HP 3325B Fail 026 repair

[LazyJack]

Hi,

I have a HP 3325b function generator, which gives a fail 026 during selftest. Which is AC gain/offset for sine is out of spec. There is output, and that is mostly ok, except for levels above about 0.8 (V, mV, whatever, as the final amplitude is done by attenuation of the generated signal), where it gives a fail 010, which is DAC out fo range.

I'm now going trough all the detailed debugging, but maybe any of you has any advice on where to start, is there any usual suspect?

Thanks.

[wn1fju]

There are three 21-pin ribbon cables in the middle of the unit connecting various boards.  Try putting pressure on them or wiggling them a bit and see if
the error goes away.  That's what happened numerous times with my HP 3325B until I finally got sick of it and replaced the cables.  Funny, I also have
the earlier HP 3325A and I've never had a cable issue with that one.

[LazyJack]

Yeah, I've already reseated all the connectors and cables, doesn't seem to help.Guess, I'll need to do some more digging.

[vindoline]

It sounds like your unit may have trouble with the output attenuator. I assume that you have the service manual? I picked up a 3325B a few years ago and it had damage to the output amplifier and the output attenuator (https://www.eevblog.com/forum/repair/hp-3325b-repair/). If you have it apart, I would do a careful visual inspection looking for components that may have overheated in the past. If everything appears to work correctly until lowering the output to 0.8V, it may be a stuck or non-functional relay on the attenuation board. Let us know and post some pics!

[LazyJack]

I think the attenuator is ok, as this error only affects the sine waveform. The other waveforms are ok.
I do have one lead now, Service Group G, item b, calls for a check on the voltages on TP4 and these are incorrect.
It suggests possible problems with the multiplier, A3U11, which I replaced, but did not fix the problem. So now I'm checking the other parts of the sine amplitude circuit.

[LazyJack]

All right, it's getting interesting. I checked the DAC circuits, comparing the serial DAC pulses with the output of the S/H circuits. The gain voltage is wrong, but it is same on the DAC output and the SH output. As suggested by the service manual, I disconnected the DAC output from the rest to see if the analog switch or something else loads it. Voltages are OK, then reconnect and voltages are still ok, for a short, then wrong again. So looks like there is some intermittent fault on around this circuit. Time to inspect all components in depth.

[wn1fju]

In an earlier reply, I suggested looking at the ribbon cables.  I mention it again because I went through what you are going through - namely, problems in and
around the DAC.

Simply reseating the cables was not enough.  In my case, I needed to apply just the right amount of pressure to the ribbon cable and then all the errors (including
#26) disappeared.  I ended up with a bunch of rubber bands to try to keep the pressure just right.  It would work for a while, then fault eventually again.

It looked to me like there were marginal connections at the point the plastic connector is pressed on the end of the ribbon.

As I said earlier, I got tired of repeating my rubber band scheme, and eventually replaced the ribbon cable with single pin-to-pin jumper wires (a lot of them!).

Maybe this is your problem, or maybe not.  But give the cables another look.

[LazyJack]

Thanks, good idea, I'll have an other look. But I already swapped the cables with each other, and doesn't seem to make any change. Bur for a test, I'll replace them with jumper wires to make sure.

[LazyJack]

Ok, problem solved. It was the multiplier. I was now able to properly adjust. For test, I swapped back the old multiplier IC (I put the new one into a socket), and it was not possible to adjust into spec, it failed the sine amplitude selftest. So somehow that part got marginal. Anyway, I'm happy now :-)

[vindoline]

  Awesome!

[JJ]

I just had the same problem - 026 Fail - but it was only on the sine wave function. After reseating the top ribbon cable in the middle of the board, the fail went away. All is good. Worked perfectly, and painlessly!
Infinite thanks to WN1FJU!