I have been collecting boat anchors for years, I think that I have met my match. Probably because no tubes are used.
Following the Section V Adjustments, I found the following:
1. Volt meter adjustments-no problems.
2. Gain adjustments-no problems.
3. Oscillator adjustments-no problems.
4. Analyzer adjustment- could not get the distortion range past 50dB without the meter hard pinning to full scale.
I then tried a sine wave with a known distortion, the 339a worked perfectly.
It was accurate in the 80dB range. I measured the 339a oscillator (with another analyzer) it was better than 80dB.
Conclusion, the instrument works great when not using the internal oscillator.
Any ideas would be appreciated, thank you.
Following the Section V Adjustments, I found the following:
1. Volt meter adjustments-no problems.
2. Gain adjustments-no problems.
3. Oscillator adjustments-no problems.
4. Analyzer adjustment- could not get the distortion range past 50dB without the meter hard pinning to full scale.
I then tried a sine wave with a known distortion, the 339a worked perfectly.
It was accurate in the 80dB range. I measured the 339a oscillator (with another analyzer) it was better than 80dB.
Conclusion, the instrument works great when not using the internal oscillator.
Any ideas would be appreciated, thank you.
Hi xenadog,
I can try to help.
I'm a bit confused by this part of your post and not sure how to interpret:
So the suspect oscillator measures better than -80dB on another THD meter but worse than -50dB when measured with its own internal analyzer? For the moment I'll assume I'm mis-construing and assume it's strictly an 339A oscillator issue, but please advise.
Again assuming only suspect oscillator, would you report observed DC voltages at TP8, TP5, TP4, TP3 and also any advise AC waveforms that are in serous conflict with the notes on the schematic? Thanks.
Steve
I can try to help.
I'm a bit confused by this part of your post and not sure how to interpret:
So the suspect oscillator measures better than -80dB on another THD meter but worse than -50dB when measured with its own internal analyzer? For the moment I'll assume I'm mis-construing and assume it's strictly an 339A oscillator issue, but please advise.
Again assuming only suspect oscillator, would you report observed DC voltages at TP8, TP5, TP4, TP3 and also any advise AC waveforms that are in serous conflict with the notes on the schematic? Thanks.
Steve
Hi Steve,
Thank you for your reply.
I am sorry that I was not more clear.
Using the oscillator output of the 339a, I tried to measure the 339a oscillator distortion with the 339a.. Turning the Adjustment Range Switch, I could not get past -50dB without the meter pinning hard up scale.
I then measured the 339a oscillator on another THD instrument. It showed distortion greater than -80dB. Therefore I concluded the 339a oscillator was good.
Finally, using the output of another oscillator (not related to the 339a), I measured the distortion using the 339a. The 339a showed the correct distortion of about -86dB. So I concluded that the filter in the 339a was working.
Using a oscillator source other than the 339a, the analyzer seems to work.
Why can't I measure the distortion of the 339a oscillator using the 339a itself?
Bob
Thank you for your reply.
I am sorry that I was not more clear.
Using the oscillator output of the 339a, I tried to measure the 339a oscillator distortion with the 339a.. Turning the Adjustment Range Switch, I could not get past -50dB without the meter pinning hard up scale.
I then measured the 339a oscillator on another THD instrument. It showed distortion greater than -80dB. Therefore I concluded the 339a oscillator was good.
Finally, using the output of another oscillator (not related to the 339a), I measured the distortion using the 339a. The 339a showed the correct distortion of about -86dB. So I concluded that the filter in the 339a was working.
Using a oscillator source other than the 339a, the analyzer seems to work.
Why can't I measure the distortion of the 339a oscillator using the 339a itself?
Bob
That is mysterious. I assume that the frequency and amplitude arrows offer no clues? They should respond when an external generator is intentionally mistuned relative to the settings on the 339A and it would be good confirm they behave so that they can be relied upon diagnosticly. Of course the frequency arrows should track automatically when using the internal generator.
Wild guess is mis-tracking of frequency setting components due to dirty switches. Do they "feel" dirty--- i.e. intermittent when manipulated? Cleaning may help.
