Hewlett-Packard 3586 A, B & C selective level meter, 3336 A, B & C synthesizer / level generator User's Guide

Using the HP 3586A, B&C ‚je Level Meter and HP 3336A,B & C MEASUREMENT /ENTAY FREQUENCY ENTRY TT E MEASUREMENT ENTRY FREQUENCY FEMTRY MEASUREMENT /ENTRY J - 1080 ve I CAL ————— on erie ei Каса AUTO LA af EA = Li > = CENTER CARRIER TO HE PANDEO Fa | вто | [recs] | || 2 || 3 "E Ea 336A SYNTHESIZER/LEVEL GENERATOR HEWLETT = PACKEANO ee IN EEE EEE EEE ": _u Ee Г. ЕЕ )Г. © [7] HEWLETT PACKARD Mile, SE Table of Contents Chapter | Introduction. ................ 1 HP Model 3586A, B & C Selective Level Meter .............. 2 HP Model 3336A, B & C Synthesizer/Level Generator... . . .. 3 Chapter II Front Panel Features. .... ...... .. 4 HP 3586A,B& C SLM :……….…_î….î….ïî…_….£…11111111111 1111111 4 HP 3336A,B& C S/LG ..... .... 8 Chapter III Specification Summary ................. 10 HP 3586A,B& C SLM ... 10 HP 3336A,B& C S/LG .......... 10 Chapter IV Operation and Applications ............ 11 HP 3586A,B& C SLM . 11 Basic Measurement Modes .......... 11 Selective Measurements ............ aa 11 SSB Channel Measurements .......... aa aaa 11 SSB Measurements withthe 3586C SLM....... 111111 13 Wideband Power Measurements . ........................ ........ 13 Convenience Features. . .......... 2.2 aaamooo 13 General Purpose Wave Analysis... .................... ... 15 Harmonic Levels... . aa aaa 15 Inter-modulation Products . .......... aaa 16 Modulation Distortion . ...... LL LL LL ALL 17 Spurious and Signals Close to the Noise............................ 17 Frequency Response Testing Using the Tracking Output. . .............. 18 Frequency Response Testing Using a Tracking Synthesizer .............. 18 Selective and Wideband Noise Measurements ................ ....... 18 Return Loss. 11200000 aa 19 Measurements on FDM Systems... ................... ... 20 FDM Channel Measurements at a Keystroke ........................ 21 Transmission Impairments Measurements with Option 003... ........... 23 Slot Noise, Noise Power Ratio (NPR) . ли, 23 Cross Talk Measurements .......... LL LL . ...... 24 HP 3336A,B&C S/LG Ce 25 Basic Operation. . ........ aaa 25 DataEntry. . aaa aananona a a arrin, 25 Using Modify to Tune Signal Parameters ........... .. . .. . . . . .. .. . .. 25 Using the 3336A, Bor Casa Sweeper. ........ 25 Using the Storage Registers ........... . .. ....... .. .. ... 26 Operating the Phase Control... . LL LL 26 Rear Panel Outputs... ...... LL LL 26 Options ........ aaa neaa raro 27 Applications .......... o 27 Wideband Sweep ....... LL AE 27 Swept Measurements Using the 3575A Gain/Phase Meter... .... 29 Phase, Dual Phase and Synchronizing ............... . . . . . . . ... . .. 29 Linearity Testingof VCO's . . 111111111111 LL 1111 30 Phase Lock Loop Testing. . ........... LL LL 31 Chapter V Remote Operation 32 Introduction ........... 32 HP-IBExtender. ............. LL 32 Remote Operation of the 3586A, B& CSIM .... 0 32 Instrument Programming Codes... ............. 32 Formats for Programming ................ EN 32 Outputting Data... 35 Measure Instructions ............ N 35 Interrogate . ooo III 35 Other Considerations... ............... ~~~ ° 35 Require Service . .................. 7 35 Typical HPL Program... .............. ~~~ 36 3336A, BE C S/LG Remote Operation .......... E 37 Control Moades ......... .... . .... 37 Learn Mode... O 37 Programming with the 9825A Calculator .. ........ ... 37 HP-IB Programming Codes ........... aaa 39 Parameter Interrogation............ LL 40 Status Byte ........ oo 40 Abridged Description of the HP-IB . . . ........ .. . . 41 Chapter VI Technical Description ........ 42 I586A, BE CSLM...... LL 42 I336A, BE CS/LG..... 7 43 Chapter VII Serviceability. ....... . 45 This User's Guide for the HP 3586A. B and C Selective Level Meter and the HP 3336A, B and C Synthesizer/Level Generator shows how these new instruments can be utilized as solutions for your wave analysis, FDM system and frequency synthesis applications. For quick review of operating features. turn to paged. HP-IB remote operation information is provided on page 32. For more information, contact your local Hewlett-Packard - Field Sales Office. Chapter | Introduction >] A ай Ч LT E — N Е 500 00.00 DERE + de — CE 115) . e. sec JT IE = & | am OR nee 8 ® | ry Ни so Л d = am a „дне EIA @ A : : (EIEN) A trims di e DU et ....... seu. ее añ A HG. seccre Le © РН № e mn .. © » РЕ = “e . ee mol, SEE pe SRY wh en ee AT SY os, SE ARE | убны ; E eu fu rh Aba var Sat = og Gen I >| És ЖЕ 1 ! à À if : ; | о ; = ] o = > = Fr 3 М a á il) = Lam 57 os E > LACIE . " ! ; Ba = ur | o eo Эа: ... ©. Jo E AS | | 88027055 © © 099 особое ARE Dota tp o lO rm —— 1 | ess DL] me mea = see с 2095000 @& <. | peón NAM 1 9 | 8“ mana Moa A | La Mia SA rt oe were E SS mas da Fn? mR ete Уи” ие 3 * EEE The 3586A, B and C Selective Level Meter and the 3336A, B and C Synthesizer/ Level Generator represent a powerful combination of precision microprocessor-based instruments for general purpose wave analysis, frequency synthesis and FDM system measurements. Three Models to Fit Your Needs 3586A Selective Level Meter Meets CCITT Requirements 3336A Synthesizer/Level Generator 3586B Selective Level Meter Meets North American (Bell) Requirements 3336B Synthesizer/Level Generator 3586C Selective Level Meter For General Purpose Wave Analysis and Frequency Synthesis 3336C Synthesizer/Level Generator 3586A, B & C Selective Level Meter 50 Hz to 32.5 MHz The HP Model 3586A, B and C Selective Level Meters are designed for use in general purpose and Telecommunication applications including audio, sonar. HF radio and Frequency Division Multiplex (FDM) systems testing. The wide frequency coverage of 50 Hz to 32.5 MHz, combined with excellent measurement accuracy and precision makes this powerful microprocessor based instrument your best choice for design, manufacturing and maintenance operations requiring highly accurate selective wave analysis or FDM voice grade and carrier level measurements. Accuracy, Precision, Selectivity HP Models 3586A, B and C combine up to + .2 dB level accuracy with a unique fractional-N frequency synthesis technique to allow highly selective level measurements with 0.1 Hz frequency resolution and .01 dB amplitude resolution and full auto-ranging and automatic calibration. 3100 Hz Channel Filter Option 003 2000 Hz C-Message Equivalent Noise Filter : 586B м) — ” +0.3dR Flatness { N e + 1740 Hz Psophometric Equivalent Noise Filter : E 3586A AA — 60 dB Carrier Rejection 1 + + + + i + + + 1 + 1 1 + t + + i i 1 1 1 + + + + + + + 1 1 1 Rejection dR 60 dB J 75 dB” \ ~ 75 dB Adjacent Channel Rejection 4 kHz Voice Channel ——al pa T T 0 1 2 3 - Frequency i | 4 kHz Filters from 20 Hz to 3100 Hz bandwidth are available with shape factors to 1.2, performance available first from Hewlett- Packard. You can resolve signals as close as 80 Hz apart with 50 dB rejection, or use the 3100 Hz filter as a "tunable channel bank filter,” virtually duplicating the performance characteristics of channel filters in an FDM system. COUNTER Use the accurate frequency counter to measure the frequency of a signal, and then tune to it with synthesizer precisio nin one keystroke. OFFSET The amplitude offset feature allows you to make amplitude level measurements with respect to any reference level to speed harmonic or tone level measurements. A measured level can be made a reference level at a keystroke. The Wideband Power measurement mode lets you make true RMS baseband power measurements down to — 45 dBm with up to +1 dB accuracy. MEASUREMENT MODE WIDE SELECTIVE SSB CHANNEL Ty BAND NOISE/ TONE LO DIST DEMOD 1010Hz CARRIER 800Hz 50V 858828 LO NOISE GHITER NOISE/TONE IMPULSE we START Fast SSB Testing The Single Side Band (SSB) measurement mode provides demodulation capability with speaker or headphone output. The demodulated audio can be output for further processing or measurement. 3586A and B models allow the user to enter a FDM carrier or tone frequency, choose an inverted (I~) or erect (1) sideband and precisely align the 3100 Hz channel filter on a voice channel. Choose from a "menu" of measurements including channel noise, power or demodulation, carrier leak or tone level measurement (plus signalling tone level in the Model 35868). Transmission Impairments Option Transmission Impairment measurements including noise- with-tone, signal to noise-with-tone ratio, phase jitter, and single level impulse noise, and a C-message or Psophometric noise filter can be added to A or B model instruments with Option 003. Use this capability to compare voice channel impairments at both voice grade and carrier level for FDM troubleshooting. Worldwide Connectors and Impedances The modular front-end design allows a selection of impedances and connectors for general purpose, CCITT or North American (Bell) requirements. Choose impedances and connectors by model or option. Most special connector requirements can be met, contact your local HP field sales office to discuss your specific needs. Model Input Impedance Mating Connector ЗЫВБА 75817 10k Unbalanced BNC 11 6/56 mm merrici * CCITT 1500, 6000/1007 Balanced | Siemens 3 prong 9REL-BAC 3586B 75817 10k§) Unbalanced WECO 4394/4404 (358A) No American 11240 Balanced WECO 443A (3724) (Вей 1350 Balanced WECO 2414 EDON/ TORN Balanced WECO 310 3586C 500/750 10k) Unbalanced | BNC General Purpose 60007 10k0? Balanced Dual Banana 75” spacing “Optional Ls O Many Convenience Features Microprocessor power and flexibility provides additional ease- of-use features including frequency stepping. amplitude units in dBm, dBV or dBpW, analog frequency control with variable resolution. and low distortion or low noise operating modes. Nine storage registers allow multiple front panel setting con- figurations to be stored for later use. This feature significantly speeds repetitive manufacturing testing. Programmable All necessary functions on the 3586A, B and C Selective Level Meter are HP-IB programmable by HP 9800 series Desktop Computer controllers, or by a mainframe computer like the HP 1000. Fast, efficient manufacturing testing maintenance and calibration, and automatic, remote FDM system surveillance are only a few applications. The HP-IB is also used for implementing frequency tracking with the 3336A. B and C and 3335A Synthesizer/Level Generators. 3336A, B and C Synthesizer / Level Generator 10 Hz to 20.9 MHz The HP Model 3336A, B and C Synthesizer/ Level Generators are designed for General Purpose, Telecommunica- tions and FDM system applications requiring a high accuracy, high resolution stable signal source with precise output level, excellent spectral purity and wideband sweep capability, The 3336A, B and C can also be used as a frequency tracking com- panion synthesizer for the 3586A, B and C Selective Level Meter up to 20.9 MHz for frequency response and FDM system testing. ¿UU 999999999 Frequency Precision and Spectral Purity The HP 3336A. B and C provide one microhertz (. 000001 Hz) resolution up to 100 kHz and one millihertz (.001 Hz) up to 20.9 MHz with frequency accuracy and stability of +5 x 10°. (+5x10% stability optional), Harmonics are down 60 dB to 1 MHz and down 50 dB to 20.9 MHz, performance not previously available in a synthesizer. + .05 dB Accuracy The 3336A, B and C combine outstanding frequency accuracy and purity with absolute amplitude accuracy of + 05 dB at 10 kHz or 50 kHz and + .1 dB flatness { 07 dB optional) at full output. Overall 75 ohm accuracy is + .22 dB with the optional high accuracy attenuator over the full frequency range, down to — 72.99 dBm output. CA A e BE ® Ц 2 = i Ze E | 5 E SUR ла ee. eo eo a TIT TT Tal dd TPL : : am e E 5 A “ОВ Аз 5 % jo EEE > e AE A = = pt i > 3 a E 1 + * mart Ann - te да Залы дур ate MD om al Зум вии оо aye Worldwide Impedance/ Connector Options Like the 3586A, B and C Selective Level Meter, impedance and connector combinations are available for CCITT. North American (Bell) and general purpose applications. Model Output Impedance Output Connector 3336A 758 Unbalanced | BNC (16/56 mm metrics" CCITT 1500 Balanced Siemens 3-prong 6004 Balanced 9 REL-6 AC 33368 75Q Unbalanced | WECO 439/440A 358A)" No. American 1244 Balanced WECO 443A (372A: Bell} 1354 Balanced WECO 241A 6000 Balanced WECO 310 3336C 500 Unbalanced | BNC General Purpose | 7581 Unbalanced “Optional a e 5 SZ Precision Broadband Sweep HP Models 3336A, B and C provide unique leveled sweep capabilities including full band phase continuous sweep, single or continuous, log or linear with +15 dB flatness at full output. X. Y and Z axis outputs allow oscilloscope or X-Y recorder display of swept frequency response with a precise TTL output frequency marker. Chapter II Front Panel Features 0 3586B for No. American (Bell) Applications Shown with Option 003 Transmission Impairments 7 FREQUENCY/ENTAY CAN wit AL Pa Ч Ч „и . > a Г! Г — NS Pi = TEN j EN NN E ms he n в y | a ы IH EN — AUTO CAL LE ро Ful LM: — 7 r fe c a dr pi e roel 1 F [CFE М —_— wading | LL LLL dti FPAN x FE = RT TOM | 1 3 | Bi Fr MEASUREMENT MODE iH SELECTA SER CHARMED TONE Mi à a | ENTRY EQUENC y TUN DA UN: En peus FES QUENCY TUNE INR, fr pan SE mA САННИ В 2500 Hz 2885 0820509 O08 (nouer RE SOL 750 iE NOISE ÉHTTER MOISEZTONE Balin BF wee STARY $ TERMINATION MEAS tga 1365 TONT 3 Ê € 513 BRIDGE D HOU o © 0 00.) вова = £ a. роны в # i EEE a .. 5 i : í ЗОНЫ неф ен : $ | u) C/(3) GB) Es | N STORE ECALL 4 2 & 3 | A \ : MIN OFFSET В í LN | 0 Í i i } 5 wa“ : ot Е E Î N # ; FT o | Ë f Be Da 3 31006 5, 5 = у Е A ww à 5 O j т / ADPHOB: - } ; ; # # / ; | 7 f Ë | ê # 7 7 - 4 3 Е E # Bopp reeling toeeez § AER eiii РОДНЫЕ Ë Ë г Ë J i A Ë J A £ Ë Z & Е # # == e o 2e © © & 9 4 3586A for C.C.LT.T. Applications (same as 3586B except as shown) SSB CHANNEL measurements include a "menu” of tests MEASUREMENT MODE | available at a keystroke. Enter this mode by choosing the FDM Sano a. ное, - om томе \ carrier or test tone frequency as an entry frequency, LSB (N) or в © в в в e a USB (A) and any one of four measurements. EEE 121 0 аи — NOISE/DEMOD, (3586A/B only), is used for idle channel (9)—C) (+) o o noise, noise power or channel demodulation. The 1740 Hz | (35864). 2000 Hz (3586B), or 3100 Hz (Option 003) bandpass x e —_ т e x filter is automatically chosen and aligned on the voice channel. —e | o 1010 Hz and 800 Hz + 15 Hz TONE levels or CARRIER leaks , \ e | / e | are measured at a keystroke on the 3586A after the SSB im A — CHANNEL mode is entered. 1004 Hz TONE or 2600 Hz Ao commie À ame ue sims 0008 SIGNALing tone or CARRIER leak measurements are available on the 3586B. The 20 Hz bandwidth is chosen automatically. @ CHOOSE UNITS for level. noise in db picowatts. (3586A/B only). or dBV referenced to .775V, (0 dBm into 6000} 3586A, B Selective Level Meters 0 ~ “AMPLITUDE DISPLAY provides 01 dB resolution in 10 dB nd range and .1 dB in 100 dB range mode. Measurement range is POWER STANDBY applies power to High Stability Reference +20 to -120 dBm. Level. FULL SCALE level. OFFSET level oven, Option 004, HP-IB interface and power supplies. and OPTION 003 impulse noise threshold level can be displayed. AUTOMATIC LEVEL CALIBRATION occurs approximately © every three minutes or when the frequency change is > 1 MHz. OFFSET allows amplitude measurements relative to a keyboard entered or measured level. FREQUENCY DISPLAY is 9 digits with .1 Hz resolution. Displays all frequencies entered or counted or Option 003 Im- pulse Noise time period in minutes and seconds. (8) COUNTER is used to measure any signal frequency within the 60 dB points of the IF bandpass (or — 100 dBm) to an accuracy of + 1.0 Hz plus the center frequency accuracy. Use with to tune to the measured frequency without “rocking.” HP-IB STATUS lights indicate bus mode. WIDEBAND measurements such as baseband power are made from 200 Hz to 32.5 MHz over + 20 to — 45 dBm with an RMS detector. In this mode the front end 15 "wide open” with no pre- selection filtering. The RMS value of all signal and noise power in the frequency is measured and displayed. SELECTIVE measurements mode is used for basic wave analysis measurements with the precise center of the bandpass at the entered and displayed frequency. LO DIST (Low Distor- tion) is used for most selective measurements such as harmonic and intermodulation and spurious levels. In the LO NOISE (Low Noise) mode, the noise floor is reduced for better accuracy for noise and low level signals close to the noise. OPTION 003 instruments add these three impairment measurements to the SSB CHANNEL menu. (3586A/B only) Set threshold level for impulse noise measurements (Option 003 only) from — 45 dBm to — 120 dBm in 1 dB steps. © Set time period for impulse noise measurements (Option 003 only) up to 99 minutes and 59 seconds. A 0 Time Entry sets up a continuous measurement. ENTRY FREQUENCY can be an FDM test tone or carrier frequency used to automatically align the channel filter on a volce channel in the SSB CHANNEL measurement mode. In this mode the center of the passband is not the frequency displayed unless the tone or carrier level is being measured. (See SSB Channel Measurement Mode). (16, Use RDNG- OFFSET to make a measured level an offset. Subsequent measurements will be made relative to that level when offset is on. "MANUAL TUNING RESOLUTION is chosen by the operator in the FREQUENCY STEP MODE in .1 Hz to 32,5 MHz steps. Resolution is 2, 4 or 20 Hz in AUTO. 3100 Hz and 3100 Hz WTD FILTERS are part of OPTION 003 on the 3586A/B. Standard filters are 1740 Hz (Psophmetric equivalent on the 3586A), and 2000 Hz (C-message equivalent on the 3586B). FREQUENCY TRACKING with the 3586A, B Syn- thesizer/ Level Generator is enabled by changing the bus switch to Remote and connecting HP-IB interfaces. The frequency of the synthesizer will always be the same as the SLM in this mode. Probe Power is compatible with HP High Impedance Accessory Probes. | + The 3586A & B are designed with a flexible output section to allow special connectors to be provided at minimum cost. MArrey Cd A E 8 Li LE ' | EN IT Fa e FL N PARTS TY ON (re CTRL LA ALE | 3586C for General Purpose Applications | MEASUREMENT MODE remo HE LECTIVE PADE | } BAND LO DIST LD MOUSE ao [== ев TERMINA TIOS \ S03: , | . ) es Em x BO FEL BRIDGED (D / | о f | / FOR ОНЕОрЕ | LEICHE i ; / E / + i : CouPies Big: cw GOW My / / ЛА / / de Te ul E a I a FI ZI ILL _ E [AE =— — E DL Fi A TO — Comte ENTRY A ES TRE зо, Foor OR SOC ни 2) 0 © 8 || | \ 13 0) 0) [9 © — ma TE Shani rEg HE ALTO STEF A Craie RESOLUTION cm? RIAS ше #ûD5Hz | SONT pe = ta te E НЕАОРНСНЫЕ 3586C Selective Level Meter ANALOG METER for "peaking” signals in the FULL SCALE ENTRY mode. 10 dB RANGE and FULL SCALE AUTO automatically chooses the proper range and full scale level for best signal to noise ratio and measurement accuracy and .1 dB resolution. AVERAGING averages five measurements and displays the result once a second to minimize “racking” during noise measurements. SSB measurement are made on the 3586C in the selective mode by manually offsetting the frequency from the carrier frequency. "CHANNEL chooses LSB {N} or USB (A) for SSB operation. 504 for general purpose RF measurements. “Measure in units of dB, dBV referenced to 1V, or dB referenced to .775V on the 3586C. Nine non-volatile storage registers can be used to store nine front panel settings for repetitive measurements. RECALL ZERO sets the front panel to the turn-on condition. Enter a frequency step from .1 Hz to 32.5 MHz and then use these keys or the frequency tune control to step the frequency. The EB keys can be used to step the full scale settirig in 5 dB steps. or the impulse noise threshold. (option 003 only) in 1 dB steps. 3100 Hz BW (flat) is standard on the 3586C. HEADPHONE OUTPUT provides 0 dBm into 6009. Also can be used for further measurements on demodulated AUDIO. FEI mn, Î FRY — f a 2407 A. i ian 3 FEES Î Ë VOLTAGE | # ssctons + N 4 AG = - .. 186Y-240V - 107 -5% Y 48-642, ¡SOVA MAX — À «о | > WARNING | FOR CONTINUED PROTECTION AGAINST FIRE HAZAHD REPLACE ORLY WITH THE SARE TYPE AMD RATING OF Е EA USE AS SPECIFIED OS THE LINE YvOL TAGE E THIZED F su UIT Arz 0 3586A/B/C Rear Panel TRACKING SYNTHESIZER SWITCH, set to TRK for talk only mode for frequency tracking. Set to REM (Remote) for normal HP-IB operation. ©) HP-IB ADDRESS SWITHCHES, Binary weighted switches to set TALK/LISTEN address. HB-IB INTERFACE accepts all metric thread HP 10631 cables. For HP-IB or frequency tracking with 3336A/B/C or 3335A Synthesizer. “10 MHz OVEN REFERENCE OUTPUT, (Option 004 only), +1 x 10-7 stability, connected to 10 MHz + N by jumper. Meter output provides 10 mV/dB on 100 dB range and 100 mV/dB on 10 dB range. Output is offset so 0 dB is 0 mV out. 10 MHz + N input is used to lock the 3586A, B or C to a 1 MHz to 10 MHz 0 to + 20 dBm reference (must be a sub-harmonic of 10 MHz, 1, 2.5, 3.33 MHz. etc). Audio output provides the demodulated audio, .75 Vrms into 1 kil at full scale level. | eur | EXT MEF Os an | LEY MAX ELL CONECTOR TEE. В с. РГ Lime 2 = + i ia CHE ha | J METER J # В г E j т / CAUTION: но сРЕВАТСЯ f SERVICEASLE PARTS meting No SEEN SEMVICINS TO f dirt M / rv aude ] 7 ALIFIED PERSONNEL. y e ÿ OPTION 003 OPTION 504 | BETTER Се) 1 | © Бо» БОМН: ¡ED a: ï amo, af e wif ая 10 MHz REFERENCE OUTPUT, +1 x 105 stability. TRACKING OUTPUT provides an output signal at the bandpass center frequency for frequency response measurements. Level is 0 dBm + .5 dB (at 10 kHz) with + .5 dR flatness. 50 MHz auxiliary and F o + 50MHz outputs are not connected to REAR PANEL. PHASE JITTER OUTPUT provides 20 to 300 Hz demodulated. Phase Jitter Spectrum, sensitivity is 166 mV/o. (3586A/B Op- tion 003 only}. A my весеннее fm TRY пра 1 | FREQUENCY GUYPLY (e) Ce) FEO PROS BOO My 20 SMH: PA HZ - 7 Cu; FOO: - 1O06 Hi: Y % 1 spe E [e] SINGLE FREG TEN AMPLITUDE рог жа O Da O Pa Pe Fea e Leena a 1 В a main , | || | 4 A a fone = i o ê % : FART WKH CF LUG bod O | ® 1 0 : : LE WEL NG Ё : E : i Figen, } } : | % | PHASE i i % nn La e ~~ © . i ; ALLI or fo) С°) 9, ®@ 10 :. NE. 1 oy ; “MAR ASK ZERO À 2 ® 3336B for North American (Bell) Applications 3336C for General Purpose Applications (same as 3336A except as shown) (same as 3336A except as shown) Le OUTPUT EE — = {1 a 1 В ТО МЕ- 20 oxy; Окна с 106 Mel 10kHz- 5 OBKHz FOOHr - 1 OK Hz / e ojo: e | 505 750 } Doa E > - | : id AV + i | Las | fons as Ë i 1 Max 7 i LOA! Ei ! ? # ——E E as MÁS + 3 FLAT Кой f TE POE BAK Je PUT na > 6 Ted MAS CT WARNING FOR CONTRLED PR - pe TT TEA РС, eas tl ST T= OL ea FE REE a ET SO OR TEN Li AFTER hy a 3336A,B & C Synthesizer / Level Generator Sweep entry, sweep start, and sweep modification keys. BLUE PREFIX keystroke required to initiate sweep modification, log sweep, or to choose fast or slow leveling. HP-IB STATUS lights indicate bus mode. LOCAL keystroke returns 3336A.B.C from remote to front panel unless local lockout is programmed. LOCAL keystroke preceded by BLUE PREFIX keystroke initiates bus address in display. OPTION 005 HIGH ACCURACY ATTENUATOR improves amplitude accuracy at > 9.99 dB attenuation from full output. ® SIGNAL PARAMETER PREFIX KEYS. PHASE keystroke when preceded by BLUE PREFIX keystroke will assign zero phase (ASGN ZERO 0) to any previously selected phase offset. ® PARAMETER Amplitude. ® Entry light is "on" when 3336A is waiting for DATA following an ENTRY selection. Data and units of frequency, amplitude, phase, or sweep time must be selected to complete the entry. DISPLAY (11 digits max.) of Frequency UNIT light is "on" when entry is complete. Output signal assumes new entered value. EXT REF light is "on" when external frequency reference or Option 004 10 MHz high stability reference is connected to the rear panel REF IN. This light flashes if phase lock not complete. MODULATION lights "on” when AM or PM is selected. LIGHT "ON" when a frequency of 21 MHz or greater is selected. Output is automatically switched to rear panel aux. output. See rear panel view. | POWER STANDBY applies power to High Stability Reference Oven Option 004, HP-IB interface circuits, and power supplies. BLUE PREFIX key used for double coding several keys. | STORE/RECALL up to 10 tull front panel settings. AM enabled {ON) when BLUE PREFIX keystroke followed by STORE; disabled (OFF) by BLUE PREFIX/RECALL. PM enabled (ON) by BLUE PREFIX/CLEAR: disabled (OFF) by BLUE PREFIX/ —. Modulation inputs on rear panel. "SOFT" AMPLITUDE BLANKING to < = 85 dBm enabled by BLUE PREFIX key followed by MHz/BLANK: disabled by kHz/ NORM. - O (©) > selects the parameter digits to be modified, then use the analog control to tune frequency, amplitude, phase offset or sweep parameters, © 3336A OPTION 001 provides 750 1.6/5 6mm coaxial connector. OUTPUT IMPEDANCES and connectors meet CCITT. NORTH AMERICAN (BELL), or GENERAL PURPOSE requirements, OPTION 001 provides large WECO on 759 and 1249. HIGH STABILITY REFERENCE OPTION 004 provides 5 x 10% per week stability. Output must be jumpered to EXT REF IN. EXTERNAL REFERENCE INPUT phase locks to external stable reference signal or Option 004 output. 