Hi all,
My HP339A distortion analyzer has a defect:
-the meter-dial allways goes full tilt, no matter what setting or function I select (osc. level. , dist level, or input level).
-On a scope, the monitor-output when set to osc. level shows a 50Hz, 5mVpeak jagged wave.
-the monitor-output, when set to osc. level, show a wideband noisely line of about 1.5mVp
Some years ago, I connected the output of the HP339A oscillator by accidence to the High Tension supply of a tube amp. The oscillator was fixed easily as only the safety zeners were shot. Maybe more then just the oscillator output zeners were effected? or something entirely unrelated is going on?
Any help or suggestions are highly appreciated.
Regards and best wishes,
Thijs Schrama
My HP339A distortion analyzer has a defect:
-the meter-dial allways goes full tilt, no matter what setting or function I select (osc. level. , dist level, or input level).
-On a scope, the monitor-output when set to osc. level shows a 50Hz, 5mVpeak jagged wave.
-the monitor-output, when set to osc. level, show a wideband noisely line of about 1.5mVp
Some years ago, I connected the output of the HP339A oscillator by accidence to the High Tension supply of a tube amp. The oscillator was fixed easily as only the safety zeners were shot. Maybe more then just the oscillator output zeners were effected? or something entirely unrelated is going on?
Any help or suggestions are highly appreciated.
Regards and best wishes,
Thijs Schrama
Hi Thijs,
I'll try to help but responses may be sporadic with time zones and Christmas. 🙂
Can you report observed waveforms at the test points on board A2?
Best,
Steve
I'll try to help but responses may be sporadic with time zones and Christmas. 🙂
Can you report observed waveforms at the test points on board A2?
Best,
Steve
Thanks for your reply. Offcourse, same here, lots of familie christmas celebrations.
With a sine presented at the distortion analyzer input? I'll have to check.
Will report back later.
With a sine presented at the distortion analyzer input? I'll have to check.
Will report back later.
Well, to my suprise, the meter is working. .. Both TP8 and TP9 have a sine wave on them. The oscilator level function works fine now, just as both the input level and relative level work fine now.
But I am afriad the nulling doesn't wok. Set to the Distortion function, it does attenuate the fundamental more than 10dB.
But I am afriad the nulling doesn't wok. Set to the Distortion function, it does attenuate the fundamental more than 10dB.
All kinds of possibilities for the problem--- perhaps something as simple as dirty frequency selection switches.
Some scatter-shot questions: I surmise null depth is limited. Is the signal at the monitor output a clean sine wave at the fundamental?
Do the frequency LEDs provide clues about possible frequency misalignment? Are you able to confirm internal oscillator frequency is accurate with respect knob settings? If you have a separate sine wave generator, you could try tuning frequency to see if deeper null depth is available. A discrepancy between applied frequency and knob setting may help distinguish between null misalignment or 339A generator frequency error.
If these tests don't point the way, we'll dig deeper.
Merry Christmas!
Some scatter-shot questions: I surmise null depth is limited. Is the signal at the monitor output a clean sine wave at the fundamental?
Do the frequency LEDs provide clues about possible frequency misalignment? Are you able to confirm internal oscillator frequency is accurate with respect knob settings? If you have a separate sine wave generator, you could try tuning frequency to see if deeper null depth is available. A discrepancy between applied frequency and knob setting may help distinguish between null misalignment or 339A generator frequency error.
If these tests don't point the way, we'll dig deeper.
Merry Christmas!
Merry Christmas!
Thanks, those are good suggestions.
Using an external oscilator, a Krohn-Hite 4400A, at 5.5kHz @ 1Vrms, setting the HP339A also to 5.5kHz, adjusting the relative input to 0dB.
-yes, the monitor output is a clean sine wave at the fundamental frequency.
-Distortion level indicates about -12.5dB on the dial, when I mismatch the external oscilator frequency about 500Hz (at 5K instead of 5.5kHz) same effect when I mismatch the HP339A frequency. Without misaligning the external oscilator frequency, it reaches -9.5dB.
-Yes, both the level indicators and frequency indicators work (staring at about a factor 2x of mis match)
I am thinking 1 of the two auto-adjustment parameters is not functioning.
Greetz,
Thijs
Thanks, those are good suggestions.
Using an external oscilator, a Krohn-Hite 4400A, at 5.5kHz @ 1Vrms, setting the HP339A also to 5.5kHz, adjusting the relative input to 0dB.
-yes, the monitor output is a clean sine wave at the fundamental frequency.
-Distortion level indicates about -12.5dB on the dial, when I mismatch the external oscilator frequency about 500Hz (at 5K instead of 5.5kHz) same effect when I mismatch the HP339A frequency. Without misaligning the external oscilator frequency, it reaches -9.5dB.
-Yes, both the level indicators and frequency indicators work (staring at about a factor 2x of mis match)
I am thinking 1 of the two auto-adjustment parameters is not functioning.