Assuming that brute-force cleaning doesn't help, arranging to monitor the test points in the oscillator mentioned earlier and, more importantly, some similar test points in the analyzer section is the approach I'd suggest. The general idea to monitor these points and observe how they behave when oscillator frequency and amplitude are manipulated experimentally.
We can delve more deeply if you wish.
I will monitor the test points as you stated, and get back to you.
The LEDs seem to work as stated in the manual.
Thanks,
Bob
The LEDs seem to work as stated in the manual.
Thanks,
Bob
Hi Bob,
I looked at manual and couldn't find what step your having a problem with?
Are you having an issue with
5-10/-11 Mechanical Meter Zero?
5-12 Gain Adjustment?
5-11 Frequency Adjustment
5-12 Output Adjustment
Double check that A2r17 and A2r37 are shown properly
in the manaul, ERROR is they are mismarked in the manual.
5-13 Oscillator Adjustment?
5-14 Amplitude Adjustment?
5-15 Frequency Adjustment?
5-16 Analyzer Adjustment
5-17 Notch Filter Null Adjust?
5-18 High Frequency Adjustment?
I would assume when you are Measuring the
internal oscillator, you are using the Oscillator Outputs
connected through BNC/BANANA Jacks to Distortion Analyzer input?
Modified here as shown in Figure 4-10?
How did Table 4-7 turn out?
Prior to your Section V tests
what were the results of your Section IV performance tests?
If you are having issues at the end of Section V,
I would assume Section IV performance tests should have
revealed something, yes?
Just trying to pin down some issue prior to the end of Section V.
Cheers,
I looked at manual and couldn't find what step your having a problem with?
Are you having an issue with
5-10/-11 Mechanical Meter Zero?
5-12 Gain Adjustment?
5-11 Frequency Adjustment
5-12 Output Adjustment
Double check that A2r17 and A2r37 are shown properly
in the manaul, ERROR is they are mismarked in the manual.
5-13 Oscillator Adjustment?
5-14 Amplitude Adjustment?
5-15 Frequency Adjustment?
5-16 Analyzer Adjustment
5-17 Notch Filter Null Adjust?
5-18 High Frequency Adjustment?
I would assume when you are Measuring the
internal oscillator, you are using the Oscillator Outputs
connected through BNC/BANANA Jacks to Distortion Analyzer input?
Modified here as shown in Figure 4-10?
How did Table 4-7 turn out?
Prior to your Section V tests
what were the results of your Section IV performance tests?
If you are having issues at the end of Section V,
I would assume Section IV performance tests should have
revealed something, yes?
Just trying to pin down some issue prior to the end of Section V.
Cheers,
Hi Bob
I also meant to ask earlier if similar behavior was observed on all ranges, amplitudes, etc.
The analyzer sections test points I alluded to earlier are A4 TP1 and TP2. They should be insightful since, for some reason, it seems that full suppression of the fundamental may not be occurring. If this guess proves correct, the gist of the next exploration is to monitor behavior of these two TPs while experimenting with an external oscillator (frequency and amplitude) in contrast with similar tweaks of the internal oscillator. The hope is to tease out differing behavior and converge on the problem.
Please keep us advised and good luck! Hi Sync!
I also meant to ask earlier if similar behavior was observed on all ranges, amplitudes, etc.
The analyzer sections test points I alluded to earlier are A4 TP1 and TP2. They should be insightful since, for some reason, it seems that full suppression of the fundamental may not be occurring. If this guess proves correct, the gist of the next exploration is to monitor behavior of these two TPs while experimenting with an external oscillator (frequency and amplitude) in contrast with similar tweaks of the internal oscillator. The hope is to tease out differing behavior and converge on the problem.
Please keep us advised and good luck! Hi Sync!
Hi BSST (Steve), I really like the SST!
Not trying to butt in here with your helping Bob and all.
Just thought it might help all of us following along if we
identified the HP Performance Checks and Steps the manual provides so that
we are all on the same page.
Ah yes the old measure the A4 board. I must have spend 6 months
pulling my hair out trying to troubleshoot that board.