1 MHz REFERENCE OUTPUT to phase lock other instruments to 3336A,B,C. HP-IB CONNECTOR. @) PHASE MODULATION INPUT 5 kHz maximum {can work simultaneously with AM). MARKER OUTPUT, a TTL (0 to 5V) negative going transition during sweep. X-AXIS DRIVE provides 0 to 10V sweep drive for oscilloscope or plotter displays. | Z-AXIS OUTPUT blanks retrace on swept display. 21 to 60.9 MHz AUXILIARY OUTPUT provides 0 dBm signal when frequency keyed in on front panel is 2 21 MHz, Output will under-range to 20 MHz. To return to main signal output, < 20 MHz must be keved in. SYNC OUTPUT TTL SQUAREWAVE OUTPUT, frequency identical to main output, zero crossing lags main output by 25 ns {nominal}. AMPLITUDE MODULATION INPUT 10 Hz to 50 kHz. Specification Summary 3586A, B & C Selective Level Meter Frequency Range: 50 Hz to 32,5 MHz ) Accuracy: +1 x 10"/Yr (+2 x 10 /Yr. optional} Resolution: .1 Hz Counter Accuracy: +10 Hz + 1 Hz 35EDA i AnRAR i РНЕ EE 37 5 MH dz ma AZ OS MI, Suyral Inpuís 35B6E TL a El ТВ Kris 16) MAS: 10 bir 1 1 MHZ 1740 3350 1500 OD ORE ТК 1 Е МНЕ BE Hr fo 1065 kHz Selectivity 3 dB Bandwidth + 10% (SERA 3586B 35500 20 Hz 20 Hz 20 Hz 400 Hz 400 Hy 450 Hz 1740 He 2000 Ну" 31060 Hz “3100 Hz and WTD optional Adjacent Channel Rejection: 75 dB Carrier Rejection: 60 dB Passband Flatness: +.3 dB Amplitude Range: + 20 to —- 120 dBm Resolution: .01 dB Level Accuracy: +20 to — 80 dBm : { 75 SE! 1240 Г 250 1500 2 dB AR dR = Ah dR SINE] + 3545 [7 1 dB 211 REE: de 18 МЫ: E kE 5 MP BA kHz ne 1 Mb; а 5 108 kHz 2 kE in je Mb + iD na E Er : Wideharnt Poway - 45 JE FRE Noise Floor (Full Scale Setting — 35 to — 120 dBm) Frequency Bandwidth Noise Level 3100, 1740, 2000 Hz! — 116 dBm 100 kHz to 32,5 MHz 20 400 Hz — 120 dBm 100 kHz ts 106 kHz Al — 105 dBm Dynamic Range Image Rejection {100-132 MHz): —80 dBc IF Rejection: 15625 Hz, — 80 dBc 50 MHz — 60 dBc Spurious Signals: > 1600 Hz offset, — 80 dBc 300 Hz to 1600 Hz offset. — 75 dBc Distortion: Harmonics: — 70 dB below full scale, Lo distortion mode IM: — 70 dB below full scale. 200 Hz to 20 kHz offset — 75 dB below full scale. 20 kHz to 1 MHz offset HP-IB Control Standard 3SB6A dE Bei 35860 |! Gil Connectors 50 168 56 mm repiaces BNC 752 WECO 1240 МЕСО replaces min: WECO UNZ Bandwidth ie 1740 Hz replaces 2000 He UCA Transmission imparments be ren a: et Adds Phase prey FUE He WT Nour usth-tone, Single Level Impulse Noise 004 High Sabilng Frequency Hel Oscillator +9 x 18 7 у In 3336A, B 8: C Synthesizer / Level Generator Frequency Range: 10 Hz to 20.9 MHz Accuracy: E5x 107 (+5 x 10° optional) Resolution: 1 pHz, < 100 kHz: 1 mHz > 100 kHz Signal Outputs я ae TTL 7 ye - Chapet 1 33364 33368 DO 33080 500 10H: 10 20 MH 75 НН to ZU 9MHz 10H; #2 10 GMHz | 10%Но то 2 09MH; | 1244 1358 ISE 10H: 2 7 NOME, HULE НН: tr 109 GRH: Amplitude Range: 500: — 71.23 to +8.76 dBm 752, 6002: — 72.99 to + 1.00 dBm 1240, 1359, 1509: — 78.23 to + 1.76 dBm Resolution: .01 dB Accuracy: + .05 dB, 20°C to 30°C, at 10 kHz for 50. 75, 600% outputs or 50 kHz for 124, 135, 150Q outputs (+ .08 dB 0° to 55°C). Flatness: +.1 dB (+ 07 Option 005). at full output Attenuator Accuracy: *.1 to +.3 dB Standard +.035 to +.1 dB Option 005 Blanking: soft blanking to < —85 dBm Spectral Purity Harmonic Levels: ~60 dB 50 Hz to 1 MHz —55 dB 1 MHz to 5 MHz — 50 dB 5 MHz to 20.9 MHz Phase Noise: Integrated, — 60 dB/30 kHz BW SSB, — 70 dB. 3 kHz BW, F. + 2 kHz Spurious: 70 dB down or — 100 dBm (- 115 dBm with Option 005, depends on output and frequency) Phase jitter:<x0.3°p-p Modulation AM: 50 Hz to 50 kHz, 0 to 100% PM: DC to 5 kHz, O to + 850° Phase Offset + 719.9° vs. arbitrary starting phase or 0° relative Frequency Sweep Sweep Range: Full range of signal output Sweep Time: .01 s to 99.9 5, depends on mode Flatness: +. 15 dB, fast level, 03 s HP-IB Control Standard 33384 TH 1 5/56 mm replaces BNC 33385 33360 Tu 174 Large WECO E £5x [07 Accuracy Cnuons fi Connector $3 High Siatnlay Prequeno; Reference UD High Accursey ! Attenuator Improved Attenuator Accuracy Chapter IV Operation & Applications 3586A,B & C Selective Level Meter Basic Measurement Modes The 3586A, B or C can make SELECTIVE, SSB CHAN- NEL, or WIDEBAND power measurements for a wide variety of general purpose wave analysis and FDM system measurements. This section describes how basic measurements are easily made in each mode. Selective Measurements Getting Started Selective is the normal "wave analyzer” mode for selective measurement of discrete frequencies, spurious, harmonics, pilot tones, carrier leaks or noise levels. In this mode the entered and displayed frequency is the precise center of the passband. * At turn-on, the 3586A, B or C is ready for selective measurements, The basic configuration is SELECTIVE, LO DIST, 10 dB RANGE, FULL SCALE AUTO. * Choose the proper input impedance for your application. * Enter your frequency by pressing [8 , your numerical frequency digits, and complete the entry by pressing = |, (ve), ог ($ )аз appropriate to your entry. If your frequency is within the 3100 Hz bandwidth, the level will be displayed with O1 dB resolution. To make more selective measurements, choose the 400 Hz or 20 Hz bandwidth. If the signal falls outside the bandwidth chosen, enable the FREQUENCY TUNE control by pressing MS and tune to the signal. Automatic Fine Tuning with the Counter The 3586 can be automatically tuned so that the passband is centered precisely on the strongest signal in the passband by turning on © to measure the actual frequency, followed by ‘ses which tunes the passband center to the counted frequency. The counter accuracy is 1.0 Hz + 1 Hz, in addition to the center frequency accuracy of the instrument. The counter will measure the strongest level within the 60 dR points of the filter, down to — 100 dBm. If two or more signals are within 3 dB of each other or less, a measurement uncertainty results. Passband aligned on signal with synthesizer precision — with two keystrokes, к т кз © Frequency Desired Frequency Entered il Low Distortion vs. Low Noise in the SELECTIVE mode, the operator has the choice of LO DIST (low distortion) or LO NOISE (low noise} operation. Low distortion is used for most selective level measurements. In this mode, RF and IF gain levels are chosen for most linear operation, providing best harmonic and intermodulation performance. When in doubt, use LO DIST. The LO NOISE selective mode adjusts RF and IF gains to reduce the noise floor by about 5 dB to provide better accuracy for noise level measurements, and small signal measurements close to the noise. Enter this mode by pressing the BLUE PREFIX key, followed by the LO NOISE key, ==: 3 НЫЕ, -— SSB Channel Measurements (3586A or B) The Single Sideband (SSB) channel measurement mode is used to make level and noise measurements on a telecommunications voice channel such as in a Frequency Division Multiplex (FDM) system, or for SSB AM demodulation to listen to audio or to use the demodulated audio for further measurements using additional instrumentation. The SSB Voice Channel A SSB voice channel is offset from a carrier on either the upper (USB) or lower (LSB). The carrier and unwanted sideband are suppressed at the transmitting end by phasing or filtering techniques. Suppressed Carrier a | ” \ / \ | 7 Suppressed \ | / Sideband \ | / Vi / \ | 300 Hz _ 3400 Hz - Offset An Upper Sideband SSB Channel In telecommunications nomenclature: LSB = "INVERTED CHANNEL” =N USB = "ERECT CHANNEL" = A The audio sideband bandwidth is usually restricted to 300 to 3400 Hz away from the carrier, so to make measurements on this channel we must place the SLM bandpass filter precisely over the channel. The power of the microprocessor allows us to do this automatically, with synthesizer precision on the 3586A or B. We can enter the SSB channel mode using the known carrier {or fest tone) frequency, choose USB or LSR operation and then choose from a "menu" of SSB channel measurements, The required frequency offsets are automatically computed and the bandpass tuned to the proper frequency depending on the measurement chosen. 3586A/B SSB Measurement ” Menu” The table shows the SSB channel measurements available on standard 3586A & B and Option 003 instruments. The offsets shown are for carrier frequency reference entry. PAE ARLE RF Rt MEO Capt eee ve $ a as | в | 50 88828 Pan e e | _” e. “A | A- e 3550A (Standard) 3586A (Option 003) SSB Channel Offset Offset Measurement From Carier‘| Bandwidth | From Carrier Bandwidth Noise/Demod {Noise or + 1350 Hz 1740 Hz + 1850 Hz | 3100 Hz demodulated audio} 1010 Hz + 15H: + 1010 Hz 20 Hz + 1610 Hz 20 Hz Carrier 0 20 Hz 0 20 Hz Tone -800 Hz + 800 Hz 20 Hz + 800 Hz 20 Hz + Hz Psophometric + 1350 Hz | 1740 Hz £1850 Hz | WTD Weighted Noise {Equivalent 3100 Hz noise BW) Phase Jitter / + 1850 Hz | 3100 Hz {on 1010 Hz tone) f Noise/ Tone + 1850 Hz | 3100 H: {Notched noise) (995-1025 Hz Notch) impulse Noise + 1850 Hz | 3100 Hz {with notch} (995-1025 / Hz Notch) "With carrier frequency entry. 800 Hz tone entry may also be made with a resultant change in offset. 3586B SSB Channel Measurement "Menu" — и | MEASUREMENT MODE nn 50 08088 9-0 © © © | | | | Ca vr pb [т | р eo. A LA EH = | 3586B (Standard) 3586B (Option 003) SSB Channel Offset Offset Measurement From Carrier” | Bandwidth | From Carrier‘ | Bandwidth Noise / Demaod {Noise or + 1500 Hz | 2000 Hz + 1850 Hz | 3100 Hz demodulated audio) 1004 Hz + 15 Hz + 1004 Hz 20 Hz + 1004 Hz 20 Hz Carrier {3 20 Hz О 20 Hz 600 Hz + 2600 Hz 20 Hz + 2600 Hz 20 Hz (Signaling tone! C-message + 1500 Hz | 72000 Hz + 1850 Hz | WT Weighted Noise {Equivalent 3100 Hz Noise BW) 1004 Hz tone) Noise/ Tone + 1850 Hz 1 3100 Hz (Notched noise! (995-1025 Hz notch: impulse Noise + 1850 Hz | 3100 Hz {with noch] (995. 1025 /) Hz noch) "With carrier frequency entry. 1004 Hz tone entry may also be made with a result ant change in offset, 12 Entering the SSB Channel Mode The 3586A/B uses either a carrier frequency entry. or a test tone frequency as a reference from which offsets are internally computed for the appropriate test. The carrier or test tone frequencies are usually available on an FDM frequency chart. 1. Choose a carrier or test tone frequency reference and press the appropriate entry frequency key. These keys are used only for frequency entry. They do not determine the measurement frequency. For example, the Channel 1 carrier frequency in a basic group for both CCITT or Bell FDM plans is 108 kHz. [The tone frequency would be 107 kHz (Bell) and 107.2 kHz (CCITT).] 2. Enter the frequency chosen in the ENTRY block. 3. Choose an erect (USB) or inverted ( SB) sideband. For a basic group. the choice is inverted or LSB. 4. Choose any one of the SSB measurements in the MEAS- UREMENT MODE block. CHOIGE / DEMO Ín this case we have chosen to measure noise on the channel, or to listen to the demodulated audio. The widest bandwidth filter, (1740 Hz, 2000 Hz, or 3100 Hz depending on model or option) is now centered over the voice channel as shown. Carrier A | | | | | ed 104 kHz 108 kHz 4 kHz G kHz Using the 3586A/B SSB Measurement ” Menu” Once the SSB CHANNEL measurement mode has been entered, the following measurements can be made at a keystroke or two. (3586A/B) (3586A) rims BOs (3586A/B) Cama Idle channel noise with C-message or Psophometric equivalent noise bandwidth, channel power or audio demodulation for listening or further measurement, Measures 800 Hz or 1 kHz tone level on 3586A instruments. 20 Hz bandwidth is chosen automatically. Measures 1 kHz tone or 2600 Hz in band signalling tone with 20 Hz bandwidth. Measures carrier leak with 20 Hz filter to reject an adjacent pilot. Instruments with Option 003, Transmission Impairments Note: Tone and carrier measurements are identical on 3586A/B standard and option 003 in- strurnents. [dle channel noise. channel power, or audio demodulation with 3100 Hz flat filter. Measures phase jitter on a 1 kHz tone in the channel with 2° p-p resolution. Compatible with BSP 41009 (3586B) or CCITT Recommendation 0.91 (3586A). GUTTER Measures noise on a channel with a 50 dB notch to notch out a 995 Hz to 1025 Hz test tone. Tone is used to activate com- panders on a telephone circuit for a valid noise measurement. (blue key) NOISE/ TONE Initiates impulse noise measurement. Enter a threshold 45 dB or more below full scale in 1 dB steps, and a time period up to 99 minutes. 59 seconds. (0 time determines a continuous measurement.) Press START to begin measurement period. Counts all impulses above threshold during the time interval entered. Compatible with CCITT Recommendation 0.91 (3586A) or BSP 41009 (3586B). {blue key) -) @ 8 PULSE we START Measures Psophometric weighted noise (3586A) or C-message weighted noise (3586B). A weighted filter is placed in series with the 3100 Hz flat filter. NOISES DEMOD WT IO o P00HMZ SSB Measurements with the 3586C The 3586C Selective Level Meter is designed for general purpose wave analysis and telecommunications measurements and has a simplified front panel. The 3586C is basically the same as the 3586A/B without the SSB CHANNEL measurement mode, carrier or tone ENTRY FREQUENCY, or provisions for impairments measurements. A 50 ohm impedance has been added and 124/135 ohm or 150 ohm removed. The 3100 Hz channel filter is standard on the 3586C. Although it has a less sophisticated front panel, the 3586C can still be used for the SSB channel measurements of pilot, carrier and tone levels, idle channel noise, channel power, and demodulation. For tone measurements, the frequency of the pilot, carrier, or test tone is entered and measured directly. Use the frequency counter as described above in SELECTIVE MEASUREMENTS and make final measurements with the 20 Hz filter. Noise and demodulation measurements require a calculation of the proper frequency offset from the carrier. 13 3586C SSB Measurements Measurement Procedure Carrier Enter the frequency to be measured directly, Pilot Tone use 20 Hz BW. Use the counter and E Test Tone with a wider bandwidth to start if frequency error prevents direct use of the 20 Hz BW. Enter the carrier frequency offset + 1850 Hz LSB or USB demodulation. Choose E or Bl for proper product detection {demodulation} of the channel. Use the widest bandwidth Signalling Tone Idie Channel Noise Channel Power Demodulation Wideband Power Measurements In the WIDEBAND mode, the 3586A, B or C becomes an RMS power meter with a power level range of — 45 to — 120 dBm and a frequency range of 200 Hz to 32.5 MHz. The RF energy entering the 50 or 75 ohm input is fed directly to the RMS detector through the RF amplifiers and attenuator without any filtering. Use this mode for measuring baseband power entering a microwave radio link, or integrated signal and/or noise power over any band within the 200 Hz to 32.5 MHz frequency range. Wideband power accuracy: 200 Hz 10 kHz | +20dB | +1.0dB 10 MHz 32.5 MHz +2.0 dB Convenience Features The power of microprocessor control is most evident in the wide variety of operator convenience features available on the 3586A, B and C. In addition tó those covered in the previous section, convenience features include: * Level Offset * Automatic Level Calibration * Choice of Measurement Units * Averaging * Stepping of Frequency, Full Scale, Time and Threshold Levels * Analog Frequency Tuning * Nine Storage Registers * 10 k0/Bridged Inputs * 10 dB/100 dB Range * Auto or Entry Full Scale Level Offset This powerful convenience feature allows the operator to enter a reference amplitude level and make level measurements relative to the entered reference. Alternatively, a measured level can be entered into offset and subsequent measurements made relatively. This feature speeds harmonic level testing. Entering an offset level or TLP (Test Level Point} in FDM systems is accomplished by: * Enter an amplitude level offset up to +199.99 dB by pressing MB in the ENTRY block, followed by the numerical entry, followed Бу (55) ог (=) [| * Turn on OFFSET in the MEASUREMENT ENTRY block. * The display will show the OFFSET entered. * Press 18, | {now flashing) to resume measurement. The LED display will add the unit suffix "0" to indicate the level display is " offset.” NOTE: If units are changed, a new offset must be entered. To make a measurement relative to a measured level: s Measure ire the level to be used as a reference. * Press © a to enter the measured level into offset. i a “LA #5 wh and | Ls) {now flashing) to resume measure ment. | Subsequent level measurements will be relative to the measured level entered into offset. Automatic Level Calibration When ON, all ranges and bandwidths are calibrated using an external voltage standard. In the LOCAL (non-HP-IB) mode, a CAL occurs once every three minutes, or when the frequency is changed > 1 MHz, at turn-on, or when the CAL is turned off and on, During remote (HP-IB) operation, CAL occurs only for > 1 MHz frequency changes, or when enabled by bus command. If a calibration error is found, a CAL error code is displayed (see Operating and Service Manual for code definition). Choice of Amplitude Level Units Use units of dBm (dB relative to 1 mw) for tone or noise level measurements in any impedance. dBpW (dB relative to 1 picowatt) is often used for noise measurements in FDM systems. 0 dBpW = -90 dBm = 0 dBrn (dB relative to reference noise} in a 3.1 kHz channel. dB .775V (dB relative to .775V) is useful for measurements in 6000 systems. 0 dB .775V = 0 dBm in 6000. The dBm and dB.775V units on the 3586C are identical to the 3586A/B. dB1V (dB relative to 1 volt) units are included in the 3586C for general purpose selective voltmeter measurements in systems using voltage instead of power units. Averaging The averaging feature changes the normally displayed five measurements per second to a display of five averaged measurements once per second. Averaging reduces LED level display "racking" caused by externally applied noise, internal noise for small signal measurements near the noise floor, or the beat frequency of closely spaced signals in the bandpass. The accuracy of the averaged measurement depends on the signal to noise ratio and to some extent, the nature of the noise. A noise impulse or spike, for example, could cause a displayed average to be considerably different than previous or subsequent averages. Stepping Frequency, Full Scale, Offset, Threshold and Time Stepping Frequency is easily accomplished by: ha e Press ES * Enter a frequency step from .1 Hz to 32.5 MHz with 1 Hz resolution. * Press £. to return to the frequency display. * Increment the frequency up or down with © ©. The frequency stepping feature is especially useful for repetitive channel to channel FDM measurements where a 4 kHz step is normally used, in harmonic measurements or frequency response measurements, Analog Frequency Control FREQUENCY STEPPING with the FREQUENCY TUNE control is accomplished by entering the frequency step as accomplished above and pressing a . The frequency will be step as the tuning knob is rotated. When = is pressed, the step is automatically chosen as .2, 4 or 20 Hz steps depending on bandwidth. Bandwidth | Auto Tune Resolution 20 Hz .2 Hz 400 Hz 4 Hz 1740 Hz 2000 Hz 20 Hz 3100 Hz The FREQUENCY TUNE control is often very useful for fine tuning and manual frequency surveillance such as spur sear- ches. STEP FULL SCALE SETTING: * Press tl to enter the manual entry mode. M to display the full scale setting. * Step the full scale setting in 5 dB steps with the ©) [5] keys. 2, to resume measurement. STEP OFFSET LEVEL: * Press MY to display the offset level. e Step the OFFSET level in 1 dB steps with (2) (>). e Press ($) to resume measurement. STEP IMPULSE NOISE THRESHOLD LEVEL: (3586A/B Option 003 only) * Press to display the threshold level. * Step the threshold in +1 dB steps with (>)(5).. MEAS e Press to resume measurement. * The range is + 199 dB. STEP IMPULSE NOISE TIME: (3586A/B Option 003 only) e Press to display the impulse noise time period. e Step the time in +1 minute steps with EE) * Range is Ô to 99 minutes, 59 seconds.(A O time entry enables a continuous measurement). HEAL * Press | 5. to resume measurement. NOTE: The impulse noise threshold and time can be set with the instrument in or out of the impulse measurement mode. Nine Storage Registers Nine different front panel settings can be stored in registers 1 through 9 for repetitive testing. The registers are non-volatile so that registers may be recalled indefinitely even after the instrument has been turned on and off. Register O permanently contains the instrument turn-on front panel settings. To store a front panel setting: * Press == followed by an integer of 1 through 9. Storing into a previously used register will write over it. To recall a front panel setting: * Press "1 followed by an integer of 1 through 9. Recalling another register does not disturb previously recalled registers. In addition to repeating different measurement configura- tions, the storage registers can be used to "step” a parameter in non-uniform steps. For example, a pilot frequency sequence in an FDM system. 10 k{ and Bridged Inputs BRIDGED SOU © 75Q unbalanced inputs (and 509 on the 35860) can be unterminated for high impedance measurements on terminated circuits. This method results in a tapping loss of less than 01 dB when used as shown below. 