Greetz,
Thijs
Last edited:
For those interested, the "Operating and Service Manual" can be downloaded from several source e.g.
http://thehistoryofrecording.com/Test_Equipment/Manuals/HP/HP-339A_Manual.pdf
http://bee.mif.pg.gda.pl/ciasteczkowypotwor/HP/339a_6.pdf
HP-339A DISTORTION MEASUREMENT SET-OP-SM Service Manual download, schematics, eeprom, repair info for electronics experts
https://bama.edebris.com/manuals/hp/339a/
The Boat Anchor Manual Archive allways makes me grin 😀
Looking at section 5-16 now, at page 5-2 now...
http://thehistoryofrecording.com/Test_Equipment/Manuals/HP/HP-339A_Manual.pdf
http://bee.mif.pg.gda.pl/ciasteczkowypotwor/HP/339a_6.pdf
HP-339A DISTORTION MEASUREMENT SET-OP-SM Service Manual download, schematics, eeprom, repair info for electronics experts
https://bama.edebris.com/manuals/hp/339a/
The Boat Anchor Manual Archive allways makes me grin 😀
Looking at section 5-16 now, at page 5-2 now...
The nulling hardly reponds to adjusting A4R16 ('Notch filter null adjust) or to A4R43 (Nothc filter freqency adjust). 🙁
Uppon close inspection:
-adjusting A4R43 (Nothc filter frequency adjust) does effect the monitor output slightly
-adjusting A4R16 ('Notch filter null adjust) does absolutely nothing
leading me to think that the Notch filter Null Adjust part of the circuit is dis-functional.
Link to a photo of the HP339A on my bench:
OneDrive
-adjusting A4R43 (Nothc filter frequency adjust) does effect the monitor output slightly
-adjusting A4R16 ('Notch filter null adjust) does absolutely nothing
leading me to think that the Notch filter Null Adjust part of the circuit is dis-functional.
Link to a photo of the HP339A on my bench:
OneDrive
Last edited:
I suggest setting frequency to 10x 10 or 10x 100 or 10x 1000, etc as there are then fewest resistor paths in play for switch contact misbehavior.
You might check A3 TP1. There should be a sine wave equal in amplitude to TP4, but shifted 90 degrees re TP4.
You might check A3 TP1. There should be a sine wave equal in amplitude to TP4, but shifted 90 degrees re TP4.
You may have a defective optocoupler. They are a common failure point in both HP and Sound Technology analyzers. You could try measuring the DCR across the photocell as its trying to tune and see if it changes. Replacements should be reasonable and the newer parts from Optocouplers | Advancedphotonix are supposed to be lower distortion. I'm sure somewhere on the web someone has figured out a cross to the NSL parts.
I think you're on the correct path re auto-null suspicion and Demian may be right about optocoupler.
My best guess is the phase null path since you observed deepest null at lower frequency than knob settings. Make sure clean sine wave at A3 TP1 and confirm phase looks shifted re TP4. A perfunctory test is to check U1 pin3 with scope. It should be a sine wave 3dB lower in amplitude re TP4. A dirty switch S1A could leave C1-C4 unconnected--- no phase shift, but seemingly correct amplitude at TP1. (BTW, I'm suspicious of dirty switches in general. They might have been the culprit mentioned in your first post.)
If all looks well with phase at TP1, on with the null trouble-shooting. Again, I suggest setting to 1.0 x1K (and/or other decade ranges) to minimize frequency switch connects in play. Reconfirm misbehavior null misbehavior, that is best null is with K-H lower in frequency than indicated on 339A.
Would you describe waveforms and behavior of test points on board A4? Of particular interest is behavior of TP2 and TP1 as K-H frequency is tuned across the null. Also of interest is Frequency arrow LED activity.
Good luck.
My best guess is the phase null path since you observed deepest null at lower frequency than knob settings. Make sure clean sine wave at A3 TP1 and confirm phase looks shifted re TP4. A perfunctory test is to check U1 pin3 with scope. It should be a sine wave 3dB lower in amplitude re TP4. A dirty switch S1A could leave C1-C4 unconnected--- no phase shift, but seemingly correct amplitude at TP1. (BTW, I'm suspicious of dirty switches in general. They might have been the culprit mentioned in your first post.)
If all looks well with phase at TP1, on with the null trouble-shooting. Again, I suggest setting to 1.0 x1K (and/or other decade ranges) to minimize frequency switch connects in play. Reconfirm misbehavior null misbehavior, that is best null is with K-H lower in frequency than indicated on 339A.
Would you describe waveforms and behavior of test points on board A4? Of particular interest is behavior of TP2 and TP1 as K-H frequency is tuned across the null. Also of interest is Frequency arrow LED activity.
Good luck.
Great information guys. Thanks a bunch! It wont be until tommorow that I can work on the unit again.
Greetings,
Thijs
Greetings,
Thijs
Hi Thijs,
Because you observe a partial null with K-H below 339A knob frequency setting, my theory/guess is that the A3 R43+S3B+E1 photocell path is presenting more than nominal resistance. A diagnostic test is to temporarily short the E1 cell leads; if the R3+S3B path is intact, the K-H null frequency should now occur above 339A setting rather than below.
