50 Hz

Calibration Performed by PME

OXYGEN CHANNEL

Sensor

(under development with

Unisense)

Range

Maximum

Depth

0 to 10E-3 grams/liter

10 meters

Accuracy

+/- 1E-3 gram/liter depending on calibration

Stability +/- 1E-3 gram/liter / 4 hours

Bandwidth Approximately 1 Hz

Filter 2 pole Butterworth @ 50 Hz

Calibration Performed by the user

PRESSURE CHANNEL

Sensor

Range

Keller PSI PAA-10

50 to 100 meters full scale http://www.pme.com/USB_smanual/specifications.htm (3 of 5) [10/5/2005 5:18:19 PM]

Specifications

Accuracy +/- .5% full scale

Repeatability +/- 0.25% full scale

Stability

Calibration

+/- 0.25% / 6 months

Performed by PME

GRADIENT CHANNEL

(up to 8 available)

Function

Gain

Computes d/dt using analog circuits

Gains of 256/N N= 2,3,4...

255

Anti-alias

Filter

6 pole Butterworth @ 45 Hz

DATA LOGGER, ANALOG

Channels

Sampling rate

Sampling resolution

Analog processing: gain

Analog processing: offset

16 direct, 16 processed

1000 samples/sec maximum

16 bit

256/N N= 2,3,4 ... 255

+/- N/256 of full scale

N= 0... 255

RMS noise

91uV RMS on direct channels,

130 uV RMS on processed channels

DATA LOGGER, DIGITAL

http://www.pme.com/USB_smanual/specifications.htm (4 of 5) [10/5/2005 5:18:19 PM]

Specifications

Memory: program

Digital ports

Memory usage

512K flash, 512K RAM

Memory: data

64M MultiMedia Card

(128K available)

Communication

USB 1.0 Data and control

RS232 housekeeping

None implemented, but available by special order.

2 bytes per channel per sample http://www.pme.com/USB_smanual/specifications.htm (5 of 5) [10/5/2005 5:18:19 PM]

Wiring Chart

Precision Measurement Engineering,

Inc.

Wiring Chart

(Note that some wiring harnesses now use only one color wire.)

5264 - uDO & 2xT (FORWARD CIRCUIT A/D SIDE)

CONNECTOR P1

PIN FUNCTION COLOR GOES TO

PIN 1 V+

PIN 2 T1_OUT

PIN 3 -

PIN 4 uDO_OUT

PIN 5 -

PIN 6 VREF

PIN 7 V-

PIN 8 T0_OUT

RED

VIOLET

-

GREEN

WHITE

GREY

5026-P3-28

5026-P3-14 5267-1-P1-1

N/C

BROWN 5026-P3-17

N/C

5026-P3-3

5026-P3-1

5026-P3-13 5267-1-P1-4

5378 - uCT - CIRCUIT 1 (FORWARD CIRCUIT CPU SIDE)

CONNECTOR P2

PIN FUNCTION

PIN 1 C_OUT

PIN 2 VREF

PIN 3 GND

PIN 4 V-

PIN 5 T_OUT

COLOR

BROWN

GREEN

-

WHITE

VIOLET

GOES TO

5026-P3-15 5267-2-P1-4

5026-P3-3

N/C

5026-P3-2

5026-P3-23 http://www.pme.com/USB_smanual/wirechart.htm (1 of 4) [10/5/2005 5:18:31 PM]

Wiring Chart

PIN 6 V+ RED 5026-P3-27

5132 - ACCURATE CT - CIRCUIT 2

(2ND CIRCUIT BACK CPU SIDE)

CONNECTOR P2

PIN FUNCTION

PIN 1 C_OUT

PIN 2 VREF

PIN 3 GND

PIN 4 V-

PIN 5 T_OUT

PIN 6 V+

COLOR

BROWN

GREEN

-

WHITE

VIOLET

RED

GOES TO

5026-P3-19

5378-1-P2-2

N/C

5378-1-P2-4

5026-P3-21

5378-1-P2-6

5550 - FLUOROMETER & PAR

(SOMETIMES INSTALLED 3 RD CIRCUIT CPU SIDE)

CONNECTOR P2

PIN FUNCTION

PIN 1 V+

PIN 2 F_OUT

PIN 3 VREF

PIN 4 P_OUT

PIN 5 GND

PIN 6 N/C

PIN 7 ENABLE

PIN 8 GNDD

PIN 9 VD+

PIN 10 V-

COLOR

RED

BROWN

GREEN

VIOLET

-

-

BLUE

BLACK

-

-

5378-P2-6

GOES TO

5026-P3-25

5378-P2-2

5026-P3-16

5002-P4-3

5002-P4-1

ORANGE 5002-P4-28

WHITE 5378-P2-4

5267-1 - DUAL GRAD, GAIN, & FILTER - CIRCUIT 1

(2ND CIRC AFTER UDO,T A/D SIDE)

CONNECTOR P2

PIN FUNCTION COLOR GOES TO

http://www.pme.com/USB_smanual/wirechart.htm (2 of 4) [10/5/2005 5:18:31 PM]

Wiring Chart

PIN 1 GND

PIN 2 V+

PIN 3 OUT1

PIN 4 OUT0

PIN 5 DACSEL\

PIN 6 DACA/B\

PIN 7 DACD7

PIN 8 DACD6

PIN 9 DACD5

PIN 10 DACD4

PIN 11 DACD3

PIN 12 DACD2

PIN 13 DACD1

PIN 14 DACD0

PIN 15 GNDD

PIN 16 V-

-

RED

BROWN

VIOLET

N/C

5026-P3-28

5026-P3-6

5026-P3-5

BLUE

BLUE

5026-P2-5

5026-P2-1

YELLOW 5026-P2-7

YELLOW 5026-P2-9

YELLOW 5026-P2-11

YELLOW 5026-P2-13

YELLOW 5026-P2-14

YELLOW 5026-P2-12

YELLOW 5026-P2-10

YELLOW 5026-P2-8

BLACK

WHITE

5026-P2-15

5026-P3-1

CONNECTOR P1

PIN

PIN 1 -

PIN 2 -

PIN 3 -

PIN 4 -

FUNCTION

-

-

-

-

COLOR GOES TO

(RECEIVES WIRE)

N/C

N/C

(RECEIVES WIRE)

5267-2 - DUAL GRAD, GAIN, & FILT - CIRCUIT 2

(3RD CIRCUIT BACK CPU SIDE - USUALLY NOT

INSTALLED)

CONNECTOR P2

PIN FUNCTION

PIN 1 GND

PIN 2 V+

COLOR

-

RED

GOES TO

N/C

5026-P3-27 http://www.pme.com/USB_smanual/wirechart.htm (3 of 4) [10/5/2005 5:18:31 PM]

Wiring Chart

PIN 3 OUT1

PIN 4 OUT0

PIN 5 DACSEL\

PIN 6 DACA/B\

PIN 7 DACD7

PIN 8 DACD6

PIN 9 DACD5

PIN 10 DACD4

PIN 11 DACD3

PIN 12 DACD2

PIN 13 DACD1

PIN 14 DACD0

PIN 15 GNDD

PIN 16 V-

CONNECTOR P1

PIN FUNCTION

BROWN

VIOLET

BLUE

BLUE

YELLOW

YELLOW

YELLOW

YELLOW

YELLOW

YELLOW

YELLOW

YELLOW

BLACK

WHITE

5026-P3-8

5026-P3-7

5026-P2-6

5026-P2-1

5026-P2-7

5026-P2-9

5026-P2-11

5026-P2-13

5026-P2-14

5026-P2-12

5026-P2-10

5026-P2-8

5026-P2-15

5026-P3-2

PIN 1 -

COLOR

PIN 2 GNDD

PIN 3 -

PIN 4 -

5026 - MAIN ANALOG

-

-

BLACK

N/C

GOES TO

(RECEIVES WIRE

FROM PIN 2)

5267-2-P1-1

(RECEIVES WIRE)

CONNECT THESE P3 PINS WITH BLACK WIRE

4, 9, 10, 11, 12, 16 (skip if PAR), 18, 20, 22, 24,

25 (skip if Fluorometer), 26

CONNECT BLACK WIRE UNDER SCREW HEAD.

http://www.pme.com/USB_smanual/wirechart.htm (4 of 4) [10/5/2005 5:18:31 PM]

O-rings

Precision Measurement Engineering, Inc.

SCAMP Seals

PME P/N

4645

4711

4760

5140

5141

5333

5334

5574

DESCRIPTION USED IN

metric, 15.6 x 1.78, buna-n, 70 dur metric, 46 x 3, buna-n, 70 dur

0.070" x 0.364" ID,

70 dur ferrule, 1/4", teflon

1.734" ID x 0.139", buna-n, 70 dur

2-008, buna-n, 70 dur

2-012, buna-n, 70 dur metric, 10.5 thru

11.5 x 3 (select), buna-n, 70 dur pressure transducer sensor cover seal release screw, fluorometer cap sealing around sensor shafts sensor and release endcap seal seals below pressurization fitting seals below I/O connector fluorometer sample tube

QTY USED

1

1

4

4

2

1

1

2 http://www.pme.com/USB_smanual/orings.htm (1 of 5) [10/5/2005 5:18:42 PM]

O-rings http://www.pme.com/USB_smanual/orings.htm (2 of 5) [10/5/2005 5:18:42 PM]

O-rings http://www.pme.com/USB_smanual/orings.htm (3 of 5) [10/5/2005 5:18:42 PM]

O-rings http://www.pme.com/USB_smanual/orings.htm (4 of 5) [10/5/2005 5:18:42 PM]

O-rings http://www.pme.com/USB_smanual/orings.htm (5 of 5) [10/5/2005 5:18:42 PM]

Sensor Connection Drawings

Precision Measurement Engineering,

Inc.

Sensor Connection Drawings

These drawings show sensor connections, component layout information, and wire harness information for SCAMP's internal circuitry.

BOARD DESCRIPTION BOARD

SCAMP's memory and digital circuits that interface to the

A/D converter

CPU 5220

SCAMP's pressure transducer circuitry, channel multiplexors, offset and gain circuits, and the

A/D converter

A/D Converter - 5026

SCAMP's accurate conductivity and temperature circuits. The board is nearly identical to the fast conductivity and temperature board and can be used for connection information for either sensor

Conductivity &

Temperature - 5132

SCAMP's dissolved oxygen,

Fast T0, and Fast T1 circuits

Dissolved Oxygen & Dual

Temperature - 5264

SCAMP's d/dt, gain, and filter circuits. These implement the programmable gradient gain

SCAMP's PAR and fluorometer circuits. The location of the fluorometer adjustment potientiometer is shown

Dual Gradient, Gain, Filter -

5267

PAR & Fluorometer - 5550

http://www.pme.com/USB_smanual/sensorsconn.htm [10/5/2005 5:18:49 PM]

A/D Input Connection Definition

Precision Measurement Engineering,

Inc.

A/D Input Connection Definition

P3 PIN #

22

24

26

N/A

25

16

18

20

N/A

5

19

21

23

13

14

15

17

6

A/D

CHANNEL

11

12

13

14

7

8

9

10

15

16

4

5

6

2

3

0

1

17

FUNCTION (IF INSTALLED)

FAST T0

FAST T1

FAST C uDO

ACCURATE C

ACCURATE T

T OF FAST CT CIRCUIT

FLUOROMETER

PAR

AVAILABLE

AVAILABLE

AVAILABLE

AVAILABLE

AVAILABLE

DEPTH

ANALOG GROUND

GRAD FAST T0

GRAD FAST T1 OR GRAD

FAST C http://www.pme.com/USB_smanual/adinput.htm (1 of 2) [10/5/2005 5:18:55 PM]

A/D Input Connection Definition

N/A

N/A

N/A

N/A

N/A

N/A

11

12

4

N/A

7

8

9

10

26

27

28

29

30

31

22

23

24

25

18

19

20

21

GRAD FAST C

GRAD ANALOG GROUND

AVAILABLE

AVAILABLE

AVAILABLE

AVAILABLE

AVAILABLE

ANALOG GROUND

VREF

CIRCT

ABAT+

ABAT-

DBAT

ANALOG GROUND http://www.pme.com/USB_smanual/adinput.htm (2 of 2) [10/5/2005 5:18:55 PM]

WINKLER.PAS

Precision Measurement Engineering,

Inc.

WINKLER.PAS

{======================================================================

WINKLER: VARIOUS A-D CARD SUBROUTINES FOR DOTCAL

=======================================================================

=======================================================================

Revision 0 21-MAY-94: Initial release

=======================================================================}

{$I+} { enable automatic I/O error traps }

{$R+} { enable range checking }

{$N+} { 8087 mode } unit WINKLER;

{----------------------------------------------------------------------} interface

{----------------------------------------------------------------------}

FUNCTION DOConc( {calc O2 conc in mg/l}

Sample :real {ml vol of 0.01 N sodium thio added}

) : real; {returns DO in mg/l}

{----------------------------------------------------------------------} implementation

{----------------------------------------------------------------------} const {all volumes in ml}

N = 0.01; {normality of sodium thiosulfate solution}

blank = 0.00; {adjustment for errors in thiosulfate solution}

B = 300.0; {DO bottle volume}

Reagents = 2.00; {volume of reagents added to fix DO bottle}

S = 50.0; {volume of sample with drawn for titration}

VolSTP = 22.414689; {volume of gas at STP, liters}

O2molewt = 31.988; {molecular weight of oxygen gas} var

DO_ : real; {DO concentration in ml/l}

{======================================================================}

{ FUNCTION DOConc } http://www.pme.com/USB_smanual/winkler.htm (1 of 2) [10/5/2005 5:19:01 PM]

WINKLER.PAS

{ Return DO concentration in mg/l }

{======================================================================} function DOConc; begin

DO_: = N * (Sample-blank) * (B/(B-Reagents)) * 5.6 * 1000/S;

DOConc := DO_ * O2molewt/VolSTP; {convert ml/l to mg/l} end;

{-----------------------------------------------------------------------} end.

{WINKLER ends} http://www.pme.com/USB_smanual/winkler.htm (2 of 2) [10/5/2005 5:19:01 PM]

Cal of Eng Units

Precision Measurement Engineering,

Inc.

Calculation of Engineering Units

OVERVIEW

An explanation of the way that engineering units are calculated from SCAMP's raw data is useful for two reasons. First, it may happen that the customer needs to perform the calculation using a customer-written program. Second, and most important, it is necessary to understand how engineering units are calculated so that the effects of the NOFF, NPOLAR, and NGAIN parameters can be understood.

SCAMP's electrical block diagram

and the Hardware Overview sections of this manual show

how the sensor information is processed within SCAMP's various circuits. SCAMP records a measurement by the following process: as SCAMP travels through the water column its sensors respond to environmental parameters such as temperature or electrical conductivity. Sensing methods vary, but every sensor, together with its electrical circuit, produces a voltage that is in some way related to the parameter sensed. These voltages are produced continually as SCAMP travels.

Every 1/100 second interval the SCAMP's digital electronics records a scan of all active channels. There are two multiplexors (electronic switches) shown on the block diagram and there are thus two paths by which channel voltages arrive at the A/D converter. Voltages from channels 0 to 15 arrive at the A/D converter after being processed by a programmable offset and gain circuit. Voltages from channels 16 to 31 are directly routed to the A/D converter. Channels are scanned in ascending numerical order. The A/D output, a 16 bit integer, is stored within

SCAMP's memory by the digital electronics. This scanning process continues while SCAMP travels.

When data collection is completed, SCAMP's MultiMedia Card memory normally contains a great number of A/D integer values in sequential order. PME's SCAMP Control Dialog is used to upload these data via SCAMP's USB port onto the host computer's hard disk drive. Two files are created, a *.RAW file and a *.TXT file. The *.RAW file contains SCAMP parameters and raw data in binary format. The *.TXT file is a simple ASCII file and contains information about the profile and also calibration information.

CALIBRATION AND ENGINEERING UNITS

SCAMP's environmental measurements flow from sensor to disk drive through the above path.

http://www.pme.com/USB_smanual/eng.htm (1 of 4) [10/5/2005 5:19:07 PM]

Cal of Eng Units

The *.RAW file contains all the information needed to translate the raw A/D integers to the actual engineering unit values that were present at the sensor when the data were measured.

The *.RAW file contains data that was acquired at the time of the measurement and also

SCAMP parameters such as calibration coefficients. This information was not acquired at the time of the measurement, but was instead acquired when the SCAMP was calibrated. Normally calibration is done before using SCAMP for measurements, but it is also possible to provide calibration information from calibrations done after the measurements. The calculation of engineering units depends upon both the A/D integers and also upon the calibration so it is useful here to discuss calibrations.

Calibration is the activity of exposing SCAMP's sensors to known values of the environmental parameters being sensed and recording the A/D integers that occur. Sets of these observations are recorded over the expected range of the environmental parameters. A 16 bit A/D can resolve about 65000 different values of the environmental parameter. However, it is convenient to calibrate the SCAMP at a very limited number of these possible points. Since any of the 65000 values may occur in the measured data, a numerical function is found (a "least square fit") that can produce an estimate of the engineering unit associated with any of the 65000 points.

Calculation of the engineering unit associated with any integer from the *.RAW file should thus be quite simple: read the integer from the file, read the numerical function from the SCAMP parameters, evaluate the function using the integer, and save the resulting engineering unit.

Things are not this simple however, due to the presence of the programmable offset and gain circuit that processes channels 0 to 15.

In order to incorporate offsets and gains for channels 0 to 15 the calibration concept presented above is modified somewhat. Calibrations are considered to be relations developed between sensed environmental parameters and the channel voltages produced by the associated circuits and delivered to the multiplexors. At this stage all channels 0 to 31 are the same. The channel output voltage is, for computational convenience, not expressed in volts but rather as a ratio of the output voltage to the SCAMP internal reference (+3 volts), times 256.

CR = 256 * (Vb/Vref)

CR is the channel ratio (no unit)

Vb is the channel output (volts)

Vref is SCAMP's internal reference voltage (3.000 volts)

This makes sense because many channels have ratiometric outputs, the A/D converts its inputs ratiometrically, and all channels can be treated the same no matter if their signal paths include the offset and gain or not.

Of course during a calibration the only information available from SCAMP is the A/D converter output. If the calibration will be parameter-vs.-CR, then CR must be computed from the A/D output. Thus the effect of the A/D conversion and also of the offset and gain block must be approximated. Note that the accuracy of the approximations used is unimportant (if the offset and gains aren't changed for actual measurements) since they will be canceled out in the final calibration path.

http://www.pme.com/USB_smanual/eng.htm (2 of 4) [10/5/2005 5:19:07 PM]

Cal of Eng Units

The A/D is modeled by:

Nad=32768*(Vad/Vref)

Vad is the voltage input to the A/D

Nad is the integer produced by A/D conversion

The offset and gain for channels 0 to 15 is modeled by:

Vad=G*(Vb+Voff) ;NOFFPOLAR=0

Vad=G*(Vb-Voff) :NOFFPOLAR=1 or for channels 16 to 31 by:

Vad=Vb

G=-256/NGAIN

Voff=Vref*(NOFF/256)

NGAIN is the HOST SETUP\CHANNEL NGAIN for the channel

NOFF is the HOST SETUP\CHANNEL NOFF for the channel

NOFFPOLAR is the HOST SETUP\CHANNEL NOFFPOLAR for the channel

A CR for a given calibration point is obtained by taking the Nad at that point and finding Vad. The offset and gain equation is used to convert Vad into Vb, and the channel ratio relation used to find the final channel ratio from Vb. An entire set of parameter-vs.-CR is used to find a numerical equation that estimates the environmental parameter from CR. HOST software presently supports only one form of this estimation:

EU=C0+C1*CR+C2*CR^2+C3*CR^3

EU is the engineering unit value of the sensed parameter

C0..C3 are the calibration coefficients

So the final computation of engineering units is somewhat more complex: Convert the Nad into the corresponding Vad. Find Vb. Find CR. Find EU.

Note that the approximate equations for A/D response and for offset and gain effectively cancel themselves out since they appear in the same way in the path from Nad to CR for both calibration and for measurement. This is not the case if NOFF, NGAIN, or NPOLAR are changed. If any of these values are different for measuring from the values used for calibration, then the engineering unit computation will still work but its accuracy will depend upon how accurately the Vad to Vb relation models the actual effect of the offset and gain block.

CALCULATION METHODS

PME provides several means to convert the information in these files into the corresponding engineering units for each channel (i.e. temperature in degrees C, conductivity in mS/cm and so on).

The S_LOAD2 mex program allows Matlab to load SCAMP channels from the *.RAW files directly into arrays. The engineering units are automatically calculated by this load http://www.pme.com/USB_smanual/eng.htm (3 of 4) [10/5/2005 5:19:07 PM]

Cal of Eng Units process.

The ENG_UNIT.C source code provided with some releases of SCAMP illustrates how customers might write their own calculation programs.

