Optimizing Cold Weather Performance Optimum Cold Temperature

Optimizing Cold Weather Performance Optimum Cold Temperature
Application Note
Optimizing Cold
Weather Performance
Date
o
1-1-2000
N
AN2000-14
By
Lars Boettern
Optimum Cold Temperature Performance From Biosystems
Portable Gas Detectors
1. Introduction
Often asked questions, which occur during the onset of colder months, relate to proper specification and use
of portable gas detectors in this condition. In the 1990's, gas detectors have evolved to become more
sophisticated, allowing for real-time temperature sensing and compensation of gas sensor readings. There
are currently four generations of portable detector designs in use with varying capabilities and proper use
conditions in colder weather. This applications note has been written to inform gas detector specifiers and
end-users on how to get optimum performance from Biosystems portable gas detection equipment for use in
colder environments.
2. Scope
This information principally applies to portable - PhD/Cannonball multisensor and Toxi Series single gas
detectors. There are some general guidelines in regards to power systems, displays, etc., which are true for
all Biosystems portable detectors. Any Detector/Model specific features will be explicitly stated as such.
3. Background/Perspective
3.1 Temperature Compensation
Rule of thumb in chemistry: "The rate of a reaction doubles for every 10°C (20°F) rise in temperature". For
galvanic-type oxygen and electrochemical type toxic gas sensors, which are small chemical reactors or fuel
cells, temperature compensation of response is certainly an area of great importance.
With the advent of "smart" sensors in the PhD Plus & Ultra design an EEPROM, or smart chip, has become
an integral part of each plug-in sensor assembly. This has led to a rapid evolution of custom features. One of
the most important of these being the sensor-specific array of temperature compensation factors for both
baseline (zero) and sensitivity (span) values.
As we will explain later, these values are not static but are refined each time the detector is fresh air and span
calibrated. The following table provides a summary of the sensor temperature compensation capabilities of
Biosystems portable gas detectors.
Detector/Model
Temperature Compensation Type
Toxi Series
PhD
PhD2
Cannonball2
PhD Plus/Ultra
PhD Lite
PhD5
Cannonball3
none
Partial-O2 Sensor Only
Partial-O2 Sensor Only
Partial-O2 Sensor Only
Full-Smart Sensor Specific
Full-Smart Sensor Specific
Full-Smart Sensor Specific
Full-Smart Sensor Specific
n/a
Fixed Polynomial for Span
Fixed Polynomial for Span
Fixed Polynomial for Span
Dynamic Array for Zero & Span
Dynamic Array for Zero & Span
Dynamic Array for Zero & Span
Dynamic Array for
3.1.1 General Operating Principles
For detector designs before the advent of smart sensors, or for those without smart sensors, - PhD/PhD2,
Cannonball2 and Toxi Series respectively, to obtain the most accurate readings the best course of action is to
zero and span calibrate the detector at the temperature of use. To perform this process successfully, it is
important that the detector be thoroughly acclimated to the anticipated use temperature by waiting an
Optimizing Cold
Weather Performance
Application Note
Date
o
1-1-2000
N
AN2000-14
By
Lars Boettern
appropriate amount of time for the system to stabilize - typically ½ to 1 hour, depending on the physical size
of the detector and the magnitude of temperature difference.
With smart sensors and on-board temperature sensing - PhD Plus/Ultra, 5, Lite and Cannonball3, this has
become partly unnecessary. The general temperature compensation values factory programmed into each
sensor serve the purpose described above. The next important consideration is how best to maintain, and
improve the accuracy of compensation values given various conditions of use. This will be reviewed in detail
in Section 5 - Operations.
4. Initial Specification & Setup
4.1 Power Systems
It is best to avoid the use of alkaline batteries, if possible, in colder weather conditions. Alkaline batteries may
give as little as approximately 10% of the run time at 0°F (-18°C) than at room temperature. In general, NiMH
and NiCad batteries perform best, followed by lead-acid types.
Ambient temperature has an effect on the recharge efficiency and life of NiCad and NiMH rechargeable
batteries. NiMH batteries should be recharged in areas where the temperature range is 0 to 40°C (32 to
104°F). Attempts to charge NiMH batteries below 0°C may cause leakage of battery electrolyte, impair
performance or shorten the operating life of the batteries. For NiCad batteries, optimum temperature range for
efficient charging is narrower at 5 to 30°C (41 to 86°F). Repeated attempts to recharge NiCad batteries at
temperatures below 0°C (32°F) will increase gas pressure in the cell at the end of the charge cycle which may
cause a safety vent to operate and battery life to shorten.
Battery systems in Biosystems portable detectors vary in terms of types and interchangeability, as battery
technology has also evolved rapidly in the past decade. For example, earlier PhD/PhD2 model detectors are
all rechargeable battery units. They are either internal lead/acid battery (units have steel belt clip on bottom
surface) or NiCad (pack drops out of a cutout in the unit bottom, thus no belt clip). In contrast PhD Plus/Ultra
and 5 models have fully interchangeable "snap-in" alkaline and NiCad battery packs, making this exchange of
battery type trivial. The following table provides a summary of battery systems and their interchangeability for
portable detectors.