Cheers,
Not trying to butt in here with your helping Bob and all.
Just thought it might help all of us following along if we
identified the HP Performance Checks and Steps the manual provides so that
we are all on the same page.
Ah yes the old measure the A4 board. I must have spend 6 months
pulling my hair out trying to troubleshoot that board.
Cheers,
Hi Bob,
Earlier today I was thinking about how to trouble-shoot the problem you described and I had a "dope-slap" moment.
I still believe the A4 TP1 and TP2 test points will prove revealing, but I forgot to mention the most readily available test point--- the Monitor Output! It should provide quick insight about the nature of the problem.
You mentioned that when using the internal osc, the meter indicated distortion in the vicinity of -50dBc which is roughly 50dB shy of typical performance. With the analyzer in this state, look at the monitor output with your scope. If the analyzer were nulling the fundamental properly, you'd see no fundamental content and only higher order harmonic distortion content. But if you still see a nice looking sine wave with no visually-obvious distortion, clearly the notch filter is not nulling as deeply as it should.
If the latter is what's observed, then observing the A4 test points will help point to the problem. Note the observed voltages, and then (without changing the analyzer frequency settings) try substituting an external generator. See what range of generator frequencies can be tracked before analyzer null depth suffers. You should see the TP2 voltage change as generator frequency is tweaked. Also try adjusting the generator frequency and amplitude in an attempt to reproduce the A4 test point voltages noted earlier. Be alert the possibility that external oscillator frequency may be at odds with indicated analyzer frequency. Significent differences may point to the frequency tracking problem. If the problem lies in null tracking between oscillator and analyzer, these experiments should help reveal the culprit.
BTW, I also recommend setting the analyzer frequency to 10.0 x 100 because there are fewer frequency determining resistors in play, so fewer possibilities for uncertainty.
Good luck, keep us posted, stay well.
Steve
Earlier today I was thinking about how to trouble-shoot the problem you described and I had a "dope-slap" moment.
I still believe the A4 TP1 and TP2 test points will prove revealing, but I forgot to mention the most readily available test point--- the Monitor Output! It should provide quick insight about the nature of the problem.
You mentioned that when using the internal osc, the meter indicated distortion in the vicinity of -50dBc which is roughly 50dB shy of typical performance. With the analyzer in this state, look at the monitor output with your scope. If the analyzer were nulling the fundamental properly, you'd see no fundamental content and only higher order harmonic distortion content. But if you still see a nice looking sine wave with no visually-obvious distortion, clearly the notch filter is not nulling as deeply as it should.
If the latter is what's observed, then observing the A4 test points will help point to the problem. Note the observed voltages, and then (without changing the analyzer frequency settings) try substituting an external generator. See what range of generator frequencies can be tracked before analyzer null depth suffers. You should see the TP2 voltage change as generator frequency is tweaked. Also try adjusting the generator frequency and amplitude in an attempt to reproduce the A4 test point voltages noted earlier. Be alert the possibility that external oscillator frequency may be at odds with indicated analyzer frequency. Significent differences may point to the frequency tracking problem. If the problem lies in null tracking between oscillator and analyzer, these experiments should help reveal the culprit.
BTW, I also recommend setting the analyzer frequency to 10.0 x 100 because there are fewer frequency determining resistors in play, so fewer possibilities for uncertainty.
Good luck, keep us posted, stay well.
Steve
Last edited:
I Found The Problem
I am a little embarrassed to say this, but the problem was in the BNC to Banana connector. I used two connectors one for an external oscillator (used for testing) and one for the 339a oscillator connection. For some reason this was a bad connection. I canned both of them.
I do not have the instruments to fine tune the filter. The best I can obtain is -92dB
from the 339a oscillator. For what I do, I don't think it is worth the effort to obtain an additional -8dB.
Thank You for all the input. I have another 339a that has been on the shelf for a while. Will use the information gained here to work on that one, if I ever get to it.
Bob
I am a little embarrassed to say this, but the problem was in the BNC to Banana connector. I used two connectors one for an external oscillator (used for testing) and one for the 339a oscillator connection. For some reason this was a bad connection. I canned both of them.