There should be no need to create additional programming for finding engineering units from raw data.

http://www.pme.com/USB_smanual/eng.htm (4 of 4) [10/5/2005 5:19:07 PM]

RS232 COMMUNICATION

Precision Measurement Engineering, Inc.

RS232 COMMUNICATION

SCAMP will communicate with a host computer via RS232 serial interface. This interface is provided mostly for use by PME. To communicate via RS232 you must set up the host computer and connect to the SCAMP.

HOST COMPUTER SETUP

The customer must supply the appropriate host computer software for RS232 communication. This can be the

Hyperterminal program in Start\Programs\Accessories\Communication.

To initialize Hyperterminal, select File\New Connection. Select the proper COM port for your computer in the box that appears:

Select OK. Select the settings shown below from the subsequent box: http://www.pme.com/USB_smanual/RS232.htm (1 of 4) [10/5/2005 5:19:16 PM]

RS232 COMMUNICATION

If a different communication program is used, SCAMP’s RS232 I/O communicates at 9600 Baud, with 8 data bits, no parity, and one stop bit. There is no flow control.

CONNECTION TO SCAMP

Open SCAMP. Remove the longer electronics cover. Connect the RS232 cable supplied with SCAMP to the connector as shown in the following picture: http://www.pme.com/USB_smanual/RS232.htm (2 of 4) [10/5/2005 5:19:16 PM]

RS232 COMMUNICATION

SCAMP COMMANDS

SCAMP will respond to host computer requests. ‘HELP’ brings up a list of possible requests. These are mostly for

PME use but a list is provided below: keyword function

------- --------

(Misc)

BANNER Present TAG-ALONG software header.

BAT Read batteries and print result.

CEQ Set channel calibration info.

CPARM Set channel parameters

DELAYT Loop NCOUNT times.

DEPTH Read depth and print result.

ENGUNIT Return 500 samp avg of chan(s) EU

GETSCB Restore SCB from block.

HELP Display this message.

INITREC Initialize event recorder.

MEASURE Return 500 samp average of channel(s) Nad

MMCERASE Erase entire MMC except PF blocks

MMCFILL Fill or display a block of MMC memory.

MMCSTAT Display MMC statistics http://www.pme.com/USB_smanual/RS232.htm (3 of 4) [10/5/2005 5:19:16 PM]

RS232 COMMUNICATION

MYREC Print current SCB.

OFF Turn power off.

PFCONTROL Display PF_Control.

PROSTAT Display profile statistics

PRCHAN Display channel current values.

SNUM Display SCAMP serial number.

STEPT Step response at NGAIN.

(TESTS)

TEST00 Cycle weight release.

TEST01 Cycle AD power.

TEST02 Increment DAC1A .

TEST03 Select A/D channels sequentially.

TEST04 Ramp offset DAC at gain = 1, bipolar.

TEST05 Ramp gain at offset = 1, bipolar.

TEST06 Step Response at G=1, G=8.

TEST07 Calibrate A/D.

TEST08 Read A/D ground channel.

TEST09 Ramp A/D and print result.

TEST10 Print T, T Grad RMS CR at 4 NGRAD

TESTALL Perform TEST00 through TEST09 in sequence

Enter any keyword without parameters to see its function.

http://www.pme.com/USB_smanual/RS232.htm (4 of 4) [10/5/2005 5:19:16 PM]

DOWNLOAD A NEW OPERATING SYSTEM

Precision Measurement Engineering, Inc.

DOWNLOAD A NEW OPERATING SYSTEM

INSTALLATION OF MINIMON

Minimon is provided in the SCAMP\Minimon folder. Select setup.exe and follow the on-screen directions.

INITIALIZATION OF MINIMON

Run Minimon program. Select Target\Configuration. Fill in boxes as shown below.

(If SCAMP does not appear in the Controller type list, select C167SR)

Select Interface\Settings. Fill in settings as shown below. If you are using COM1 instead of COM2, fill in COM1.

http://www.pme.com/USB_smanual/download_new_OS.htm (1 of 6) [10/5/2005 5:19:24 PM]

DOWNLOAD A NEW OPERATING SYSTEM

Select Interface\Terminal Settings. Fill in as shown below.

http://www.pme.com/USB_smanual/download_new_OS.htm (2 of 6) [10/5/2005 5:19:24 PM]

DOWNLOAD A NEW OPERATING SYSTEM

This completes initialization of the Minimon program.

DOWNLOADING NEW OPERATING SYSTEM TO SCAMP’S FLASH

Downloading of SCAMP’s operating system to flash occurs in five steps. The first step is to connect SCAMP to the host computer and initialize SCAMP for downloading. In the second step the operating system is downloaded into SCAMP’s RAM memory. In the third step a short additional program is downloaded. In the fourth step this short program is run. This program copies the operating system from SCAMP’s RAM into SCAMP’s flash memory. The fifth step is to restore SCAMP to its normal operation status.

STEP 1 – Connect SCAMP and initialize it for downloading.

Insure that SCAMP's batteries are sufficiently charged to operate SCAMP for 30 minutes or more. Open SCAMP. Remove the longer electronics cover. Connect the RS232 cable supplied with SCAMP to the connector as shown in the following picture: http://www.pme.com/USB_smanual/download_new_OS.htm (3 of 6) [10/5/2005 5:19:24 PM]

DOWNLOAD A NEW OPERATING SYSTEM

Jump SCAMP’s bootstrap jumper. This is just visible in the picture. It is directly above the 4 circuit USB connector with black, red, white, green wires. There should be a jumper slid over one post of the connector. Use this jumper to jump both posts.

Run the Minimon program. After the program appears, press SCAMP's reset button (the white circle visible at lower left in the picture above). SCAMP will print a banner display that gives information about the SCAMP hardware and the software revision date.

Now jump SCAMP’s bootstrap jumper. This is just visible in the picture. It is directly above the 4 circuit USB connector with black, red, white, green wires. There should be a jumper slid over one post of the connector. Use this jumper to jump both posts.

Press SCAMP’s reset button again. Select Target\Connect. Minimon should connect to the SCAMP and display a few lines of connection information.

STEP2 – Download the operating system.

Select File\Load and browse to the directory containing the operating system to be loaded. An example

: http://www.pme.com/USB_smanual/download_new_OS.htm (4 of 6) [10/5/2005 5:19:24 PM]

DOWNLOAD A NEW OPERATING SYSTEM

Select PF_DEPLOY xxx.H86 (where xxx indicates the revision number, example '103'). Minimon loads this program into the host (not the SCAMP!) computer after you select OK

Select Target\Download Selection. Minimon downloads PF_DEPLOY.H86 to SCAMP’s RAM memory. It is not placed into

SCAMP's FLASH memory at this time, only downloaded to RAM. The next steps move it from RAM to FLASH

STEP 3 – Download the flashing program

Repeat step 2 except load PFlash.H86

STEP 4 – Run Pflash.H86 to program SCAMP’s flash

Select Target\Disassemble Debug. The following screen will appear. Select ‘GO’ The Pflash.H86 program (now within

SCAMP) copies the operating system from SCAMP's RAM memory to SCAMP's flash memory and will print its status as it programs the flash. Select Cancel to exit the screen. (Selecting Cancel will not influence SCAMP – it is performing the programming using its own internal CPU.) When copying is completed SCAMP forgets about Pflash.H86.

http://www.pme.com/USB_smanual/download_new_OS.htm (5 of 6) [10/5/2005 5:19:24 PM]

DOWNLOAD A NEW OPERATING SYSTEM

STEP 5 – Restore SCAMP to normal operation

Remove the jumper that you installed in Step 1. Place it back over one post only. Press the reset button. SCAMP should print its banner, now showing a new software revision in the banner.

This completes installing a new operating system into SCAMP’s flash memory.

http://www.pme.com/USB_smanual/download_new_OS.htm (6 of 6) [10/5/2005 5:19:24 PM]

LED Indications

Precision Measurement Engineering, Inc.

L.E.D. Indications

INTRODUCTION

SCAMP’s CPU circuit has a Light Emitting Diode (L.E.D.) installed on it. This L.E.D. can be seen by

opening SCAMP

and removing the longer electronics cover . The L.E.D. is located to the right of the Reset Button. See the picture below.

SCAMP flashes the L.E.D. at various times to indicate internal circuit conditions. Some flashes indicate normal operation, some do not.

http://www.pme.com/USB_smanual/led_Indications.htm (1 of 2) [10/5/2005 5:19:32 PM]

LED Indications

NORMAL OPERATION

Reset Response

Immediately after the reset button is pressed, or after the last AA battery is installed into the digital battery pack, and if the USB is not connected, SCAMP will flash the L.E.D. twice. The L.E.D. will then remain off. This indicates normal reset response.

USB Response

Approximately 3 seconds after SCAMP is connected to the host computer via USB, SCAMP will light the L.E.D. continuously for the duration of the USB connection.

RS232 Response

SCAMP illuminates the L.E.D. immediately upon RS232 connection for the duration of the connection.

Data Acquisition Response

SCAMP illuminates the L.E.D. briefly whenever the supervisor is active. Supervisor activity occurs about once per second and may be too brief to see. However, when a 512 byte block of data is acquired, it is written to the MMC by the Supervisor. This event takes longer and can be seen periodically as a brief flash of the L.E.D. Upon conclusion of a profile the supervisor performs clean-up chores which takes somewhat longer. The end of a profile can be seen as a longer L.E.D. flash, followed by no further flashing.

ABNORMAL OPERATION

If SCAMP’s software determines that there has been an internal error it flashes the L.E.D. a certain number of times in a repeating pattern. If any of the following are observed, then remove all SCAMP’s batteries and contact PME.

Repeating

Flash Pattern

Cause Notes

2 Unable to initialize MMC

3

4

Unable to read MMC

Can sometimes be corrected by removing/reinstalling AA batteries

5

6

Unable to initialize profiler control block

NMI interrupt

Stack overflow

7

8

9

10

11

Undefined op code

Protection fault

Illegal word operand

Illegal instruction address

Illegal external buss access

Low voltage at AA batteries. Replace http://www.pme.com/USB_smanual/led_Indications.htm (2 of 2) [10/5/2005 5:19:32 PM]

MatLab Software

Precision Measurement Engineering,

Inc.

Discussing the Matlab Software

SCAMP is supplied with software that allows the user to load and process profiles within the

Matlab 5.1 or later environment on a Windows PC platform. The software is supplied in three files:

SCAMPGUI.DOC

– Overview and installation instructions

SCAMPGUI.ZIP – Matlab scripts to implement the GUI

SCAMPTOOL.ZIP – Matlab Mex files to perform specific data processing

These files should be installed according to the instructions in SCAMPGUI.DOC.

http://www.pme.com/USB_smanual/Matlabsoft0.htm [10/5/2005 5:19:38 PM]

SCAMP Library

Precision Measurement Engineering,

Inc.

SCAMP Library

THERMISTOR RESPONSE

Fozdar, F.M.., Parker, G.J. and Imberger, J., Matching Temperature and Conductivity Sensor

Response Characteristics . J. Physical Oceanography, Vol 15, pps 1557-1569 (1985)

Gregg, M.C. and Meagher, T.B. The Dynamic Response of Glass Rod Thermistors . J

Geophysical Research, Vol 85, No C5, pp 2779-2786 (1980)

SEGMENTATION

Atal, B.S. and Hanauer, S.L. Speech Analysis and Synthesis by Linear Prediction of the Speech

Wave . J Acoustical Society of America, Vol 50, No 2, Part 2, pp 637-655 de Souza, P.V., Statistical Tests and Distance Measures for LPC Coefficients . IEEE

Transactions on Acoustics, Speech, and Signal Processing, Vol ASSP-25, No. 6, pp 554-558

(1977) deSouza, P.V. and Thomson, P.J., LPC Distance Measures and Statistical Tests with Particular

Reference to the Likelihood Ratio . IEEE Transactions on Acoustics, Speech, and Signal

Processing, Vol ASSP-30, No. 2, pp 304-315 (1982)

BATCHELOR SPECTRA

Caldwell, D.R., Dillon, T.M., Brubaker, J.M., Newberger, P.A. and Paulson, C.A., The Scaling of

Vertical Temperature Gradient Spectra . J. Geophysical Research, Vol 85, No. C4, pp

1917-1924 (1980)

Dillon, T.M. and Caldwell, D.R., The Batchelor Spectrum and Dissipation in the Upper Ocean. J.

Geophysical Research , Vol 85, No. C4, pp 1910-1916 (1980)

Dillon, T.M., Vertical Overturns: A Comparison of Thorpe and Ozmidov Length Scales, J.

Geophysical Research , Vol 87, No. C12, pp 9601-9613 (1982) http://www.pme.com/USB_smanual/library.htm (1 of 4) [10/5/2005 5:19:44 PM]

SCAMP Library

BATCHELOR FITTING TECHNIQUE

Krocsis, O., Prandke, H, Stips, A., Simon, A., and Wuest, A.(1999). Comparison of

Dissipation of Turbulent Kinetic Energy Determined from Shear and Temperature

Microstructure. J. Marine Systems, Volume 21, Pages 67-84.

Luketina, David A. and Imberger, J., Determining Turbulent Kinetic Energy Dissipation from

Batchelor Curve Fitting . Submitted to the J.. Atmospheric and Oceanic Technology (1998)

Ruddick, Barry, Anis, Ayal, and Thompson, Keith . Maximum Likelihood Spectral Fitting: the

Batchelor Spectrum . Submitted to J. Atmospheric and Oceanic Technology. (1999)

SCAMP MEASUREMENTS

Boashash, B. and Imberger, J., Application of the Wigner-Wille Distribution to Temperature

Gradient Microstructure: A New Technique to Study Small-Scale Variations , J. Physical

Oceanography, Vol 16, pp 1997-2012 (1986)

Chen, H. , M. Hondzo, A. Rao, Segmentation of temperature microstructure , Journal of

Geophysical Research-Oceans, 107(C12), 2002.

Etemad-Shahidi, A. and J. Imberger, Anatomy of Turbulence in Thermally Stratified Lakes,

Limnology and Oceanography, 1998. This paper presents a comparison of the dissipation obtained from shear with dissipation obtained from thermal gradients. View the abstract in

Adobe format.

Gregg, M.C., Uncertainties and Limitations in Measuring e and Xt, Journal of Atmospheric and

Oceanic Technology, 16, 1483-1490, 1999.

Hondzo, M., and Z. Haider (2004), Boundary mixing in a small stratified lake, Water Resour.

Res., 40, W03101, doi:10.1029/2002WR001851.

Jef Huisman, Jonathan Sharples, Jasper M. Stroom, Petra M. Visser, W. Edwin A. Kardinaal,J olanda M. H. Verspagen, and Ben Sommeijer (2004), Changes in Turbulent Mixing Shift

Competition for Light Between Phytoplankton Species, Ecology, Vol. 85, No. 11, pp. 2960–2970.

View the abstract in Adobe format.

Imberger, J. and G. Ivey, On the Nature of Turbulence in a Stratified Fluid. Part II: Application to Lakes, Journal of Physical Oceanography, 21, 659 - 680, 1990. This paper (together with

Part I) is the fundamental work upon which SCAMP is based. View the abstract in Adobe format.

Ivey, G. and J. Imberger, On the Nature of Turbulence in a Stratified Fluid. Part I: The

Energetics of Mixing, Journal of Physical Oceanography, 21, 650 - 658, 1990. This paper

(together with Part II) is the fundamental work upon which SCAMP is based. View the abstract in Adobe format.

Krocsis O., H. Prandke, A. Stips, A. Simon, and A. Wuest, Comparison of Dissipation of

Turbulent Kinetic Energy Determined from Shear and Temperature Microstructure, J. Marine http://www.pme.com/USB_smanual/library.htm (2 of 4) [10/5/2005 5:19:44 PM]

SCAMP Library

Systems, 21, 67-84, 1999. This paper presents an excellent comparison of the Batchelor temperature microstructure method with the velocity shear microstructure method for determining turbulent dissipation.

Luketina, D. and J. Imberger, Determining Turbulent Kinetic Energy Dissipation from Batchelor

Curve Fitting , Journal of Atmospheric and Oceanic Technology, 1998. This paper presents a method for fitting temperature gradient spectra to the Batchelor form. View the abstract in Adobe format.

MacIntyre, S., Jellison, R., , Hydrobiologia, 466: 13-29, 2001.

Nutrient fluxes from upwelling and enhanced turbulence at the top of the pycnocline in Mono Lake, California

MacIntyre, S.,Turbulent Mixing and Resource Supply to Phytoplankton, Physical Processes in

Lakes and Oceans Coastal and Estuarine Studies, 54, 561-590, 1998. This paper presents an example SCAMP use. View the abstract in Adobe format.

MacIntyre, S., K. Flynn, R. Jellison, and J. Romero, Boundary mixing and nutrient fluxes in

Mono Lake, California , Limnology and Oceanography, 44 (3), 512 - 529, 1999. This paper presents an example SCAMP use. View the map of Mono Lake on Sally MacIntyre's web site.

Piera, J., E. Roget, J. Catalan, Turbulent patch identification in microstructure profiles: A method based on wavelet denoising and Thorpe displacement analysis, J. Atmos. and Oceanic

Tech., 19 (9), 1390 -1402, 2002.

Piera, J., J. Catalan, Non-local turbulent mixing parameterization derived from microstructure signal processing: Potential application to small scale biological-physcial data analysis, J.

Atmos. and Oceanic Tech., 2004. View the abstract in Adobe format.

Rehmann, C., J. H. Hwang, Small-Scale Structure of Strongly Stratified Turbulence , Journal of

Physical Oceanography, 151 - 164, 2005.

Ruddick, B., A. Anis, and K. Thompson, Maximum Likelihood Spectral Fitting: The Batchelor

Spectrum, J. Atmos and Oceanic Tech., 17 (11), 1541 - 1555, 2000. This paper presents a method for fitting temperature gradient spectra to the Batchelor form. View the abstract or the submitted draft in Adobe format.

Sharples, J., M. Coates, J. Sherwood, Quantifying turbulent mixing and oxygen fluxes in a

Mediterranean-type, microtidal estuary, Ocean Dynamics, (53) 126 - 136, 2003. View the abstract in Adobe format.

Sharples, J., Tidal-Cycle Surveys, 2002. Boat-based work will be carried out over single tidal cycles at neap and spring tides during each of the mooring deployments. For more information view his web site .

Sharples J., M. Moore, and E. Abraham, Internal tide dissipation, mixing and vertical nitrate flux at the shelf edge of NE New Zealand, Journal of Geophysical Research, 106 (C7), 14,069 -

14,081, 2001. View the abstract in Adobe format.

http://www.pme.com/USB_smanual/library.htm (3 of 4) [10/5/2005 5:19:44 PM]

SCAMP Library

Sharples J., J. Waniek, and C. Ribeiro, Quantifying The Variability Of Vertical Turbulent Mixing

And Its Role In Controlling Stability And Mass Transport In A Partially-Mixed Estuary , 2001.

Smith, M., C. Stevens, J. McGregor and C. Lemckert, Physical Processes at the Air-Sea

Interface During an Impulsively Initiated Wind Event, 2002. Poster Session at AGU Ocean

Sciences in Hawaii. The poster explains the NEar- Surface Turbulence EXperiment (NESTEX) conducted in Wellington Harbour during the passage of a southerly front. Equipment deployed included acoustic Doppler velocimeters (ADV), a temperature microstructure profiler (SCAMP) and a microwave radar. View the graphic poster .

Soga, L.C.M. and Rehmann, C.R, 2004. Dissipation of Turbulent Kinetic Energy near a Bubble

Plume . J. Hydraul. Eng., 130, 441-449.

Stevens, C. L., Turbulence in an estuarine embayment: Observations from Beatrix Bay, New

Zealand , J. Geophys. Res., 108(C2), 3030, doi:10.1029/2001JC001221, 2003.

Stevens, C., M. Smith, and A. Ross, SCAMP: measuring turbulence in estuaries, lakes, and coastal waters , NIWA - Water and Atmosphere, 7 (2), 20-21, 1999.

Wain, D.J. and Rehmann, C.R. 2005 Eddy diffusivity near bubble plumes , Water Resour. Res.,

41, W09409, doi:10.1029/2004WR003896. http://www.pme.com/USB_smanual/library.htm (4 of 4) [10/5/2005 5:19:44 PM]

SCAMP manual limited warranty

Precision Measurement Engineering,

Inc.

Limited Warranty

1.

PME warrants that the hardware and sensors, (not including the fast oxygen sensor), shall be free of defects in workmanship and materials, under normal use, for a period of 90 days from the date of shipment. This warranty is made only to the original purchaser. This warranty does not apply to expendable parts such as batteries and fuses nor to underwater connectors supplied by manufacturers other than the warrantor. Modification and servicing of the sensors and electronics is not recommended. If modifications are made to the instrument or sensors by persons other than the manufacturer, then the 90-day warranty is void.

