HP 33120A waveform is distorted. Video inside. What is going on?

[fran]

Even though I pulled them and they showed "ok" in a tester?

Although I don't have the metal can transistors, I'm sure I'd have something I can substitute in to test at least.... I'll have a go and report back.

[fran]

For info - the two output transistors are 2n5109 (npn) and 2n5583 (pnp)

Long obsolete now in both cases - what would be recommended as a replacement - the signal generator goes to 15MHz.

[fran]

OK, scrap that last post.

I put in a pair of B139/BD140 that I had in the stash - the clipping is still there..... so I think its not the output transistors.

[nctnico]

Get an oscilloscope and trace the signal path. That will be more fruitful than randomly replacing parts. Before anything else: check the power supply rail voltages. I assume there is a service manual available for this unit; start by going through the fault finding flowchart.

[fran]

Good advice.

The signal is OK all the way to the output amplifier - so this is where the fault lies but it is complicated, hence me scratching around looking for info and help.

Voltages are all OK.

[nctnico]

Which points did you check in the output amplifier? It can be anywhere in the driver stage as well.

Maybe simply replace all transistor as free_electron suggested in his post 10 years ago.

[strawberry]

CDIL makes TO-18 TO-39

[fran]

I get back to the notes I took tomorrow night - but the basic premise was that up to the entry to the amp inputs, the signals are clean. After that the distortion is there.

I'll go looking at what each of the transistors are and see what is available, but I also note that even though free-electron replaced all of them, the DC offset still wasn't working at his last post.

Thanks for the direction and suggestions.

[free_electron]

Even though I pulled them and they showed "ok" in a tester?

Although I don't have the metal can transistors, I'm sure I'd have something I can substitute in to test at least.... I'll have a go and report back.

You need to run them on a curve tracer. They degrade. It's not just work/no work. Just replace oll of them. they work hard in that design.
if you have the version 1.0 of this design : i have the schematics and actual part lists. It is not available as PDF. There are many differences between the "pdf" schematic and the older generation.

[fran]

Ugh - although there are only a handful of transistors involved, there are a few out of stock even if I was to do 2 or 3 separate orders. So I will try to source them but only as I'm ordering other stuff over time.

The parts shortages are quite the pain.... I think it's time for a new generator.

[strawberry]

TTC015B

[fran]

If the schematic is to be believed - and it "looks" like it matches the board, the transistors are:

AD711
PMBTH10
MMBTH81
MMBT3906
MMBT3904
PZT3906
PZT3904
2n5901
2n5583

Some things are in stock at mouser (AD711, MMBT3906, MMBT3904, PZT3904) but others are not, nor are they in RS or farnell. I must try digikey..... but this is looking increasingly like a dead duck.

As outlined above, I did swap in some BD139/140 as a quick check on the output transistors, but still had exactly the same waveform problem.

[free_electron]

If the schematic is to be believed - and it "looks" like it matches the board, the transistors are:

AD711
PMBTH10
MMBTH81
MMBT3906
MMBT3904
PZT3906
PZT3904
2n5901
2n5583

Some things are in stock at mouser (AD711, MMBT3906, MMBT3904, PZT3904) but others are not, nor are they in RS or farnell. I must try digikey..... but this is looking increasingly like a dead duck.

As outlined above, I did swap in some BD139/140 as a quick check on the output transistors, but still had exactly the same waveform problem.

let me check what i have laying around. i'm sitting on three of these generators with various faults. (one has a failed sram) . i know i have some o f the transistors and i want to buy the remaining ones as well.

-sent pm-

[fran]

I'e continued looking at this now and again over the last week or so.

I found a faulty zener on the output transistor supply - CR710. I replaced that but it didn't fix the issue.

I also found that this board (which I cannot see any version marking on) might be a little newer than the one in the service manual. It uses motorola SO-8 transistors for the driver stage of the amp circuit:

Q715/716 were PZT3904 are MRF5943 (NPN)
Q711/714 were PZT3906 are MRF5583 (PNP)

Diode junctions measure fine on each of these.

When I measure DC on Q712 and Q717 - there is a difference - voltage on the emitters of both is spot on +/-13V or so, base again is equal at around 12.x volts on each. The collector though shows some difference - about 2V less on the 3904 compared to 3906

[free_electron]

nope, that is version 1.0 with the so8 transistors. i have repaired two of that generation. i don't know if i still have spares for those , back then they were already extremely difficult to find ( i'm talking 10+ years ago ... )

The schematic for that machine is print only. none of the pdf's out there have the schematic. There are other differences too. the main DAC is a totally different animal. it uses a raytheon dac, where the new machines are an SPT ECL dac with associated level shifting.

check cr703 and cr704. cr710 should have a counterpart on the other rail as well

[fran]

From memory (I don't have my notes in front of me) the voltage drops across the others were all ok and as expected. I'll check again though and report back with real measurements.

The feeling here is that it's not looking good for getting this one fixed .....might have been better 5 years ago but with current parts shortages it's more difficult for sure.

[free_electron]

this has nothing to do with shortages. these parts went out of production 10+ years ago. the 33120 is old. it is a design from the late 1980's, in production in the 90's. that's 30 years ago.