750 ео < dass, ——J 0 ¿Bm A 7501 $ Load o Y © Г e Ш | 10k | { HP 3586 . | dBm and dBpW measurements will be referenced to the impedance of the input used, 6009 balanced inputs are unterminated or "bridged” by 10 kQ and 50 pF when in the BRIDGED mode. Since the impedance ratio is smaller, the "tapping loss” 5 — .26 dB. If desired, the tapping loss can be entered as an offset to correct the measurements. High-frequency, high-impedance measurements often require an active probe to prevent capacitive loading by the test lead. 10 dB and 100 dB Range The 10 dB range results in the most accurate level measurements since all signals are automatically ranged to be detected on the most linear region of the Detector/Logger operating range. Resolution is .01 dB on the 10 dB range. This range is most often used. The 100 dB range selects the entire 80 dB operating range of the internal Detector/Logger, resulting in faster tuning and easier " peaking” of signals when manually tuning. Accuracy is substantially reduced and the resolution is .1 dB in this mode. The 100 dB label for this control refers to the meter scale. This mode is used only when there is an established need for it. Full Scale Auto or Entry In the FULL SCALE AUTO mode, the internal gain levels of the 3586 are automatically adjusted for best signal to noise ratio. This mode is used most often. Full Scale Entry can be chosen for specific operatina cir- cumstances such as to eliminate fluctuations caused by a varving signal (use 100 dB range), to eliminate full scale autoranging time in automated production, and to improve signal to noise ratio by over-driving the instrument past a fixed full scale. The latter resultsin increased harmonic and inter-modulation distortion but can be useful in some applications. General Purpose Wave Analysis The innovative nature of our customers makes an exhaustive presentation of SLM applications impractical. Here are a few that may be helpful, or suggest additional applications to you. It is suggested that the preceeding section, Basic Measurement Modes, be read prior to this section. Harmonic Levels ta 101 202 303 404 ® Measure the fundamental output level of the source (Ex- ample, 1.01 MHz, — 10 dBm). * Press Counter followed by to measure and tune to the precise fundamental frequency. Turn the Counter off. * Enter the fundamental frequency as a frequency step by pressing MES and entering 1.01 MHz (or the precise counted frequency). * Press EXA and successively for the 2nd, 3rd, 4th, etc., harmonic levels. * To measure harmonics relative to the fundamental, add to step 2 above: Press [25] followed by OFFSET and Each of the subsequent harmonic levels will be measured in dB relative to the fundamental level. High Impedance Accessory Probes The 3586A, B and C Selective Level Meters are designed to operate with the HP High Impedance probes shown below. Probe power is provided by a front panel connector on the 3586A, B or C. The use of a high impedance probe allows measurement of levels in terminated unbalanced lines without loading down the circuit significantly. Other probes may be used with external power supplies if the frequency ranges and connectors are compatible. Probe SIM Probe Frequency Tapping Output Model Model Range Loss (750) Gain Impedance Connectors 3586A | 15880A 20 kHz to < 15 dB O0 + 2 dB 759 BNC 3586B | 15880A/004 25 MHz (50 kHz to 20 MHz) | (50 kHz — 20 MHz) 754 WECO 440A 3586A | 158818 < .25 dB 20 dB + 2 dB 759 BNC 3586B | 15881B/004 (50 kHz to 20 MHz) (50 kHz — 20 MHz) 759 WECO 440A 3586C | 1124A DC to 100 MHz input Impedance X10, X100 508 BNC Input 10 MQ/10 pF +5% Probe Tip Output 15 Inter-modulation Products Inter-modulation products are the result of the mixing of two or more signals in a non-linear element. Since all active net works exhibit non-linearity to some extent, inter-modulation products are always present with two or more signals. Inter- modulation products will occur at sum and difference of the fundamental and harmonic frequencies, i.e., at mf; + nf when two signals are present. The two most common sets of inter modulation products of concern are 2nd order (fi + 1), and 3rd order (2f; — £2) and (267 — fi). M2f; + £) and (26 + f1) are normally too high in frequency to be of concern] Ind Order IM Produc fs A | : | | 05 25 Levei —m | a FE В + Frequency ————. 3rd Order IM Products Usually out of Band f: в 2-1, AA 26+ 6 26 - fy Higher ot à 4 À Order ¡Ms A +1; A af af af A A | A i i Frequency ——————— fun Levei ——— The relative level of IM products compared to the fundamental levels is an important measure of the linearity and filtering properties of a test network. Note that fi and fo do not have to be at the same level, although they usually are for measurement convenience, Relative inter-modulation Levels {No Harmonic Cancellation} Intercept Point / DER Non-Linear 7 Region Output Level dB —W Input dB The input/output characteristics and relative IM levels of a typical active network is shown above. Inter-modulation products are often specified in terms of an intercept Point output level, or in a specified IM level for specified fundamental levels. (If harmonic cancellation is present in a network then the 2nd and 3rd order intercept points can be at different levels, for example, in a balanced mixer.) Measuring Inter-modulation Products with the 3586A/B/C Inter-modulation measurements are made quickly and easily by taking advantage of the frequency step, storage registers, counter, and offset features of the HP 3586 SLM. 2nd Order IM Measurement: * Connect as shown. * Enter the frequency of the fundamental signal, f;, used as a reference. * Choose a bandwidth smaller than the spacing between the two test signals so only one at a time will be measured. 16 inter modulanon Measurement : Pad î PA nette 3336 Synthesizer ri Device Linde: Test | tr : Figs + ; | SLM iaa a: 3336 Synthesizer CHI ERES * Use the COUNTER and | to count and tune precisely to the reference signal frequency. * Change bandwidth to 20 Hz (or wider if the frequency stability is > +3 Hz). * Press [157 and turn on OFFSET for measurements relative to the fundamental. (Delete if readings are desired in absolute level.) * Press and | to store the f; settings. * Enter the frequency of fo and (sw) the fo setting in register 2. Use [77 if required. * Enter the frequency of fi + fe and (sw) in register 3. * Enter the frequency of f, — fs and in register 4. * Each of the frequencies may be measured again by pressing and the appropriate storage register. e Note that if the reference frequency level OFFSET is changed, the new OFFSET value must be re-entered into each storage register if the registers are used for repetitive measurements, 3rd Order IM Measurement This measurement is simplified by the fact that inter- modulation products are spaced AF from the fundamental signals, where AF is the spacing between the two fundamental signals. * Enter the frequency of fi, use the widest bandwidth. e Press MT and , and reduce bandwidth to 20 Hz if signal stability permits. ® For relative measurements, press fax: and turn on OFFSET, * Enter a EA equal to the difference frequency. * Step up or down sequentially to measure each IM product or fe, step to each product and/or [se] in separate registers. Measurement Limitation * Internally generated IM products in the 3586 are specified at 75 dB below full scale for AF 20 kHz to 1 MHz, and 70 dB for AF 200 Hz to 20 kHz. To maintain these specifications, use FULL SCALE AUTO and stay within the 3586 SLM amplitude range. * Sufficient isolation between test sources must exist to reduce IM products caused by cross talk between test sources. * Fundamental signal levels must be set accurately to obtain best measurement accuracy since IM levels change by 2 or 3 times the fundamental level change. * Test source harmonic levels must be 10 dB lower than harmonic levels generated in the device under test for accurate results. Use low pass filters, if necessary. Modulation Distortion It is often desirable to measure the harmonic levels generated by AM or SSB modulation of a transmitter. This is most easily accomplished by modulating the transmitter at rated modulation {usually 80 to 100 preent) with a test source with harmonic out put well below the desired specification. (An HP 239A Low Distortion Oscillator is an ideal audio test source since harmonic levels are 90 dB down to 20 kHz.) The 3586 SLM can then measure the modulation tone fundamental and harmonic levels. À i ROA - A kHz = 2 kHz ~ 1 kHz i +] Test Setup for AM Modulation Distornon HP 239A Oscillator * Set the output of the test oscillator to the audio level required for rated modulation at 1 kHz. * Enter the RF frequency plus 1 kHz into the 3586 SLM for AM or USB modulation, or minus 1 kHz for LSB modulation. * Measure the 1 kHz sideband level using the 20 Hz filter. Manual tuning may be required to peak the 20 Hz filter on the sideband. * Use the COUNTER and f- to tune precisely to the sideband. * Press E and turn on OFFSET to reference readings to the fundamental sideband level. i HP 3586 fo SLM Transmitter AM in * Entera 4 of 1 kHz. * Measure modulation harmonic relative levels using to step to each harmonic. * % modulation (envelope) distortion can be calculated as follows: % distortion = 100 (antilog A; 10 + antilog Az, 10 + .. An 10) Where Ay is the relative level of the Nth harmonic in dB (power), Measurement Limitation: * The inter-modulation suppression of the 3586 SLM is 70 dB below full scale. * The harmonic suppression of the test oscillator should be 10 dB better than the desired spec level. * The modulation level for rated % modulation must be accurately known to achieve accurate results. Spurious and Other Signals Close to the Noise Spurious signals from an active network are generally unwanted signals that are not harmonically related to the output signal and may exist at the output with or without a signal present. Spurious with signal are often caused by modulation of the output by the 50/60 Hz line frequency and harmonics or by modulation or inter modulation of the output with other inter nally generated signals such as 20 kHz power supply switching frequencies or local oscillator fundamental and harmonic out puts. Spurious outputs without signal are the same internally generated signals, not mixed with or modulating the output signals. Another type of spurious is caused by an oscillation of a part of the test network which may or may not require a signal to ex- cite if. In a well-designed network, spurious signals are generally close to the noise and, therefore, require careful measurements for accurate results. Guidelines for spurious testing: * Store known spurious frequencies in storage registers for quick recall. * Use the LO NOISE mode to reduce the noise floor by up to 5 dB when the FULL SCALE setting is greater than — 35 dBm. * Use the narrowest bandwidth possible consistent with the spurious frequency stability. This will usually establish the lowest noise floor and best selectivity. * Use AVEraging to reduce LED display racking. Note that the averaged reading can be higher than the actual signal level due to the addition of averaged noise when the signal is close to the noise. * Note that the noise floor can exceed minimum levels if the FULL SCALE setting is too high. For example, with 20 Hz bandwidth, the noise floor above 100 kHz is — 120 dBm for a FULL SCALE setting of — 35 dBm or lower, but is specified at 80 dB down (— 90 dBm in for — 30 dBm full scale) for higher FULL SCALE settings. Noise Floor (Full Scale setting — 35 to ~ 120 dBm): Frequency Bandwidth Noise Level 3100, 1740, or 2000 Hz | - 116 dBm 100 kHz to 32.5 MHz | 20 Hz, 400 Hz — 120 dBm 10 kHz to 100 kHz All Bandwidths — 105 dBm The noise floor for full scale settings of — 30 to +25 dBm is 80 dB below full scale for > 100 kHz, and 60 dB below full scale for < 100 kHz. The FULL SCALE setting is determined in AUTO by the total power entering the 3586 front end. If this setting results in the noise floor exceeding the desired level, some improvement is possible by allowing the input to be over- driven. To achieve this improvement, the FULL SCALE setting in AUTO must be < —50 dBm, > — 105 dBm if the 10 dB range is used, or < — 35 dBm if the 100 dB range is used. (The 100 dB range sacrifices measurement accuracy.) If the above conditions are met, choose the 10 dB or 100 dB range and follow these steps: — Find the FULL SCALE in AUTO by pressing ES . — Change to FULL SCALE ENTRY mode and enter a FULL SCALE 5 dB lower. This reduces the noise floor 5 dB {for FULL SCALE settings > — 105 dBm). — Continue reducing the FULL SCALE SETTING as long as the reading drops. The lowest level reading occurs when the signal to noise ratio is the highest. Measurement Limitation When using the above overdrive procedure, the inter- modulation and harmonic performance of the 3586 SLM will be degraded. This will only be a problem when inter-modulation or harmonic signals fall near a desired measurement frequency, in- side the bandwidth chosen, Frequency Response Testing Using the Tracking Output The 3586 SLM includes a rear panel tracking output of 0 dBm + 0.5 dB, at the passband center frequency. The flatness is + 0.5 dB referenced to 10 kHz and harmonics are typically 30 dB down. The output frequency has the same accuracy and stability and resolution as the center frequency specifications. The tracking output can be used for frequency response testing of active or passive networks using the test set-up shown below. Variable 7 Attenuator ¿Optional Input 1 DUT Variable Lo A Attenuator {Optional The optional variable attenuators can be used to adjust the input and output levels to acceptable ranges, if required. The attenuation level required can be entered into OFFSET for display of the actual gain or loss of the network under test. Since the frequency resolution is .1 Hz, high-Q filters and other selective networks can be tested using this approach. Frequency Response Testing Using a Tracking Synthesizer Frequency response testing of active or passive networks can be performed with an external tracking synthesizer, HP Model 3336A/B/C or 3335A Synthesizer/Level Generators. The 3586A/B/C SLM and the tracking synthesizer are put in the frequency tracking mode by connecting their HP-IB interface connectors with an HP 10631 Series cable and setting the REM/TRK switch on the rear panel of the 3586 SLM to TRK, and the 3336 SLG switches to listen only. This puts the SLM in the TALK ONLY mode so that the tracking synthesizer frequency will automatically be programmed to the SLM frequency. While frequency response measurements can be made with the built-in tracking output as described in the preceding section, the tracking synthesizer offers a number of measurement advan- tages including: * Amplitude flatness and accuracy with full range control (No external attenuator required) * Substantially better spectral purity * Remote frequency tracking capability for telephone circuit loop or end-to-end testing. 3586 SLM A336 Synthesizer or 33354 Synthesizer ро РСС 18 Use the 3336A, B or C Synthesizer/ Level Generator for frequency tracking to 209 MHz, or choose the 3335A Suyn- thesizer/Level Generator for full frequency coverage to 32.5 MHz. While any model or option of the two synthesizers will frequency track with any 3586 SLM model. care should be taken to insure connector/impedance compatibility between the two, Selective and Wideband Noise Measurements The true RMS detector used in the 3586A, B and C Selective Level Meter allows accurate, selective noise or channel power measurements in the SELECTIVE measurement mode. or WIDEBAND noise or power measurements. RMS vs. Average Detection Average detection has been traditionally used for noise detection since it is a less costly approach. For purely gausian noise, this works out fairly well providing a 1.05 dB correction is added to compensate for the lower average noise power reading. Un- fortunately, this correction factor changes for complex waveshapes, — for example, the noise in a telephone channel. Consequently some error will result, an error that is unfor- tunately not predictable. The RMS detector used on the 3586A, B and C SLM will measure the RMS value of any waveshape with a crest factor up to 571, or noise characteristics, so it is not subject to the unpredictable error of the average detector and is therefore a more accurate approach to noise measurement. Selective Noise Measurements Noise or channel power measurements can be made in any bandwidth, but are most often made in the telephone voice channel bandwidth of 3100 Hz, with or without C-message or psophometric weighting, or in 1740 Hz or 2000 Hz equivalent weighted noise bandwidths. (See page 21 for a C-message and psophometric weighted noise measurement and equivalent noise bandwidth discussion.) Generally speaking, the SLM measures the total power in the bandwidth chosen, both noise and discrete signal power. Therefore, the guidelines provided above in the BASIC MEASUREMENTS section generally apply. Units of dBpw (dBrn) may be used. (0 dBpw = 0 dBrn = — 90 dBm in 3.1 kHz BW) Specific guidelines for selective noise measurements are: * Use AVE (averaging) to reduce display racking. * Use the LO NOISE (low noise) mode. * Use the spurious testing guidelines on page 17 for using full scale overdriving to further reduce the noise floor if necessary. * Remember that the noise power measured is directly proportional to the noise bandwidth of the filters used. The noise bandwidth of the 3100 Hz, 1740 Hz. 2000 Hz or 400 Hz filters used on the 3586A, B or C SLM can be consider- ed to the 3 dB bandwidth for most applications. This is true only for very selective (low shape factor) filters. To convert noise power measured in a bandwidth to that measured in another bandwidth, use the relationship: BWy N/BW1) in dBm = N(BWo) + 10 LOG BWo dBm + For additional information on noise measurements, see page in the FDM Systems Measurements section. Wideband Noise and Power Measurements The 3586A, B or C becomes an RMS power meter in the WIDEBAND measurement mode with a frequency range of 200 Hz to 32.5 MHz. All noise and signal power within the frequency range is measured, with an amplitude range of — 45 to +20 dBm. Power entering the input passes through the RF input attenuator and RF gain stages and is connected directly to the RMS detector, bypassing all frequency conversion and frequency selective circuitry. The WIDEBAND measurement mode is most often used for baseband power measurements between the output of a Fre- quency Division Multiplex system and the input to a microwave radio. 7 ave Microwave Radio 284 kip FEAT за kHz Baseband Wideband Power Accuracy: After calibration, 100 dB range, averaging on —45 to + 20 dBm. 200 Hz | +2.0 dB 20kHz 10MHz 32.5 MHz +10dB | +2.0dB | Return Loss Accurate return loss measurements can be made using an external return loss bridge and the 3586A, B or C SLM. The SLM is used for a frequency selective, accurate detector. The signal source can be the tracking output of the SLM (for unbalanced measurements), or an external precision level generator, such as the HP 3336A, B or C or the 3335A Synthesizer/ Level Generators, for improved accuracy and for all balanced measurements, Return Loss Bridge (Simplified) input 3586A, B or C | Г SLM ; Optional | Variable Attenuator Tracleng Output Test Load for 50/758 only) | 33364 Bor C or P3330A Test Source The following return loss bridges are available from HP: impedance | Return Loss Bridge 500 SIZ1A 759 S721A Opt. 008 1240 5061-1136 1509 5061-1135 The signal from the test source is delivered to the test load through the return loss bridge with typically 6 dB loss. Any impedance difference between the bridge balancing impedance and the impedance of the test load results in an unbalanced condition. Reflected power from the test load mismatch is then directed to the 3586 SLM. The difference between the reflected power from the test load and the power reflected from a short or open reference impedance is the return loss. The 3586A, B or C SLM is often the best choice for return loss testing since harmonics and spurious signals generated by the test network outside the bandwidth are rejected by the highly selective bandpass filter characteristics. Procedure: * Set the test source output to a level compatible with the test load dynamic range. * Connect a reference short to the test port of the return loss bridge. The test signal will be reflected back to the SLM less the path loss through the bridge. *.Set a 0 dB return loss reference by measuring the reflected level in the SELECTIVE mode with the narrowest bandwidth compatible with the test source stability. * Set the measured reference level to a 0 dB return loss reading on the 3586 SLM by pressing and turning on OFFSET. * Replace the reference short with the test load. The return loss in dB will be displayed by the 3586A, B or C SLM. Measurement limitations: * The directivity of the return loss bridge must be significantly greater than the return loss measured. For example, the measurement uncertainty of a 30 dB return loss measurement with a bridge directivity of 40 dB is approximately +3 dB. The input test signal must be compatible with the dynamic range of the test load. The return loss (or input inpedance) of an active network such as a transistor amplifier or mixer changes with input power if tested at levels exceeding linear operation. Use a bridge with the same connector type and impedance as the test load to prevent mismatch uncertainties. Make sure that the reflected level measured is sufficiently above the SLM noise floor for accurate measurements, Measurements on FDM Systems The 3586A/B Selective Level Meter and the 3336A/B Syn- thesizer have been designed with many features to meet exac- ting FDM system measurements. Worldwide impedances and connectors are available by model, option or special order to meet CCITT and North American (Bell) requirements. Microprocessor design allows many measurements to be made at a keystroke with operator convenience such as level offset. choice of units, frequency counter, analog tuning/ stepping and storage registers available to reduce measurement time. A significant contribution is the ability to make transmission impairment measurements on 3586A/B Option 003 models at both voice grade and carrier frequency for troubleshooting com- parisons. 