2.

PME MAKES NO WARRANTIES, EITHER EXPRESSED OR IMPLIED, THAT THE

SOFTWARE WILL BE ERROR FREE, OR THAT IT WILL PERFORM ITS FUNCTION IN

AN UNINTERRUPTED MANNER. PME MAKES NO WARRANTY WITH RESPECT TO

THE MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR USE OF THE

SOFTWARE. IN NO EVENT SHALL PME BE LIABLE FOR ANY DIRECT, INDIRECT,

SPECIAL OR CONSEQUENTIAL DAMAGES CAUSED, DIRECTLY OR INDIRECTLY, BY

THE SOFTWARE.

3.

Should PME make improvements to the software within a period of one (1) year following the date of delivery to the customer these improvements will be provided to the customer without charge.

4.

PME MAKES NO WARRANTIES, EITHER EXPRESSED OR IMPLIED, THAT THE

SENSORS WILL BE OPERABLE AFTER THEY ARE EXPOSED TO ADVERSE

ENVIRONMENTAL CONDITIONS SUCH AS BIO-FOULING, OIL FOULING, OR

OTHERS.

5. In the event an instrument covered by this warranty fails to operate according to our published specifications, PME will repair the instrument if it is returned with shipping prepaid within 90 days of purchase. If it is determined that the failure occurred through other than normal use, repairs will be billed at a nominal rate. In such cases, an estimate will be submitted for approval before repair work is started.

http://www.pme.com/USB_smanual/warranty.htm (1 of 2) [10/5/2005 5:19:51 PM]

SCAMP manual limited warranty

6. The cost of shipping the instrument for repairs will be paid by the purchaser. No COD shipment will be accepted by the warrantor. Returned instruments shall be shipped in the original shipping container and packing material. PME accepts no responsibility for damage in returning shipments.

http://www.pme.com/USB_smanual/warranty.htm (2 of 2) [10/5/2005 5:19:51 PM]

Revision History

Precision Measurement Engineering,

Inc.

Revision History

DATE

12-SEP-00

09-APR-04

22-DEC-04

PAGE REVISED

SCAMP SEALS

ALL PAGES THAT REFER

TO THE USB INTERFACE

"FREQUENTLY ASKED

QUESTIONS" SECTION

ACTION TAKEN

ADDED PICTURES OF SEALS

ADDED THE USB INTERFACE

INFORMATION

ADDED THIS TO THE MAIN USB

MANUAL NAVIGATIONAL MENU http://www.pme.com/USB_smanual/revision.htm [10/5/2005 5:19:57 PM]

Home

SCAMP USB MANUAL

http://www.pme.com/USB_smanual/about.htm [10/5/2005 5:29:04 PM]

http://www.pme.com/USB_smanual/images/SCAMPGUI1.gif

http://www.pme.com/USB_smanual/images/SCAMPGUI1.gif [10/5/2005 5:51:34 PM]

NTC THERMISTORS:

TYPE FP07/10/14

FASTIP PROBE THERMISTOR

DESCRIPTION:

The FASTIP THERMOPROBES consist of small diameter glass coated thermistor beads hermetically sealed at the tips of shock resistant glass rods. The small bead thermistor has a very thin glass coating which allows for relatively flat frequency response for flow applications. As much of the bead as possible is exposed at the tip of the glass rod to provide the fastest response times. The units are rugged and unaffected by severe environmental exposures including high density nuclear radiation.

DIMENSIONS:

LEAD-WIRES

.012” (.3MM) DIA.

TINNED DUMET

APPLICATIONS:

The FASTIP THERMOPROBES are ideally suited for high speed measurement and control of fluid temperatures, fluid level or flow. They offer the ease of handling associated with large glass probe thermistors as well as ultra-fast response times of small glass coated bead thermistors. These units exhibit relatively flat response to flow input from 200 H z to

1000 Hz.

CODING:

The code number to be ordered may be specified as follows:

FP07 D A 103 N

Tolerance code letter per Table B(see NOTE 1)

Zero-power resistance as 25°C (see NOTE 2)

Material System Code Letter (see TABLE A)

PROBE LENGTH D = .5" (12.6mm) Other lengths available on special order.

Type FP07, FP10, OR FP14

NOTE 1: Special tolerances are available on request. To specify a non-standard tolerance, use the letter “S” followed by the desired tolerance (i.e., S7= ±7%).

NOTE 2: The zero-power resistance at 25°C, expressed in Ohms, is identified by a three digit number. The first two digits represent significant figures, and the last digit specifies the number of zeros to follow.

Example: A Fastip Probe with a 7msec response time in water, 100k ohms ± 25% at 25°C would be coded FP07DA103N.

1.0 / 1.1 / 1.2 / 1.3 / 1.5 / 1.6 / 1.8 / 2.0 / 2.2 / 2.4 / 2.7 / 3.0

3.3 / 3.6 / 3.9 / 4.3 / 4.7 / 5.1 / 5.6 / 6.2 / 6.8 / 7.5 / 8.2 / 9.1

BOWTHORPE THERMOMETRICS

Crown Industrial Estate, Priorswood Road

Taunton, Somerset TA2 8QY UK

Tel +44 (0) 1823 335200

Fax +44 (0) 1823 332637

THERMOMETRICS, INC.

808 US Highway 1

Edison, New Jersey 08817-4695 USA

Tel +1 (732) 287 2870

Fax +1 (732) 287 8847

KEYSTONE THERMOMETRICS CORPORATION

967 Windfall Road

St. Marys, Pennsylvania 15857-3397 USA

Tel +1 (814) 834 9140

Fax +1 (814) 781 7969

TABLE A: THERMAL AND ELECTRICAL PROPERTIES:

The following table lists the THERMAL and ELECTRICAL properties for all LARGE RUGGEDIZED THERMOBEADS. All definitions and test methods are per MIL-PRF-23648.

THERMISTOR TYPE:

BODY DIMENSIONS:

Max. Diameter:

Body Length: lead-wires:

Nom. Diameter:

Lead Length:

Lead Material:

MATERIAL SYSTEM:

CODE

LETTER

R-vs-T

CURVE

A

A

A

A

A

A

E

A

B

B

B

B

B

B

B

B

D 16

D 17

8

9

10

11

12

13

14

15

6

7

4

5

2

3

0

1

THERMAL TIME CONSTANT:

Still Air at 25°C:

Plunge into Water:

DISSIPATION CONSTANT:

Still Air at 25°C:

Still Water at 25°C:

POWER RATING: (in air)

Maximum Power Rating:

25/125

RATIO

29.4

30.8

32.3

35.7

38.1

45.0

48.1

56.5

75.6

81.0

5.0

11.8

12.5

14.0

16.9

19.8

22.1

22.7

FP07

.085" (2.2mm)

.5" (12.6mm)

.012” (.30mm)

.875" (22mm)

Tinned Dumet

Nominal

Resistance

Range @ 25°C

(Ohms)

300 – 680

680 – 1.6 k

1.6 k – 3.6 k

3.6 k – 6.8 k

6.8 k – 27 k

27 k – 75 k

75 k – 130 k

130 k – 240 k

240 k – 360 k

360 k – 820 k

820 k – 1.3 M

1.3 M – 3.3 M

3.3 M – 6.8 M

6.8 M – 10 M

0.10 sec

7 msec

.05 mW/°C

.25 mW/°C

.006 Watts

FP10

0.85" (2.2mm)

.5" (12.6mm)

.012” (.30mm)

.875" (22mm)

Tinned Dumet

Nominal

Resistance

Range @ 25°C

(Ohms)

300 – 680

680 – 1.6 k

1.6 k – 3.6 k

3.6 k – 6.8 k

6.8 k – 27 k

27 k – 75 k

75 k – 130 k

130 k – 240 k

240 k – 360 k

360 k – 820 k

820 k – 1.3 M

1.3 M – 3.3 M

3.3 M – 6.8 M

6.8 M – 10 M

0.12 sec

10 msec

.09 mW/°C

.45 mW/°C

.010 Watts

FP14

.085" (2.2mm)

.5" (12.6mm)

.012” (.30mm)

.875" (22mm)

Tinned Dumet

Nominal

Resistance

Range @ 25°C

(Ohms)

300 – 680

680 – 1.6 k

1.6 k – 3.6 k

3.6 k – 6.8 k

6.8 k – 27 k

27 k – 75 k

75 k – 130 k

130 k – 240 k

240 k – 360 k

360 k – 820 k

820 k – 1.3 M

1.3 M – 3.3 M

3.3 M – 6.8 M

6.8 M – 10 M

0.15 sec

16 msec

.10 mW/°C

.50 mW/°C

.014 Watts

RESISTANCE -VS- TEMPERATURE CHARACTERISTICS: The nominal resistance range for the zero-power resistance at 25°C is shown for each THERMOBEAD Type and each available Material System. Each Material System is denoted by an ordering Code Letter, a referenced Curve number and the nominal 25°C/125°C resistance ratio..

TABLE B: STANDARD TOLERANCES:

Tolerance Code Letter

± % Tolerance at 25°C

F

1

G

2

J

5

K L M N P Q R S

10 15 20 25 30 40 50 Non-standard – consult factory

BOWTHORPE THERMOMETRICS

Crown Industrial Estate, Priorswood Road

Taunton, Somerset TA2 8QY UK

Tel +44 (0) 1823 335200

Fax +44 (0) 1823 332637

THERMOMETRICS, INC.

808 US Highway 1

Edison, New Jersey 08817-4695 USA

Tel +1 (732) 287 2870

Fax +1 (732) 287 8847

KEYSTONE THERMOMETRICS CORPORATION

967 Windfall Road

St. Marys, Pennsylvania 15857-3397 USA

Tel +1 (814) 834 9140

Fax +1 (814) 781 7969

SERIES 10

PRESSURE SENSORS

+ + APPLICATIONS + +

Ih’ D USTRIAL:

Hydraulic Oil, Refrigerant,

Air and Combustion

Pressures

MILITARY/

AEROSPACE:

Altitude, Sonobuoys,

Turbine Gases, Fuel Oil, and Hydraulic Oil Pressures

GEODETIC:

OCEANOGRAPHIC:

PROCESS

CONTROL:

Down Hole Pressures

Seismic Streamer, Sea

WaterDepth

Liquid Level, Steam,

Corrosive Liquids, Pharmaceuticals, Sanitary Liquids L

+ +FEAI’URES & BENEFITS + +

I-IIGII STATIC ACCURACY & REPEATABILTY:

Insures reproducible measurements

HOUSING MATERIAL SELECTION:

Easily integrated into user’s product

GAGE, ABSOLUTE & DIFFERENTIAL VER-

SIONS: Allows design flexibility

100% COMPUTER TESTED & CALIBRATED:

Minimizes cost to utilize

UNAMPLIFIED OUTPUTS UP TO 1000 mV:

Simplifies or eliminates user supplied signal conditioning

Series 10 pressure sensors are designed for a wide variety of industrial applications. These transducers provide excellent accuracy and are available in ranges from 1.5 to

15,000 PSI.

All Keller-PSI sensors feature a four active arm Wheatstone Bridge strain gage diffused directly into a silicon diaphragm. The resultant monolithic structure exhibits no measurable hysteresis, withstands overpressures to

200% of rated range without damage, and has a life expectancy in the tens of millions of cycles.

The Keller-PSI media proof Series 10 sensors are specifically designed for use with hostile fluids and gases. A conventional Keller-PSI silicon pressure sensor cell is fitted into a stainless steel housing with an integral, compliant stainless steel barrier diaphragm, providing impervious media isolation.

Spring rate ratios in excess of

1OOO:l between the silicon sensing diaphragm and the metal isolation diaphragm preserve the inherent accuracy of the silicon sensor.

The Series 10 all media sensor provides the OEM user with the basic building block necessary to construct his own custom pressure transducer tailored to his own unique requirements.

Series 10 sensors are available in absolute, sealed gage, gage and differential versions in standard ranges from 1.5

PSI through 15,000 PSI.

Standard case and barrier diaphragm material is 316 SS (Hastelloy B/, Tantalum,

Platinuflantalum Optional). Other options include gold plating of the transducer front end for corrosion protection, true flush diaphragm construction for applications where crevice corrosion is a concern, and special designs for sanitary and explosion proof applications.

Consult the Sales Office for your specific needs.

Each Keller-PSI pressure transducer is shipped with a calibration card specific to that transducer. The card specifies compensation resistors required to achieve optimum performance over the compensated temperature range.

+ + SPECIFICATIONS + +

CONFIGURATION

PA-10 =

PAA- =

Sealed Gage

Absolute

PR-10 = Gage

PD-10 =

Differential

NOTE: For applications where measurand can be subatmospheric use PA4 configuration.

M E D I A

Gases & liquids compatible with Stainless Steel and

Buna “N” (Nitrile) [“O-Rings” available in ethylenepropylene, silicone, neoprene, viton or fluoro-silicone]

R A N G E S

S E R I E S

PA10 PM10 PRlO I’D10

PSI B A R

0

0

0

0

0

0

0

0 0 0 0

0 0 0 0

0 0 0 0

0.0. 1

0 0 0 0 30

@ 0 a I 75

0 0 0

0 a

0

150

300

1 . 5

3.0

5

7.5

750

1500

3000

6000

9000

15000

2 0

5 0

100

200

5

10

2

1

400

600

1000

.l

.2

.5

Series

10

250

250

250

250

250

250

2 0

3 5

7 5

100

150

200

250

250

F.S. OUTPUT [mV) P R O O F

(@ I mA Excltalion) PRESSURE [PSI)

+‘ -

1125

2250

4000

7500

10500

38 15

38 15

38 15

45 15

60 30

110 50

225 75

400 75

LONG TERM STABILITY

Ranges 1.5 to 15: * 1% F.S. per 6 months Typical

Other Ranges: f 0.2% F.S. per 6 months Typical

OPERATING RANGE

-10’ to 80°C - Standard

-55’ to 150°C - Optional

THERMAL COMPENSATION RANGE

-10’ to 80°C

THERMAL STABILITY

ZERO SHIFT:

4.01 mVPC typ.: 20.04 rnVPC max.

COEFF. OF SPAN: ~O.Ol%~C typ.: *O.O25%fC max.

1 mA Excitation

ZERO OFFSET

5 mV @ 1 mA Excitation

STATIC ACCURACY*

<* 0.5% FSO BFSL (includes nonlinearity, hysteresis and nonrepeatability at 25’C per ISA S51.1)

LINE PRESSURE

3000 PSI max.

LINE PRESSURE SENSITIVITY (PD-10 only)

0.003 mV/PSI

BRIDGE IMPEDANCE

3.5 K Q Nominal

INSULATION RESISTANCE

100 Megohms at 50 VDC

EXCITATION

0.5 to 5.0 mA Constant Current

WEIGHT

PAA-10, PA-lo, PR-10 - 24 grams

PD-10 - 38 grams

PD-10

PR-10

(Performance specifications stated with compensation resistors installed. All values measured in reference to 25°C at specified constant current.)

liigh Pressure

PA-10

K~PSI

SALES OFFICE: 503 VISTA BELL& SUITE 15A OCEANSIDE, CA 92057 (619) 967.6966/TELM 36094WFAX (619) 9674563

CONTENTS

1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 SPECIFICATIONS. . . . . . . . . . . . . . . . . . . . . . . . . .

3 OPERATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.1 OBS Sensors. . . . . . . . . . . . . . . . . . . . . . . . . . .

3.2 Driver/Signal Conditioner Circuits. . . . . . . . .

4 PREPARING OBS SENSORS FOR USE . . . . . . . .

4.1 Mounting Circuit Boards . . . . . . . . . . . . . . . .

4.2 Electrical Connections . . . . . . . . . . . . . . . . . .

Power Supplies (OBS-1). . . . . . . . . . . . . . . . . .

Power Supply (OBS-3) . . . . . . . . . . . . . . . . . . .

Sensor and Data Logger Connections . . . . . .

4.3 Bench Tests . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.4 Gain Adjustment . . . . . . . . . . . . . . . . . . . . . . .

4.5 Calibration with Turbidity Standards . . . . . .

4.6 Calibration with Sediment . . . . . . . . . . . . . . .

4.7 Deployment . . . . . . . . . . . . . . . . . . . . . . . . . . .

Mounting Sensors and Housings . . . . . . . . . .

Sampling Considerations . . . . . . . . . . . . . . . .

Batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Master/Slave Operation. . . . . . . . . . . . . . . . . .

5 TROUBLE SHOOTING . . . . . . . . . . . . . . . . . . . . . .

6 MAINTENANCE . . . . . . . . . . . . . . . . . . . . . . . . . . .

6.1 OBS Sensors. . . . . . . . . . . . . . . . . . . . . . . . . . .

Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sensor and Circuit Replacement . . . . . . . . . .

6.2 Housings . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6.3 Antifoulant Coatings . . . . . . . . . . . . . . . . . . .

7 OPTICS AND TURBIDITY MEASUREMENTS

7.1 Transmissometers . . . . . . . . . . . . . . . . . . . . . .

7.2 Nephelometers . . . . . . . . . . . . . . . . . . . . . . . .

7.3 OBS Sensors. . . . . . . . . . . . . . . . . . . . . . . . . . .

7.4 Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Power-up Transients . . . . . . . . . . . . . . . . . . . .

Temperature Effects . . . . . . . . . . . . . . . . . . . .

Hydrodynamic Noise and Sampling Errors

7.5 Suspensions with Sand and Mud . . . . . . . . . .

High Sediment Concentrations . . . . . . . . . . .

6 TERMINOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . .

9 USEFUL REFERENCES . . . . . . . . . . . . . . . . . . . . .

......

......

......

......

......

......

......

......

......

......

......

......

......

......

......

......

......

......

......

1 9

2 1

2 2

1 7

1 8

1 9

1 9

1 0

1 1

1 2

1 3

8

8

7

7

3

3

4

2

1

23

24

24

24

24

25

25

33

33

34

35

28

29

29

30

37

38

39

4 1

OD & A Instrument Co.

Rev. 3191

FIGURES

1 5

1 6

1 7

1 8

1 9

20

2 1

2 2

9

1 0

1 1

1 2

1 3

1 4

1

6

7

8

4

5

2

3

Mechanical Dimensions . . . . . . . . . . . . . . . . . . . . . . .

BeamPattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

OBS-1 Circuit Board . . . . . . . . . . . . . . . . . . . . . . . . . .

OBS-3 Gain Adjustment Switch and Potentiometers

Circuit Board Mounting Dimensions. . . . . . . . . . . . .

OBS-1 Wiring Diagram . . . . . . . . . . . . . . . . . . . . . . . .

OBS-3 Connections . . . . . . . . . . . . . . . . . . . . . . . . . . .

OBS-3 Connector Assembly . . . . . . . . . . . . . . . . . . . .

Formazin and Sediment Calibrations . . . . . . . . . . . .

OBS-3 Housing Assembly . . . . . . . . . . . . . . . . . . . . .

Battery Capacity Requirements . . . . . . . . . . . . . . . .

OBS-1 Master/Slave Wiring Diagram . . . . . . . . . . . . .

Optical Particle Detectors . . . . . . . . . . . . . . . . . . . . .

Volume Scattering Functions . . . . . . . . . . . . . . . . . .

Calibrations with Formazin . . . . . . . . . . . . . . . . . . . .

Calibrations with River Mud . . . . . . . . . . . . . . . . . . .

Turbidimeter Calibrations . . . . . . . . . . . . . . . . . . . . .

Power-up Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Water Temperature Effects . . . . . . . . . . . . . . . . . . .

Particle Size and Concentration Effects. . . . . . . . . .

Noise Versus Sand Concentration . . . . . . . . . . . . . .

Response to High Sediment Concentrations . . . . . .

TABLES

1 Solution Volumes for Calibrations with Formazin

.

1 8

2 Sampling Schedules for Monitoring Sediment Transport . . . . 2 1

.

. . .

3

4

.

. . . .

5

6

. .

7

. . . .

9

. . . 1 0

. . . . 1 3

. . .

1 4

1 5

.

2 1

. . 2 2

28

.

.

29

. . . . 3 1

3 1

33

. . 34

.

. . . . 34

35

36

. . . . 38

OD & A Instrument Co.

F&v. 3/91

1 INTRODUCTION

This manual describes the operation and general use of OBS-1 and OBS-3 series turbidity monitors. It contains the most current information available, summarizes relevant contents of past editions of TECHNOTE (our technical news letter), and supercedes all previous versions of OBS Instruction

Manuals.

The heart of OBS monitors is an optical sensor for measuring turbidity and suspended solids concentrations by detecting infrared (IR) radiation scattered from suspended matter. With its unique optical design, OBS sensors perform better than most in situ turbidity sensors in the following ways: l l l l

Small size and sample volume

Linear response and wide dynamic range

Insensitivity to bubbles and organic matter

Ambient light rejection and low temperature coefficient

OBS sensors can be calibrated to measure turbidity or suspended solids directly. Sections 4,5, and 6 explain how to use and maintain them for most applications. Additional information and assistance with problems can be obtained by calling the Technical Assistance number given on the inside front cover.

