HP 3312A Square wave issue

[belzrebuth]

Hi,
My newly acquired HP 3312A has a problem with its square wave.
The output is very similar to figure (c) of the attached picture (actual pics to follow).
The issue is noticeable in the less than MHz range.
At the MHz range the square seems far more normal.
Any ideas on how I should solve this?
I've tried different BNC cables and probes too but it's the actual generator that has the issue.

[Armadillo]

Did you calibrated your oscilloscope probe?

Try also using BNC to BNC cable from FG to OSCilloscope.

[belzrebuth]

Yes the probes are calibrated and I've also tried using a standard 50ohm (not terminated) bnc to bnc coax cable.
I think my generator has a fault or maybe needs a little adjustment..
From what I've seen the problem regarding my square output could be described as "reduced low frequency amplification".
I've briefly checked the service manual and there is no adjustment procedure for anything similar except from a high frequency compensation variable capacitor on the top board..
I don't know if this will affect the square or it's something else so I haven't touched it yet.
All the other waveforms seem fine but since the square is derived from a comparator stage after the triangle wave is generated I think that doesn't really matter.

[duak]

Belzerebuth, do you have an oscilloscope?  You say you have a manual so look at sheet 6 of the schematics.

K601 is a two pole relay that switches either the square wave or the other waveforms to the output circuitry. (+/- FUNCTION signals)  Check that the square wave is being generated correctly on K601 pins 4 & 7 and that it is passing them correctly on pins 3 & 8.  Most likely, this relay is not working correctly.

If the waveforms are not correct, then the problem is further upstream.  U620 is a comparator that converts the sine wave into a square wave.  You should see clean square waves at both low and high frequencies on pins 7 & 8, the output pins.  If these are both good, then you will have to test the circuitry around Q603 to Q605.  Look for strange deposits, burnt or swollen parts.

Cheers,

[belzrebuth]

I do have an oscilloscope..
I'll check those when I get home and report back.
Thanks for your help duak.

[Armadillo]

Hi, it looks like overcompensated.
Then you should Adjust A1C505, see page 5-12 of the SM, using a BNC to BNC [oscilloscope set for 50ohm load].

Yes the probes are calibrated and I've also tried using a standard 50ohm (not terminated) bnc to bnc coax cable.
I think my generator has a fault or maybe needs a little adjustment..
From what I've seen the problem regarding my square output could be described as "reduced low frequency amplification".
I've briefly checked the service manual and there is no adjustment procedure for anything similar except from a high frequency compensation variable capacitor on the top board..
I don't know if this will affect the square or it's something else so I haven't touched it yet.
All the other waveforms seem fine but since the square is derived from a comparator stage after the triangle wave is generated I think that doesn't really matter.

[belzrebuth]

Belzerebuth, do you have an oscilloscope?  You say you have a manual so look at sheet 6 of the schematics.

K601 is a two pole relay that switches either the square wave or the other waveforms to the output circuitry. (+/- FUNCTION signals)  Check that the square wave is being generated correctly on K601 pins 4 & 7 and that it is passing them correctly on pins 3 & 8.  Most likely, this relay is not working correctly.

If the waveforms are not correct, then the problem is further upstream.  U620 is a comparator that converts the sine wave into a square wave.  You should see clean square waves at both low and high frequencies on pins 7 & 8, the output pins.  If these are both good, then you will have to test the circuitry around Q603 to Q605.  Look for strange deposits, burnt or swollen parts.

Cheers,

I know it's been a while but I thought about taking a look at this generator today.Still the same problem so that's good I suppose.
Can't find K601 nor U620 on the schematics..
I'm looking at this service manual
https://literature.cdn.keysight.com/litweb/pdf/03312-90004.pdf?id=734381

[belzrebuth]

Here's what TP3 and TP4 look like when the generator is in operation mode A.
TP4 looks way off than it should be (should be a clean square with 50% duty cycle about 6 divisions peak to peak @2V/div)
TP3 should be lower in amplitude (about 2 1/3 divisions peak to peak max @2V/div) from what I see in the service manual.

I also attached pictures of what the 10k and 100k square look like.

[belzrebuth]

It seems that's my scope causing this.
I never noticed before maybe because I work with lower frequencies most of the time (audio mostly).
A 10kHz square from my rigol dg1022z has the same effect on the 2235.
Also, the 3312a seems to function okay when I test it with my DS1054z.
I thought that my scope was okay so naturally I suspected the function generator..
Time for a new thread I guess

[dom0]

Are you sure those probes are compensated? You have to do it every time when moving the probe from one scope to another (and sometimes when moving probes between the inputs of the same scope!).

Your scope shot of TP4 indicates that it / you aren't triggering properly, because you see the waveform in-phase and 180° out of phase, so the display isn't stable.

[belzrebuth]

I don't use any probes all the pictures were taken with bnc cables.
Both channels of the scopes exhibit the same behavior.
It's like the scope itself is not "compensated" after the 10kHz mark ..
Here's how a 10k square from the dg1022z looks on the 2235..
A 50ohm terminated bnc cable used.(ch2)

[belzrebuth]

I went through the adjustment procedure and a bunch of stuff actually needed a bit of trimming.
My problem was actually the MF/LF Compensation.
Now the scope is fine with both channels depicting flat top square waves.
I could probably continue the adjustment procedure until is full.

[shakalnokturn]

That's looking much better! (Set aside the 90° rotation  )
When in doubt always question your tools first...