3586A CCITT | | © y WRB ES | о = The 3336A, B & C Synthesizer Level/Generator brings a combination of frequency resolution, spectral purity, and amplitude level precision not previously available. Additional features include worldwide impedances and connectors, wideband phase continuous sweep, phase offset, AM/PM modulation and frequency tracking compatibility with the 3586A, B and C SLM, This section is a synopsis of basic FDM measurements methods using the 3586A, B and C Selective Level Meter and the 3336A, B and C Synthesizer/ Level Generator. The reader should be familiar with Chapter IV, Basic Measurements, and optionally, General Purpose Applications starting on page 27 before proceeding with this chapter. E eH - 11 [Er hr U ULLI Shown with Option 003 Transmission Impairments E. FDM Channel Measurements at a Keystroke The following measurements can be made by simply entering the SSB CHANNEL measurement mode and pressing the appropriate key. 35868 3586A CCITT No. American (Bell) 1 kHz tone level and | kHz and 800 Hz 2600 Hz signalling tone level tone level * Idle channel noise, audio demodulation, channel power * Carrier leak * Baseband Power Psophometric C-message equivalent equivalent weighted noise weighted noise {1740 Ha {2000 Hz) with Option 003 Transmission impairments ® Phase Jitter on 1 kHz tone * Noise-with-tone (Signal to noise with tone ratio) * Single level impulse noise with or without 1 kHz tone NOTE: Start all procedures from the turn-on condition. Use RECALL 0 to set this condition. Entering the SSB Channel Measurement Mode (See Basic Measurements, page 11 for details) Choose an 800 Hz (3586A) or 1004 Hz (3586B test tone frequency or carrier frequency reference from an FDM plan chart that applies to the channel you wish to measure. For example, 108 kHz is the carrier frequency for channel 1 of a basic group, or 107 kHz is the 1 kHz test one frequency (3586B) ог 107.2 kHz is the 800 Hz test tone frequency (3586A). Enter this frequency and press CARRIER or TONE ENTRY as appropriate. This enters the frequency as a reference frequency for SSB channel alignment. It is not necessarily the frequency the passband is tuned to. * Choose an erect (~) or inverted (4) channel as required. * Press any one of the four keys in the SSB CHANNEL measurement mode, The channel filter will be aligned on the voice channel if NOISE/DEMOD is pressed, or the 20 Hz filter will be aligned on the tone or carrier frequency chosen. Channel Filters provided: 3586A 35868 Standard Option 003 Standard Option 003 1740 Hz 3100 Hz, 2000 Hz 3100 Hz, (Psophometric Wid. 3100 Hz | (C-Message Wid. 3100 Hz Noise Equivalent] Noise Equivalent) 21 Precise Channel Alignment with the Frequency Counter Frequency errors between the measured tone and the center frequency of the 3586A/B SLM can be eliminated by using the COUNTER and [777 keys. The counter measures and displays the frequency of the strongest signal in the 3588 SLM passband to +10 Hz accuracy. The "counter to center frequency” key tunes the precise center of the passbass to that frequency. Use this feature when entering the SSB CHANNEL measurement mode by counting and tuning to either the test tone or carrier frequency used for SSB channel mode entry. Procedure: * After entering the SSB CHANNEL mode as described in the section above, measure the carrier leak or test tone level used for entry by pressing CARRIER or TONE as appropriate. * Press COUNTER and the precise frequency of the carrier leak or tone used will be measured and displayed. * Press 77) to tune the center of the passband precisely to the measured frequency. MELE . * Press| 9, | to resume previous measurement. The channel filter is now precisely tuned on the channel. No further tuning is required. Test Tone Level: * Press 800 Hz or 1010 HZ (3586A). or 100 Hz (3586B) the test tone level will be displayed. 2600 Hz Signalling Tone Level (3586B): * An in-band signalling tone at 2600 Hz is measured by pressing the 2600 Hz key. The 20 Hz filter is used. Carrier Leak * The carrier leak level is measured by pressing 840. The 20 Hz filter is used. An adjacent pilot is rejected 50 dB. Transmission Level Point (TLP Reference) A TLP or any other level within the range can be used as a level reference. The TLP is entered as an OFFSET and subsequent measurements are made in dBm. * Enter the TLP by pressing UB and entering the TLP level (between + 199.99 dBm). Turn on OFFSET . The display will read 00.00 dBmo. * Press 1) to resume previous measurement. Level is displayed in dBmO. Noise and Demodulation Idle channel noise, channel power, demodulated audio on speaker output or 600 ohm front panel output are all enabled by pressing . The 1740 Hz filter (3586A), 2000 Hz filter (3586B) or 3100 Hz filter (3586A/B Option 003) is aligned precisely on the channel. Voice, tones or noise can be heard on the speaker by turning up the volume control. Noise level can be measured in dBm or dBpW. (Remember O dBpW = 0 dBrn = -90 4Вт та 3100 Hz bandwidth) Equivalent Weighted Noise Psophometric equivalent noise is measured in a 1740 Hz bandwidth automatically on the standard 3586A when 1740Hz is pressed. The same key chooses the 2000 Hz C-message equivalent bandwidth on the standard 3586B. The 1740 Hz filter (standard on the 3586A) provides psophometric equivalent weighted noise measurements for CCITT FDM systems. Ît is offset + 1350 Hz from the carrier in the SSB mode and can also be used for general purpose demodulation with reduced bandwidth. The 2000 Hz filter (standard on the 3586B) provides C-message equivalent weighted noise measurements for systems meeting North American (Bell) requirements and is offset + 1500 Hz from the carrier in the SSB mode. The 2000 Hz bandwidth, combined with the RMS detector used, provides a more accurate C-message noise equivalent bandwidth than the conventionally used 1740 Hz equivalent noise bandwidth. The 2000 Hz filter also has the additional advantage that more of the channel is demodulated. The 1740 Hz filter is available as an option on the 3586B; some users will want to maintain histor- ical continuity with past data. Direct weighted noise measurements can be made with the weighted response filter and a 3100 Hz channel filter available as part of Option 003. Equivalent Weighted Noise Bandwidth Psophometric or C-message weighted noise measurements are often made with a filter having noise bandwidth equivalent to the noise bandwidth of a filter having the precise frequency response for psophometric or C-message requirements. This approach has the advantage that a conventional bandpass filter can be used for the weighted noise measurement instead of a more costly noise weighting filter with its special frequency response. The disadvantage is that the equivalent weighted noise filter is accurate for white noise only and does not completely cover the voice band of 300 to 3400 Hz. The equivalent noise bandwidth of a filter is the bandwidth of an ideal rectangular filter that passes the same noise power as the non-ideal filter. Filters with relatively little selectivity will have an effective noise bandwidth (ENBW) that is wider than the 3 dB bandwidth. ~ ENBW — A 1 Guassian Filler #3 dE BY J L : р e ENBW -— Fiat top Filter A flat top filter response more closely approximates an ideal filter and so have an ENBW closer to its 3 dB bandwidth. If a flat top filter has sufficient selectivity, (i.e., is almost ideal} then the 3 dB bandwidth and ENBW are virtually identical. This is the case with the 1740 Hz or 2000 Hz equivalent noise filters on the 3586A/B. 22 1740 Hz vs. 2300 Hz vs. 2000 Hz ENBW Older selective level meters using average detection used a 2300 Hz bandwidth filter with an ENBW of 1740 Hz — the ENBW of a psophometric weighted noise filter. These filters were often on the instrument front panel as "2300 (1740) Hz." For purely white noise and psophometric weighting, this approach yielded reasonable, but somewhat erroneous results. Measurements of C-message equivalent weighted noise often are made with the 1740 Hz ENBW filter since it approximates C-message weighting, although a — 0.5 dB error results. The 3586A/B uses RMS detection and a 1740 Hz ENBW filter for psophometric equivalent weighted noise measurement. À 2000 Hz ENBW filter is used for C-message equivalent noise to precisely provide the proper noise weighting, thereby eliminating the 0.5 dB error. The ENBW of the 1740 Hz and 2000 Hz filters used in the 3586A/B are the same as their 3 dB bandwidths as a result of their excellent selectivity. The 2000 Hz C-message noise equivalent filter has another advantage over using the 1740 Hz filter - more of the telephone channel is covered by the filter so there is less chance of high frequency noise affecting the measurement. For those needing to correlate with past test data, the 1740 Hz filter is optional on the 3586B. Frequency Stepping for Repetitive Channel Measurements Repetitive channel measurements such as carrier leak or tone level measurements can be quickly made on a channel to channel basis by simply entering the SSB CHANNEL measurement mode as described above and using the FREQUENCY STEP feature with 4 kHz steps. (Any step from .1 Hz to 32.5 MHz may be entered.) Step tuning is accomplished with incremental step keys , or the FREQUENCY TUNE CONTROL. Procedure: * Press EJ and enter the frequency step desired. * Press * Use [>][5] to step incrementally up and down in frequency. * Alternatively, press 3 in the FREQUENCY TUNE block ESTER and use the tuning knob. The step keys are also used to step several other functions. See page 14. Pilot Tone Level Pilot tones are measured with respect to TLP, in the SELEC- TIVE MODE with the 20 Hz bandwidth. The adjacent carrier leak 80 Hz away is rejected at least 50 dB. Pilot frequencies can be entered directly, or nine pilot frequencies can be stored in the storage registers for later recall. Alternatively, an 80 Hz frequency step can be used to step to a pilot frequency from a carrier measurement. The 80 Hz step can be stored in one storage register and the 4 kHz carrier step in another, 6 Baseband Power The WIDEBAND measurement mode is used to measure the baseband power input to a microwave radio. In this mode the 3586A/B is an RMS power meter measuring the total power in- stantaneously over the 200 Hz to 32.5 MHz frequency range with an amplitude range of — 45 to + 20 dBm. See page for additional information. Transmission Impairments Measurements with Option 003 instruments with Option 003 included make all these additional impairments measurements at voice frequency and carrier frequency. All are made through the 3100 Hz channel filter included in Option 003. * Phase jitter * Weighted Noise with 3100 Hz channel filter and direct weighted noise filter. e Noise-with-tone (Notched noise) * Signal to Noise-with-tone Ratio -* Single level impulse noise For the first time, impairment measurements can be made on voice frequency circuits and compared with measurements made at carrier level. This allows more complete troubleshooting — and with one instrument. Phase Jitter Phase jitter measurements can be made on any signal up to 32.5 MHz, —65 dBm minimum (or lower with reduced accuracy}, that can be demodulated to 990-1030 Hz in the voice channel. The resultant phase jitter is filtered with a bandpass of 20 to 300 Hz as required by CCITT Recommendation 0.91 and BSP 41009. Accuracy is + 10% plus a residual phase jitter of 0.5° p-p. Example: Measure the phase jitter on a 1010 Hz tone in channel 1 of a basic group. * Press TONE FREQUENCY ENTRY . * Enter a tone frequency of 106.990 kHz, (the frequency of a 1010 Hz tone in channel 1). e Choose an inverted channel . Press PHASE JITTER . * Read display in degrees p-p To measure phase jitter on a carrier simply enter the carrier frequency as if it were a TONE FREQUENCY. This method is used to measure phase jitter on any frequency up to 32.5 MHz. Weighted Noise Measurements Psophometric (3586A) or C-Message (3586B) weighted noise measurements are made by superimposing the weighted filter characteristic over the 3100 Hz channel filter. The measurement is made in the single sideband mode by entering the channel carrier or tone frequency, choosing inverted or erect channel and pressing 7 . The 3100 Hz channel filter is automatically chosen in NOISE/DEMOD. Use units of dBpW for noise measurements in dBpWP or dBrnC, use the offset feature for measurements in dBPWPO or dBrnCo. 23 Noise with Tone Noise measurements on a channel with a 995 to 1025 Hz tone present, (notched noise) can be made by following the procedure for weighted noise except the blue key [©] is pressed followed by << to activate the notch filter. The filter is compatible with CCITT Recommendations and BSP 41009 requirements and provides 50 dB rejection. The measurement can be made with or without noise weighting. Signal to noise-with-tone ratio measurements are easily made by using the amplitude offset feature. Example: Measure the tone level, press OFFSET = . and ke. This references subsequent measurements to the signal level. Measure noise-with-tone as described above. The display will read the signal to noise-with-tone ratio directly in dB. i : | ; ods | CRIE AE бр В Aa heya 0 hi Le Noise with Tone | Impulse Noise Measure single level impulse noise at any frequency up to 32.5 MHz with the 3100 Hz bandwidth. The measurement can be made with or without a 1010 Hz + 15 Hz tone present. The notch filter is automatically inserted. The threshold level can be set in 1 dB steps from 0 to 50 dB below full scale, or greater than — 80 dBm. Counting accuracy is 1 dB, and the time period can be set from 1 to 99 minutes, or continuous. To make an impulse noise measurement, enter the SSB CHANNEL measurement mode by entering the carrier or tone frequency and or @ for the channel chosen and press the blue key and IM- PULSE . Press THSHLD and enter a threshold level from 0 to 50 dB below full scale. Press and enter 1 to 99 minutes, or enter 0 minutes for a continuous time duration. Press Sar to begin the measurement. Pressing START a second time will terminate the measurement prior to the completion of the time duration entered. Impulses greater than the threshold level will be counted up to a maximum of 999 counts. Impulses wider than 125 ms (3586A) or 143 ms (3586B) will result in multiple counts. Slot Noise, Noise Power Ratio (NPR) Slot Noise Measurements Noise measurements are often made in unused frequency bands or " slots” between groups or supergroups. This measurement is often called "intergroup” or "intersupergroup” noise. This measurement provides a measure of the noise performance of an FDM system under loaded conditions. Slots are typically 48 to 56 kHz wide. One or two channel widths on both sides of the slot are not used. Unweighted noise levels are measured in ten “channels” in the center of the slot and the results averaged, A typical slot is shown below. Master Gruun 2 Ra E e к = = , 2 É о 5 Во This measurement is easily made with the 3586A/B in the SSB CHANNEL NOISE/DEMOD mode using 4 kHz steps. Noise Power Ratio (NPR) Since an FDM system contains a very large number of signals, it is not practical to measure inter-modulation products on a signal-by-signal basis. The NPR test was developed to provide a measure of inter-modulation and thermal noise levels by simulating system loading with white noise. NPR is defined as "the ratio, expressed in decibels, of the noise in a test channel with all channels loaded with white noise to noise in the test channel with all channels except the test channel fully noise loaded.” It can be shown that white noise loading the FDM system will cause noise simulating inter-modulation noise to appear in the test channel (in addition to thermal noise) with the noise load input rejected in that channel only. The test is usually performed on a baseband-to-baseband basis with the white noise generator providing a noise power of P{dBm0) = — 15 + 10 Log N for N > 240 channels. High pass and low pass filters allow only noise in the FDM frequency range to be applied. The selective level meter is used to measure the noise in the test channel under the two conditions. The test channel frequencies are specified in CCIR REC. 399-1 and according to number of channels and the band limits. High Low Pass Pass Filter Filter White Noise 4 1 wy (3 * Co 5 * Generator Reject Filter 3586A/B | | AUN SLM Radio —— Radio Baseband | Baseband Output input After the proper noise generator output is determined according to the number of channels and band limits, the 3586A/B SLM is used as follows: * Use CARRIER or TONE FREQUENCY ENTRY and SSB CHANNEL NOISE/DEMOD measurement mode to set the channel filter on the test channel with the band reject filter on the noise generator enabled. * Press and о establish the measured level as a reference. * Disable the noise generator reject filter. The display now reads the NPR in dB. Cross Talk Measurements ? Crosstalk in a telephone system can be defined as "the presence in a telephone receiver of unwanted sounds from another telephone conversation.” There are three causes of both intelligible and non-intelligible crosstalk in an FDM system. The first is non-linear performance resulting in inter-modulation products. Second is poor frequency response caused by poor filter selectivity or alignment. The third is direct coupling between transmission media. Generally, noticeable crosstalk should be present in less than 1% of phone circuits randomly connected to a subscriber. ‘Freeman, Roger L., Telecommunication Transmission Handbook, 1975, John Wiley & Sons. Inc. “Members of the Technical Staff, Bell Telephone Laboratory, Trans- mission Systems for Communications, Western Electric Company, Inc. 1971 Non-linear performance is measured by inter-modulation or NPR measurements. Both poor filter selectivity and direct coupling can be measured by introducing a tone in a test channel and then measuring the level in adjacent channels where adjacent may mean in the same group (close in frequency); or adjacent physically, such as capacitive or inductive coupling of circuits in a common cable. AN ; inductive and Capacitive Coupiing Crosstalk of Adiacent Cir- cults on a Cable Adjacent Channel Crosstalk from Poor Filter Selectivity. НА ТОМЕ CARRIER intermodulation Crosstalk in Adjacent Channels The precise procedure used for crosstalk depends on the type of crosstalk being measured. The convenience features and excellent filter selectivity of the 3586A/B SLM combined with the high signal purity and accuracy of the 3336A/B Syn- thesizer/Level Generator make an ideal combination for crosstalk measurements. Crosstalk measurement guidelines: * Use OFFSET to make measurements referenced to test tone level or TLP, * Use 7 to step to desired measurement channel: or alternatively STORE various test channel frequencies in the nine available storage registers and RECALL as desired. * identify crosstalk signals in the test channel with the freguency counter * Use the 20 Hz bandwidth to isolate the crosstalk test signal from other possible close-by signals. 3336A,B & C Synthesizer / Level Generator Basic Operation Operation of the HP 3336A B,C is quite straight forward. This section is intended to help you take full advantage of its versitility and capability. A quick reference summary of features is shown on pages8 and 9. Frequency Range of Outputs The overall frequency range of the 3336A, B& C is 10 Hz to 20.999 999 999 MHz; however, only Ше 509 ог 759 outputs are specified over this range. The 124Q, 1350, 150Q and 600Q balanced outputs are fully specified over frequency ranges normally required by equipment specifications. Any impedance output can be used over the full 10 Hz to 20.9 MHz frequency range with reduced performance outside the specified range. The 600Q output performance is probably not useful above 1 MHz. The frequency range for each output is actually somewhat higher when the full resolution is included as shown below: Output Frequency Range 509, 759 10 Hz to 20.999 999 999 MHz 1247 10kHz to 10.999 999 999 MHz 1352, 1508 10kHzto 2.099 999 999 MHz 6000 200 Hz to 109.999 999 999 kHz Frequency Resolution Frequency resolution is 1 microhertz (.000001 Hz) up to 99.999 999 999 kHz and 1 millihertz (.001 Hz) from 100.000 000 kHz and above. Data Entry Frequency, amplitude and phase offset can be entered by pressing the chosen ENTRY key, the numeric keys and then the entry suffix (MHz, dBm, degrees, etc.). DATA 7 EXT ACT : CLEAR + = 3 wf y & Tam = un 141 © 1 EC че FA ci Y Ë sir Tin Bas GEE i For example, a —~ 10 dBm, 20 MHz signal is entered as foliows: FSECUENCY . |— 7 |» | 0 | pn [i | AMPLUTUDE If an entry error is made, "ERROR" is displayed and the previous operating state is retained. As long as the same entry (frequency, amplitude, etc.) is retained, numeric and suffix en- tries can be continuously made without again selecting the entry parameter. If a frequency outside the specified frequency range of an output is chosen, "F LIMIT ERROR" is momentarily displayed as a warning but the entry will be accepted. 25 Amplitude Range, Resolution Amplitude resolution is always .01 dB over the ranges shown: Output Amplitude Range 500 — 71.23 to +8.76 dBm 759, 6000 — 72.99 10 + 1.00 dBm 124, 135, 1500 | - 78.23 ю + 1.76 аВт Using Modify to Tune Signal Parameters The MODIFY key and analog tuning control is a convenient means for tuning any one of seven signal or sweep parameters being displayed. The parameters are: FREQUENCY SWEEP START FREQUENCY AMPLITUDE SWEEP STOP FREQUENCY PHASE SWEEP TIME SWEEP MARKER The modification is "real time,” that is, any changes entered will immediately modify the signal output. Here's how it works. You simply select the digit you want to modify with these keys (>). The digit selected will flash brightly. The analog tuning control will increment the digit up or down with automatic carry-over. Using the 3336A, B or C as a Sweeper The sweep capability of the HP 3336 is one of its outstanding features. You can be sweep logarithmically and linearly over the full frequency range of any waveform with synthesizer precision and + .15 dB leveling. BAKA SINGLE FREC TIME RAT MF Ce CE Lea LEVEL farc Not only is the sweep fully synthesized, but it is phase continuous over the full range. This means that it sweeps like a VCO without the phase discontinuities commonly found with sweeping or stepping synthesizers. This is an important feature when testing phase lock loops or other phase responsive deivces. Sweep time can be set from .01 to 99.9 seconds, depending on the parameters chosen as shown below: Sweep Time Range | Sweep Parameter 01 — 99,99 sec. Linear .1 — 99.99 sec. Continuous Log 2 — 99.99 sec. Single Log Internally Leveled Output The output level of the HP 3336A is internally leveled to +.15 dB in the NORMAL LEVELING mode for a .5 second sweep time, and in the FAST LEVELING mode for a .03 second sweep time. The NORMAL LEVELING mode is normally used for the lower frequency range of 50 Hz to 1 MHz where a slower loop response time is required. The FAST LEVELING mode is used for the 10 kHz to 20 MHz frequency range. Note that using the LEVELING modes in the incorrect frequency range will result in degraded output level accuracy, flatness and distortion. To change from one leveling mode to the other, press the BLUE shift key followed by fast leveling. External Amplitude Leveling An external amplitude leveling input {located on the rear panel! allows regulation of the amplitude at a remote point, This input, marked EXT LEVEL. has a nominal input impedance of 1 kilchm. A +1 volt change at this input causes a + 25 dB change in the output. Sweep Marker The HP 3336A also includes a sweep marker. Unlike some sweepers which provide a marker birdie superimposed on the main signal output, the 3336 provides a negative going TTL transition signal on a separate rear panel output. The position of the marker on a swept display can be chosen anywhere between the sweep end points by defining the marker frequency with the MKR FREQ key. A swept passband filter with the marker connected to a second channel is shown in Figure ñ "EEE EE Ea a НЙ Passband Filter Response with Marker Using the Storage Registers The HP 3336A has ten storage reigsters, 0 through 9, that are accessible from the front panel DATA group, another time- saving feature. DATA AMPLD DOOM or жж „+ = Lan o > Г STORE” ê-8 ; и Bi Ane res | - | in к г % = Е FERS OOO DOME Any combination of front panel parameters except phase offset can be stored by pressing and then a numeric of 0 or any number up to 9. When needed, the parameter set can be recalled by pressing and the register number. The parameters stored in that register (except phase) will maintain their integrity unless a new parameter set is stored into the register or the power is turned off. This feature is particularly useful when a number of repetitive tests need to be made. Operating the Phase Control The PHASE key allows selection of up to 719.9 degrees of absolute phase offset. When phase offset has been im- plemented, the phase of the main output signal (any waveform) will have been advanced or retarded depending on the polarity of the specified offset. Here is how it works. Simply press the PHASE key followed by the numerical entry and then terminate the sequence with the DEG key. An immediate change in phase at the output will be initated. For example: PHASE Le) Le) Lo) Le) will cause the output phase to advance 90 degrees. The modify function can be used to continuously "tune" the phase. 