Like other optical turbidity monitors, the response of OBS sensors depends on the size, composition, and shape of suspended particles. For this reason,

OBSsensors must be calibrated with suspended solidsfrom the waters to

be monitored. Because there is no “standard” turbidimeter, comparisons

of turbidity data acquired with an OBSsensor to datafrom another tur-

bidimeter require in&-calibration with a turbidity standard. The procedures for calibrating OBS sensors are explained in Sections 4.5 and 4.6.

Section 7 gives a summary of the principles governing optical particle detection and recommends ways to reduce errors in OBS measurements.

OBS sensors can be immersed in dilute acid and alkaline solutions. They have been used for a wide variety of monitoring tasks in industrial, laboratory, riverine, estuarine, and oceanic settings and can be integrated in water quality monitoring systems, CTD’s, laboratory instrumentation, flow meters, and sediment transport monitors. Technical terms used in the manual are defined in Section 8.

OBS is a registered trademark of D & A Instrument Company.

U.S Patent 4,841,157

OD & A Instrument Co.

1

Rev. 3/91

2 SPECIFICATIONS

OBS-1 OBS-3

DYNAMIC RANGE

Turbidity 0.5

Mud(maximum)’ 5

Sand(maximum)2

- 2,000 FTU

- 5,000 mg/l

0.02 - 2,000 FTU

0.1 - 5,000 mg/l

100 -100,000 mg/l 2 - 100,000 mg/l

FREQUENCY RESPONSE 10 10 Hz

AMBIENT LIGHT REJECTION Optical filtering and synchronous detection

TEMPERATURE COMPENSATION Solid state temperature transducer

NONLINEARITY

3

Turbidity (formazin, O-2,000 FTU) 2.0%

Mud (O-4,000

Sand (O-60,000 mgl) 2.0% mgl) 3.5 %

2.0%

2.0%

3.5%

DRIFT

Time

4

_.

Temperature

_.

_.

_,

_.

_.

_.

_. -3.5%

_. O.O5%/OC

SUPPLY VOLTAGE COMPLIANCE N/A

SETTLING TIMES

Power-up . . . . . . . . . 10s

25 o C Step Change in Water

Temperature 3 0 s

OUTPUT SPAN

5

(MAXIMUM) 0 - 5 V

OUTPUT IMPEDANCE < 150 Ohms

RMS NOISE AT 0 FTU < 1OOfiV

P O W E R R E Q U I R E M E N T S +9930V/57mA

OUTPUT FILTER 20 Hz

PHYSICAL DIMENSIONS [Inches (mm)/pounds (kg)]

Sensor . . . . . . . . . . . . . . . . . . . . . . . . . ..2(51)x 0.7(18)

Housing Length 9 (230)

Housing Diameter 2.4 (60)

Weight (air) 3 (1.3)

Weight (submerged) 0.7 (0.4)

WORKING DEPTH . 2000 meters

- 3.5 % per decade

O.O5%/~C

250 pV/V

<l s

15 s

0 - 5 v f.s.

< 300 Ohms

< 50 fiv

+ 6 - 15V/12mA

20 Hz (- 3dB)

2 (51) x 0.7 (18)

4.5 (115)

1.2 (31)

0.4 (0.17)

0.16 (0.07)

2000 meters

A %Ep

83-S

nti s*‘yo Q 2

3 5

em

1 Amazon River Mud, D50 = 10Fm.

2 Beach Sand, Dss = 200am.

3 Maximum deviation of response from a least-squares straight line, expressed as a percentage of the calibration range.

4 The output will not drift more than -3.5%, continuous operation, or more than - 3.5% x (duty cycle), burst operation, in the first 2,000 hours.

5 Output span depends on adjustable gain settings.

6 9-12 Vi32 mA for units with 4-20 mA current loop.

OD & A Instrument Co.

2

Rev. 3191

3 OPERATION

3.1 OBS Sensors

OBS sensors consist of a high intensity infrared emitting diode (IRED), a detector (four photodiodes), and a linear, solid state temperature transducer; mechanical dimensions are shown on Figure 1. The IRED produces a beam with half-power points at 50 o in the axial plane of the sensor and 30 o in the radial plane, Figure 2. The detector integrates IR scattered between 140 o and 160 9 Visible light incident on the sensor is aborbed by a filter (< 1% transmission below 790 nm). Sensor components are potted in glass-filled polycarbonate with optical-grade epoxy.

r

~ m O B S - I / - 3 T

1.2(31mm) OB(20mm)

0.8(20mm)

32 mkro-inches

Chamfer 0.04010.0201n

(1.0/0.5mm)

OBS-IC

I

Figure l-Mechanical Dimensions (inches [mm])

OD & A Instrument Co.

3

Rev. 3191

Figure 2-Beam Pattern a

3.2 Driver/Signal Conditioner Circuits

Driver/signal conditioner circuits perform the following functions: l

Switching IRED power. (OBS-l/3) l l l l

Synchronous detection of backscattered radiation (OBS-l/3)

Temperature compensation (OBS-l/3)

Analog temperature signal generation (OBS-1)

Master/Slave synchronization (OBS-1)

A CMOS oscillator provides timing for switching and synchronous detection. IRED radiant power is controlled by an operational amplifier with a reference voltage from the temperature transducer to compensate for the temperature coefficients of the IRED and photodiodes Synchronous detection of photocurrent is accomplished with a operational amplifier, a unitygain inverter, and 180 o phase-shifting switch. The output of the detector is

OD

& A Instrument Co.

4

Rev. 3/91

C O N T A C T N O .

1 IF-1

B-VOLT

REGU\ATOR

Tb3

(PD OUTPUT)

W A V E F O R M S

TPl

‘s-----L

TP3 6

OD

&

A

Instrument Co.

Figure 3 - OBS-1 Circuit Board

5

Rev. 3/91

amplified and lowpass-filtered (-3 dB point: 20 Hz, rolloff: 20 dB per decade). An adjustable, zener-diode reference in the output amplifier allows nulling of offset voltages. The temperature signal fromaBS-1 is buffered with a voltage follower. Clock (OSC) and not-clock (0%) outputs are provided for slaving one or more OBS-1 circuits to a master circuit. The

OBS-3 has a wien-bridge amplifier which provides better noise rejection than the OBS-1 and switch-selectable gain with a trimmer fine adjustment.

The OBS-3 adjustment points are shown on Figure 4. Test point and cornponent locations on OBS-1 circuits are shown on Figure 3. Output impedances are nominally 150 Ohms.

,

Figure 4-OBS-3 Gain Adjustment Switch and Potentiometers

OD & A Instrument Co.

6

R e v . 3191

4 PREPARING OBS SENSORS FOR USE

4.1 Mounting Circuit Boards

WARNING

CMOS components in OBS circuits are susceptible to damage from electrostatic discharge. Use standard CMOSprecautions when handling them.

OBS-1CB

Refer to Figure 5. OBS-1CB circuit boards can be mounted in system enclosures on a flat surface with four 4-40 x l/2-inch plastic screws and l/8 plastic spacers using the holes in the corners of the board or on a bulkhead with two L-brackets (Bracket Kit, Part No. OBS-1B). Select a location for the circuit that will prevent shorts to other system components and mount the board so that electromagnetic interference (EMI) will not be coupled to the board or interconnecting leads. In systems prone to EMI, mount the circuit board in a metal box grounded to the system chassis and use short leads between the sensor and circuit.

Figure 5 - Circuit Board Mounting Dimensions

OD & A Instrument Co.

7

Rev. 3/91

OBS-3B

See Figure 5a. OBS-3B circuits come in an aluminum enclosure that can be mounted with two 4-40 x 314.inch, flathead screws. Select a low- EM1 location for installation and ground a corner of the enclosure to the system chassis.

4.2 Electrical Connections

Power Supplies (OBS-1)

The standard circuit will operate on 12 to 30 Volts and draw 57*2 mA through an onboard &Volt regulator. A voltage inverter on the circuit board provides negative supply voltage to analog circuits. Refer to Figure 6 when connecting leads to the edge connector C-l (TRW 50-12A-30 or equivalent).

Use l/&inch heat shrink on all power-lead connections to C-l.

~-F

D&A INST. CO.

a MADE IN U.S.A.

3.9(98mm1

3.38(85 9mm)

It0.51(13mm)_~

0.16(4.6mm)

~*

1 l(27mm)

I

~

@

I

1) Coax shield

2) Common

3) (IRED anode)

4) Coax center (Iphoto)

5) Temp. Sensor

G Gain

Z Offset

6) Brown (IRED Cathode) T Temp. Coefficient

Figure 5a - Circuit Board Mounting Dimensions

OD

&

A

Instrument

Co.

8

Rev. 3/91

There are three ways to power the OBS-1:

1)

Standard Solder the (+) power lead to contact A (+%) of C-l and connect the (-) lead to contact 5 (COM).

2 ) Low Power regulated 7-9.5 Volt supply without onboard regulator.

3) Lowest Noise split, regulated supplies.

Low Power - For a 7% power savings, bypass the regulator by removing jumper Jl and connect the (+) lead of a regulated 6 to 9.5.Volt supply to contact B and connect the (-) lead to contact 5. An inline, l/8-amp fuse is recommended in the (+ ) supply.

Figure 6 - OBS-1 Wiring Diagram

I

WARNING

Do not connect a negative voltage to pin C

(C-l) unless the unit is cofigured for split supply operation.

OD & A Instrument Co.

9

Rev. 3191

Split Supplies

- Circuit operation from regulated, split supplies results in lowest noise levels. For this option, remove Jl and 53, and move 54 from position A to B. Connect the (+ ) lead to contact B and the (- ) lead to contact C (-V,). Put inline, l&amp fuses in the (-) and (+) supplies and connect common to contact 5.

+&15 VDC (PIN 1)

POWER COMMON (PIN 21

SGNAL (PIN 41

SffiNAL COMMON (PIN 31

Figure 7 - OBS-3 Connections

Power Supply (OBS-3)

See Figure 7 for the pinout of the XSJ-4-BCR bulkhead connector The color code for the XSJ-4-CCP cable assembly is as follows:

RED

Regulated/battery supply (+)

BLACK :::::::::::: PowerCOMMON

WHITE SIGNAL (+ sig)

GREEN Signal COMMON (- sig)

WARNING

Serious damage will occur if more than 15

Volts is applied directly to eitherPin 1 ofthe

XSJ4 bulkhead connector or to the red lead from the circuit enclosure (Figure 7).

OD

&

A

Instrument Co.

10

Rev. 3/91

The power leads for the OBSSB are red (+Vo) and brown (Power Common).

Batteries must provide 6 to 15 VDC to operate the OBS-3. The supply current will vary from 16 to 32 mA, for units with 4-20 mA current loop transmitters, as the sensor output swings from zero to full scale. Power supply compliance must accommodate this range of loads. When power consumption is not critical, a voltage regulator: LM7806C, LM78L82AC, ICL7663, or

MAX663, can be used in the supply circuit.

Sensor and Data Logger Connections

OS-1

Common-mode voltages produced by supply current flowing in 30.meters

of 18 to 24 AWG wire can produce offsets of 30 to 125 millivolts at data logger inputs if it shares a common lead with the power supply. When cables longer than 2.5 meters are used to connect data loggers and power supplies to OBS sensors, differential measurement (see page 39) of the output is recommended (Steps 3 and 4). Follow the wiring and connector-pinout diagrams on Figure 6, page 9 when connecting sensors and data loggers to edge connector C-l.

1. Connect the IRED anode (red wire from OBS-1T sensor) to contact B

(+ 5V) of C-l, the cathode (brown wire) to contact D (Itred), and the detector signal (yellow wire) to contact F (Iphoto).

2 . Connect the power supply (orange wire from OBS-1T sensor) and signal

(blue wire) leads from the temperature transducer to contacts B and

E (Itemp). The sensor cable shield and detector cathode (green wire from

OBS-1T sensors) are connected to contact 5.

3 . Connect the low differential input (- sig) of t,he data logger to contact

5 (COM) and operate the logger from a separate battery or isolated power supply.

4 . Connect the high differential input (+ sig.) to contact 6 (V,,,). The buffered output from the temperature sensor is on contact 1 (Vtemp).

OBS-3

Connect the green wire in the interconnect cable from the OBS-3 to the low differential input( - sig) of the logger and the white wire to the high input (+ sig). Figure 7 shows the connection points for the OBS3B’s. The coax lead (Iphoto) and shield should be soldered to terminals 1 and 4 with short leads. Connect the other color-coded sensor leads as show on the diagram. Connect the flying leads from the OBSSB as follows: yellow (V,,,) and green (Signal COM) to logger (+ sig) and (- sig).

OD

&

A

Instrument Co.

11

Rev. 3/91

4.3 Bench Tests

This section explains tests that were performed prior to shipment. We recommend that you perform them before unattended use of the instrument in the field for long periods. For routine use, a simple “finger-wave test” will tell you if the sensor is working. Test points and wave forms are shown on Figure 3. If a fault is found during bench testing, the section on trouble shooting will help you find the problem.

OBS-1

[email protected]:

DMM with diode tester and dual trace oscilloscope.

Procedure

1. Check all connections.

2 .

Plug the OBS-1PT sensor into the bulkhead connector and plug the circuit board into the edge connector; align the black arrow on the connector with the number (1) stenciled on the upper-lefthand corner of the circuit board. Connect power leads to the unit. On first power up, check the supply current with the DMM and verify that the unit is drawing about 57mA. TP5 is common for the circuit.

3 .

With the oscilloscope, check the clock at TP2 (250 f 50 Hz square wave,

5 V,,) and IRED driver at TPl (- 1.2V square wave in phase with clock).

Check photodiode (PD) output at TP3 (square wave 180 o out of phase with IRED driver and variable amplitude when you wave your finger across the sensor).

4 .

Check vt,, at TP6 with a DMM. It should be about 1.2 V and should increase slowly when the sensor is gripped in the palm of your hand or submerged in warm water.

5 .

Connect the DMM to TP4 (V,,,); repeat the finger-wave test. The output should vary.

OBS-3

Equipment:

DMM.

Procedure

1.

Check all connections.

2 .

Plug the interconnect cable into the bulkhead connector, Figure 8. Connect the power and common leads to the cable, Figure 7, and verify that the unit draws 12 f 2 mA with the DMM.

3 .

Connect (+) DMM probe to the white wire and the (-) probe to the green wire. Do the finger-wave test; the output should vary.

OD & A Instrument Co.

12 Rev.

3191

If all tests are successful, the sensor is ready for gain adjustment and calibra tion. Units ordered with preset gain do not require any additional preparation or adjustments.

LOCKING SLEEVE

Figure 8 - OBS-3 Connector Assembly

4.4 Gain Adjustment

Turbidity

For monitoring applications requiring data in nephelometric turbidity units

(NTU’s), measurements are referenced to a turbidity standard and information about the particles causing backscatter is not essential. Calibration with turbidity standards is explained in Section 4.5; an example calibration is shown on Figure 9. Direct conversion of turbidity in NTU’s to sediment concentration can be very misleading because NTU’s are relative optical units whereas sediment concentration is measured in mass and volume units. Potential problems that can arise in making such conversions are explained in Section 7.3.

Sediments

Sediment or suspended solids concentration is the dry weight of sediment divided by the weight of sample (expressed in ppm) or by the volume of sample in liters (expressed as mg/l). The purpose of adjusting the sensor gain for monitoring suspended sediments is to match the highest output voltage expected from the OBS in the field with the input span of your data logger. One of two undesirable results will be obtained if the gain is not

OD

& A Instrument Co.

13 Rev. 3/91

TURBIDITY- FTU

A M A Z O N M U D - . , ,YELLOW SEA SILT

,

10

IIII

100

1 I /II

1000

, , ,,I

1 0 0 0 0

S E D I M E N T C O N C E N T R A T I O N - m g / L

, 1,

1 0 0 0 0 0

Figure 9 - Formazin and Sediment Calibration Curves correctly adjusted. When the gain is too high, data will be lost because the sensor output is limited by the supply voltage and will “saturate” before peaks in sedimant concentration are detected. If the gain is too low, the full resolution of your data logger will not be utilized. The general procedure and equipment for adjusting gain are the same for OBS-1 and OBS-3 sensors.

Field experience has shown that sensor gain (Volts per mg/l) varies by a factor of 200 with particle size, Figure 9. Particle size varies from one environment to another and with time. Practical consequences of size variations are discussed in Sections 7.3 and 7.5. A rule of thumb is that the relative gains will be about 1.0, 0.3, and 0.08 for mud (<x10 urn), silt (~10 to 62 urn), and sand (> ~200 urn) respectively. Knowledge of the particle size and maximum concentration is therefore critical when monitoring suspended sediments.

OD

& A Instrument

Co.

14

Rev. 3191

Before Starting (OBS-3)

Disassemble the housing by removing the plastic locking strip and withdrawing the end cap (Figures 4 and 10). Slide the circuit card out of the housing to access the gain-selector switch. Be careful not to strain the leads connecting the circuit board to the end cap and sensor. Adjust potentiometer R20 full cw or ccw, then back eight turns to midrange.

O-RING (2.OW)/

Figure 10 - OBS-3 Housing Assembly

Equipment and Materials (Sediments):

Bottom sediment or suspended matter from the monitoring site, two black, matte-finished, plastic buckets with 25-cm I.D.(minimum), l-liter, Class A volumetric flask, 2 gallons filtered distilled water (purified water from the super market works fine), hand-drill motor, paint stirrer, and DMM (data logger or analog recorder are other options).

Procedure

1. Prepare a five-liter mixture of sediment and distilled water with the highest sediment concentration expected in the field. Use the definition on page 39 to estimate sediment concentration. Accuracy is not critical in this step.

2 . Secure the sensor vertically in the bucket so the beam “looks” across the diameter and is at least 5 cm from the bottom: see Figure 2. Add the mixture from Step 1 until the sensor is submerged at least 5 cm.

OD

&

A

Instrument Co.

15

Rev. 3/91

3.

4.

5.

6.

7.

Mix the suspension with the drill motor/paint stirrer, keeping the propeller out of the sensor beam.

OBS-1

sensors: Adjust R16, Figure 3, until the voltage at TP-4 is about

90% of the data logger span. If R16 does not provide enough adjustment, other techniques described below can be used.

OBS-3 sensors: Select the range with SWl, Figure 4, and make fine adjustments with R20.

Fill the second container with distilled water and secure sensor as in

Step 2.

Adjust R20 (OBS-1) or R16 (OBS-3) to obtain an output of kO.005 Volts.

This is the offset.

The gain and offset adjustments interact so it may be necessary to repeat Steps 4 and 5 until the exact desired gain and offset are obtained.

WARNING

The OBS-3 has a single o-ring seal in the end cap. Use extra care when opening and closing the housing. Clean, inspect and grease the o-ring seal before submerging the instrument.

To reassemble the OBS-3 housing:

1. Clean o-ring and mating surfaces; coat lightly with silicon grease.

2. Tuck the connector leads under the circuit board

3. Align the end cap to mate the locator pin with the milled cup in the end of the housing (Figure 10) and press the parts together using a rolling motion. If resistance is encountered during assembly, withdraw the end cap and inspect the o-ring for cuts and nicks; replace if necessary.

4. Insert the plastic locking strip into the guide hole.

Other Gain-setting Techniques (OBS-1)

The gain of OBS-1 sensors is determined by factory-installed resistors RlO and R12; see Figure 3, page 5. These resistors can be increased or decreased to change sensitivity. The standard values are 36.5K for R10 and 665K for

R12. To increase the sensor gain by 200%) replace RlO with a 73.2K, RN55C,

1% resistor To reduce the gain by 50%) replace R12 with a 332K, RN55C,

1% resistor. Use the required combination of R10 and R12 to give the desired gain. RlO should not exceed 200K.

OD &

A

Instrument Co.

16 Rev. 3191

4.5 Calibration with Turbidity Standards

If specified in an order, sensors are factory calibrated with formazin and the sensor gain is shown on a calibration report in mV per FTU. This section explains how to perform the calibration procedure.

Two commercially available turbidity standards are formazin and AMCO-

AEPA-1. The recipe for making formazin is in: Standard Methods

for

the

Examination of Water and W&water (1989 Edition), ISBN O-87553-161-X.

4000-FTU formazin can be purchased from Hach Co., Loveland, CO and many other scientific supply vendors. AMCO-AEPA-1 is supplied by Advanced Polymer Systems, Redwood City, CA. Calibration standards are made by diluting stock turbidity solutions with filtered distilled water. Formazin is stable for about 12 hours. AEPA-1 is stable for six months, but has a maximum turbidity of only 40 NTU. Hereafter, turbidity units are expressed in FTU’s (Formazin Turbidity Units) to indicate the most commonly used standard. In practise, NTU’s and FlU’s are equivalent and interchangeable.

