HP 8643A Frequency Generator Won't Power Up

[Chris K]

I bought this about 6 months ago and the only thing wrong was the memory NiCad battery being dead. I did a bunch of things with it including trying to replicate the sweep generator output of my HP8600/8601 combo for aligning vintage TVs only to find in the 25MHz range, all it could do was sweep 2x per second. Not exactly useful if you're trying to tweak an alignment trace on a scope. Anyway, it's totally dead. I know these units have problems with bad caps on the power boards and I watched Bob's repairs on his 8662 but my unit has a differently designed power section and my God, is it user UNFRIENDLY for service. I saw a YT video that the place to start is the safety cap on the AC line and mine is the suspect, epoxy encased brand (Rima???) referred to. Although the design is different, anyone have any ideas on what else to look for or what the most common power section fail component is? None of the boards look damaged or burned and there's no obviously damaged or vented cap on the top. In fact, there're not a lot of non-electrolytic caps in this circuit anywhere. I don't see anything like the capacitor Bob found that was open and the key to powering up the 8662. BTW, the cooling fan never worked on it either. The fan is good...I tested it with a 24V DC power supply but hearing the fan running noise for the first time, I don't think it was ever running.

[bandersen]

These are newer and very different than the 8662.
HP never published service info for these beasts. You had to be an authorized service center to get your hands on it. It's mostly sealed modules that are factory calibrated.

That's a RIFA cap and, yes, it should be replaced.

Here's my 8664.

[Chris K]

Thanks. I did review your videos.

[bandersen]

Have you seen these?
https://youtu.be/5btPjheYEjs?si=8XnWDIZQB7ZPyPm1

https://youtu.be/Q8K2l9SKUCw?si=XTs5gWyotmdaRctp

[Chris K]

I pulled out the power supply board off the inside back of the chassis that’s directly hooked up to the AC line. The 120V is split, one line going to the board I pulled and another to the 2 boards mounted to the plate above. There are no RIFA caps on the board, just the safety cap on the AC line which I replaced. Maybe this is a later model because the RIFA caps have been switched to WIMA capacitors and I don’t see any evidence they have been changed afterwards from the soldering. Anyway, I gave the board 120V AC and measured the output from the 6 pins of the J2 connector output on the other side of the board. I got nothing. No DC or AC which I checked just in case I wasn’t reading the HP sketch correctly. No schematic from HP of course. Hopefully someone knows the board I’m talking about. It’s the one with the big 100uf 450V line filter cap on it. I guess if I’m not getting any of the low volt DC outputs from it my initial focus should be on this power supply. I did find a list from a YouTubber who tore one of these down of the most suspect caps that should be replaced regardless of failure so I’ve ordered them. Do you guys think I’m troubleshooting this in the right way?

[Chris K]

Probably the culprit. On one of the power supply boards.

[bandersen]

Damn RIFA caps strike again

[Chris K]

Yep...took out the fet too. Quick question...any harm replacing this .047uf safety cap with a .1uf safety cap? It's all I have in hand at the moment. I also replaced a 2200pf RIFA safety teeny tiny cap with another brand. It hadn't gone bad yet but...not a good track record.

[Oldperson1]

Yes, using a 0.1µF instead of 0.047µF is generally okay as long as it's a proper safety-rated cap (X2 or Y2 depending on placement). The tolerance on those values is wide anyway. More important is making sure it's rated for across-the-line use, not just general purpose. Good call preemptively replacing that tiny 2200pF too... once you've seen one go off like a smoke bomb, you don't forget.

[Chris K]

OK thank you. Yes it is an across the line safety cap. There are several videos on YouTube about this failure and the .047 is the most common capacitor on this power supply board to go and it usually goes like this…catastrophically. This cap is charred internally and split open. I ordered the main mosfet that got shorted in fact, I ordered replacements for all the transistors and all of the electrolytic capacitors. There’s an online owner who has posted on forums and created videos on YouTube and he recommends replacing all of the caps on the board even if they have not failed so I’ll do that as well.

[Chris K]

The carnage so far. Transistors coming tomorrow. The electrolytic caps are being replaced proactively. They've tested OK so far but it's been recommended they get switched out with new.

[Chris K]

Finished the work but still no joy. I made a mistake in saying the output of the module is several low power DC voltages. It's a straight 24V DC coming out of pins 1,2 and 3 on the J2 connector. I know I'm getting DC from the rectifier transistor because there's a huge buildup of DC in the filter cap. Ask me how I found out! The transformer that splits the board seems to be OK but after that, I can't find where the voltage gets stopped. There's no IC chip on it to start or stop voltage production based on voltage thresholds so this should just be an AC power on/DC power out board. The board is a non HP product, an Artesyn NFS40-7624 unit but I can't find a schematic for it. I did order the CLIP manual and schematic package but there's no schematic for this board because it's not HP.

Found a NOS one on eBay for $99 so I'll se if I can figure out what's wrong with mine by using it. I'll then put the NOS unit in my 8643 and sell the repaired one. That's the plan anyway.

[Chris K]

This is the board I'm talking about in the above post. The second board that had the blown RIFA and was completely recapped and all transistors changed out is the second power board in line. The one below is the first. There's yet another board that processes power but I have not touched that one yet. All of these replacements are from suggestions from an owner who had the exact same issue and there's a list of proactive replacements he has published.

Power Board #1:

[bandersen]

Hey, guess what? Curious Marc is tackling one of these beasts! Should help us both out.
https://youtu.be/w4sdCkcI9hA?si=SdVoiI6J0i1eZFx1

[Chris K]

Yes I saw your Patreon post. I will definitely check it out. I did get the schematic file but the power boards 1 and 2 are just labeled blocks on the print out. Not HP parts…I did figure out the issue with the 24V DC supply board is with the transformer. When it’s powered up there’s 150V DC on the main filter cap but nothing downstream of the transformer. The electrolytic is holding that voltage for an insane amount of time. It only dropped 10 volts in 24 hours of sitting. Strange…never had a cap do that