26 A selected phase is relative fo an initial output phase. A second phase selection of 10 degrees, for example, following the 90 degree offset, would return the signal phase to 10 degrees relative to the initial phase. ASGN ZERO 0 is a convenient feature for defining any phase offset previously selected as zero phase (the display 15 set to read zero degrees). Two keystrokes are all that is necessary to set your new phase reference. See page 29 for more information on dual channel phase operation and phase locking instruments together. External Amplitude and Phase Modulation The HP 3336A can be amplitude modulated up to 100% from 50 Hz to 50 kHz. The output can also be phase modulated to a depth of + 850° from dc to 5 kHz. Both AM and PM can be applied simultaneously and both require a modulation amplitude of +5 Vdc for full modulation depth. x + иди A aly IN ¡NN ПОЙ Ца | 4 TT | il I fil! Will” "iu Pa \ === * AM is activated by pressing the blue prefix key and the (4985) key and turned off with the blue prefix and keys. The same sequence with the and (—] keys control the PM. Whenever a modulation control key is pressed, the display momentarily indicates the ON, OFF state of both AM and PM. Sync Output À rear panel SYNC OUT connector provides a squarewave output at a frequency identical to the main output with an amplitude of > 1.2Vy, < 0.2V,, into 502. The squarewave lags the main output crossover by approximately 25 ns and the output impedance is 500. Rear Panel Outputs 21-61 MHz Auxiliary Output This output provides an ac coupled squarewave at a fixed level of nominally O dBm over the frequency range of 21.000 000 000 MHz to 60.999 999 999 MHz with under-ranging down to 20.000 000 001 MHz. The signal output automatically switches to the AUX output whenever frequencies greater than 20.999 999 999 MHz are entered. The front panel "21-60 MHz REAR” annunciator indicates that the AUX output is active. Once the AUX output has been activated, frequencies as low as 20.000 001 MHz can be entered. The output will automatically switch back to the main output when a frequency of 20 MHz or lower is entered. While this output retains the accuracy, stability, phase noise and resolution of the main output, the harmonics are down typically 10 dB, the output amplitude cannot be changed, nor can the frequency be swept. X-Axis Drive This output provides a 0 to 10 V ramp proportional to the sweep frequency for any span width entered. The linearity of the range is < 0.1%, and the ramp is always positive with fran = 0 V and fuen = 10 V. However, fran can be higher than fac for linear sweeps. Sweep Marker Output This output provides a TTL compatible voltage transition at the keyboard selected marker frequency for línear sweep only. See below for ways to use this output for sweeper applications. Z-Blank Output The Z BLANK output voltages are TTL compatible, and the output logic levels are as follows: Linear Sweep: Single: Goes LOW at start of sweep, HIGH at stop, whether the sweep is up or down. Remains HIGH until start of next sweep. Continuous: LOW during sweep up, HIGH during sweep down. Log Sweep: Goes LOW at start frequency, HIGH at stop. In single sweep, remains HIGH until start of next sweep. In continuous sweep, is HIGH momentarily at stop frequency. When the Z BLANK output is low, it is capable of sinking current through a relay or other device. The maximum ratings are: Maximum current sink: 200 mA Allowable voltage range: OV to +45 Vdc Maximum power (voltage at output x currents): 1 W Reference Outputs The 1 MHz reference provides a nominal 0 dBm squarewave output to phase lock other instruments to the 3336A. The 10 MHz oven output provides a nominal O dBm output from the high stability frequency reference oscillator in Option 004 instruments only. This output is connected to the reference input. External Reference Input The HP 3336A may be operated with an external reference to control the standard 30 MHz internal reference oscillator frequency. The external reference level must be greater than 0 dBm (50 ohms), and the frequency must be within 1% of 10 MHz or a sub-multiple thereof, down to 1 MHz (10, 5, 3.33, 2.5 or 1 MHz). The front panel EXT REF annunciator will light to indicate that an external reference is being used. The internal reference oscillator is phase locked to the external reference, and a phase lock detector circuit causes the EXT REF light to flash if synchronization is lost. Options Option 001 3336A — 751) 1.6/5.6 mm metric connector replaces the standard BNC. 3336B — 75Q mates with WECO 358A and 1244 mates with WECO 372A, ("LARGE WECO”). Option 001 should be ordered with the instrument, or retrofitted at the factory for an additional charge. 27 Option 004 3336A/B/C High Stability Frequency Reference Option 001 is a temperature controlled 10 MHz oscillator which provides increased frequency stability over the stendard instrument. The aging rate is 5 x 10% per week or 1 x 107 per month and the accuracy is increased to 5 x 10° over 0° to 50°C. The oven will remain on as long as ac power is connected to the instrument, whether the power switch is on or off. The reference will be within + 1 x 107 of final value 15 minutes after turn-on at 25°C if the oven has been off less than 24 hours. This option can be easily retrofitted into standard instruments. An installation kit, HP P/N 11477A, is available. Option 005 3336A/B/C Precision Attenuator This option significantly improves the level accuracy, flatness and spurious level of the 3336A, B or C (see the Specification Summary, page 10). Option 005 should be ordered with the instrument, or retrofitted at the factory {additional charge). Precision Attenuator Accuracy Attenuator Range 19 30 to ; +. to . + Accuracy Applications See the 3586A, B & C Basic Measurements Mode Section, page 11 for 3336A, B & C frequency tracking applications with the 3586A, B & C Selective Level Meter. The HP 3336A provides excellent wideband, leveled sweeping capability in addition to its performance as a synthesizer and function generator. Because of the single phase lock loop fractional-N design, the sweep is entirely phase continuous and the frequency can be swept over the full range of each waveform with sweep times from Ol to 99.9 seconds per sweep. Since the 3336A is a synthesizer, very narrow frequency bands can be swept, less than .2 mHz for fast sweep times and less than 1 Hz for slower sweep times. The minimum linear sweep width is 2 na/sec x {sweep time) sec. The sweep in the linear continuous mode is from the start to stop frequency and back. The start frequency can be higher than the stop frequency. When used with a plotter or oscilloscope swept display, the Z-blank output blanks the retrace. Log sweep and single sweep are from start to stop only, with no return sweep. The log sweep ramp is a piece-wise linear approximation of a two-piece approximation per decade as shown, The single sweep range is a ten-plece approximation. The more accurate approximation in log single sweep is necessary for accurate swept presentations on a plotter. It is important to note that log sweeps can be made in decade multiples only. [f a start-stop entry other than a decade multiple is made, the sweep will stop at the last decade multiple stop frequency within the sweep width, Another important point to remember is that the actual stop frequency in log sweep will be slightly higher than the programmed stop frequency by no more than 44% as a result of the approximation calculation. The hesitations noted in log sweep occur while the microprocessor is calculating the next approximation. 1 Decade 4 Frequency —— Time -— Bü Two-Piece Log Ramp Approximation, Continuous Sweep 1 Decade § Frequency dm Time ——— Ten-Piece Log Ramp Approximation, Single Sweep Swept Frequency Response of Networks The block diagram shows a typical test set-up for displaying a swept response on an oscilloscope. Marker _ "” X-Drive RE 3336A.B,C Z-Drive Test Unit CH CH A E Swept Response Test Set-Up A typical filter response display including the frequency marker is shown . The vertical position has been set so that only the positive peaks of the waveform as shaped by the filter are shown. This filter has a center frequency of 7.98 MHz and a bandwidth of 8 kHz. The display was achieved by using the convenient marker and zoom {AFx2, AF + 2) functions. к © сш i — — Typical Filter Response with Marker The START and STOP frequencies are set to encompass the passband, 7 to 10 MHz for the example shown. The marker can be set to the passband frequency with the MODIFY function and the passband moved to the center of the display with MKR — CF (marker to center frequency). Successively using AF + 2 reduces the span to obtain the desired response. Obtaining Desired Swept Response a. 7-10 MHz Sweep b. ff | a A c. MKR—CF e — d. AF +2 e. AF +2 Sweep Time Too Fast Note that if the sweep time is too fast (.01s in this case), the display will be distorted as shown. This happens when the fast sweep time causes rapid amplitude changes that are faster than the settling time of the filter. This is especially true with very narrow filters. To set the sweep time to the proper value, start with à very slow time and increase it until distortion is evident and then reduce if to an acceptable value. > Sweep Time is Set Too Fast for Filter to Settle Other Considerations The accuracy of swept response measurements can be adversely affected by other factors such as: Fast or slow leveling incorrectly chosen impedance mis- matches between source or display and the unit under test. Matching networks, pads and terminations must be used so that the instruments used as well as the unit tested are properly terminated. Excessive amplitude causing saturation of active devices such as amplifiers. Swept Measurements Using the 3575A Gain/Phase Meter The 3575A Gain/Phase Meter is an excellent detector to use with the 3336 in the swept mode since it can provide log amplitude as well as phase information from 1 Hz to 13 MHz. The test set-up shown below is configured to measure the amplitude characteristics of a 1.0073 MHz crystal and record the display on an HP 7004B X-Y Plotter. In this case, the sweep time must be very slow, 99 s, because of the slow settling time of the narrowband high-Q crystal. Plotter response time is also considerably more limited than that of the oscilloscope. The sweep width is 4 kHz. A log frequency plot of a 10 kHz low pass filter is shown with a sweep span of 1 kHz to 1 MHz and a sweep time of 99s. Single sweep is used, providing a more accurate log ramp. Care must be taken when making log plots to make sure that the sweep rate is not excessive. This can be more of a problem with log plots because the sweep rate {Hz per second) increases as the sweep progresses. Thus, distortion problems due to settling time would most likely occur at the high frequency end of the sweep, rather than at the low end. X-Drive ее b/a Analog Output 3336 e A wr у a Cup RF Outout REF Test X.Y Plotter Test finn gy a ! Amplitude and phase measurements with the HP 3575A Gain; Phase Meter. 29 $ a = im = «< 1.006 1.007 1.008 1.009 1.010 Frequency MHz Amplitude response of 1 0073 MHz crystal, 704 = 604 5 Чи a + S 40 «< 20 4 1 kHz 10 kHz 100 kHz 1 MHz Frequency Log frequency swept response of 10 kHz Low Pass Filter. Phase, Dual Phase and Synchronizing Phase Offset The phase can be changed with 0.2° increment accuracy and .19 resolution with respect to an arbitrary starting phase, or an assigned zero reference. The phase range is +719.9%. The modify function can be used to continuously step the phase in 0.19 increments. Dual Phase Operation Two Hp 3336 instruments can be synchronized or phase locked together as shown, providing two phase variable outputs. 1 MHz Reference 3336A.B.C о 3336A,B,C À Ÿ А В Dual Phase Test Set-Up At turn-on, both instruments are phase locked together, but are not at the same phase. They can be synchronized in phase by connecting both outpuis to a dual channel oscilloscope or phase meter as shown and changing the phase of one with respect to the other until they are synchronized. The MODIFY function is convenient to use for this purpose. Once synchronized, both phase registers can be assigned zero phase and then either output can be changed with respect to each other, or the zero reference. Synchronizing Other Instruments The 3336 can be used as a phase lock reference to another instrument or a stable reference can be used to synchronize the HP 3336A, B or C. Reference Output The 1 MHz reference output is a 1 MHz squarewave at nominally OU dBm output. Option 004 instruments also have a 10 MHz stable reference which can be used for synchronizing to other instruments. However, this reference must also be connected to the 3336 reference input. A number of instruments may be locked in parallel as long as enough amplitude is available for each instrument. Reference Input The reference input will accept 1 or 10 MHz or any sub- harmonic of 10 MHz (2.5, 3.33, 5 MHz, etc.) for synchronizing the 3336. The frequency must be within 1% of 10 MHz, or 1% N, if a sub-harmonic is used. The typical amplitude locking range is from 0 dBm to + 20 dBm. The front panel REF annunciator light is on when this input is used and it blinks if lock is lost. Linearity Testing of VCO’s The sweeping capability of the 3336 combined with an X-drive output linearity of better than .1% can be used for rapid VCO linearity measurements. The measurement is made by connecting the VCO under test to a phase lock loop and using the 3336 as a linear ramp source with a tracking frequency output as shown. As the ramp (X-drive output) tunes the VCO, the output frequency is compared to the tracking 3336 output in a phase detector (double balanced mixer). The error voltage output is directly proportional to the non-linearity of the VCO compared to the 3336. The error voltage can be plotted on an X-Y plotter for precise measurements or continuously displayed on an oscilloscope for calibration and adjustment. Any VCO can be tested whose direct or divided frequency is within the 10 Hz to 20.9 MHz frequency range of the 3336A, B ог С. v Optional { | inverter To Horizontal tte X. Double Display о Balanced = " Mixer | à To 3336 To Vertical $ X-Dríve Display Output VCO Under Test Connected in Phase-locked Loop The Loop Design A typical VCO from 3336 RF output under test is illustrated below. In this example, the tune input of the VCO requires that the positive sweep output of the 3336 be inverted and offset. This inversion is provided by the operational amplifier. Since the proper polarity of the VCO tune input is obtained from the output of the operational amplifier, the optional inverter is not required. The ratio Rs to Ry scales the sweep while Rg, adjusts the offset. Since the 8 to 12 MHz output of the VCO ia in the frequency range of the 3336, no divider is necessary. To insure proper operation of the phase-locked loop, several guidelines should be followed. The ratio of R: to Ra sets the loop gain or pull range so that the loop remains locked over the entire sweep range. The network consisting of R, and C, may be needed to stabilize the loop. Selection of these components is unique to each loop design. The cutoff frequency of the 4-pole low pass filter should be less than one-fourth of the lowest output fre- +5 o | ra Ren Hs, 1 | у м ; Res ААА ААА м AA | | | ; Under Test м № о 2 ON lid VCO | Вю 18 МН: q & 74307 Tune input E Tex | Eire [*— (20) Balonce T J ica! LA HP 10534 [he | 50 kHz — 150 MHz Display | | 1 333pA RC --- JA = 7 dBm Y J B10 12 MH: To Horizontal Sweep Display 154, 10V К; = 149 КЮ К; = 7Ю0, Во = 5 КО. Вы = 5 К@, Ho. = 10 КО. Ka = 14 КО Phase locked Loop Design for Testing VCO quency of the VCO. An oscilloscope trace of the error voltage of the network is shown below. An Oscilloscope Display of Error Voltage The error curve calibration factor CF in terms of Hz per volt for the display may be determined as: CF = Ri Frune range Hz/ Volt Rz Vpre-tune range For the example above, where Frune range = 4 MHz, Vore. tune range == 10.00 V, К, = 14 k, and Ro = 7 К, СЕ 15 800 kHz/volt. For the display shown, the vertical sensitivity of the oscilloscope was set to 0.05 volts/division. Therefore, the CF for this display is 40 kHz/division (.05 V/division x 800 kHz/V = 40 kHz/division). The non-linearity error in the voltage-to-frequency transfer of the VCO may be easily determined at any frequency. At 8.8 MHz, for example, the output frequency is approximately 50 kHz low. This represents an error of 1.2% of the 4 MHz sweep range. Improper selection of offset, tune range, or loop gain may allow the loop to drop out of lock over part of the sweep range. The display below illustrates the results of an improperly adjusted loop. An Oscilloscope Display of Error Voltage for an Improperly Adjusted Phase-locked Loop Phase Lock Loop Testing Closed loop testing of frequency response up to 10 kHz and transient response of phase lock loops (PLL's) is made possible by the phase modulation capability of the 3336. À number of techniques can be used for these measurements: these two have been chosen for their simplicity. For more information on these two techniques and other phase lock loop measurements, see HP Application Note AN164-3, "New Techniques for Analyzing Phase Lock Loops.” Closed Loop Frequency Response, Using an RF Spectrum Analyzer This method of obtaining frequency response is extremely useful because few instruments are required and the test set-up is extremely simple (figure below). This method can be used to test loop frequency response at rates up to 10 kHz and is limited at low frequencies only by the resolution of the spectrum analyzer. In order to test loop frequency response above 5 kHz, it is necessary to calibrate the phase modulation rate response between 5 kHz and 10 kHz. The response is typically 6 dB down at 10 kHz. Operation In this set-up, the 3336 is phase modulated using a test oscillator, a second 3336 and the PPL frequency response is displayed by the level of the first sideband as the modulation rate is varied. This technique is possible because in phase modulation the level of the sidebands is independent of the modulation rate. Also for small modulation deviations (i.e., less than 30°), the amplitude of the first sideband is almost directly proportional to the phase deviation. Thus the first sideband amplitude essential- ly tracks the closed loop frequency response. RE Spectrum Analyzer 3336A.B.C Synthesizer PLL Under Test [ro TTT mm Tm mm Sn ие В | Phase | Detector [7* \ an. | | Filter VCO i Lo —— x 3336 Synthesizer —| PPL Analysis Set-up Using an RF Spectrum Analyzer 2dB/DIV О 20 kHz PLL Frequency Response Measurement Procedure * With the 3336 phase modulation on and adjusted for a deviation of less than 30°, adjust the spectrum analyzer to the appropriate scan width and resolution. Tune the analyzer such that the carrier is on the left edge of the display. 31 Since the first sideband carries the frequency response information, adjust the spectrum analyzer's IF attenuation so that this sideband is conveniently positioned vertically on the display. * If the spectrum analyzer has variable persistence, turn it to maximum and slowly sweep the frequency of the second 3336. The resulting display is the PLL frequency response. Closed Loop Transient Response In many applications it is desirable to measure a phase lock loop’s response to transients in order to determine such parameters as rise time, overshoot, damping factor, etc. The test set-up shown (figure below) can be used to obtain PLL response for step changes in phase. With calibrated phase modulation the size of the phase, steps can accurately be controlled to ensure operation in the linear region of the PLL under test and of phase detector #2. Operation In this test set-up, 3336 #1 is phase modulated with a squarewave which causes the RF output to switch back and forth between two discrete phases. The VCO output is then demodulated and the resulting response to step changes in phase is displayed on the oscilloscope. Phase Demodulator ГО ЗЗЗ6А ВОС №2 1 PLL Transient Analysis Test Set-up | Synthesizer | 3336A,B,C No. 1 | Synthesizer | | da | PLL Under Test E | ? Pr ee 1 | | | | : | | Phase Phase Detector \ | | Detector | VCO a pe 3325A or Square Wave Generator PLL Response to Step Changes in Phase. Measurement Procedure * Connect the equipment as shown (figure above). Connect both 3336's to the same frequency standard so that they are frequency coherent. * With the modulation off, step 3336 #2 up and back down 1 or 2 Hz until phase detector #2 is operating in the center of its range. * Turn on 3336 #1 phase modulation. Set the phase deviation small enough to ensure linear operation of the PLL and phase detector #2. Then observe the PLL transient response on the oscilloscope. Chapter V Remote Control Introduction 3586A, B & С Selective Level Meter (SLM) 3336A, B & C Synthesizer/Level Generator (SLG) Both instruments can be operated remotely on the Hewlett- Packard Interface Bus (HP-1B*), using a desktop computer controller such as the HP9825A, 9835A, 9845A/B, or mainframe computer such as the HP 1000 series. Instrument operation can be automated either singly or integrated with two or more instruments to form a system. The 3586 Selective Level Meter and 3336 Synthesizer/ Level Generator are often operated from a controller as a measurement set or with additional instruments such as printers or plot- ters for data manipulation and output of permanent records. Both instruments can be interrogated to output measurement data or instrument status. 3336 SLG El — 3586 SLM | { 3} HP.IB € > { | MP 10631 series cables | X HP9825A , | Computer lorter Controller Typical Automated Test System HP-IB Extender Up to 14 instruments may be operated from a single controller with up to 20 meter separation if HP 10631 series cables are used alone. Remote operation with up to 1000 meters separation is possible using the HP 37210A Bus Extender. The addition of modems allows operation over telephone circuits to virtually any distance. This capability is of particular im- portance for surveillance of FDM systems where the controller, selective level meter and level generator may be separated by thousands of miles. | 3586A/B : Test Path | 333648 | SLM в : Lu Synthesizer o | | | HP.IB | | ; | Lo bor HP 372014 НР 372014 po | HP-IB N НРАВ Extender Tuan Pair Extender or Modem Link Remote Frequency Tracking LOCAL SITE HEB | HP 372014 | MP ZA HE Extender HP 372014 HPIR Extender > >... Remote Site Measurements with Controller HP 372014 | REMOTE SITE #1 KT ! —— >. 