Equipment and Materials:

4000-FTU formazin, Class A, loo-ml TD volumetric pipet, 25-ml TD measuring pipet, one gallon filtered, distilled water, and other equipment from previous procedure.

2.

3.

Procedure

1.

Scrub the sensor, container, stirring mechanism, and glassware with detergent and water and rinse everything twice with filtered water.

4.

Perform the calibration under fluorescent lighting.

Mount the sensor in the container as in Step 2, page 15 and add filtered water with the l-liter volumetric flask until the sensor is submerged at least 5 cm.

Wait one hour for the water to equilibrate to room temperature, and tap bubbles off the container wall.

5.

6.

7.

Monitor the output for one minute. If the output is fluctuating more than a few mV, check the sensor mount and electrical connections, before proceding.

Follow the schedule in Table 1, modified as required for the initial volume used and turbidity range to be monitored.

Stir the container continuously during this step. Add increments of the stock solution with a pipet, recording volumes added and outputs for

OD & A Instrument Co.

17

Rev. 3191

each increment. Allow one minute between increments for mixing. Use the turbidity values in Table 1 or calculate them with:

T

s t d = Tstk

Where:

T

std =

Turbidity of the standard solution; Tstk = Turbidity of the stock solution; Vtot = Cumulative volume of stock solution at each calibration point; vdw = Initial volume.

8 . For very precise measurements, use a polynomial to calculate turbidity from output voltages.

Table 1

SOLUTION VOLUMES FOR FORMAZIN CALIBRATIONS

vdw =

1.0

LITER

and Tstk = 4,000 FTU

Volume to Add

(ml)

Cumulative Volume, Solution Turbidity,

V,, (ml) T,,, (F-TU)

0.25

2.3

23.1

27.0

58.5

65.4

73.5

83.3

2 6 6 . 7

4 0 0 . 0

0.25

2.5

25.6

52.6

111.1

176.5

2 5 0 . 0

333.3

600.0

1000.0

1 0

1

1 0 0

200

4 0 0

600

8 0 0

1000

1500

2000

4.6 Calibration with Sediment

The procedure for sediment calibration is more involved than for turbidity. For a modest charge, we will precalibrate OBS sensors with sediment provided by users. Call us for a quotation to perform this service.

Equipment and Materials:

Analog chart recorder or data logger, scale

(Range: 150 g, Resolution: +O.OOl g), 250 cc plastic syringe, and equipment from gain-setting procedure, page 15.

Procedure

1. Clean containers and glassware with detergent and rinse with filtered water.

OD & A Instrument Co.

18

Rev. 3/91

2 . Do the calibration under fluorescent lighting

3 . Mount sensor as in step no. 2, page 15.

4 . Add filtered water until the sensor is submerged at least 5 cm; allow one hour for temperature equilibration; tap bubbles off container walls.

5 . Check drift as in Step 5, page 17.

6 . Mix the bulk sediment thoroughly, adding distilled water if necessary, and determine the total solids content of the mixture

(Standard

Methods, Method 2540 B with wet weight substituted for sample volume in the calculation). Weigh 5 to 10 equal increments of the mixture with the total dry weight needed to produce the maximum concentration expected at the monitoring site (see definition on page 40).

7 . Monitor and record the output during this step. Turn thestirringmotor on and add sediment increments, allowing 3 minutes between each for mixing.

8 . After each increment, withdraw samples from the suspension with the syringe for suspended solids analysis (Method 2540 D)

9 . Fit output voltages and total solids data to a polynomial curve. The equation for this curve is used to convert voltage output to total solids concentration. See examples of sediment calibration curves on Figure

9, page 14.

4.7 Deployment

Mounting Sensors and Housings

The OBS-1T sensor can be mounted in CTD housings, water quality probes, and other monitoring systems in a l/2-inch hole through a bulkhead

(Figure 1). Pick a location where the sensor is vertical and “looks” away from reflective surfaces (Figure 2). Rotate the sensor so the beam “looks” across the direction of flow. For bottom-mounted systems, housings and sensors should be at least three diameters above the sediment bed to avoid scouring sediments during periods of strong flow. OBS sensors will not operate in water less than 25 cm deep on sunny days because the detector will saturated with by IR from the sun. When monitoring in shallow water, record OBS data at night. Insulate housings from mounting brackets with plastic washers and fasteners.

Sampling Considerations

A well-designed sampling scheme collects the minimum number of data points needed to understand processes relevant to the objectives for a specific monitoring program. lb do this, the high and low-frequency variations must be sampled so that the hourly or daily fluctuations as well as the mean and

OD & A Instrument Co.

1 9

Rev. 3191

maximum values will be captured in the data. Burst sampling with programmable data loggers is a proven way to meet most sampling objectives in rivers, estuaries, and the ocean. To optimize burst sampling programs, a lot must be known about a site before monitoring begins, so when in doubt, oversample until uncertainties about the environment are eliminated.

The program parameters you will have to select are the sampling frequency, burst duration, and burst interval. Sampling frequency, the number of recordings made per second, determines the most rapid fluctuations you can detect, typically 18th to l/4 the sampling frequency. For example, if turbidity fluctuates at 1 Hz, sample at 4 to 8 Hz to resolve the fluctuations. The second parameter is the length of time the logger records data during each burst. The number of samples, n, is thus proportional to burst duration and the uncertainty in the burst-sample mean values is proportional to l/&r.

Smaller errors can therefore be obtained with longer bursts at the expense of space in logger memory. The burst interval is the time between the start of consecutive bursts. It should be about 1/16th to 18th of the period of low frequency fluctuations in turbidity. lb sample turbidity in an estuary with semidiurnal tides, for example, sample at least once per hour

With this brief background, two general recommendations are: l

Use the shortest burst duration and longest burst interval needed to characterize the time variation of the process under study and to conserve battery power l

Synchronize the sampling schedules for turbidity, flow velocity, and pressure.

The first recommendation requires knowledge of flow conditions and sediment transport regimes at the monitoring site. The penalties for bursting too long or frequently are that excess data will be recorded and battery capacity will be wasted. If the sampling frequency is too low or the burst interval too long, critical information about sediment transport and the flow regime will be missed.

The second recommendation is important because correlations between flow and turbidity can provide valuable information about particle transport, turbidity control strategies, and process optimization. Table 2 recommends sampling schedules for natural waters. Use these schedules when local knowledge is not available.

IMPORTANT NOTE

Program the logger to start burst sampling [email protected] thesensor has equilibrated with water temperature andproduces stable readings.

See Figure 18, page 34. For OBS-1 ‘s, allow 10 seconds aJ?erpower is switched on. Allow 1 second for OBS-3’s.

OD & A Instrument Co.

2 0

Rev. 3191

‘hble 2

SAMPLING SCHEDULES FOR MONITORING

SEDIMENT TRANSPORT

Environment

Burst Duty

Frequency Duration Interval Cycle

River Estuary

Beach

Continental

Shelf

2 2 Hz Hz

2 H z

1 H z

30 30 sec. sec.

30 15 min. min.

3.2 1.6 % %

5 min.

30 min.

14.3 %

7 min.

60min.

11.7 %

Batteries

Battery life depends on water temperature and duty cycle (ON time/total time). Once a sampling schedule is chosen and water tempemture is known, the length of time the sensor will operate on a particular battery can be calculated with the formula:

DeploymentTime =[‘:’ x::GbTs)]I1 - (30- tw) x Dl

Where:

Tg = burst duration in minutes; T, = burst interval in minutes; tw = water temperature in OC; Cb = battery capacity in Ah; and i, is 0.057 A for OBS-l’s and0.012 A for OBS3’s. D = Temperature denting factor ( 0.007 for lithium

(LiSOCLz), alkaline, NiCad and Gel cells). Figure 11 shows the required capacities for operating OBS-1 and OBS-3 sensors from alkaline cells.

The battery for the (-) side of a split supply must have at least 5X of the positive supply capacity. Use the same type of battery for both supplies.

1

I 1111

10

I IllI I

100

IllI

1000

1

T I M E - h o u r s

Figure 11 - Battery Capacity Requirements

OD

& A Instrument Co.

21

Rev.

3/91

Master/Slave Operation (OBS-1 series only)

When two or more OBS-1CB circuits are powered from a single supply, master/slave operation is recommended to suppress spikes in outputs caused by asynchronous switching of sensor emitters. Lowest noise levels will be achieved in systems with an even number of circuits. In master/slave opera tion, the master circuit provides clock for the slave circuits. Each OBS-1 circuit has a CLK input the (contact 2 of C-l, Figure 6) for connecting an external clock and two clock outputs, OX and OSC (contacts 4 and 3); the

OSC is the OSC signal phase-shifted 180 9

Configure a system with four OBS-1 sensors as follows (Refer to Figure 12):

1. Select one circuit for a master.

2 . Remove the 52 jumpers from the slave cicuits (Figure 3).

3 . Connect the OSC output of the master to the CLK inputs of circuits two and three,

4. Connect the OSC output of the master to the CLK input of circuit four.

Each pair of circuits will trigger on alternate half cycles of the master clock and the system will draw 114 mA DC when configured in this way.

Figure 12 - OBS-1 Master/Slave Wiring Diagram

OD & A Instrument

Co.

22

Rev. 3/91

5 TROUBLE SHOOTING

The purpose of the procedures explained in this section is isolate problems that can be fixed easily (replacement of fuses and resistors, cleaning contacts, etc.) from serious ones you will need help with. If the information in this section is not sufficient to solve your problem, call us at the Technical

Assistance number given on the inside front cover.

OBS-1 The OBS-1CB is thru-hole technology which can be repaired by a knowledgeable person with I.C. extraction tools and replacement parts, however, I.C. replacement can damage printed circuit traces and void the warranty If you must repair an OBS-1CB to meet operational commitments, call us for assistance first.

Test point locations and wave forms are shown on Figure 3. TP5 is common for the circuit.

Fault Procedure

No power to the I.C.‘s.

Check battery and sensor connections. Check

(Pin 4 of I.C. U6 should fuses Fl and F2. If either is blown, locate be 6 Volts; pin 11 of U6 and fix the short circuit, and replace the should be about 5.5

fuse(s).

Volts.)

Clock waveform at TP2 Check connections (circuit board/sensor and but no square wave at sensor/bulkhead connector).

TPl

Square wave at TPl Check connections (circuit board/sensor and but no inverted square sensor/bulkhead connector).

wave at TP3 during finger-wave test

Sensor connections and circuit board are

OK, but unit does not pass the finger wave test

Something may be wrong with the sensor that you can not fix. To isolate the problem, remove the circuit board from the edge connectar; set DMM to the microampere scale; connect the red (+) test lead to contact F of

C-l and the black (- ) lead to contact 5. Shine a flashlight on the sensor; the DMM should read 100 to 500 uA. Switch the DMM to diode test and connect the red (+ ) test lead to contact B and the black (-) lead to contact D.

The DMM should read 1.1 to 1.3 Volts. If the sensor fails either test, the sensor or underwater connector must be replaced.

OBS-3 There are no user-servicable components on the OBS-3 circuit board. If a sensor fails the “finger wave” test (Step 3, page 12), check the sensor lead connections to the terminals on the circuit enclosure and the bulkhead connector (Figure 7). If they are correct, something is wrong that we must fix.

OD

&

A

Instrument Co.

23

Rev. 3191

6 MAINTENANCE

OBS sensors require little maintenance. The pressure housing and o-ring seals require routine maintenance and should be carefully inspected every six months and serviced before every long-term deployment.

6.1 OBS Sensors

Cleaning

Sensors can be cleaned with mild detergent applied with a soft cloth or sponge. If thick biofouling has developed:

1. Scrape the material off with a flexible knife, taking care not to scratch the epoxy.

2 . %pe a strip of 400 to 600-grit wet/dry abrasive paper on the edge of a bench top.

3 . Add a few drops of water and work the face of the sensor on the wet abrasive, using the edge of the bench for a guide.

4 . Continue until the sensor is shiny and pit-free.

Polishing with abrasives can be repeated as needed until about 1 mm of epoxy has been removed. Deeper polishing may damage the IRED. Check the calibration of the sensor with formazin before and after cleaning with abrasives; see page 17.

WARNING

Do not usesolvents or abrasives coarser than

400 grit to clean OBS sensors.

Sensor and Circuit Replacement

OBS sensors and circuits are matched by serial number and factory-adjusted to compensate for temperature effects. Replacing either component, without readjusting the compensation circuit will degrade data quality. The adjustment is straight forward but requires temperature-controlled turbidity standards. We will make the adjustment when processing orders for new sensors or circuit boards if the defective units are returned to us.

A simple test to check if your sensors are operating within specification is as follows.

1. Prepare a lOO-FTU standard, at room temperature as described in Section 4.5.

2 . Measure and record the temperature of the standard to +_ 0.1 OC.

3. Place the sensor in the standard; see step 2 page 15, and record the output.

OD & A Instrument Co.

24

Rev. 3191

h

v

.

4. Chill the standard to about 10 OC. Don’t freeze it; you will have to start over again,

5. Stir the chilled standard and repeat step 3; record the results.

6. Compute the temperature coefficient using the formula:

T.C. =

(vhi - Vlo)l(Vhi f Vlca) X 0.5

Thi - TI,

1 x 100%

Where:

Vhi = output in Volts (warm standard); VI, = output (cold standard), and

Thi and TI, are the warm and cold standard temperatures.

7. If T.C. is less than 0.05 %. the sensor is operating within specifications.

6.2 Housings (OBS-3)

l

Disassemble o-ring seal and inspect mating surfaces for pits and scratches.

l l l

Inspect o-ring for cuts and nicks; replace if necessary.

Clean o-rings and mating surfaces with a cotton swab and alcohol.

Remove fibers from groove and mating surfaces. Regrease o-rings with

DOW Compound 111 and reassemble.

Check exterior surfaces of housing for pits, chips, and corrosion. If needed, recoat damaged areas with metal primer and marine-grade polyurethane paint.

6.3 Antifoulant Coatings

Small quantities of clear TRTA antifoulant coating will be supplied, free of charge, for monitoring biologically active waters. Preliminary tests with

TRTA indicate that: l

It prevents most marine algae and encrusting animals from growing on optical surfaces for up to two months.

l l

When applied in a uniform, 0.1 mm-thick coating, the material reduces visible light transmission through optical windows by about 0.4 %. No tests have been made with IR.

The transmittance of TBTA coatings decreases by as much as 6% when immersed in unfiltered sea water for two months.

l

TRTA does not prevent algal growth in some areas. Algal growth can cause large offsets and output drift in as little as 35 hours under extreme conditions.

OD &

A

Instrument Co.

25

Rev. 3/91

Y

These tests and limited use of TBTA on OBS sensors indicate that it will initially reduce the sensitivity of OBS sensors by about 1% and that sensitivity may decline by as much as 12 % after immersion in sea water for two months. Before applying TBTA to sensors, consider the certain decreased optical performance that will occur, the cost of lost data that could result from thick biofouling, and that coated sensors may not meet our specifications. For these reasons, and because TBTA is illegal in some areas, we do not precoat new sensors for our customers.

It is illegal to use TBTA in many places.

Check the water quality regulations coated OBS sensors. Use

of

TBTA is the sole

The following procedure is recommended for coating OBS sensors with

TBTA:

Equipment and Materials:

Oven and organic vapor hood, glass microscope slide, applicator stick, acetone, and alcohol

Procedure

1. Before starting, set the TBTA vial out for one hour to release air bubbles.

2. Preheat the oven to 5O’C.

3. Clean and polish sensors, if necessary. The procedure is on page 24.

4. Swab the sensors with alcohol, ONLY, and let them dry.

5. Clean the applicator stick and microscope slide with acetone.

Perform the Next Five Steps Under a Hood.

1. Open the vial and slowly dip the applicator stick about 5 mm into the

TBTA.

2. Remove the stick and deposit two drops of TBTA on the face of the OBS sensor near the steel sleeve.

3. Press the microscope slide onto the TBTA and smear the material across the sensor face in a single, smooth motion.

4. Put the coated sensor in the oven for ten minutes.

5. Complete the cure at room temperature for 8 hours before handling the sensors.

OD & A Instrument Co.

26

F&v. 3/91

Coated Sensors

l l l

Do not store coated sensors in plastic or foil wrappers.

Calibrate coated sensors before and after use in the field.

It is recommended that water samples be taken in the field so that calibration equations can be corrected for optical degradation of the coating.

OD & A Instrument Co.

27 Rev. 3191

7 OPTICS AND TURBIDITY MEASUREMENTS

Practical applications of optical theory in fluid testing were recognized long ago. Optical theory is the starting point for most turbidity monitoring systems, but practical features and theory are usually commingled to make systems that function reliably for specific applications. Although many turbidimeter designs have traits in common with “black-art”, theory provides a good basis for understanding design approaches and what turbidimeters actually measure. We begin this section with some basic principles of light transmission and scattering in water.

A simple model of light transmission in water is: a + b = c which equates the attenuation coefficient (c) with the absorption and scattering coefficients (a and b). In words this means that light transmission in water is attenuated by scattering (deflection by water molecules, dissolved solids, and suspended matter) and absorption which converts light to energy with a different wavelength. Attenuation, absorption, and scattering are inherent properties of water which are affected only by impurities and have precise mathematical definitions. Water, completely free of impurities, is optically pure and is not commercially available and must be made or obtained from a local laboratory. Deiononized water that has passed through a 0.2 urn filter is adequate for most practical purposes. As optical in struments improve, better pure-water standards will be essential for intercalibration.

Figure 13 - Optical Particle Detectors

OD & A Instrument Co.

28 Rev. 3/91

.

Most turbidimeters are configured in one of the ways shown in Figure 13.

These include: transmissometers (A), forward-scatterance (B), 90 o scatterance (C), and backscatterance nephelometers (D). Some instruments combine two or more of these configurations and blend signals to produce a useful output.

7.1 Transmissometers

These instruments measure the ratio, T = Et/E,, of light at the ends of a beam with length, L. T equals 1.0 (100%) when all light (E,) at the near end of a beam exits the far end (Et). Attentuation equals 1-T; it is the only optical property “inherently” easy to measure with off-the-shelf instruments. An ideal transmissometer measures only light that transverses the entire length L by producing a very narrow, collimated beam. Most transmissometers, however, detect some stray light, shown by the + symbols on Figure 13, and overestimate transmission.

7.2 Nephelometers

Nephelometers measure scattered light, and respond mainly to the firstorder effects of particle size and concentration. For example, water containing 10 ppm of suspended clay scatters more light than water containing

10 ppm of suspended sand and both materials produce more scattering as particle concentration increases. Most laboratory turbidimeters are con- figured to measure 90’ scattering; OBS senors operate in configuration D.

10 -

*

I -

FORWARD SCATTER

I

*

-

I I I

In w z lo-'-

! ?

4 IO-Zt

=:

: c" IO-S-

IO-G-

1 -

10-S -

IO

I I III

IO"

I

SCATTERING ANGLE- +

I I I II I II I

1000 180”

Figure 14 - Volume Scattering Functions

OD & A Instrument Co.

29

Rev. 3/91

7.3 OBS sensors

OBS sensors are miniature nephelometers that integrate infrared radiation

(IR) scattered at angles between 140Oand 165 9 Infrared radiation is used because it is strongly attenuated by water, decreasing in intensity by more than 63 %, for every 5 centimeters traversed in clear water. As a result, the

IR beam emitted from an OBS sensor does not penetrate very far in water and IR from the sun is strongly attenuated with water depth. Other advantages of backscatter configurations are illustrated by the scattering functions on Figure 14 which show how scatterance varies with scattering angle inferred from data and model calculations. Collectively, the curves represent light-scattering properties of waters from nearshore, coastal, and open-ocean regions, and suspensions of particles with various sizes. The

X-axis is divided into forward and backscatter domains at 90 9 The Y-axis represents relative scatterance signal.

Three points are illustrated by the scattering functions. First, scatterance signal levels decrease exponentially with scattering angle. Second, the signal range, clear to muddy water, increases with scattering angle. Third, the signal range representing the independent effect of particle size indicated by the solid bars is largest at high scattering angles. By optimizing response to high-angle backscattemnce, signal strength is traded for greater sensitivity to changes in particle concentration and size. Some practical benefits and problems are discussed in the following sections.

Which is best, a trsnsmissometer or an 06s sensor?

Both instruments have well-documented advantages and disadvantages.

The main advantage of a transmissometer is that it measures attenuation, an inherent optical property. OBS sensors have superior linearity in turbid water The merits are illustrated by the response of the two instruments to formazin and Amazon-River mud on Figures 15 and 16. For both materials, the transmissometer is more sensitive at low concentrations than the OBS sensor. Above about 10 FTU and 50 mg/l, large changes in turbidity produce disproportionately small changes in transmissometer output, whereas the OBS sensor responds linearly over the full range of test conditions.