Detector
Model
Toxi Series
Ex Chek
Cannonball2
Cannonball3
PhD2
PhD Plus/Ultra
PhD Lite
PhD5
Lead
Acid
n/a
n/a
X
n/a
X
n/a
n/a
n/a
Nicad
NiMH
n/a
n/a
n/a
n/a
X
X
n/a
X
n/a
X
n/a
X
n/a
n/a
X
n/a
Alkaline Interchangeable
X
X
X
X
n/a
X
X
X
n/a
No
Yes
Yes
No
Yes
No
Yes
4.2. Display (LCD)
Liquid Crystal Displays (LCD's) lose contrast and their refresh rate slows at cold temperatures. Biosystems
standard design, extended temperature range LCD's normally become illegible at temperatures less than 5°F
(-15°C). Refresh rate is the time required for a new screen to appear upon command. In all cases, decline of
LCD function has nothing to do with any other aspect of detector performance. Sensors will continue to
respond properly and the detector chassis audible and visible alarms will continue to function normally. The
display is a completely separate system designed strictly for the convenience of the end user.
Application Note
Optimizing Cold
Weather Performance
Date
o
1-1-2000
N
AN2000-14
By
Lars Boettern
There are provisions for contrast adjustment on certain detectors. For the PhD Plus/Ultra models there is a
potentiometer (pot) inside the detector that actually serves to darken the display more than it adjusts the
contrast.
In the PhD Lite and 5, with their more advanced digital displays, there are two ways in increase contrast. One
way is through the menu options. Please refer to the advanced functions section of your detector manual.
If the PhD Lite or 5 display has already become illegible, it is also possible to enhance contrast upon powerup by simultaneously holding down on the up arrow key which enables the contrast adjust mode. The "up"
arrow enhances, and "down" arrow decreases, contrast. Pressing the mode button locks in the setting and
exits from this feature. There is one difference in this routine for the PhD5 versus the PhD Lite. In the PhD5
upon entering into the contrast adjust mode the display automatically starts at the lowest contrast setting. This
can be disconcerting for the uninformed as it looks as if the display has "blanked out" initially. Contrast will be
restored by pressing the up arrow key.
On Cannonball2, PhD/PhD2, PhD Plus/Ultra multisensor detectors there are channel/sensor specific visible
red alarm LED's on the front face of the detector. It is worth taking the time to become familiar with
gas/sensor order on the display, since position denotes which gas type - and associated hazard is in the
alarm state, should it occur.
4.3 Low Temp Alarms
Oxygen and toxic gas sensors contain fluid (water-based) electrolytes and membrane systems that are
adversely affected in the event of freezing at very cold temperatures.
PhD Plus/Ultra, PhD Lite/5 and Cannonball3 models have in each sensor, factory programmed low
temperature alarm limits. Below the preset limit, the detectors' sensor channel display alternates between the
numerical reading and a "T" denoting this alarm condition. Sensor specific low temperature alarm setpoints
are shown in the following table.
Sensor Type
Low Temperature
Alarm Point (F)
Low Temperature
Alarm Point (C)
O2 Oxygen
-4
-20
LEL Combustible
CO Carbon Monoxide
-20
5
-29
-15
H2S Hydrogen Sulfide
SO2 Sulfur Dioxide
-4
-4
-20
-20
NO2 Nitrogen Dioxide
Cl2 Chlorine
-4
-4
-20
-20
NH3 Ammonia
PH3 Phosphine
-4
-4
-20
-20
HCN Hydrogen Cyanide
-4
-20
If performing monitoring sessions for limited times at temperatures anticipated to be colder than those listed, it
may be acceptable to temporarily disable low temperature alarm setpoints (allowable option is either to
enable or disable, not to alter the values). For specific details on how to access this feature, please consult
the "advanced features" section of your detector manual. For advice on your specific end-use application,
please contact Biosystems Technical Support Department.
Optimizing Cold
Weather Performance
Application Note
Date
o
1-1-2000
N
AN2000-14
By
Lars Boettern
5. Operation
5.1 Temperature Compensation & Sensor Accuracy
The following is a brief review of how "smart" sensor specific temperature compensation works and how it
differs between detector models.
For each sensor, typical temperature compensation factors are programmed into an array on the "smart chip"
in the temperature range of -20°F to +140°F, for both zero and span. There are sets of zero and span
correction factors for every 10°F increment within this range. These "typical" values represent average/default
values based on new sensor performance. With each detector fresh air (zero) and gas (span) calibration, both
values are refined.
In the case where small changes occur over time, for example as sensor output gradually changes with age,
localized corrections as a function of unit temperature are made to the array. Specifically, 100% of the
refinement to the correction factor is made at the nearest 10F° increment of detector temperature, smaller
refinements are made at the next ±10°F values, and still smaller refinements are made to values at ±20°F,
etc.
Where there is a large shift in sensor performance, for example as may occur when beads shift in an LEL
sensor subjected to an extreme mechanical shock, the same refinement, upon subsequent calibration is
applied over the entire array of values. This effectively shifts the entire set of values or compensation curve(s)
by the same amount.