2586A/B.C E E ¥ ЗЕ EE ENTE EDU - = : Controller fu, —} PB с HEB a SLM |, м. Extender Modern Extender Do] Link “ i $ на Po = o 39064 /B € a Synthesizer REMOTE SITE *2 privées, I58LA ВОС ЗЕМ ahs FDA BC Synthesizer ADDITIONAL one vaa ENSTRUMENTO Remote Operation of the 3586A, B& C SLM Instrument Programming Codes All of the HP 3586 programming codes and their binary. octal, decimal and hexadecimal values are presented in Table |. Each programming code is an instruction to the instrument. In most cases, sending these instructions corresponds to pressing front panel controls during local operation. For instance, receiving the ASCII characters CH1 during Remote operation has the same effect as pressing Bl during local operation. There are ex- ceptions to this one-to-one relationship. All of the "on/off" controls, the dB and volume controls, all controls in the Fre- quency Tune group and instrument functions not controllable from the front panel. Formats for Programming The format for instrument programming codes depends on the sophistication of the instrument function being controlled. A unique two or three ASCII character code is sent to the instrument to activate functions controlled by front panel keys in Local mode. For example, the instruction E1 programs the SELEC- TIVE tuning mode. While the characters comprising each code must be sent in a certain order, the codes themselves can be sent in any order within a group. Sending E1, T1, CH2 selects SELECTIVE tuning, the BRIDGED 75 Ohm Input and the upper sideband CHANNEL in that order. Sending CH2, T1, El will set the same instrument functions in reverse order. Note that the HP 3586 ignores commas. They are included in the data string examples for clarity. When the HP 3586 is in the Local mode, certain instrument functions are set using several front panel controls. For instance, to enter the Entry Frequency, the control is pressed, the appropriate digits are entered and then , or is pressed. This method is used because the Entry Frequency can assume so many different values that individual switches for each value are impractical. Obviously the order in which the controls are actuated is important. When operating in the Remote mode, almost the same method is used to set the Entry Frequency except that ASCII characters are sent over the HP-IB to activate the instrument functions instead of pressing front panel controls. The ASCII character group "FR” actuates the function controlled by [GRY . ASCII digits correspond to the digit controls and the functions controlled by [Ji] , [1%] , Les) and (Fi ) are actuated by the ASCII character groups "HZ." "KZ" and "MZ" respectively. For example, to enter an Entry Fre- quency of 250 kHz, the ASCII character group "FR,250,KZ" is sent. As before, the order within the group is important; however, this ASCII character group can be placed anywhere in a largr group of instrument instructions, Observe that the groups E1.FRZ50KZ. T1 and FR250KZ,E1,T1 and T1,E1 FR250KZ all result in the same instrument settings. Other functions of the HP 3586 set by this method are Frequency Step, Full Scale, Offset, Threshold and Time. HP 3586A, B & C Programming Codes ASCH Binary Octal Decimal Haxadecimal inatraction Churactars Cara Cods Cada Coda MEASUREMENT Widshand \у 1010118 127 87 B 010000170 107 a8 47 Selective LOs DISTartion LA 04001101 115 77 40 1 00410001 8! 48 31 LOw NOISE (3588. ses MI м 010011701 115 77 403 8 00110119 56 54 38 558 Channel ROSE/DEMODulaton ña 1001101 115 77 40 {Low Nome, 35860 oniy] 2 00110010 52 50 az 1O10Mz, TONE 10048, M 0165011601 115 77 40 3 00110011 83 51 33 CARRIER ha 1001101 115 77 45 4 GOT10160 84 52 34 TONE BOOMz, 280083 a 1001101 118 77 45 в 00110101 65 53 35 9 JITTER M | 01001101 115 77 40 7 QO1IO111 67 ER 37 NOISE/TONE M 01001101 115 77 40 8 00111060 70 58 38 IMPULSE м S1001101 145 77 40 g 001119013 71 57 39 Impulse START $ г О1010071 123 83 53 ] ; 1001001 171 73 44 MEASUREMENT/ENTHY Range 1098 A 01010010 122 82 52 1 001 10001 81 ag 33 10048 R 31010010 122 82 52 z 00110010 87 80 32 Full Scala AUTOmatic Е 01000110 108 70 46 1 G01 10001 61 4% 31 ENTRY F 01000110 198 70 46 2 004110010 82 50 32 AVErage OH À 21500001 101 65 41 8 001 10000 80 48 30 AVErage On A 01000001 101 85 41 1 50110001 61 49 31 UNIT dem U 01010101 125 85 55 Î 00119001 81 49 31 dBpw {dBvy TV, 35EBC) U 01010101 125 85 E 2 001 100170 82 80 32 dB 775 UU 01010101 125 85 55 3 00110011 83 51 33 OFFSET Cf 9 01001111 137 79 4F 8 01010011 123 83 8 501 10000 80 48 30 OFFSET On o 01001111 117 79 4F 5 01010011 123 83 53 1 501 10001 81 49 31 TERMINATION A в © 1ükı SOpHTEM 10k BOs TRO son Y 1010100 124 84 54 1 20110001 51 449 31 780 759 760 T 91019100 124 84 54 2 00110010 82 50 32 1500 1240 TOO T 1010100 124 84 54 3 00110011 63 51 33 +388 теор УВ Y 1010100 124 84 54 4 00110100 54 57 34 Andgad-S00N Bridget BOCH Bridged- 8000 T 01610100 124 ña 54 5 301 19101 ES 53 35 ECON soon é2on T 1010100 124 54 54 6 SC110110 86 84 38 FREQUENT Y ENTRY Entry Freguency 558 CARFIER E 21060101 168 ES 45 1 20116061 61 48 Zi SSA TONE E iO 105 59 45 2 20110010 EZ 5U 32 Chancal e 1000011 103 87 43 Нее H 51001060 118 72 48 3 SHIT 10001 81 39 31 © 1000011 103 87 43 113587 54 1001000 110 77 48 2 20110010 82 50 32 COUNTER Of c 1000011 103 87 43 N 1001110 118 78 4E # 301 15000 8 48 30 COUNTER On & G10060 41 183 £7 43 Hi GI 1110 116 78 aL 1 501 1000] £1 #9 31 TABLE | 33 HP 3586A, B & C Programming Codes ASCH Binary Чета! Üecimal Haxadecimal instruction Charactar: Code Code Cade Cade ENTRY FREGuency F IO00 lo 108 FG 45 KR SICIONIO 127 BZ EZ FRE Cuency STEP E DI010401 123 Ba £3 Р 10100080 120 BO 7 FULL SCALE F 1000110 106 70 46 5 1010011 123 83 £3 OFFSET © 0410014111 117 7% AF F 01000116 106 70 48 STORE 8 DIDIOOTT 123 83 £3 T OICIDIDO 124 84 54 RECALL Я 1010010 122 Bz £2 с 010000171 103 67 43 TRESH (Threshole T 01010100 124 ña £4 H 01001000 110 77 48 TIME T 01010100 124 #4 54 Г 01001001 111 73 48% 0 # 00110000 80 48 30 1 1 00110001 61 48 31 2 2 00110010 62 50 32 3 3 001 10011 63 51 33 4 4 00110100 64 52 34 E 5 00110101 65 53 35 6 в 0110110 66 54 36 7 7 00110111 67 55 37 8 8 00111000 70 56 38 g g 05111001 71 57 39 . decimal) 00111100 74 80 IC Н у G1010101 125 85 55 Р 01010000 120 80 EL | Р 01000100 104 88 44 N 01001110 116 78 4E Hz H 01001000 110 72 48 z 01011919 132 95 BA kHz к 01001011 113 75 4B HH 01001000 110 72 “В or K 01001011 113 65 48 z 31011010 132 80 BA MHz M 010011701 115 77 45 H 01001000 110 72 48 or MA 01001101 115 77 4D * 01611810 132 20 БА dB В 01009105 104 68 44 В 01000010 102 56 47 MEASure CONTinue M G1001101 115 77 45 с 01000011 103 67 43 RONG — OFFSET R 01010010 127 BZ 52 {Reading Offset) 9 01001111 117 79 4F CNTR — FREQuency с 01000011 103 67 43 {Counter — Frequency) F 9010200110 108 70 48 BANDWIDTH 2000 B 01000010 102 86 47 1745 1 51 10001 81 4% 31 3100 400Hz B 01000010 102 £6 42 2 051 10010 82 50 3z Z0hZ В 01606010 102 88 42 3 00110011 63 51 33 WTO (Weighted! B 10000140 102 gg 42 4 00110100 64 52 34 ALDO VOLUME Off Y 01010118 175 En BE @ (044 10000 5 #5 35 VOLUME On Y 1010110 178 BE 58 1 OO 10001 Gi 45 31 MISTCELLANECUS interograte i 01061061 111 73 #8 ki 1001110 1718 78 4F CALIDAate Df © 0155056011 103 87 43 A 1000001 101 85 41 g 00110000 80 48 35 CALbrare On с 01000011 103 87 43 A 21550061 10% ES 41 1 01 10001 fi 48 31 Fast Calibrate È О100001 103 87 43 L 1004106 114 78 AL | TABLE | (continued) 34 Outputting Data The Data Message is used to transfer the results of measurements, or the value of any entered parameter from the HP 3586 to another device on the HP-IB. Usually, the device receiving the data is the controller. Entered parameters are those instrument functions, such as Frequency and Threshold, that are set by entering numerical values. The instructions sent to the instrument before it is instructed to send data determine which type of data will be transferred. If a measure instruction is sent, measurement data will be transferred. Likewise, if an interrogate instruction is sent, the value of the entry parameter designated in the instruction will be sent. Measure Instructions The results of each measurement can be transferred from the HP 3586 only once. A measure instruction must be sent to the instrument before each measurement data transfer to make new data available. There are two instructions that will trigger a measurement in the HP 3586. One is the Trigger Message. It should be used only when simultaneous response from the HP 3586A and another device on the bus is required. The other measure instruction is the programming code TR. This instruction actually directs the instrument to wait and then measure. The duration of the time delay depends on the bandwidth selection. The delay is inserted to allow time for the IF amplifiers in the instrument to adjust to any new signal conditions that might have been programmed. The ASCII instruction is sent to the HP 3586 using the Data Message just like other instructions ac- tuating instrument functions. It can be included in a group of instructions as long as it is the last instruction in the group. It must be the last instruction so that all of the instrument functions can stabilize during the time delay. If the signal being measured is not within the dynamic range of both the Input Amplifier and the IF Amplifiers, the measurement data will not have the normal instrument accuracy. Likewise, the frequency measurement is invalid when the counter is not locked to the input signal. When any of these conditions occur, it is indicated in the measurement data output string. Interrogate The value of any Entry parameter can be output over the HP- IB. This is useful whenever a routine in the program does a search that involves an entry parameter. For example, consider a routine that finds the threshold level that permits ten impulse counts per minute. The threshold is varied using the t | functions until the desired level is found. Once the desired level is found, the threshold is read using the interrogate instruction. Normally, the HP 3586 will output measurement data when it is addressed to talk. If Entry parameters are to be output, the instrument must be instructed to send the value of the selected parameter in place of the measurement data. This is done by sending an "interrogate” instruction to the instrument. An interrogate instruction consists of the ASCII characters IN followed by the ASCII instruction for the prefix of the selected parameter. For example, to interrogate the Frequency Step, the ASCII character group INSP is sent. The interrogate instruction is sent using the Data message like all other programming instructions. It can be sent in a group of instructions as long as a measure instruction does not follow it in the group. If a measure instruction follows an interrogate instruction, the interrogate instruction is negated. Once the parameter has been interrogated, its value will appear in the appropriate display until it is output. The selected Entry parameter will be output when the instrument is addressed to talk. Calibration The instrument automatically calibrates itself approximately every three minutes when it is in the Local mode and AUTOmatic CALibration is on. During remote operation, the three minute calibration is disabled. This is done because 45 pseudo-random calibrations would make the execution time of the program statements unpredictable. If the instrument specifications are to be maintained every three minutes, the controller must direct the instrument to calibrate itself. Done this way, the calibration is predictable and cannot interrupt other programming statements. Fast Calibration When a Fast Calibration is executed, the instrument is only calibratéd on its current range and in the widest bandwidth. This Calibration mode can be used in any of the Selective Measure- ment Modes. It was designed for use during automated surveillance of telecommunications systems. Other Considerations If possible, lock the instrument to the frequency reference of the signal source. This will simplify the tuning routine in the controller program. In an HP 3586 not equipped with Option 004 High Accuracy Frequency Reference, tuning errors of 200 Hz are possible at higher frequencies. When the 20 Hz bandwidth in one of these instruments is used, it is possible that the bandpass of the instrument will not include the Entry Frequency. This is not really a problem when the instrument is operated in a local mode, since the operator can quickly search for the signal and verity that the instrument is tuned to the proper signal using the Frequency controls. Locking the HP 3586A to the frequency reference of the signal source eliminates the need for a search and verify routine in the controller program. When this is done, the tuning procedure is reduced to simply programming the En- try Frequency. If it is not possible to lock the signal source and the HP 3586A together, use a high accuracy frequency reference. This will reduce the frequency error which, in turn, usually simplifies the required search and verify routine in the controller program. Require Service The Require Service Message is a request for service which is sent from a device on the HP-IB to the active controller. Any of the following conditions in the HP 3586A will generate a Re- quire Service Message: * Received an unrecognizable string a Unable to calibrate * Local oscillator not locked * Tone not present for S/N or Phase Jitter Measurements * Attempt to enter Full Scale level while in AUTOrange The Require Service Message is completely independent of all other bus activity. It is sent on a single line (wire) called the SRQ Line, whose state is either true or false. This line is shared by all devices on the HP-IB. When a Require Service Message is received, the controller must determine which device or devices are requesting service. It does this by conducting a Serial Poll. Each polled device responds by sending a Status Byte which indicates, among other things, whether or not the instrument requested service. À Status Byte Message is sent by a device on the bus to the active controller. The individual bits of the Status Byte indicates the status of various device (instrument) functions and whether or not the instrument requested service. The definition of each bit in the HP 3586 Status Byte Message is presented in Table I. Once the Status Byte of an instrument is in the controller, the status of the instrument functions assigned to the bits can be determined by examining the truth state of each bit. The controller then takes appropriate action. For example, if bit 3 of the HP 3586 Status Byte is true, the controller might print a message advising the operator that a tone is required during S/N and Phase Jitter Measurements, * AE MEME MO Mar ENE DIE LG A к ее ен нк кр 00 7 я кл Pal rs PES wa E. 3 Г Tk Fol Ta Poy ad Oo 0 Tad nd £4) LT CA A tan Las 4 Fa LAN EMULE EL La Le Fu FU E. «Las Jin + CE A AT EA Cy CR CREE Oy CPC CR CT C7 © Oy End TRE ve CE Og CCE FOP ex » - E LADA e CE TE EX AE CPS EE Lie же RU we vu ww vi e Bit True State Definition Received unrecognizable string of ASCII characters. Unable to calibrate. Local oscillator unlocked. Tone not present for S/N or Phase Jitter measurements. Attempt to enter Full Scale level while in AUTOrange. Reference not locked to external standard. This instrument requested service. Not used. EO OY OB 0 Во нос Table Il True State Definitions of the Bits in the HP 3586 Status Byte. ЗОНЕ routine arcup FE app EE nt nb "LS duly FEC dim DEBIDA. TL AIM] EDO MALA a fa Typical HPL Program Programming the 3586A,B,C Selective Level Meter requires a working knowledge of the specific controller used. See the in. srtuction manual for your particular controller. This section provides a typical program in HPL language (9825A) to illustrate remote operation. Specific measurement programs similar to those shown here can be used as sub-routines for more extensive test programs. Both the HP9825A and HP9835A Desktop Computer require an HP Interface Card for HP-IB operation. The HP-IB address setting for the instrument can be changed using the address switch on the rear panel. See page 7. FILOT-CA H OTEST deu "385868": 716 3raup measurement”: cll ‘arour information? GEOR FILOT cll Tarour pilot measure’ TT ” Pull Terour carrier leak neuztirement? EP SE dema 5 "SP information , - “nn : ent frei lon BELL = CCITT=1"27{[ 53 -4i., E 2H, : ent "channel 1 corrier trea in bhæz=7"1[I518TI 19 1412 5—T[ 1] : ent "Is erect chonnelzs -isinverted chonnelz"a T[Z2] ; . : ent "TLR in dbe=T07, TE 4] Ei i oi tr ея = " >” Ta пи TE re da e — ооо ENT impedance” yom, TEN, i7d 2 TIA] Td , я = В i Вот 1 see Giert C3SE6A:E FPILOT-CARNIER LEAN TEST I£pO 2 2 16 TL 3 I=Biprt "РЕ 75 зря” 15 TZIziiert "22 123% ohne" if TLSisiiprt "Z= 124 char” — - : if T[2]J=<1tert “INVERTED CHANNEL" LCHERIER LEAL : 1f TCZ2isiiprt "ERECT CHAHNELS® 1 fat CTLP=E 4515" d6n "ture 18.7047 N 1 EpC Gifmt ch ОР и Em t ret * “SroUF pilot measure”! 1 -77, = = 55, З 1 daB "srourp pilot nencurements - - - == 1. : clr VISEERTIurt CISGSH , RIE] = a =. 8 OTLII4TLZI+5 S2e3-Fiir TIS T=13T0 1 14702 ]e0 Burg dF =, UE om &e гыг "ЭбеБАТЬ ТРЕТЬЕ, "НТ" | = Da 09.4 à CÙl *impedance write' TDI]: 4 = ЕН, 5 = 5, 1 5 wait 200 = “o mA wy urt "I5BEÑ", “CHI” нот, О-В. * eli 'ZUBEH amp Pres reed PSs EEE TIS TILA]: - a, 9 Sb = Port TERERHT THR = ” 7 ” = = pas - и TU œ = 4 : #Ert CGEOUE PILOT ! TF ta ld “04, i tee liert > ЭР dba” i ape | i 77 7 HE 1 fat xa FÉ, La x£6,1 _ Poe га © à fxd Ziurt 16:1091,1781 104] = 7, - Ч, 4 har liert "pilot irea = ied Lidar #9, 0. FRICISPE 16:T0 101 1605 in ms eo 1 frmt TUF oe od TE i ret 11 -74,6 -52.É 15 т = ЗВ, 3 FUOUE COFFier Lack EINEN deb Arouk corryiesr Рес! measurement Sir CSS el] *ieeedorcr write’ T0037 Ber CBS": EIB 5 tor 1x8 te 11 Ez] PTL I THT 2 Mensf I vrt USERS ERY BL THD 258 1 211 3 3ÉBER reodt:1,27 10h - CE! SURER read 2 211 3506 read!:5, 28.5, 11] tor Est tea eiléat | ETA CHER red SEHR crt Ine e E : next Y CED Yopéifior Кар обо Рона РР нех Pina pisos Pozec Hiern “CARRIER LE CGT 14 obscpd-rl> pit sup sbz Lip, ch dE EAU Lae | GB: РЁ pd PHBE ap for T=8 to ti lf 14 pd“ -SUGI-2SGa; 1 fm TE FE, De tE ita ps Teds BUT Ja BET 3 {43 Эс: ret next Tiarr 36.1 23: yore ar ОО) VISESR sep trea seal BES for Fai to elifuart estisa DIA rad DA srt arb eh inet + Po leredancte write $ ES: Gspáse Elmer Fe] €. lindo rdiptariha BoA Eirmdoni-páie?d wien 1 11 pisBierr CIP Pf, Bue” Tf CA] SOI Y abztma-riiteEloE - Rf pizalart "USSR 03587: GB 1 4154 1—T1 203 89: 14 ВЧ УЗО 959.pd A 12 IT: BOIS ER TL 50 FEE Lf p4C-300I-SET+; 1 1 res Vis Fei : 321 end DANI 36 v REMOTE SRQ FTN LISTEN LOCAL BUS ADRS 3336A, B& C S/LG Remote Operation Control Modes All necessary functions on the HP 3336 can be controlled from the HP-IB. This not only includes the programming of frequency, amplitude and phase, but it also includes modulation on/off, all sweep parameters, amplitude calibration, and self- test. Because the MODIFY, sweep modification (Afx2, Af+ 2), MKR ~ CF and CLEAR-ENTRY keys are strictly for bench use, they are not programmable. In addition, store and recall operations can be initiated by program command through the bus. Although the storage locations 0 through 9 cannot be loaded directly from the bus, operating and signal parameters may be programmed first on the front panel and then stored in memory via a bus command. Using internal memory when the 3336 is controlled via HP-IB can simplify programming and speed execution time for a series of tests where a large number of parameters are changed for each test. The contents of registers 0 through 9 cannot be directly interrogated by the bus. However, if the register contents are recalled to the front panel of the 3336 , the instrument state may be communicated by using the TALK mode. This feature is ex- plained next. [Learn Mode The 3336 has a TALK mode which allows the instrument to output one ASCII coded parameter at a time upon interrogation by the controller. Some products which have a talk mode similar to this refer to it as a "learn" mode. Some of these "learn” modes output the full instrument condition by sending an uninterrupted string of ASCII characters. For signal sources, where often the only parameter which may change from one test to the next is frequency, the complexity of system programming is not so great as to require the full ” learn” mode capability Local lockout can be initiated via the bus to protect from ac- cidental front panel operation. Programming with the 9825A Calculator The following basic examples are provided to assist the operator in developing programs for the Model 3336 in an HP-IB system which uses the HP Model 9825A Calculator as the system controller. The calculator must be equipped with a General I/O ROM and an HP-1B Interface set to select code 7. The calculator (controller) normally holds the REN line true, unless the "lc! 7” (local) command is sent. REN may be returned to the true state by the "rem 7” (remote) command. Example 1: This is a basic program statement which ac- complishes the following: Address the controller to talk Address the 3336A, B or C to listen Send Program Data: Frequency: 123456783910 MHz Output Impedance: 750 Output Level: - 71.01 dBm AM Modulation: ON This portion places the Bus in the command mode, addresses the calculator to talk and the 3336 to listen Frequency pin 750 - Amplitude AM Modulation The last parameter programmed can be changed without sending the parameter mnemonic. For example, following the program string above, AM may be turned off by using MAG Example 2: This program sets up sweep parameters and initiates a single sweep. Address the controller to talk Address the 3336A to listen Send Program Data: Function: Sine Amplitude: 10 dB Start Frequency: 64 kHz Stop Frequency: 108 kHz Marker Frequency: 104.08 kHz Sweep Time: 2 seconds Start Single Sweep NOTE: To start a single sweep, the mnemonic "SS" must be sent twice. The first "SS” sets the 3336 to the Start frequency, and the second "SS" starts the sweep. Amplitude Start Frequency Stop Frequency Marker Frequency Sweep Time | Start Sweep lo A = EA AP TL TOUS 41 LEO D Ei REIT ox i ML ét IEE de wrt ¿dia PEI Ted HOPliBSEh МЕНА, HEEM TIÍSSE 5535 Example 3: This example checks the "Require Service” status of the 3336 and if it did request service, determines the reason. 1. Enables all service request condi- Br wrt 744; H20R, tions. M°DESTIERZFSSHH 2. Program data contains an error. ore ; 118 es s ane y Stop frequency (SP15KH) is too high. di rdzs{76841—45 4 in = E В : ; i т | 5 3. Wait statement allows time for Service" lack 5 sweep to start before reading status. 4: der “Proceed 4. Read status byte from the 3336 с, : h Froaran + J6 and place in the calculator variable Si if bitif.SI=N) 5 Et CF rogram dy 5. lf bit 6 of the status byte = 1, the Error iwrt rBás 3336 did request service. Go to , lER +red 7941E, subroutine to determine the reason. E: if E=liert \ “Parameter out 6. Programming continues at this of boundz" point if the 3336 did not request ser- PI if E=2iert vice or upon returning from the Invalid delime subroutine. ter с я E=tigrt 7. If service request resulted from a Frea too hiab" program string error, interrogate the 3: if E=diert 3336 to determine the error code and = CE ED time place in the calculator variable “E.” malic ta {+ o eri & €=6 Determine the nature of the pro- “Hatd error” gram error. 