The upper limit of linear response of transmissometers can be extended by shortening the path length (L on Figure 13). A rule of thumb is that L should equal the reciprocal of the attenuation coefficient (l/c). For a turbidity level of 10 NTU, c is 5 mm ‘, and 20 cm is about the right path length.

A suspension containing 100 mg/l Amazon River mud has a attenuation coefficient of about 10 m-‘, and 10 cm is about the right length. In very turbid water, this is impractical because the source and detector get too close together for undisturbed flow through the transmissometer. When sediment concentration and turbidity vary over a wide range, multiplepath transmissometers are required; these are expensive and not readily available.

OD

& A Instrument Co.

30

Rev. 3/91

1

TURBIDITY - FTU (X METER )

1 0 100

1

I

TURBIDITY - FTU COBS)

Figure 15 - OBS Sensor and Transmissometer

Calibrations with Formazin

SEDIMENT CONCENTRATION - mg/ L (XMETER)

1 0

I

100

I

I’ 30

)OO

SEDIMENT CONCENTRATION - mg/L (OBS)

Figure 16 - OBS Sensor and Transmissometer Calibrations with Amazon River mud

OD & A Instrument Co.

31

Rev. 3191

Recommendation: use transmissometersfor monitoring water with

of

turbidity or particle concentrations is expected.

Can turbidity be converted to suspended solids concentrations and viseversa?

In most situations, conversions between turbidity and suspended solids concentrations will give misleading results because the conversion equates an optical property, in relative units, with one precisely defined in terms of mass and volume; these are “apples and oranges”. The definitions follow.

Turbidity

is the “cloudiness” in water produced by light scattered from suspended particles, colloidal material, and other impurities. It is an a~parent optical property that depends on the characteristics of the scattering particles, external lighting conditions and the instrument used to measure it. Turbidity is usually measured with a 90° -scatterance nephelometer and reported in nephelometric turbidity units (NTU’s or

FTU’s depending on the standard).

Suspended solids

can be separated from a water sample, and accurately weighed to determine concentration in parts per million (weight of solids/weight of sample in ppm) or in milligrams of solid per liter of sample (mg/l).

Data from turbidimeters made by different companies should be compared cautiously. Inconsistencies between instruments result from variations in light sources, detectors, and optical configurations. Detailed treatment of these factors is beyond the scope of this manual, but an example of the problems that can arise is illustrated by the response of two different turbidimeters, to 85 nm particles shown on Figure 17. The meters were used to measure water samples with particle concentrations from 0 to 50 mg/l.

Although calibrated with the same turbidity standard, the meter outputs differ by 35% at a concentration of 35 mg/l.

Recommendation: convert turbidity to suspended solids concentration onlg when:

1. Measurements are made with the same turbidimeter.

2. The turbidimeter is intercalibrated with a turbidity standard and suspended matter from the waters to be monitored.

3. Particle size and composition do not change over the monitoring period.

Compliance with the last condition is crucial but virtually impossible to verify in the field because it is difficult to sample particles in their natural state and preserve them for laboratory analysis in a meaningful and consistent way.

OD &

A

Instrument Co.

32

Rev. 3/91

25C

200

T U R B I D I M E T E R A

0 z

2 15c

IL

2

E 100

TURBIDIMETER B

50

(

I

10

I I

20 30

I

4 0

I

50

P A R T I C L E C O N C E N T R A T I O N - m g / L

Figure 1 7 - Turbidimeter Calibrations

7.4

ERRORS

The most common cause of errors in OBS data is improper calibration. Errors can also result from: 1) power-up transients; 2) rapid changes in water temperature; 3) hydrodynamic noise; and 4) sampling errors. In some situations the errors are very difficult to quantify and can degrade data quality.

This section describes the causes of errors, gives estimates of their relative magnitudes, and recommends ways to avoid them.

Power-up Transients (Self-heating effects):

It takes 1 to 10 seconds for the internal temperature of OBS sensors to stabilize and for the temperature transducer to track and compensate for heat from the emitter after power up. This can result in significant errors when OBS sensors are burst sampled. Figure 18 illustrates these effects. The shaded areas under the output signals show the exponentially-decaying errors that occur after power is switched-on. For OBS-1 sensors, the initial error is + 6%; after 8 seconds, it drops to 1%. The sensor reaches thermal equilibrium and gives a stable reading in 10 seconds. The settling error is initially much larger for the

OBS-3, but the output stablizes in less than a second.

Recommendation: Allow 10 seconds for OBS-1 sensors to stabilize merpower-up before sampling. Allow one second for OBS-3 sensors to stabilize.

OD & A Instrument Co.

33 Rev. 3191

- - - _ -

O - -

ON

OFF O N fO- ’

5-

0

I

1 0

I I

I

20 30

TIME - SECONDS

I

Figure 18 - Power-up Errors

I I

4 0

I

I I

5 0 o-

-5-

-1Ok ,

0

I

3 0

I

6 0

I

9 0

-I

T I M E - s e c o n d s

Figure 19 - Water Temperature Effects

Temperatute effects:

Water temperature can change a few degrees per hour in lakes, rivers, and estuaries, and sensors must track variations of a few degrees C per minute when used to profile stratified lakes and estuaries.

Figure 19 shows how OBS sensors respond to a stepwise, 25 OC change in water temperature. Because the sensor is an insulator it takes about 6 minutes for the transducer to produce a stable output. The initial error in

OD & A Instrument Co.

34

Rev. 3191

.

turbidity produced by a 25 OC, stepwise change in water temperature is about 8% for OBS-l’s and about 3.5% for OBS-3’s. The turbidity reading of the OBS-1 stablizes in about 30 seconds whereas the OBS-3 stablizes in

15 seconds. Normal water temperature fluctuations in the ocean will produce errors less than 0.05% per ‘C.

Recommendation: When making vertical profiles in strong thermal gradients, reduce instrument lowering speeds and stop theprofiler at sampling depths to allow the OBSsensor to equilibrate with water temperature. Do not rely on the temperature output

of

OBS-1 sensors when water temperature changes by more that 6OC per hour.

Hydrodynamic noise and sampling errors:

These are inherent problems with sensors that are about the same size as the length of gradients in properties being measured. Hydrodynamic noise is an artifact of the turbulent flow around the sensor which redistributes particles in the water via inertial effects and increases the variation of sediment concentration above natural levels. The volume sampled by OBS sensors depends on how far the IR beam penetrates into the water This decreases as sediment concentration increases and so the sample volume is constantly varying with concentration.

T U R B I D I T Y - F T U

‘(SILT) 1

I

2

I

3

I

4

I

(SAND) 10 2 0 3 0 4 0

S E D I M E N T C O N C E N T R A T I O N - g/L

5

I

5 0

Figure 20 - Particle Size and Concentration Effects

OD

& A Instrument Co.

35

Rev. 3/91

It is virtually impossible to determine the independent effects of hydrodynamic noise and changing sample volume with field data because they usually interact and are correlated with each other and flow velocity. For this reason, the relative magnitude of the errors from these effects is usually unknown. In applications involving coarse sediments, hydrodynamic noise and sampling errors lead to considerable uncertainty about the real variance of the monitored process. Deciding what portion of the signal variance is an artifact of the OBS sensor and what part is real variation in sediment concentration is a problem best resolved by laboratory tests.

Some light has been shed on the problem by trends in calibration data for particles with different sizes. Figure 20 shows the results of tests with sand, silt and formazin conducted in a vigorously stirred tank. The graph indicates the peak-to-peak noise in the band from 0 to 5 Hz as a function of concentration. Formazin consists of very small particles (< 0.1 urn) and produces very low noise levels. The curves for silt and sand indicate strong correlations among noise level, mean concentration, and particle size. For example, at a concentration of 5 grams per liter, the noise level for silt is about

80 mV peak-to-peak on a mean signal of 1000 mV; for sand, it is about

320 mV peak-to-peak on a mean signal of 1000 mV.

From these limited tests, it appears unlikely that peak-to-peak noise would exceed 30% (5% RMS) of the mean signal level in situations with high concentrations of coarse sediment. Also, hydrodynamic effects and variations in sample volume seem to produce random noise; see Figure 21.

100

4 2g/L

> *O

F F 2.89/L

2 6 0 -

:2

4 0 -

2. -

2

OO

I I I I I

20 40 60 80 100

O.Y2g/L

-

- 0.389/L

- O.OYg/L

I I

120 13

T I ME - seconds

Figure 21 - Noise Versus Sand Concentration

OD &

A

Instrument

Co.

36

Rev. 3/91

0.679/L

Recommendation: Postprocess thedata with a lowpassfilter to reduce hydrodynamic noise. When doing this, bear in mind that somepotentiallg useful information will by removed with the noise.

Noise in OBS outputs:

Noise in OBS measurements comes from four sources: l l l l

The electronics

Environmental conditions

Hydrodynamic effects

Sampling errors.

Electronic noise in the output that can be measured with a true-RMS voltmeter while the sensor is in clear water. The RMS noise in the band from 10 Hz to 100 kHz is typically less than 100 PV for the OBS-1 and 50 PV for the OBS-3. This level is insignificant for most applications.

Environmental noise results from factors unrelated to the processes you want to observe. The signatures include: spikes or offsets several times larger than the mean signal level and drift uncorreated with changes in flow conditions. Some causes are: biofouling, excess suspended sediment resulting from scour around instrument structures, and mooring cable, line, or fish moving in front of the OBS sensor with the currents.

Recommendation: With the exception of biofouling andjish, environmental noise can be avoided bggood sgstem design. There is no completely acceptable cure for biofouling, however, clear antifoulant coatings described in Section 6.3 alleviate theproblem to some degree.

The best remedy is regular maintenance and cleaning of the sensors.

7.5

SUSPENSIONS WITH MUD AND SAND

As mentioned earlier, backscattering from particles is inversely related to particle size, on a mass concentration basis, page 14. This causes the sensitivity of OBS sensors to change with particle size by more than an order of magnitude and can lead to serious difficulties when monitoring flow regimes where particle size varies with time. For example, when sandy mud goes through a cycle of resuspension and deposition during a storm, the ratio of sand to mud in suspension will change. An OBS sensor calibrated for a fixed ratio of sand to mud will therefore indicate the correct concentration only part of the time. Also, the calibrations discussed above indicate that statistics of sampling errors and noise can change with the particle size distribution. So, in addition to uncertainty about mean sediment concentration resulting from the unknown variation in the slope of the calibration curve, there will also be uncertainty about the natural variance of concentration as explained on page 36.

There are no simple remedies for these problems. The obvious thing to do is take a lot of water samples and analyze them in the laboratory. This is not only impractical during storms when the errors are likely to be largest, but also defeats the main purpose of having a sensor.

OD & A Instrument Co.

3 7

Rev. 3191

> tl

2

I-

2

2

1

Recommendation: Do not relg solely on OBS sensors to monitor suspended sediments when particle size or composition are expected to change with time at a monitoring site.

High Sediment Concentrations:

At high sediments concentration, particularly in suspensions with high clay and silt contents, the infrared radiation from the emitter can be partially blocked and backscatter decreases with increasing sediment concentration above a critical level. For river mud, this occurs at concentrations greater than about 5,000 mg/l. Figure 22 shows a calibration in which this occurs. The slope of the curve is positive at low concentrations, decreases to zero at the critical level, then goes negative.

This presents problems when an OBS sensor is calibrated for the linear range but concentration exceed the critcal level in the field. In this situation, the output voltage can represent two concentrations depending on the previous history of the signal. For example, when the concentration reaches the critical level, a subsequent drop in output could means that sediment concentration continues to rise, or that it has begun to fall. Deciding which is the case is not easy.

Recommendation: Do not exceed the specified turbiditg orsuspended sediment ranges unless calibrations extend over the full range expected in thefield and a well-tested data interpretation scheme to resolve signal ambiguities is available.

I

I I I

I I

1

3

.

0

' 0

2

I

4

I

6

I I

8.+

I

IO

SEDIMENT CONCENTRATION - g/L

I

1 2

Figure 22 - Response at High Sediment Concentrations

I

1 4

OD & A Instrument Co.

38

Rev. 3/91

8 TERMINOLOGY

AC Coupling: Transmission of AC electrical signals between circuits, usually done with a capacitor which passes AC and blocks DC.

Back/Forward Scattering: The deflection of radiation by particles away from a beam axis at angles between 0 and 90 o (forward scattering) and 90 to 180 o (back scattering).

Burst Sampling: A schedule of evenly spaced sampling intervals (bursts) during which a large number of measurements of turbidity, velocity, pressure, etc. are taken at high frequency. A way to get a large amount of information about a time-varying process with a small number of samples.

CommonMode

Noise: A signal containing no information that appears at both input terminals of a detector or data logging device.

Detector:

An electronic device that responds in a predictable way to electromagnetic or mechanical parameters such as light, current, voltage, or force.

[email protected] Measurement: A technique for detecting an electrical signal using a detector with high input impedance and two wires connected to only one part of the circuit being measured. Differential measurements are immune to some types of common-mode noise

Drifi:

An undesireable andsometimes unpredictable change in the output of an electronic device caused by ambient temperature fluctuations, deterioration of materials, humidity changes, and heating of circuit components.

Duty Cycle: The percentage of time a device is turned on. An important factor in determining batttery life.

DynamicRange:

The difference between the highest and lowest values of a parameter that can be detected by an instrument.

Emitter: A device or material that radiates energy when excited by current, voltage, or external radiation.

Frequency Response: The highest frequency of a time-varying parameter that can be detected by a particular instrument. Usually about l/8 to l/4 of the sampling frequency.

Gain: The value of a system output divided by the value at its input. For example, if the voltage at the output terminal of an amplifier is 10 Volts and the voltage at its input is 1 volt, the gain of the amplifier is 10.0

Half-power Points: The points in a beam or cone of light, radiation, or sound where the radiant energy is half the value at the axis of the beam or cone; see Figure 2, page 4.

LinearBesponse/Linearitg:

The degree to which the output of a detector is a linear function of the parameter detected. A detector is perfectly linear if the correlation coefficient, r, for a regression of parameter values

OD

& A Instrument Co.

39

Rev. 3/91

on output levels is 1.0. Also expressed as the maximum deviation of calibration data from a least-square straight line.

Master/Slave Operation: A system of coupled electronic devices in which one device, the master, provides timing signals for operating one or more slave devices; see Figure 12, page 22.

Nephelometer:

An opto-electronic instrument that measures particle concentrations in fluids by detecting scattered light.

NTU: Nephelometric turbidity units A numericalscale for the cloudiness of liquids that is applied with a nephelometer calibrated with turbidity standards.

Offset: An undesirable constant signal at the output terminal of an electronic device that contains no information.

Scattering Angle: The angle between the acceptance cone of a light detector and a beam of light passing through a scattering medium.

Split Supplg: A power supply with one positive terminal and one negative terminal referenced to a third common terminal.

Suspended Solids @article) Concentration: The mass of filterable solid matter suspended in a known volumne of fluid divided by the mass of the fluid plus the mass of the solids. For example, a suspension of one gram of mud in one liter of fresh water will have a particle concentration of

999.0 ppm by mass.

Sgnchronous

Detection: A technique for sensing the time variation of a parameter with synchronous excitation and detection that is immune to interference and noise.

Temperature Coefficient: The percent change in the response of a sensor or device per unit change in temperature. For example a pressure sensor with a temperature coefficient of +l% per OC that reads psi at 25 OC with read 10.1 psi at 26 “C without a change in pressure.

Threshold:

The lowest value of a parameter to which a sensor will respond predictably.

Turbidity: The cloudy appearance of a liquid produced by light scattered from suspended, colloidal, or dissolved matter.

OD & A Instrument Co.

40 Rev. 3191

9 USEFULREFERENCES

Absorption and Scattering of Light by Small Particles. Bohren, C.F

and D.R. Huffman. 1983. John Wiley & Sons. ISBN 0-471-05772-X.

530 pages.

An Optical Instrument for Monitoring Suspended Particles in

Ocean and Laboratory. Downing, J.P. 1983. In: Proceedings of

Oceans ‘83 (San Francisco). IEEE & MTS. pages 199-202.

A Transmissometer for Profiling and Moored Observations in

Water. Bartz, R., J.R.V. Zaneveld, and H. Pak. 1978. SPIE Volume 160

Ocean Optics V. Bellingham, Washington. pages 102-108.

Understanding Nephelometric Instrumentation. Vanous, R.D. 1978.

American Laboratory. Reprinted by Hach Company, Loveland,

Colorado. 8 pages.

Understanding Turbidity Measurement. Hach, C.C., R.D. Vanous, and J.M. Heer. 1985. Technical Information Series-Booklet No. 11.

Hach Company, Loveland Colorado, 11 pages.

Turbidity Standards. Hach, C.C. 1985. Technical Information Series-

Booklet No. 12. Hach Company. Loveland, Colorado. 9 pages.

Optical Properties of Turbidity Standards. Zaneveld, J.R.V.,

R.W. Spinrad, and R. Bartz. 1980. SPIE Volume 208 Ocean Optics VI.

Bellingham, Washington. pages 159-168.

The Effect of Particle Size on the Light Attenuation Coefficient of

Natural Suspensions. Baker, E.T. and J.W. Lavelle. 1984. Journal of

Geophysical Research. Volume 89. pages 9197-9203.

Laboratory Apparatus for Calibrating Optical Suspended Solids

Sensors. Downing, J.P. 1989. Marine Geology. Volume 89.

pages 1-7.

Ocean Optics Principles. 1984. SPIE Volume 489 Ocean Optics VII.

Bellingham, Washington. pages l-54.

OD

& A Instrument Co.

41

Rev. 3/91

g

GENERAL PURPOSE SENSORS

STRAIGHT METAL TUBE BODY WITH EXTENSION LEADS

TYPE T1000:

Features

• Rugged, fast response sub-assemblies for a variety of uses.

• Small bead or chip thermistor types <0.050” OD.

• Small diameter metal tubing, open end.

• Magnet wires or small ribbon cables.

• Maximum temp. ratings from 60°C up to 260°C.

Options

• Stainless Steel hypodermic tubing is standard.

• Other small diameter metal tubing is available.

• Fully encapsulated (Fig. 1) or partially exposed (Fig. 2).

• Diameter “D” : from 0.022” to 0.125”.

• Tubing length “B” : from 0.375” to 1.50”.

• Lead length “X” : 12 inches (typical).

• Lead-wire gauge sizes from #32 to #40 AWG.

TYPE T1100:

Features

• Rugged, fast response sub-assemblies for limited immersion in liquids or gases.

• Miniature glass probe thermistor types <0.040” OD.

• Small diameter metal tubing, open end.

• Magnet wires or small ribbon cables.

• Maximum temp. ratings from 60°C up to 260°C.

Options

• Stainless Steel hypodermic tubing is standard.

• Other small diameter metal tubing is available.

• Fully encapsulated (Fig. 1) or partially exposed (Fig. 2).

• Exposed probe tip dimension “A” : from 0.020” to 0.060”.

• Diameter “D” : from 0.032” to 0.125”.

• Tubing length “B” : from 0.375” to 1.50”.

• Lead length “X” : 12 inches (typical).* Lead-wire gauge sizes from #32 to #40 AWG.

TYPE T1200:

Features

• Rugged, fast response sub-assemblies with maximum protection for use in harsh environments.

• Small to medium size thermistor types <0.100” OD.

• Small diameter metal tubing, closed end.

• Magnet wires or small ribbon cables.

• Maximum temp. ratings from 60°C up to 200°C.

Options

• Stainless Steel hypodermic tubing is standard.

• Other small diameter metal tubing is available.

• Spherical point (Fig. 1) for all OD.

• Flat, plug welded point (Fig. 2) for OD > 0.050”.

• Tapered point (Fig. 3) for OD > 0.093”.

• Diameter “D” : from 0.022” to 0.125”.

• Tubing length “B” : from 0.375” to 3.00”.

• Lead length “X” : 12 inches (typical).

• Lead-wire gauge sizes from #18 to #32 AWG.

Crown Industrial Estate, Priorswood Road

Taunton, Somerset TA2 8QY UK

Tel +44 (0) 1823 335200

Fax +44 (0) 1823 332637

808 US Highway 1

Edison, New Jersey 08817-4695 USA

Tel +1 (732) 287 2870

Fax +1 (732) 287 8847

967 Windfall Road

St. Marys, Pennsylvania 15857-3397 USA

Tel +1 (814) 834 9140

Fax +1 (814) 781 7969

g

GENERAL PURPOSE SENSORS

STRAIGHT METAL TUBE BODY WITH EXTENSION LEADS

TYPE T1300:

Features

• Rugged, medium to large sized, multipurpose sub-assemblies.

• Medium to large bead, chip or disk types <0.100” OD.

• Medium to large diameter metal tubing, open end.

• Hook-up wire extension leads.

• Maximum temp. ratings from 60°C up to 260°C.