I'm also surprised that that the null depth is so shallow if the K-H is carefully adjusted for best null. I wonder if the amplitude loop is also misbehaving. If A4 U1 opamp is not working, both amplitude and phase loops would be inoperable. That's a huge leap for me to make with scant evidence. In any case if the experiment with shorted cell shifts null Freq as imagined, poking around A4 test points might be insightful.
Good luck.
Because you observe a partial null with K-H below 339A knob frequency setting, my theory/guess is that the A3 R43+S3B+E1 photocell path is presenting more than nominal resistance. A diagnostic test is to temporarily short the E1 cell leads; if the R3+S3B path is intact, the K-H null frequency should now occur above 339A setting rather than below.
I'm also surprised that that the null depth is so shallow if the K-H is carefully adjusted for best null. I wonder if the amplitude loop is also misbehaving. If A4 U1 opamp is not working, both amplitude and phase loops would be inoperable. That's a huge leap for me to make with scant evidence. In any case if the experiment with shorted cell shifts null Freq as imagined, poking around A4 test points might be insightful.
Good luck.
Ah, finally time to work on the unit again...
-Using 10x 10 or 10x 100, instead of 1x 100 or 1x 1k, makes no difference.
-A3 TP4 has a 9Vpp sine, infase with the input, clean sine wave, and does not repond at all to mis tuning the analyzer
-A3 TP1 has a 9Vpp sine about 90 degrees fase shifted, clean sine wave, and does not reponds to mis tuning the analyzer, by a somewhat more or lesser phase shift
Last edited:
-Controll module A4E2 starts off at 3.42k, as drifts off to 2.20K or about
-Controll module for the phase controll, also drifts, as if it is trying to tune. With power off, is it 27K (or about), upon powering on, is reaches 6.6k slowly.
Looks like the unit is trying to tune, not?
Lower null is at about -14.8dB, when the HP339a is set to 4.2kHz, while the KH sine generator is set to 5.00kHz. So the HP339a is set lower(!) than the KH generator.
A4 TP3 is a 9Vpp clean sine wave , fase shifted 90 degrees, and hardly reponds, to tunig the freqeuncy by hand around. It slightly changes fase shift, very little.
A4 TP4is a 4Vpp clean sine wave ,inphase with the input, that reponds to mis tuning the analyzer. it quickly becomes very large if I mistune the analyzer . Allong with the neelde pegging to the max.
My latest thought:
I have a low Q bridge-T filter, wich is 15dB deep. Photo Resistors E1 and E2 do influence thier networks (powered on, multimeter measurement).
Maybe the bridge-T components are somehow mis-aligned? by a mechanical defect?
PS
Cann't tell you enough how much(!) I appreciate your input and help!
I have a low Q bridge-T filter, wich is 15dB deep. Photo Resistors E1 and E2 do influence thier networks (powered on, multimeter measurement).
Maybe the bridge-T components are somehow mis-aligned? by a mechanical defect?
PS
Cann't tell you enough how much(!) I appreciate your input and help!
A3 TP3, the output of the T-filter FB opamp, has a 2.0Vpp sine wave, in phase with input, but has some very HF in it, making the scope trace thick. It's about 10MHz, 0.3Vpp
A3 TP2, has a clean sine, inphase with the input, about 20mVpp.
A3 TP2, has a clean sine, inphase with the input, about 20mVpp.
Last edited:
Re the oscillation observed at A3 TP3, make sure you're using a 10x or 100x scope probe. The load capacitance from a 1x probe can cause opamp oscillation. If oscillation persists with 10x probe, try measuring opamp output through a resistor (say 220 ohm) in series with probe tip. If you're seeing a clean signal at TP2, the TP3 is probably fine.
Of course A3 TP2 would be fully nulled if the analyzer were working properly, but it's encouraging that it's a clean sine wave. TP2 is seen much amplified at A4 TP4, but should also be nulled with only noise and distortion remaining when all is proper. The waveforms you report on A4 TP3 and TP5 sound reasonable.
A4 TP1 and TP2 are the amplitude and phase loop error integrators. Their outputs should be between ground and about 8 volts (somewhat less than the 9V zener clamps.) If either is "stuck" near either limit, that spells trouble. I suggest tuning the K-H generator through the null and observing behavior of the integrators. Also monitoring TP4 nulling behavior versus TP5 on dual trace scope should be insightful, as both amplitude and phase nulling activity will be displayed.
Let me know if what you observe and we'll try to figure it out. Good luck!
Steve
Of course A3 TP2 would be fully nulled if the analyzer were working properly, but it's encouraging that it's a clean sine wave. TP2 is seen much amplified at A4 TP4, but should also be nulled with only noise and distortion remaining when all is proper. The waveforms you report on A4 TP3 and TP5 sound reasonable.
A4 TP1 and TP2 are the amplitude and phase loop error integrators. Their outputs should be between ground and about 8 volts (somewhat less than the 9V zener clamps.) If either is "stuck" near either limit, that spells trouble. I suggest tuning the K-H generator through the null and observing behavior of the integrators. Also monitoring TP4 nulling behavior versus TP5 on dual trace scope should be insightful, as both amplitude and phase nulling activity will be displayed.
Let me know if what you observe and we'll try to figure it out. Good luck!
Steve