I do not have the instruments to fine tune the filter. The best I can obtain is -92dB
from the 339a oscillator. For what I do, I don't think it is worth the effort to obtain an additional -8dB.
Thank You for all the input. I have another 339a that has been on the shelf for a while. Will use the information gained here to work on that one, if I ever get to it.
Bob
Hi Bob, No worries, it happens to us all.
I just got through dissecting my variac,
which suddenly quit working.
I had it connected through a dim bulb tester.
Once I verified that the amp I was testing
wasn't drawing excess current I removed the bulb.
When I went to apply power to my variac again,
it was just dead, nada, nothing.
I think I burned up a week pulling it apart and taking measurements.
Everything seemed ok, which was puzzling.
I even pulled apart the Dim Bulb Tester to check it's fuse etc.
It didn't have one? WTF- Over!
Then I realized what I had done.
It won't work without a bulb in it.
You can read about my embarrassment here: LINK
I'm just glad I figured it out before I replaced the switches.
"I've found Check to insure it's plugged in." in more than
one troubleshooting manual.
Cheers,
I just got through dissecting my variac,
which suddenly quit working.
I had it connected through a dim bulb tester.
Once I verified that the amp I was testing
wasn't drawing excess current I removed the bulb.
When I went to apply power to my variac again,
it was just dead, nada, nothing.
I think I burned up a week pulling it apart and taking measurements.
Everything seemed ok, which was puzzling.
I even pulled apart the Dim Bulb Tester to check it's fuse etc.
It didn't have one? WTF- Over!
Then I realized what I had done.
It won't work without a bulb in it.
You can read about my embarrassment here: LINK
I'm just glad I figured it out before I replaced the switches.
"I've found Check to insure it's plugged in." in more than
one troubleshooting manual.
Cheers,
Last edited:
Hi Steve,
That is interesting, I think either Richard or (RIP for DickMoore & Davada) when they were still with us,
working on the 339A. They mentioned to me they got better distortion figures when using the 10 x 100,
switch configuration. The posit was the pots/switches were less warn there. Which may be true.
And now you confirmed another reason for the lower distortion number also:
And the less uncertainty, the better. It would have been nice to have you around when we
were working on it. I documented the simplest changes to make in the wiki.
One think I don't think anyone has gotten around to, or if they did, they aren't sharing
are finding suitable replacements for the Opto-Isolators. I don't know enough about them
to make a determination about them.
Food for thought any way.
Cheers,
That is interesting, I think either Richard or (RIP for DickMoore & Davada) when they were still with us,
working on the 339A. They mentioned to me they got better distortion figures when using the 10 x 100,
switch configuration. The posit was the pots/switches were less warn there. Which may be true.
And now you confirmed another reason for the lower distortion number also:
And the less uncertainty, the better. It would have been nice to have you around when we
were working on it. I documented the simplest changes to make in the wiki.
One think I don't think anyone has gotten around to, or if they did, they aren't sharing
are finding suitable replacements for the Opto-Isolators. I don't know enough about them
to make a determination about them.
Food for thought any way.
Cheers,
Last edited:
Hi Sync
I'm embarrassed to admit I messed up my suggestion about frequency setting: I intended to advocate 1.0 x 1000 because of the fewer-resistors-in-play argument. I still recommend that strategy when there's any suspicion of possible R-C tuning uncertainty clouding diagnosis.
But I also do not doubt that the 339A might exhibit lower distortion when set to 10.0 x 100. I believe that's because the single-phase oscillator amplitude detector has lower ripple amplitude at the top of each decade range.
Best,
Steve
I'm embarrassed to admit I messed up my suggestion about frequency setting: I intended to advocate 1.0 x 1000 because of the fewer-resistors-in-play argument. I still recommend that strategy when there's any suspicion of possible R-C tuning uncertainty clouding diagnosis.
But I also do not doubt that the 339A might exhibit lower distortion when set to 10.0 x 100. I believe that's because the single-phase oscillator amplitude detector has lower ripple amplitude at the top of each decade range.
Best,
Steve