Options

• Stainless Steel tubing is standard.

• Other metal tubing is available.

• Fully encapsulated (Fig. 1) or partially exposed (Fig. 2).

• Diameter “D” : from 0.072” to 0.250”.

• Tubing length “B” : from 0.500” to 2.00”.

• Lead length “X” : 48 inches (typical).

• Lead-wire gauge sizes from #18 to #32 AWG.

TYPE T1400:

Features

• Rugged, medium to large sized thermistor sub-assemblies for limited immersion in liquids or gases.

• Medium to large glass probe types <0.100” OD.

• Medium to large diameter metal tubing, open end.

• Hook-up wire extension leads.

• Maximum temp. ratings from 60°C up to 300°C.

Options

• Stainless steel tubing is typical.

• Other metal tubing is available.

• Fully encapsulated (Fig. 1) or partially exposed (Fig. 2).

• Exposed probe tip dimension “A” : from 0.030” to 0.250”.

• Diameter “D” : from 0.083” to 0.250”.

• Tubing length “B” : from 0.500” to 3.00”.

• Lead length “X” : 48 inches (typical).

• Lead-wire gauge sizes from #18 to #32 AWG.

TYPE T1500:

Features

• Rugged, medium to large size thermistor sensors in a closed end metal tubing for maximum protection.

• All thermistor types <0.200” OD.

• Hook-up wire extension leads.

• Maximum temp. ratings from 60°C up to 300°C.

Options

• Stainless Steel hypodermic tubing is typical.

• Other metal tubing is available.

• Spherical point (Fig. 1) for all OD.

• Flat, plug welded point (Fig. 2) for all OD > 0.050”.

• Tapered point (Fig. 3) for OD > 0.093”.

• Diameter “D” : from 0.072” to 0.250”.

• Tubing length “B” : from 0.375” to 6.00”.

• Lead length “X” : 48 inches (typical).

• Lead-wire gauge sizes from #18 to #32 AWG.

Crown Industrial Estate, Priorswood Road

Taunton, Somerset TA2 8QY UK

Tel +44 (0) 1823 335200

Fax +44 (0) 1823 332637

808 US Highway 1

Edison, New Jersey 08817-4695 USA

Tel +1 (732) 287 2870

Fax +1 (732) 287 8847

967 Windfall Road

St. Marys, Pennsylvania 15857-3397 USA

Tel +1 (814) 834 9140

Fax +1 (814) 781 7969

g

GENERAL PURPOSE SENSORS

STRAIGHT METAL TUBE BODY WITH EXTENSION LEADS

TYPE T1600:

Features

• Rugged, large sized thermistor sub-assemblies for limited immersion in liquids or gases.

• All thermistor types <0.200” OD.

• Large diameter metal tubing, open end.

• Shielded / Jacketed Cable extension leads.

• Maximum temp. ratings from 60°C up to 200°C.

Options

• Stainless steel tubing is standard.

• Other metal tubing is available.

• Fully encapsulated (Fig. 1) or partially exposed (Fig. 2).

• Exposed probe tip dimension “A” : from 0.030” to 0.250”.

• Diameter “D” : from 0.125” to 0.250”.

• Tubing length “B” : from 0.750” to 6.00”.

• Cable length “X” : 48 inches (typical).

• Cable construction: RG; Coaxial; 2, 3 or 4 conductor.

• Shielded, jacketed or shielded and jacketed cable.

• Conductor gauge sizes from #20 to #32 AWG.

TYPE T1700:

Features

• Rugged, large sized thermistor sensors in a closed end metal tubing for maximum protection.

• All thermistor types <0.200” OD.

• Large diameter metal tubing, closed end.

• Shielded / Jacketed Cable extension leads.

• Maximum temp. ratings from 60°C up to 200°C.

Options

• Stainless Steel tubing is standard.

• Other metal tubing is available.

• Spherical point (Fig. 1).

• Flat, plug welded point (Fig. 2).

• Tapered point (Fig. 3).

• Diameter “D” : from 0.125” to 0.250”.

• Tubing length “B” : from 0.750” to 6.00”.

• Cable length “X” : 48 inches (typical).

• Cable construction: RG; Coaxial; 2,3 or 4 conductor.

• Shielded, jacketed or shielded and jacketed cable.

• Conductor gauge sizes from #20 to #32 AWG.

Crown Industrial Estate, Priorswood Road

Taunton, Somerset TA2 8QY UK

Tel +44 (0) 1823 335200

Fax +44 (0) 1823 332637

808 US Highway 1

Edison, New Jersey 08817-4695 USA

Tel +1 (732) 287 2870

Fax +1 (732) 287 8847

967 Windfall Road

St. Marys, Pennsylvania 15857-3397 USA

Tel +1 (814) 834 9140

Fax +1 (814) 781 7969

LbCOR Radiation Sensors

Instruction Manual

Y

LbCOR Underwater Radiation

Sensors, Type SA

Instruction Manual

Publication No. 8609-57

November, 1986

Revised December, 1990

LI-COR, inc.

4421 Superior Street

P.O. Box 4425

Lincoln, NE 68504 USA

Telephone: (402) 467-3576

TWX: 910-621-8116

FAX: 402-467-2819

0 Copyright 1986, LI-COR, Lincoln, Nebraska USA

How to Use this Manual

This manual contains the operation and maintenance information for ti

LI-COR underwater, type SA sensors.

The first section of the manual contains general information which relates to all LI-COR underwater sensors (i.e. operation, recalibration, etc).

After the general information you will find specific information

about

each sensor.

When reading through the manual you should first read the general information and then read the specific information for your sensor (i.e. the

LI-192SA Underwater Quantum or LI-193SA Spherical Quantum Sensor).

NOTICE

The information contained in this document is subject to change without notice.

LI-COR MAKES NO WARRANTY OF ANY KIND WITH REGARD TO

THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO THE

IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS

FOR A PARTICULAR PURPOSE. LI-COR shall not bc liable for errors contained herein or for incidental or consequential damages in connection with the furnishing, performance, or use of this material.

This document contains proprietary information which is protected by copyright. All rights are reserved. No part of this document may be photocopied, reproduced, or translated to another language without prior written consent of LI-COR, Inc.

0 Copyright 1986, LI-COR Inc.

Table of Contents y

Section 1. General Information

Type “SA” sensors ......................................................................

1

Sensor Recalibration ....................................................................

1

Operation ...................................................................................

2

Calibration .................................................................................

4

Cleaning Information ...................................................................

2009s Lowering Frame.. ..............................................................

Section 2.

LI-192SA Underwater Quantum Sensor

Use of the Underwater Quantum Sensor ...........................................

1 0

Immersion Effect .........................................................................

1 0

Cosine Response .........................................................................

11

Cosine Correction Properties .........................................................

1 2

Spectral Response .......................................................................

1 2

Specifications .............................................................................

1 3

Section 3. LI-193SA Spherical Quantum Sensor

Use of the Spherical Quantum Sensor.. ...........................................

1 4

Immersion Effect .........................................................................

1 4

Angular Response .......................................................................

1 5

Azimuth Response ......................................................................

1 5

Spectral Response .......................................................................

1 5

Terminology

..............................................................................

1 5

Errors ........................................................................................

1 6

MathematicalDefinitions

..............................................................

1 7

Bibliography

..............................................................................

20

Specifications..

...........................................................................

2 1

Appendix

ii

;

TYPE ‘ISA” SENSORS

LI-COR SA type sensors are characterized by having the underwater cable terminated with a BNC connector. Figure 1 shows a typical SA sensor.

“SA” type underwater sensors include the LI-192SA Underwater Quantum

Sensor, and the LI-193SA Spherical Quantum Sensor. The type SA terrestrial sensors include the LI-190SA Quantum Sensor, the LI-191SA

Line Quantum Sensor, the LI-200SA Pyranometer Sensor, and the

LI-210SA Photometric sensor.

Figure 1. “SA” type sensors are terminated with only a

BNC connector on the end of the cable.

SENSOR RECALIBRATION

Recalibration of LI-COR radiation sensors is recommended every two years.

The LI-192SA Underwater

Quantum

Sensor may be returned to LI-COR for recalibration or recalibrated using the LI-COR 1800-02 Optical Radiation

Calibrator. The LI-193SA Spherical Quantum Sensor must be returned to

LI-COR for recalibration.

1

k

OPERATION

The 2222UWB Underwater Cables used with LI-COR Underwater Sensors are terminated with a BNC connector. This connector allows the sensors to be used with the LI-189 Quantum/Radiometer/Photometer, the two current channels of the LI-1000 Datalogger, or with older LI-COR integrators, including the LI-5lOB and the LI-550B.

To use a type SA sensor with the LI-189 Quantum/Radiometer/Photometer, the calibration multiplier is entered by using the two calibrate keys and dialing in the calibration multiplier using the calibration screw (see LI-189 manual). The calibration multiplier is given on the certificate of calibration and on the sensor calibration tag.

When using the LI-1000 DataLogger, the calibration multiplier must be entered into the LI-1000 software as described in the LI-1000 Instruction

Manual.

To use a sensor with LI-COR light meters such as the LI-185B, LI-188B or

LI-1776, a factory installed calibration connector is required. Other LI-COR light meters and integrators including the LI-170, LI-185, LI-185A, LI-188,

LI-510, and LI-550 require the use of the 9901-014 connector conversion cable. Contact LI-COR for further details.

When a LI-COR light meter or data logger is not used, your sensor can be used with other millivolt recorders or data loggers by connecting a millivolt adapter. Table 1 lists the millivolt adapters required for each underwater sensor and the resistance of each adapter.

Sensor

LI-192SA

LI-193SA

Table 1. Millivolt adapters for “SA” type sensors.

Millivolt

Adapter

2291s

22915:

Resistance

1210 Ohm

1210 Ohm

The millivolt adapter connects to the BNC connector of the underwater cable, and the wire leads of the adapter are connected to the data logger.

Sensor output (in millivolts) when using the millivolt adapter can bc computed using “Ohms law” (Voltage = Current x Resistance).

2

I i

I

Example:

s

Calculate the millivolt output of an LI-192SA Quantum

Sensor which has a calibration constant of 3.0 pA / 1000 pmol s-l me2 (in water). Assume the 2291s millivolt adapter is used with the sensor.

3.0 pA

1000 l.tmol sm1rnm2

1A x lo6 lrA x 1210 Ohm =

0.00363 volts

1000 pm01 se1rne2

= 3.63 mV / 1000 ln-r~ol s-l m-2

IMPORTANT:

When using the sensor under water, the “in water” calibration constant should be used to calculate the millivolt output of the sensor. The “in water” sensor calibration includes an immersion effect correction.

The use of the millivolt adapter with a recorder or data logger other than

LI-COR instruments is often acceptable for radiation levels down to 10% of full sunlight. Below lo%, the recorder must be very sensitive to pick up the small voltage signal. The recorder should have a high impedance input

(>l megohm, such as potentiometric types), and the range adjustment should be O-10 mV, or a more sensitive range. For low light levels, the

Sensor should be connected directly to a LI-COR readout device (without using the millivolt adapter).

In LI-COR underwater cables the white wire is positive and the black wire is negative. The center pin of the BNC connector has a negative signal. This is done because the trans-impedance amplifier used in LI-COR light meters requires a negative signal.

For data logger or millivolt applications where the millivolt adapter is needed, the positive (green) lead should be connected to the low impedance

(common terminal) when plus or minus signal capability is available on the data logger or recorder. This will minimize noise.

If plus or minus capability is not available on the data logger or recorder, the green lead should be connected to the positive input and the blue lead to the negative input. If noise difficulties are encountered, consult LI-COR for special wiring instructions.

It is recommended that the LI-COR 2009s Lowering Frame (or equivalent) be used with the sensor for underwater applications.

IMPORTANT:

Do not use LI-COR 2222UWB Underwater Cable to support the sensor and lowering frame, as damage to the cable can result.

An auxiliary cable should be used to support the lowering frame and sensor.

In addition, the 2222UWB cable should not be bent sharply near the sensor.

3

u

NOTE:

For cable lengths over 75 m (225 ft.), care should be exercised in its use since movement of the cable within the water can cause excessive signal noise.

CALIBRATION

LI-COR quantum sensors are calibrated using a standard radiation source which has been calibrated against a National Bureau of Standards lamp. The photon flux density from the standardized lamp is known in terms of micromoles s-t m-*, where one micromole = 6.023 x 1017 photons. The uncertainty of the calibration is f 5%.

The lamp used in LI-COR’s calibration is a high intensity standard of spectral irradiance (G.E. 1000 Watt type DXW quartz halogen) supplied with a spectral irradiance table.

The following procedure was used to calculate the quantum flux output from the lamp. The lamp flux density (AE) in watts me*, in an increment at wavelength L\3L can be expressed as

AE = E(h)& where E(h) is the spectral irradiance of the lamp at wavelength h.

The number of photons s-lrn-* in & is

Photons s-‘rn-’ = where h is Planks constant and c is the velocity of light. This can be summed over the interval 400-700 nanometers (nm) to give

L

The result is adjusted to micromoles s-*m-* by dividing by 6.023 x 1017.

4

CI

Y

CLEANING INFORMATION

DO NOT

use alcohol, organic solvents, abrasives, or strong detergents to clean the diffusor element on LI-COR light sensors.

The acrylic material used in LI-COR light sensors can be crazed by exposure to alcohol or organic solvents, which will adversely affect the cosine response of the sensor.

Clean the sensor only with water and/or a mild detergent such as dishwashing soap. LI-COR has found that vinegar can also be used to remove hard water deposits from the diffusor element, if necessary.

2009s &ERING F R A M E

The 2009s Lowering Frame provides for the placement of two cosine sensors, one each for upwelling and downwelling radiation, or a single underwater spherical sensor (Figure 2). Each LI-COR underwater sensor has three 6-32 tapped mounting holes on the underside of the sensor for connection to the mounting ring. Corrosion resistant mounting screws are included with each sensor.

Suspension Ring

Q

Shaft

Downwelling or

Spherical Sensor

Mounting Ring

L

Fin

0 m

Weight Ring

Upwelling Sensor

Mounting Ring

I

Figure 2. 2009s Lowering Frame.

When two sensors are used, the frame is well balanced and will work well in mild currents without twisting the cables. The sensor for downwelling radiation is always attached using the mounting ring on the fin. Likewise, the sensor for upwelling radiation is attached to the opposite mounting ring.

Depending on the speed of the current the frame will tilt a few degrees, but this can be minimized by hanging a compact weight from the weight ring.

Moderate weights will often suffice (4 kg). Weights over 25 kg should be avoided

6

II

The use of a single cosine sensor will require a small weigr(0.2 kg) attached at the empty mounting ring or a moderate weight from the weight ring, or possibly both, depending upon the speed of the current.

The underwater cable(s) should be attached to the frame such that approximately 25 cm of cable forms a smooth arc to the underwater sensor connector and is restrained from being flexed or supporting any weight.

LI-COR underwater cable is not recommended as a support cable, although it can be used as a lowering cable providing it is properly attached and the attached weights do not exceed 5 kg. The cable(s) must be attached as described above. Additionally, the cable must be securely attached to the shaft of the lowering frame at multiple points so that the cable does not slip and put strain on the sensor connector.

However, the cable cannot be clamped so tightly as to damage it. Possible methods to use are numerous nylon cable clamps along the length of the shaft, or a tight wrap of light cord around the shaft and cables, starting at the suspension ring and extending downward at least 25 cm.

Underwater Cable for Sensor

Tight,

Non-slip

Wrap of Cord

Cosine Sensor

Cosine Sensor

Oceanographic Cable

1

Supporting the Frame

Underwater for Sensor

Cable

LI-193SB Spherical

Quantum Sensor

Moderate Weight (dense)

For long-term immersion or use in heavily ionic water, it may be necessary to provide electrical insulation between the underwater sensor(s) and the lowering frame to prevent galvanic corrosion.

This is accomplished by slipping an insulating flat washer over the mounting screws down to the heads, followed by a l/2” (13 mm) length of thin tubing over the screw threads. This tubing insulates the screws from the mounting ring.

Next, place a large flat insulating washer between the sensor and the mounting ring (with three holes for the screws). Use the “insulated” screws to attach the sensor in place. In this way neither the screws nor sensor have electrical contact with the frame.

8

t

"Insulated Screw" (one of three)

I

Large Insulating Washer

Cosine Sensor

I

I

9

(;,.,,,,,I

USE OF THE UNDERWATER QUANTUM SENSOR

The LX-192SA Underwater Quantum Sensor is used for measuring

Photosynthetically Active Radiation (PAR) in aquatic environments. With its 400-700 nanometer (nm) quantum response it is a valuable tool for researching primary productivity or other projects of environmental concern.

The sensor can be used in the air with accuracy similar to that of the

LI-190SA Quantum Sensor. Prior to obtaining atmospheric readings, the sensor

must

be dried.

The sensor connector should be lubricated with a silicone grease before installing it in the mating connector of the underwater cable. The yellow dots on the connector and the underwater cable should be aligned before pushing them together in order to obtain the proper pin connection. If the dots are not aligned this can result in a negative reading on the readout device due to the change in polarity of the conductors. The connector pins are small and care should be taken when mating the connectors.

The quantum sensor has three 6-32 tapped holes on the underside of the sensor which are used for mounting the sensor to the 2009s Lowering

Frame.

To maintain appropriate cosine correction the vertical edge of the diffuser must be kept clean. Periodically inspect the sensor for foreign deposits on the upper surfaces during prolonged submerged operation. See page 5 of this manual for detailed cleaning instructions.

IMMERSION EFFECT

A sensor with a diffuser for cosine correction will have an immersion effect when immersed in water. The radiation entering the diffuser scatters in all directions within the diffuser with more of the radiation lost through the water-diffuser interface than in the case where the sensor is in air. This results

because the

air-diffuser interface offers a greater ratio of the indexes of refraction than the water-diffuser interface. Thus, a greater percentage of radiation entering the diffuser in air reaches the photodiode than in the case where the LI-192SA is in water. Therefore, a normal underwater reading would need to be multiplied by this effect if the sensor is used in water.

1 0

v

The LI-192SA calibration certificate contains calibration mxpliers for both in air and in water operation. The in water multiplier includes the immersion effect correction.

COSINE RESPONSE

Measurements intended to approximate radiation impinging upon a flat surface (not necessarily level) from all angles of a hemisphere are most accurately obtained with a cosine corrected sensor.

A sensor with a cosine response (follows Lambert’s cosine law) allows measurement of flux densities through a plane surface. This allows the sensor to measure flux densities per unit area (m2). A sensor without an accurate cosine correction can give a severe error under diffuse radiation conditions within a plant canopy, at low solar elevation angles, under fluorescent lighting, etc.

The cosine relationship can be thought of in terms of radiant flux lines striking a flat surface. Lambert’s Cosine Law is explained by illustrating radiant flux lines impinging upon a surface normal to the source (Figure

3A) and at an angle of 60” from normal (Figure 3B). Figure 3A shows 6 rays striking the unit area, but at a 60” angle only 3 rays strike at the same unit area. This is illustrated mathematically as

S = (I) (cosine 60”) per unit area

3 = (6) (0.5) per unit area where S = vertical component of solar radiation; I = solar radiation impinging perpendicular to a surface and cosine 60” = 0.5.

Figure 3. Lambert’s Cosine Law.

11

COSINbORRECl+ION PROPERTIES

A comparison of the sensor’s cosine response curve in air and in water can be found in the “Immersion Effect of LI-COR Underwater Sensors” Report

(available from LI-COR). Engineering requirements result in different correction characteristics for air and water. Overcompensation in air and undercompensation occurs in water. The better response was selected for air because in water the direct incident solar radiation does not exceed the critical angle of 48.6” (a result of the air-water interface).

SPECTRAL RESPONSE

The spectral response is similar to that of the LI-190SA Quantum Sensor

(Figure 4).

\

Figure 4. Spectral Response of the LI-190SA Quantum Sensor.

1 2

Y c”

.

*

The spectral response of the quantum sensor is obtained by use of a light source and a monochromator. A thermopile which has a known spectral response over the spectral range of interest

is

used to determine the monochromator output in energy flux density, W(h). at the wavelength setting h. If Q(h) is the sensor output at wavelength h when exposed to the monochromator output, W(h), then Q(h) can be approximated by

Q(h)

= R(h) WV where R(h) is the sensor spectral response at the wavelength setting 1. The above approximation assumes that the monochromator bandwidth,

Ah, is much less than the wavelength setting 1. The normalized sensor spectral response r(h), is determined by r(h) = R(X)/Rm where Rm is the maximum value of Q&)/W(h) over the range of wavelengths measured.

SPECIFICATIONS

Absolute Calibration:

+ 5% in air traceable to NBS.

Sensitivity:

Typically 3 pA per 1000 l.trnol s-l me2 in water.

Linearity:

Maximum deviation of 1% up to 10,000 pmol s-l rnw2.