We shall discuss shortly the consequences of each refinement strategy in terms of how the detector will
perform based on external conditions and habits of use.
Detector Models/Versions differ in regard to which refinement strategies are accessed. For PhD Plus/Ultra,
PhD Lite and PhD5 models the situation is mixed depending on the software version (shown in the initial
screen upon power-up). The table below provides a convenient summary of which versions support which
refinements, and how they are set-up at the factory.
Enabling small refinements to correction factors in the current PhD Plus/Ultra, PhD Lite and 5 can only be
enabled via entry into a special mode. Call Biosystems Instrument Service Dept. for details. Note that in prior
versions of the PhD Lite and 5 the detector program version may be updated over the internet (Biosystems
website: www.biosystems.com) via modem and an IrDA for the PhD Lite and datalogger cradle & cable for the
PhD 5.
Detector/Model
Software
Version
PhD Plus/Ultra 1.84 & higher
Small Refinement
Large Refinement
(Temp. Localized)
(Universal Shift)
Yes or No(Shipped
Disabled)
Yes
PhD Plus/Ultra 1.83 & lower
Yes
PhD5
Yes
Yes or No(Shipped
1.43 & higher
Disabled)
PhD5
1.42 & lower
Yes
Yes
PhD Lite
1.21 & higher
Yes or No(Shipped
Disabled)
Yes
PhD Lite
1.20 & lower
Yes
Yes
Yes
5.1.1 Conditions & Habits of Use, Effect on Temperature Compensation
Consider the following two situations in regard to PhD Lite – software version 1.10 detector use:
Case 1 Large multiple user setting with centralized storage of many "common use" detectors. Only (few)
authorized personnel are allowed to do both fresh air and span calibrations. Calibrations are always done at
an indoor calibration station.
Application Note
Optimizing Cold
Weather Performance
Date
o
1-1-2000
N
AN2000-14
By
Lars Boettern
Case 2 Delocalized individually assigned unit type setting. Units normally come back to the shop after each
work assignment. Calibrations, both fresh air and span, are often done by trained/authorized personnel in the
field under a variety of conditions.
For Case 1 - this version PhD Lite allows both refinement types. In this pattern of use, however, over a long
time there will develop a large temperature-localized correction to zero and span correction values. If these
units are then taken into the field under very different temperature conditions, both zero and span will revert to
original (largely unrefined) values. This change can be quite abrupt, which may result in significant positive or
negative shifts in baseline depending on sensor type and the direction of temperature change. Largely
affected sensors are those with temperature dependency and very low alarm setpoints - SO2, Cl2, ClO2,
NO2. In extreme cases (very long history of being calibrated at one temperature) the LEL sensor may also be
affected similarly.
Solutions for this situation are as follows.
a.) Short-term solution would be to re-initialize the sensors. This will reestablish factory default values of
span/zero correction for all sensors throughout the entire temperature range. Performing zero and span after
this should shift the entire compensation curve, resulting in a better overall correction. In the few cases where
re-initialization and re-cal does not seem to work, arrangements may be made through Biosystems Instrument
Service Dept. for the return of a unit.
b.) Long-term solution would be to update software to a more current version. Then, as above, reinitialize
sensors and do some zero and span calibrations. With the new software having localized temperature
corrections disabled, as the default setting, continuing in this pattern of use should not result in temperature
localized corrections.
For Case 2 - in this use condition, units which permit small refinements as well as larger curve shifts would
work well here. A long history of zero and span calibrations in many different temperature environments, will
result in constant (normally small) upgrades in unit accuracy. As the sensors age, they will become closely
accommodated to the detector.
5.2 Sample Draw Systems
Be aware any potential for condensation of moisture when remote sampling from below grade to above
ground if topside conditions are less than 40ºF. Moisture condensation and possibly freezing within draw
systems may cause filters to clog and/or scrub out water soluble gas(es) passing through them.
5.3 Additional Power Demands
Any use conditions that result in greater power consumption should be taken into consideration when
determining detector run-times. These are: dim light (use of backlight); chronic high level exposures (frequent
activation of audible/visible alarms); and use of motorized pump in PhD Plus/Ultra and later PhD designs - as
the pump is powered by the detector battery system.
5.4 Storage Conditions
Gas detectors are designed to be generally unaffected if kept in areas where people can be made to feel
relatively comfortable. Low temperature alarms only work on units that are powered-up and in the hands of
knowledgeable end-users. It should be understood that if units are kept in very low temperature
environments, even when powered off, sensors could freeze. As such, it is generally considered poor practice
to store units in vehicles that are left outdoors in cold climates during the colder months. In addition to the risk
of freezing sensors, if vehicle chargers are used, it is also possible to short charge batteries that are too cold
for proper charging, and possibly to shorten battery pack life, or cause leakage of battery electrolyte.
For any questions or concerns which are not addressed by this applications note, and for further guidance on
any aspect of low temperature operations, please contact Biosystems' Technical Support Department at tel.
(860) 344-1079, or via email at [email protected]
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