11: if E=6iprt DHEED RAramet sg r error” 12: if E=m igprt “Unrecosnizanpie énénoniec” 131 if E=Siprt ‘Gnrecosnizable data choractese А 181 if E=Ÿlert “Gptinon dass not exist" ’ 37 1: у rit {14 4 =i=limert "Suggs strapped’ 181 18 Bit ize Si=ltert "Lueer started о 171 14 BILLIG. SE)“ Ert "Zysten failure" 181 1+ bit {671 prt "Susepins’ 19: 6 DILIFT:El: prit CEus >" J car ret 10 Requezt Service * Frosros Errar 11 sueco Faraocmetar Error J Yates Failure y REAUEIÍ Zervice lo ses Started Suespina J 9. Determine other reason for service request and if “Sweeping” or “Busy” flags were true. 10. Return from subroutine. 11. Printer records the results of the serial poll. 12. If the program string were corrected to make all data valid, this printout would result from the above program. Example 4: The 3336 can be set up manually to the optimum parameters needed for the test to be performed, then the calculator can interrogate the 3336 to determine and record these parameters. This example program interrogates Frequency: IFR Amplitude: [AM =" a Fe HP“ read: TEM" Са Я “Amplitud mo A sk dm dk E DE Line 0 Write statement interrogates Function; read statement address 3336 to talk, calculator to listen, and places data in variable W: "fxd 6" fixes six decimal places. Line 1 Because only numerical data can be placed in the variables, print statements may include in quotes the parameter interrogated. Lines 2-7 Other parameters are interrogated. Frequency is always returned in Hz and amplitude in dBm. This printout results from the above program. If the calculator is equipped with a String Variable ROM, the interrogate program may be changed to the following. Because string variables accept both alpha and numeric characters, the resulting printout includes the mnemonics and delimiters (units). reo, EN ory нед EE e ME A ry E E . Ty Fon нь ha hae pr po E e E 2+ от oo TE oe TT] To ia EW aio “ty EC heb hin TE ne tr TE x # wt 4 Lis, Li wi 65 ELE TE ia % Ty Le m >. г da CE e - = + Wb prt ME Tmt ant Tw Fa HA-ORORO1O, ARNO 1. Dimension a string variable for each parameter you want to inter. rogate. The dimension number (in brackets) is the number of spaces assigned to the variable. 2. This printout results when string variables are used. 38 LAO pera Est i NE MO od Tm LA re “m TE x Lf Ee Example 5: The 3336 can be made to sweep amplitude (in steps) if a for/ next statement is used in the calculator program. It is recommended that the upper and lower amplitude limits selected be on the same range because irregularities in the sweep will occur if the attenuator relays are switched. Line 1 The sweep limits (-3 to 0 dBm) are on the same range. The sweep increment is in ‚1 V steps. Because amplitude was the last parameter programmed, the write statement does not require the "AM" mnemonic. Line 2 The calculator returns to Line 1 until I=7 then proceeds to Line 3. Line 3 The sweep decrement is also in .1 dB steps. Line 5 Return to Line 1 to continue sweeping. The sweep speed is determined by calculator and 3336 data transfer and processing times. If a slower sweep time is desired, wait statements may be added before the "next |" statements. HP-IB Programming Codes Model 3336A/B/C Synthesizer/Level Generator HP-1B Programming Codes (ASCH Characters) FRequency HertZ Kilo-Hertz Mega-Hertz AMplitude dBm PHase DEgrees Sweep STart Frequency Sweep StoP Frequency Sweep Marker Frequency Sweep Time SEconds Sweep Mode Linear Logarithmic Data Transfer MoDe Normal 48 Character Buffer Output Impedance Select 3336A 75 ohm unbal 150 ohm bal 600 ohm bal 33368 75 ohm unbal 124 ohm bal 135 ohm bal 600 ohm bal 3336C 50 ohm unbal 75 ohm unbal Assign Zero Phase Start Single Sweep Start Continuous Sweep FR or FF HZ or HH KH MH AM DB PH DE ST SP MF T SE SM 1 2 MD 1 2 O! Na AY a maa AP 55* se Amplitude Blanking Off On Modulation, Amplitude Off On Modulation, Phase Off On Fast Leveling Off On StoRe Program Location REcall Program Location Mask Service Request Mask Code Interrogate FRequency AMplitude PHase Sweep STart Freq Sweep Stop Freq Sweep Marker Freq Sweep Time Sweep Mode Fast Leveling Output impedance Amplitude Blanking Amplitude Modulation Phase Modulation ERror Codes 1. Entry out of bounds 2. Invalid delimiter SR O to 9 HE G to 9 MS © thru O IFR IAM РН IST ISP IMF iT ISM (FL 1e; IAB IMA IMP ¡ER 4. Sweep time too long or too short 6. Sweep bandwidth too small: start frequency greater than stop frequency (Log sweeps) 7. Unrecognizable mnemonic 8. Unrecognizable data character * Start Single code must be sent twice S558 the sweep, —. The first SS” resets re sweep to start conditions and the second “SS” starte 39 Parameter Interrogation When the 3336 is interrogated, the data returned is formatted as follows: Parameter Interrogated Program Code Format E Frequency IFR, IFF FR DDDDDDDD.DDD HZ CR LFREO! or ® FR ODDDDD.DDDDDD HZ CR LFAEO!I Amplitude ¡AM AM 000000DD.DDO DB CR LF&EO! or AM -00000DD.DDO DB CR LF&ED! Phase PH PH OOCCODDD.DOO DE CA LFAEO! or PH -00000DD.DO0 DE CR LF&EO! Sweep Time FT] TI COOOOODD.DOD SE CR LFAFO!I Output Impedance 101 O DCRILFAEO! Sweep Mode ISM SM D CR LF&FOI Fast Leveling IFL FL DCRILFAEO!I Amplitude Blanking ¡AB АВ О CR LF&EO! Amplitude Modulation IMA MA D CR LF&EQ! Phase Modulation IMP MP D CR LF&EOI Error Codes ¡ER ER D CR LF&EQ! Codes used in these examples: All other characters are the actual ASCII characters returned. D = ASCII digits O thru 9 Spaces are not sent but are inserted in these examples for CR = ASCII Carriage Return operator clarity. LF&EOI = ASCII Line Feed concurrent with End of Identify Other frequencies interrogated are formatted exactly like the ex- Line True ample except that "FR" changes to the program code of the interrogated parameter ("ST" for Sweep Start Frequency, for example). Status Byte 7 6 5 4 3 2 1 O Status Byte Bits F R X S S S Flag; В = REQUEST; x = not used: S = Status | = Program String Error 1 = Sweep Stopped 1 = Sweep Started 1 = System Failure (Main Oscillator or External Reference Unlock- ed) ——1 = Sweep in Progress ——1 = Service Requested (SRQ Line is True) 1 = Busy Flag (-hp- 3336 is processing Instrument Programming Codes) The instrument conditions associated with bits O thru 3 can, when enabled, cause a service request. The instrument programming codes to mask or unmask these bits are: Instrument Programming Bit 3 Bit 2 Bit 1 Bit 0 Codes System Fail Sweep Start Sweep Stop Program Error MS + @ Mask Mask Mask Mask A Mask Mask Mask Enable В Mask Mask Enable Mask С Mask Mask Enable Enable D Mask Enable Mask Mask + Mask Enable Mask Enable F Mask Enable Enable Mask G Mask Enable Enable Enable H Enable Mask Mask Mask i Enable Mask Mask Enable J Enable Mask Enable Mask K Enable Mask Enable Enable L Enable Enable Mask Mask M Enable Enable Mask Enable N Enable Enable Enable Mask о Enable Enable Enable Enable 40 Abridged Description of the HP-IB The Hewlett-Packard Interface Bus (HP-IB) consists of sixteen active signal lines that are used to interconnect up to fifteen devices (e.g., instruments). The sixteen signal lines are categorized according to function. The categories are DATA, HANDSHAKE and GENERAL INTERFACE MANAGEMENT lines. DATA LINES Eight DATA lines are used to carry instrument addresses, instrument control instructions, measurement results and instrument status information in bit parallel, byte serial form. Ordinari- ly, a seven bit ASCII code represents each byte of DATA. The eighth bit is available for parity checking. Data is sent over the DATA lines in both directions. HANDSHAKE (DAV, NRFD, NDAC) Data is transferred between devices using an interlocked “handshake” technique. This method causes the data to be moved at a rate determined by the slowest device involved in the transfer. The HANDSHAKE LINES coordinate the asyn- chronous data transfer by communicating the status of the transfer to the device sending the data (Talker), the device receiving the data (Listener) and the device controlling the transfer (Controller). GENERAL INTERFACE MANAGEMENT LINES These five lines operate independently and in conjunction to send Bus Management Messages to the devices connected to the HP-IB. Each line has a precise definition that is either sent or not sent depending on the truth state of the line. The lines are defined as follows: * Attention (ATN) — Identifies ASCII characters on the DATA lines as a command (command mode) or as data to be transferred (data mode). * Remote Enable (REN) — In conjunction with the ATN Line, places the instrument in the Remote mode. * End or Identity (EOI) — Indicates the last character of a multi-byte data message. Also used with Attention Line to conduct a parallel poll. * Service Request (SRQ) — A device on the bus uses this line to request service from the controller. * Interface Clear (IFC) — Halts all bus activity. 41 DEVICE A Able to talk, listen, and contro! Data lines DEVICE B ll Tr e Pe ет Able to talk and listen | Data Byte Transfer 1 Control DEVICE C Only able to listen General Interface Management DEVICE D Only able to talk NRFD NDAC {FC ATN SRO REN ЕО! Chapter VI Technical Description 3586A, B & C SLM The 3586A, B, or C Selective Level Meter is a very sensitive, highly selective, dual conversion receiver with microprocessor control. Fractional-N synthesized local oscillator circuitry, unique to HP, provides .1 Hz resolution and synthesizer accuracy and stability. (See page 43 for a theoretical discussion of the fractional-N concept.) An RMS detector combined with an automatic level calibrator and accurate input level attenuator allow +.2 dB accuracy over most of the frequency and level range. Input MUX The input multiplexer provides impedance selection. balancing transformers for balanced input impedances, and sets the impedance into the attenuator for all inputs. Autoranging The input level is detected and both RF gain or loss and IF gain or loss are automatically set for best signal to noise ratio. The IF gain establishes the "FULL SCALE." or top of range level. In 10 dB RANGE, the most linear 10 dB portion of the logger range is chosen for best amplitude accuracy. In 100 dB RANGE, the entire logger dynamic range is used. The FULL SCALE setting can be manually entered: however, RF gain or loss is always chosen automatically. Automatic Calibration The first local oscillator is mixed back to the programmed input frequency in the tracking mixer and is accurately set to — 40 dBm + .05 dB to provide a calibration signal at three minute intervals or when programmed. This continual updating compen- sates for level drift to provide + .2 overall level accuracy without the need for manual calibration. $ & HLS Marroprocessas Tracking Output The tracking mixer output is also amplified to 0 dBm + 5 dB to provide rear panel output at the input signal frequency for frequency response measurements of networks. Dual Conversion Double conversion assures maximum image and inter- modulation performance combined with optimum selectivity and sensitivity. The first local oscillator tunes the Selective Level Meter and is actually a 50-82 MHz synthesizer using HP's unique fractional-N technology (see page 43). 0.1 Hz resolution precise frequency setting and excellent frequency stability is the result. The second local oscillator is fixed near 50 MHz and con- verts the signal to the 15,625 Hz second IF frequency. IF crystal filters provide the selectivity including 60 dB carrier rejection, 75 dB adjacent channel rejection, and 50 dB carrier rejection in the 20 Hz pilot filter. All of the filter responses are flat-topped for best level accuracy. RMS Detector/Logger The true RMS detector and logger circuitry is used for detection of the selective and SSB channel mode signals and noise, and also wideband mode noise for baseband testing. True RMS detection allows accurate measurements of both signal level and all types of noise with crest factors up to 5:1. The log amplifier takes advantage of HP's integrated circuit technology to provide excellent log accuracy at low cost. SSB Channel Demod and Impairments Upper or lower sideband demodulation is provided to either a speaker or 600 ohm headphone audio output. The headphone output can be used for additional voice frequency measurements with external instruments. Optional voice frequency transmission impairment measurement circuitry includes a selection of impairment measurements not previously available on a selective level meter. The use of direct weighting filters in Option 003 provide more accurate weighted noise measurements when compared with equivalent noise measurements. 25 Hz 41%) Hy 31 kHz HF-IE Microprocessor £37 я 3 i 17484 Hy от ZONE Ho IF Falter Figure 6-1 3586A,B 8: C Selective Level Meter 3336A, B& C S/LG Single Phase-lock Loop Replaces Multiple Loops Fractional-N synthesis. a technique first used in the 3335A Synthesizer/Level Generator and later the 3325A Syn- thesizer/Function Generator, is the largest single cost reducing factor in the 3336A, B. & C. In fact, fractional-N made it possible to build the 3336 with one phase-lock loop. And one loop costs significantly less than the many used in conventional synthesizers. Further, integration of the fractional-N control circuitry resulted in more cost savings. To see how fractional-N works, we will review the traditional + N Phase Lock Loop method of indirect synthesis. The phase of the VTO outputin Figure 2 is compared to the phase of the reference. Any difference is corrected by generating a de correction voltage which retunes the VTO. To obtain different frequencies, a +N Counter is added as shown in Figure 3. The N number is programmed such that the counter output is always equal to the frequency of the reference. The only limitation is that we can lock only to integral multiples of the reference frequency. To get more resolution, more loops are added. The synthesizer shown in Figure 4 uses four phase lock loops, each providing two digits. The outputs are then sequentially summed to achieve the desired resolution. Costs are also sequentially summed! Let's assume the desired output is 1.01 MHz. What we would really like to do is divide by 10.1 so the phase detector would still see 100 kHz at both inputs. Conceptually this is what fractional-N does. Let's open the loop and assume the VTO is operating at 1.01 MHz. The task of fractional-N will then be to create a dc feedback voltage that will cause the VTO to continue to operate at 1.01 MHz when the loop is closed. Voltage Tuned | DC Tune Voltage Oscillator VTO Frequency fs = 100 kHz Fase 100 kHz 7 Detector Filter Reference Frequency Figure 6-2 Basic Phase Locked Loop Used to Produce Single Frequencies “Рони VIA Voltage Тапей DC Tune Voltage Oscillaror Fons = N * Fon: ыы 30 kHz Frequency Control N Code Phase Lou-Pass „” Detector Filter Reference Frequency Foe = 100 kHz Figure 6-3 À Phase Lock Loop With a Divide-by-N Element to Produce a Range of Frequencies in Steps Equal to Integral Multiples of the Reference Frequency Sth, th DIGITS PLL + 100 bth, 7th DIGITS PLL Si + 100 4th, 5th DIGITS PLL y SL + 100 ist, 2nd, 3rd DIGITS PLL SL fo = 12.345678 MHz Figure 6-4 Synthesizer Using Four Phase Locked Loops y : Control oltage Tuned me СО! © Oscillator Voltage < i] MHz «N= 0 <> A Ie Cum Relerenco | 300 kHy L Phase Detector | yd Detector pr ‘вирег de Figure 6-5 The Basic Block Diagram of an N Step Loop in an Open Loop Condition 43 Fue Removal One Cycle Removed Command Every 10 Relvronce у Periods Ld Ld Ld ZII En J Average Frequency = 1 00 MHz = = it | r Reference Phase si % (106 kHz} Detector Outpin Figure 6-6 The Basic Block Diagram of a Modified N Step Loop With a Pulse Remover Added to Allow the VTO to Operate at a Fractional Frequency The + N counter is set to the closest integer (10), as shown in Figure 5. Since the phase of the 101 kHzis constantly advanc- ing on the phase of the REF, the phase DET output is ever in- creasing. We want the output of the + N counter to be 100 kHz, but this will require the input to the counter to be 1 MHz. A circuit element is added to do this. In Figure 6 we see the pulse train out of the VTO, the new circuit element (a pulse remover) and the pulse train out of the pulse remover. Upon command from the microprocessor, the pulse remover will remove one of the pulses. If a pulse is removed after every 100 cycles of the VTO, the average frequency, as seen on a counter, will be 1 MHz. The average frequency out of the phase DET will in turn be 100 kHz. However, the instantaneous frequency will be 101 kHz. foe = 101 MH: | | — 1 1 MA REF 100 kHz DET Figure 6-7 Adding an Opposing AC Voltage From Processor Controlled D/A The phase detector output will increase with time as it sees the fins of 101 kHz. When the pulse is removed, it will suddenly be back in phase with the reference. The de value of the phase DET output is the correct de to lock the VTO to 1.01 MHz. The only remaining task is to remove the ac component. Filtering would not remove enough ac without reducing switching speed to an unacceptable level. What is done is to add an equal and opposing ac voltage from a processor controlled D/A converter as shown in Figure 7 The output of this is then filtered to further "clean up” the dc correction voltage. This will then lock the VTO to 1.01 MHz This is a brief description of fractional-N. More detailed descriptions are found in the 3336A, B. C Operating and Ser- vice Manual. Amplitude Leveling The extremely accurate amplitude regulation of the 3336 Syn- thesizer/Level Generator is accomplished using a unique internal A leveling technique. The block schematic is shown in Figure 6-9. 1508 6001 Cascading a power detector for high accuracy and a peak detector for fast responses in a negative feedback loop is the heart of the AU Models 758 Unbalanced Balanced 33364 Outputs Peak Detectar r 30 MHz Amphiude Contra! Power Detector rien, Crystal - чи Fractional N Reference Fr N Output ‘équency | Amplifier " Synthesizer High MHz Stabilty A Available by Option Г НР Е Keyboard Control — Te KAARST & LED Display 44 Macremprocessor Output Attenuator 3336A, B & C. It has two distinct leveling modes: Fast Leveling OFF bes tts ts mss meer mst rior a ienns Figure 6-8 3336A,B & C Synthesizer/Level Generator FH MEL eee HE} ORALE gon МН, | Ш "оне 69 3336A,B & C Synthesizer Level Generator Amplitude Regulator , e and ON. In Fast Leveling OFF, on- Unbelanced— ly the power detector is in the leveling loop and the amplitude settling $3308 need time of the loop is = 250 ms. Г When Fast Leveling is ON e Ц + (switches set opposite to that shown on the block diagram), the power detector and the peak detector are in the leveling loop and the amplitude settling time is reduced to 1 ms, without sacrificing any amplitude accuracy. The output power is compared to, and driven until it is equal to a reference voltage. If the reference 12450 1350 mE pos 44 voltage changes, the output power must also change, The reference voltage is set to specific values, over a 9.99 dB range, that are proportional to the programmed output amplitude. For output ampli- tudes more than 9.99 dB below full output, attenuations of 10 dB through 70 dB in 10 dB steps are added, resulting in a 79.99 dB dynamic range. Aftennzafiar © о Chapter VII Serviceability Serviceability has been designed into both the 3336A, B & С, and 3586 from their concept. Serviceability features include amplitude auto-cal, internal SELF-TEST function, digital signature analysis (SA) for troubleshooting the controller and other digital circuitry, easily accessible printed circuit boards and test points and extensive component labeling. Functional Verification Tests As a result of today's complex instruments, the cost of performance testing has increased substantially. To reduce costs. one may elect to use the functional verification tests provided in the Operating and Service Manuals. These tests. when combined with self-test capability, will give the operator a high degree of confidence that the instrument is working properly. while con- suming little test time. Complete performance test information is still included for 100% performance verification when required. Signature Analysis Signature Analysis (SA) is another time-saving service feature. The "signature” are the residue of lengthy data streams, measured at logic test modes, The basic ingredients of Signature Analysis (SA} are "data compression” and “circuit generated.” Both of these features exist, to some degree. in transition counting, but in Signature Analysis they are refined in a manner that affords greater overall performance in terms of locating faults in complex digital circuitry, "Data compression” is achieved in the Signature Analyzer by probing a logic test node from which data is input for each and every circuit clock cycle that occurs within a circuit controlled time window, Within the Signature Analyzer is a 16-bit feedback shift register into which the data is entered in either ifs true or complement logic state, according to previous data-dependent register feedback conditions. In all, there are 2'* = 2%. and become a “signature.” This signature is then a characteristic number representing time dependent logic activity during a specified measurement interval for a particular circuit node. Any change in the behavior of this node {e.q.. even a transition that is one clock cycle late or skewed with respect to the clock) will produce a different signature, indicating a probable circuit malfunction. A single logic state change on a node is all that is necesary to produce a meaningful signature. And, because of the compression algorithm chosen, measurement intervals exceeding 2'° clock cycles will still produce valid, repeatable signatures. A signal that causes the node to produce a signature is the “stimulus.” In SA, the stimulus is supplied by the product itself. By doing this, a controlled environment can be created wherein selected circuit portions can be tested independently of others, while maintaining full dynamic operation. Additionally. synchronization and measurement intervals for the Signature Analyzer can be controlled by the product under test. In microprocessor systems. the stimulus is nothing more than a program {generally in ROM) that exercises the rest of the system. Taking advantage of the data manipulative capabilities of microprocessors, generating good stimulus patterns that exer- cise individual devices in the product is usually not very difficult. Indeed, it is often true that the more complex system, the greater the benefit derived from using SA. The technique can take much of the uncertainty out of servicing microprocessor and other bus structured products. For more detailed information, see HP Application Note AN 222, "A Designer's Guide to Signature Analysis,” and HP Jour- rai, May 1977, "Signature Analysis: A New Digital Field Ser- vice Method.” 1288-2568 | ae Ne O 6L6T 9 999O 111 SAUVE С° ‘§91 ono} ‘ny-pomaBng Cswoys-¢ nyseby ope 17-67 ‘pr4 presse z-Tapnauy -emebons A :uedep uf puejazums ‘влацес) “7 чибауу TT HS xo E; 3 D'd'ueTnp-siog np an LYS prpeg-pajma(4 doing ui "bOEb6 VD ‘ONV oped “Peoy ур эбед ТОСТ ‘рлехоед-иораан эта 10 sófes ¡uewnasUr 10) 10181900 BU) SY `ОБРб-В7О (ЭТ) черреиегу » ‘2821-48 (£18) чле15длу, « `0051-956 (ФОР) шещиос + 'ООВБ-592 (218) WAISIMPIN « (0002-852 (TOE) waised » "230 [Puoibay волеаи 16 ЭэЭО 59195: ДН [820 0A [23 ‘порипицори элош 103 — анужота LALFIMEH

Hewlett-Packard 3586 A, B & C selective level meter, 3336 A, B & C synthesizer / level generator User's Guide

Assistant Bot

Key Features

Frequently Answers and Questions

What is the frequency range of the HP 3586A, B & C selective level meter?

What is the frequency range of the HP 3336A, B & C synthesizer/level generator?

What is the level accuracy of the HP 3586A, B & C selective level meter?

What is the amplitude accuracy of the HP 3336A, B & C synthesizer/level generator?

What is the adjacent channel rejection of the HP 3586A, B & C selective level meter?

Related manuals

Hewlett-Packard 3586 A, B & C selective level meter, 3336 A, B & C synthesizer / level generator User's Guide

Assistant Bot

Key Features

Frequently Answers and Questions

What is the frequency range of the HP 3586A, B & C selective level meter?

What is the frequency range of the HP 3336A, B & C synthesizer/level generator?

What is the level accuracy of the HP 3586A, B & C selective level meter?

What is the amplitude accuracy of the HP 3336A, B & C synthesizer/level generator?

What is the adjacent channel rejection of the HP 3586A, B & C selective level meter?

Related manuals

Hewlett-Packard

3336 C

Hewlett-Packard

3586A

Hewlett-Packard

3586 A/B/C

TTC

T-BERD 224

Agilent

3336 A, B, C

CTC Union

Handy Communication Tester ALT-1500

Advent Instruments

AI 240

Wandel & Goltermann

SPM 16

TMG

ATS 2

ST Group

PIRANHA ST 500

General Signal

1125