Stability:

< * 2% change over a 1 year period.

Response Time:

10 pS.

Temperature Dependence:

+ 0.15% per “C maximum.

Cosine Correction:

Optimized for both underwater and atmospheric use.

Azimuth:

< f 1% error over 360” at 45” elevation.

Detector:

High stability silicon photovoltaic detector (blue enhanced).

Sensor Housing:

Corrosion resistant metal with acrylic diffuser for both saltwater and freshwater applications. Waterproof to withstand 800 psi (54 bars).

Size: 3.18 Dia. x 4.62 cm H (1.25” x 1.81”).

Weight: 227 g (0.50

lb.).

Mounting:

Three 6-32 holes are tapped into the base for use with the

2009s Lowering Frame or other mounting devices.

Cable:

Requires 2222UWB Underwater Cable.

1 3

USE OF THE SPHERICAL QUANTUM SENSOR

The LI-193SA Spherical Quantum Sensor is used for measuring Photosynthetically Active Radiation (PAR) in aquatic environments, and specifically the Photosynthetic Photon Flux Fluence Rate (PPFFR). The

LI-193SA gives an added dimension to underwater PAR measurements in that it measures PAR from all directions. The LI-193SA Sensor can also be used in air.

Because PPFFR can be defined as those photons having a wavelength between 400 and 700 nm that are incident per unit time on the surface of a sphere divided by the cross-sectional area of the sphere, the LI-193SA

Spherical Quantum Sensor is designed to respond equally to photons between 400 and 700 nm. Because the energy of a photon is inversely proportional to its wavelength, a sensor which responds equally to photons will have a linear energy response with wavelength. Therefore, an ideal

PPFFR sensor would have a linear energy response between 400 and 700 nm, and would have a slope of 1% per 7 nanometers (nm) if it were normalized to 100% at 700 nm.

IMMERSION EFFECT

Because of the difference of index of refraction between air and water, the calibration constant of the LI-193SA when used in water will be different than the calibration constant when used in air. This phenomenon is known as the immersion effect. The air/water ratio of the calibration constants is equal to the sensor output in air divided by the sensor output in water for the same PPFFR. This ratio is greater than one, and the approximate air/water ratio for normal incident radiation (0’) can be calculated by dividing the “in water” cal constant (listed on the calibration certificate) by the “in air” calconstant. The ratio for an 180” incident radiation is about 5% higher than the ratio for 0” incident radiation.

For an explanation of the immersion effect as well as methods that can bc used to determine it, a report entitled “Immersion Effect and Cosine

Collecting Properties of LI-COR Underwater Sensors” is available from

LI-COR.

14

. .

)I w

ANGULAR RESPONSE

The LI-193SA sensor uses an acrylic diffuser to obtain an angular response error of less than f 4% for angles of incidence up to 90” from the normal.

Testing is done with a collimated beam of radiation to verify these limits.

AZIMUTH RESPONSE

With a collimated beam of radiation at an angle of incidence of 90” from normal, the sensor is rotated about its normal axis. The maximum acceptable variation in response under these conditions is f 3%.

SPECTRAL RESPONSE

The spectral response is comparable to that of the LI-190SA Quantum

Sensor, (Figure 4), as both use computer-tailored filter glasses to closely approximate the ideal linear energy response (flat photon response) from 400 to 700 nm. This response ideally produces an equal output for equal

PPFFR even if the spectral irradiance varies within the cutoff points of 400 and 700 nm.

Measurement of the spectral response requires a stabilized light source, monochromator, and calibrated reference detector. Measurements taken with the test sensor and reference detector at many wavelengths yield data points used to plot a relative spectral response. For details, the “Description of

Calibration Procedures” applications note is available from LI-COR.

TERMINOLOGY

The terminology associated with radiation measurements has not been consistent. The following terms have been used to describe the same physical quantity:

1) Photon Flux Density

2) Photon Irradiance

3) Quantum h-radiance

The physical quantity described is the number of photons incident on an element of a surface in an element of time. This physical quantity is measured by a cosine-corrected quantum sensor such as the LI-COR

LI-192SA Underwater Quantum Sensor. When the wavelength range of the photons measured is limited to the 400-700 nm range, the term

Photosynthetic Photon Flux Density (PPFD) has been used.

1 5

The follozg terms have been used to describe the integral of the Photon

Flux Radiance at a point over all directions

about the point:

1) Photon Flux Fluence Rate

2) Scalar Photon hradiance

3) scalar Quantum Irradiance

4) spherical Photon Irradiance

5) Spherical Quantum b-radiance

6) Photon Flosan

This physical quantity can also be described as the number of photons incident on the outer surface of a spherical element of volume in an element of time divided by the cross-sectional area of the sphere. When the wavelength range of photons measured is limited to the 400-700 nm range, the term Photosynthetic Photon Flux Fluence Rate (PPFFR) can be used.

The LI-193SA Spherical Quantum Sensor measures PPFFR.

An ideal PPFFR sensor placed in a uniform radiance distribution (perfectly diffuse radiation) would indicate a PPFFR that is four times higher than the

PPFD measured by an ideal cosine-corrected sensor also placed in such a uniform radiance distribution.

An ideal PPFFR sensor placed in a uniform radiance distribution (perfectly diffuse radiation) would indicate a PPFFR that is four times higher than the

PPFD measured by a spherical collecting surface which exhibits the properties of a cosine-corrected collector at every point of its surface when such a sensor is also placed in a uniform radiance distribution.

An ideal PPFFR sensor placed in a spatially uniform collimated beam of radiation will indicate a PPFFR that is the same as thePPFD measured by an ideal cosine-cone&d sensor.

ERRORS

The spatial error of the LI-193SA Sensor is due to variations in the diffusing sphere, (negligible), and the sphere area “lost” because of the sensor base. This error is less than -10% for totally diffuse radiation, but is usually smaller than this because the upwelling radiation is smaller than the downwelling radiation.

In highly turbid waters the sensor will indicate high quanta

values

due to the displacement of water by the sensor sphere volume. This is because the point of measurement is taken to be at the center of the sphere, but the attenuation which would have been provided by the water within the sphere is absent. This error is typically +6.3% for water with an attenuation coefficient of 3 m-l.

1 6

L w

.w

MATHEMATICAL DEFINITIONS

The mathematical definition of photon flux fluence rate (PFFR) is

PFFR= LdQ

I

4R where L is the photon flux radiance and CJ is the solid angle. Since d = [email protected] d6 d$, this can be rewritten as

The mathematical definition of photon flux density (PFD) as measured by a cosine-corrected sensor is

PFD =

L( +, Q) [email protected] [email protected] d#

If 0’ = 20, then [email protected] = 1/2sinO’ and [email protected] = l/2 dQ’. Also, the limits of

6’ are 0 to 7r. Then

PFD=1/4 L( qb, 63’) [email protected]’ dO’ dq?r

In a uniform radiance distribution, L($,O) = L($,O’) = L (a constant). Then

PFFR = 4xL

PFD=?rL

or PFFR = (4)(PFD)

A small spherical collecting surface which exhibits the properties of a cosine collector at every point of its surface would measure the limit of the ratio of total photon flux onto a spherical surface to the area of the surface, as the radius of the sphere tends toward zero. Mathematically, the “photon flux I(@,$) per unit solid angle” in the direction (O,$) that is intercepted by a spherical surface using the cosine law is where w is the angle between the normal of dA and the direction (O,Q).

Now,

dA = nr2 where r = radius of the hemisphere (hemi).

Therefore, the total photon flux (F) intercepted by the sphere is

I7

This spherical collecting surface would then measure that is, PFFR = 4 times the associated quantity measured by a small spherical collecting surface which exhibits the properties of a cosine collector at every point of its surface in a uniform radiance distribution.

From this fact and also the fact that the cross-sectional area of a sphere = l/4 the surface area, one could define PFFR as the limit of the ratio of total photon flux onto a spherical surface to the cross-sectional area.

In a uniform collimated beam of radiation, the following conditions of photon flux radiance hold

L(@,O<O*) = L (a constant)

L(@,o>o*) = 0 whcrc O* is small such that [email protected] E 0 and co& z 1.

Then

PFFR = h

18

bd

Also,

Therefore, PFFR z PFD

Photon flux fluence rate = photon flux density in a uniform collimated beam if the beam is normal to the cosine collector. One might also note that if the beam is perfectly collimated (O*=O), then the radiance L must be infinite in order for flux to be transmitted.

One could use an alternate approach. The total flux (does not need to be collimated) impinging onto a sphere is

F = rrr2 PFFR where r is the radius of the sphere, and PFFR is the photon flux fluence rate. If the flux is collimated and covers the entire sphere, then the flux density of the beam would be F/rrr2, where xr2 is the cross-sectional area of that portion of the beam that is intercepted by the sphere, and F is the total flux in the beam that is intercepted by the sphere. If the beam is uniform, then the flux density is F/rrr2 every-where in the beam. If a cosine-corrected collector is put into the beam, it will measure the flux density times the cosine of the angle between the beam and the normal of the collector. If that angle is zero, then the cosine-collector will measure the flux density

(F/rrr2) even if its cross sectional area is less than rrr2. Therefore, the cosine-collector will measure the photon flux density to be equal to the photon flux fluence rate measured by the sphere in a uniform collimated beam of radiation.

1 9

Combs, W.S., Jr. 1977. The measurement and prediction of irradiance available for photosynthesis by phytoplankton in lakes. University of

Minnesota Ph.D. Thesis, Limnology.

Incoll, L.D., S.P. Long and M.R. Ashmore. 1977. SI units in publications in plant science. Commentaries in Plant Science (No. 28). Published in:

Current Adv. Plant Science 9: 33 l-343.

Jerlov, N.G. 1968. Optical Gceanography.Elsevier.

McCree, K.J. 1979. Radiation.

NBS Technical note 910-1, 1976. Self-study manual on optical radiation measurements.

Shibles, R. 1976.

Committee Report: Terminology pertaining to photosynthesis. Crop Sci. 16: 437439.

Tyler, J.E., and R.W. Preisendorfer. 1962. Light in the sea, p. 399-400.

In M.N. Hill (ed.), The Sea, V.I. Interscience.

,

5

20

SPECIFICATIONS

Absolute Calibration: k 5% in air traceableto NBS.

Sensitivity: Typically 3 pA per 1000 t.tmol s-l mm2 in water.

Linearity: Maximum deviation of 1% up to 10,000 mmol s-l mm2.

Stability: < 912% change over a 1 year period.

Response Time: 10 p.s.

Temperature Dependence: f 0.15% per “C maximum.

Cosine Correction: Acrylic diffuser.

Angular Response: < + 4% error up to f 90” from normal axis.

Azimuth: < f 3% error over 360” at 90” from normal axis.

Detector: High stability silicon photovoltaic detector (blue enhanced).

Sensor Housing: Corrosion resistant metal for both saltwater and freshwater applications with an injection molded, impact resistant, acrylic diffuser. Units have been tested to 500 psi (34 bars) with no failures.

Size

Globe: 6.1 cm Dia. (2.4”).

Housing: 3.18 cm Dia. (1.25”).

Overall Height: 10.7 cm (4.2”).

Weight: 142 g (0.31 lb.).

Mounting: Three 6-32 mounting holes are tapped into the base for use with the 2009s Lowering Frame or other mounting devices.

Cable: Requires 2222UWR Underwater Cable.

21

War&ty

Each LI-COR, inc. instrument is warranted by LI-COR, inc. to be free from defects in material and workmanship; however, LI-COR, inc.% sole obligation under this warranty shall be to repair or replace any part of the instrument which LI-COR, inc.‘s examination discloses to have been defective in material or workmanship without charge and only under the following conditions, which are:

1 .

2 .

3 .

4 .

5 .

6 .

The defects are called to the attention of LI-COR, inc. in Lincoln,

Nebraska, in writing within one year after the shipping date of the instrument.

The instrument has not been maintained, repaired or altered by anyone who was not approved by LI-COR, inc.

The instrument was used in the normal, proper and ordinary manner and has not been abused, altered, misused, neglected, involved in an accident or damaged by act of God or other casualty.

The purchaser, whether it is a DISTRIBUTOR or direct customer of

LI-COR or a DISTRIBUTOR’S customer, packs and ships or delivers the instrument to LI-COR. inc.at LI-COR inc.‘s factory in Lincoln,

Nebraska, U.S.A. within 30 days after LI-COR, inc. has received written notice of the defect. Unless other arrangements have been made in writing, transportation to LI-COR, inc. (by air unless otherwise authorized by LI-COR, inc.) is at customer expense.

No-charge repair parts may be sent at LI-COR, inc.% sole discretion to the purchaser for installation by purchaser.

LI-COR, inc.% liability is limited to repair or replace any part of the instrument without charge if LI-COR, inc.‘s examination disclosed that part to have been defective in material or workmanship.

There are no warranties, express or implied, including but not limited to any implied warranty of merchantability of fitness for a particular purpose on mderwater cam or on expendables such as battertes, lamDs, thermocouDles, and

.

Other than the obligation of LI-COR, inc. expressly set forth herein, LI-COR, inc.

disclaims all warranties of merchantability or fitness for a particular purpose.

The foregoing constitutes LI-COR, inc.‘s sole obligation and liability with respect to damages resulting from the use or performance of the instrument and in no event shall LI-COR, inc. or its representatives be liable for damages beyond the price paid for the instrument, or for direct, incidental or consequential damages.

.

L

2 2

I

The laws of some locations may not allow the exclusion or limitation on implied warranties or on incidental or consequential damaged, so the limitations herein may not apply directly. This warranty gives you specific legal rights, and you may already have other rights which vary from state to state. All warranties that apply, whether included by this contract or by law, are limited to the time period of this warranty which is a twelve-month period commencing from the date the instrument is shipped to a user who is a customer or eighteen months from the date of shipment to LI-COR, inc.‘s authorized distributor, whichever is earlier.

This warranty supersedes all warranties for products purchased prior to June

1.1984. unless this warranty is later superseded.

DISTRIBUTOR or the DISTRIBUTOR’s customers may ship the instruments directly to LI-COR if they are unable to repair the instrument themselves even though the DISTRIBUTOR has been approved for making such repairs and has agreed with the customer to make such repairs as covered by this limited warranty.

Further information concerning this warranty may be obtained by writing or telephoning Warranty manager at LI-COR, inc.

IMPORTANT: Please return the User Registration Card enclosed with your shipment so that we have an accurate record of your address. Thank you.

23

http://www.pme.com/USB_smanual/images/scampmanual41.jpg

http://www.pme.com/USB_smanual/images/scampmanual41.jpg [10/5/2005 6:06:17 PM]

SCAMP MATLAB PROCESSING AND DISPLAY SOFTWARE

Precision Measurement Engineering, Inc.

SCAMP MATLAB PROCESSING AND DISPLAY SOFTWARE (

P/N 5530)

REVISION 27-FEB-2004

INTRODUCTION

SCAMP measures and records CTD and other parameters at roughly 1 mm intervals in water column profiles. A typical profile will produce a large volume of data. SCAMP’s Matlab software provides a software environment that customers may use to process and display the collected data.

PME provides software in to forms: in commented script files (*.m) that the customer can read or modify, and in mex files (*.dll) that can not be read or modified. In general the script files implement processing steps and data display while the mex files do actual processing steps.

The customer must supply Mathworks Matlab software, revision 5.1 or later. PME supports only the Windows P.C. platform but the software will operate on other platforms if the mex files are re-compiled. PME will in some cases provide source code to enable re-compilation.

Processing of microstructure measurements is an evolving science. PME’s has implemented the complete path beginning at sensor measurements and ending with dissipation. Various methods used within this path are the subject of on-going development by PME and by the microstructure community. PME’s software is designed as a shell that enables customers to understand the path and substitute their own methods at appropriate points.

In addition to processing, PME provides a display GUI that allows the customer to easily review the measured data and also processed results. The display is shown below. Click to enlarge.

DISTRIBUTED SCRIPT FILES

Script files may be read and modified by customers. The important files s_process.m, and s_plot.m may be used as stand-alone activities without use of the GUI. Distributed as SCAMPGUI.ZIP.

FILE FUNCTION

SCAMP.M

SCAMP.MAT

S_SERVICE.M

S_PROCESS.M

S_BATFIT.M

S_PLOT.M

GUI display

Data file for GUI display

Call-back file for GUI display activities

Loads and processes SCAMP data

Implements the Batchelor fitting required by S_PROCESS.M

Implements plotting within the GUI http://www.pme.com/USB_smanual/scampgui.htm (1 of 3) [10/5/2005 6:10:38 PM]

SCAMP MATLAB PROCESSING AND DISPLAY SOFTWARE

DISTRIBUTED MEX FILES

Mex files are provided for operation on the P.C. Windows platform only. Each mex file is provided with a corresponding *.m file that gives a description of the file’s use and references upon which the computation method is based. Mex files are distributed in a

SCAMPTOOL folder. Distributed as SCAMPTOOL.ZIP.

FILE FUNCTION

S_AVG.DLL

Find the average of a segment

S_BLIMIT.DLL

S_BSPECT.DLL

S_BV.DLL

S_BWCOMP.DLL

S_C11.DLL

S_DDT.DLL

S_FILTER.DLL

S_GETSEG.DLL

S_GPRIME.DLL

S_LC.DLL

S_LOAD1.DLL

S_LOAD2.DLL

S_NORDER.DLL

S_PSD.DLL

S_REPVEL.DLL

S_SAL.DLL

S_SEGMEN.DLL

S_SEGPLT.DLL

S_SIGT.DLL

S_SMEAN.DLL

S_SNOISE.DLL

Find array index for frequency limit

Find theoretical Batchelor PSD

Find Brunt-Visala frequency

Smooth and Sharpen data

Find max likelyhood C11 parameter

Find DDT array

Low-pass filter an array

Extract a segment from an array

Compute gravity anomaly

Computed centered length scale

Load a old SCAMP data channel with name and units

Load a USB SCAMP data channel with name and units

Normalize array into downwards mode order

Compute power spectral density of array segment

Find representative velocity in a segment

Compute salinity

Define segment boundaries from DDT

Prepare a segment value array for plotting

Compute density, sigma-T

De-trend an array

Estimate SCAMP electronic noise http://www.pme.com/USB_smanual/scampgui.htm (2 of 3) [10/5/2005 6:10:38 PM]

SCAMP MATLAB PROCESSING AND DISPLAY SOFTWARE

S_SORTD.DLL

S_THORPE.DLL

S_VEL.DLL

S_X.DLL

Sort depth and sigma-T into density stable profile

Compute Thorpe displacements

Compute velocity

Compute Chi

INSTALLATION UNDER MATLAB ON P.C. WINDOWS PLATFORM

The script files within SCAMPGUI.ZIP may be placed at any convenient location. These files use files within the SCAMPTOOL folder.

S_PROCESS.M contains the statement: addpath ( 'c:\matlabr11\toolbox\scamptool' ); %holds SCAMP mex files

The SCAMPTOOL folder may be placed at this location, or at any convenient location within Matlab’s path, or at any convenient location if the addpath statement above is suitably modified.

The GUI is designed to operate at 1024 X 768 screen resolution. The GUI must be MAXIMIZED for proper operation.

OPERATION OF THE GUI

The GUI is begun by starting Matlab, then changing directories to the directory that contains the SCAMPGUI files. SCAMP.M is the central file and accesses the other files. After SCAMP.M is run the GUI display above appears, except that no data are displayed.

The first activity is to click the LOAD button and browse to a SCAMP data file. The GUI will load old SCAMP files as well as files collected with the USB SCAMP. After file selection a brief period elapses while data are processed. When processing is completed the results can be viewed. Example profile 1820 is supplied for testing the GUI.

The left two graphs present full-profile results. These include both directly measured data such as CTD and also full-profile processing results such as salinity. Results are selected from the drop-down boxes directly above the graphs.

The right two graphs present processing results from segments of the profile. Segments of both directly measured data and processed data can be viewed. These are essentially ‘magnified’ views of the data presented in the left (full-profile) two graphs. Additionally other processed results that apply only to the current segment can be displayed. Results are selected from the drop-down boxes directly above the graphs.

Segments within the data may be selected sequentially by clicking the <SEG and SEG> buttons. The horizontal dotted red lines on the left two graphs indicate the currently selected segment.

Data may be plotted vs time or vs depth by selection of the appropriate radio box in the upper right corner of the GUI.

New data may be loaded by repeating the LOAD process. Old data are replaced after processing.

GUI operation is terminated by selecting the X button at window upper right corner.

GENERAL NOTES

One additional note: the *.dll programs use arrays of rows rather than the more normal Matlab arrays of columns. The *.dll programs also count array elements beginning with index 0, not the Matlab convention of beginning with index 1. This is not a problem in most cases but beware of this feature. It is clearly stated in the *.m files where appropriate.

http://www.pme.com/USB_smanual/scampgui.htm (3 of 3) [10/5/2005 6:10:38 PM]

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement