Help me "unfix" an HP 8116A signal generator

[monkeysuncle]

Not long ago, I purchased an HP 8116A signal generator/pulse generator on the 'Bay. My scope is currently on the fritz awaiting a donor kidney being shipped from Israel, so I tested it as best I could by wiring up an audio amplifier and speaker. The machine seems to make all of the expected beeps and boops. But while playing around, I was made nervous by the sound of something rolling around inside the case.

So I opened it up, and what should fall out, but one of the output transistors! There are eight TO-5 cans, all with little clips that it looks like were supposed to screw into a heat sink. The heat sink is missing. All eight cans are inserted into sockets that hold them VERY TENUOUSLY. So tenuously, that one of them had fallen out of its own weight.

There is quite a bit of evidence of previous repair by someone not very skilled. Deep digs and scrapes in the solder mask. Ugly solder joints with flux left all over the place. Etc.

So my question is, if anyone has seen the inside of one of these things, are the output transistors supposed to be socketed, or were the sockets an "improvement" added by the monkey who "fixed" this thing? Should I leave them, or should I just rip them out and resolder the transistors directly to the board?

As an aside, I managed to find the manual along with schematics online. The missing heat sink appears to be no more than a sheet of 1/8" Al with a few holes drilled. I think I can fabricate a reasonable replacement.

[johnh]

Didn't look very hard   

ko4bb site is the place that I look first for manuals

http://www.ko4bb.com/manuals/download.php?file=11_Stuff_Not_Sorted/4_Miscelaneous/PULSE/8116/8116Aopsm.PDF

[MadTux]

I you need a relatively cheap (25€), replacement mainboard for the 8116A, contact me. I bought several broken units (8112A,even better since 5 black magic ASICs & 8116A) quite cheaply a while ago and made a few less working units from them, which I'm currently selling.

 Since all the magic is in those damn HP-ASICs that really like to overheat and die, all you have to do is to get one working unit and use it to test the rest of the pulled ASICs. Then you can redistribute all working ASICs back into the nicest looking units and sell them as working for good money.

Thereby I now have one spare 8116A and 8112A mainboard without ASICs that I'd like to get rid of.

[monkeysuncle]

MadTux, my board seems OK, even if someone's been a bit sloppy with a soldering iron. But if you've got one of the main signal boards out (the one with the 3 ASICs), can you tell me if the eight output transistors are socketed or soldered directly to the board? These are 9mm diameter metal cans, near the front of the machine, on the right-hand side as you look at the control panel. All of them should be attached to a common heatsink.

--Dan

[MadTux]

Mine are soldered directly onto board (both the ones on small and big heat sink), as far as I can see (since big heat sink is on top). Most seem to have a ferrite core around their base.

[monkeysuncle]

Thanks, MadTux. That's what I suspected: the sockets were an "improvement." The transistors on my example have a little bead on one lead. It's not a ferrite--just a plastic spacer to maintain some air space underneath.

I'll resolder the transistors, and make a new heat sink.

--Dan

[rastro]

I don't think these are spacers.  The schematics/component location show L508 on the leg of Q510.  This is probably a ferrite used as a inductor on the leg of the transistor.  On the parts list "A1 L508" is described as a "CORE-SHLD BEAD".  According to the schematics/component drawing all 8 heat sink transistors have a respective "bead" on one of their legs.

Q505 & Q505 share a heat sink and are soldered to the PCB.
Q508-13 (six transistors) share a common heat sink and are soldered to the PCB.

I recall seeing a warning in the manual not to run the unit with out the heat sink on these transistors.

[blacknoise]

Here you'll find some pictures of the 8116A PCBs to compare with your unit:


[url=http://www.amplifier.cd/Test_Equipment/Hewlett_Packard/HP_generators/8116A/8116A.html]http://www.amplifier.cd/Test_Equipment/Hewlett_Packard/HP_generators/8116A/8116A.html
[/url]

and a bunch of other equipment:


[url=http://www.amplifier.cd/Test_Equipment/Messtechnik_Test_Equipment.htm]http://www.amplifier.cd/Test_Equipment/Messtechnik_Test_Equipment.htm
[/url]

Kind regards

[monkeysuncle]

Thanks, rastro and MadTux, you appear to be correct about the ferrites.

MadTux, I think I'm interested in that replacement board of yours. The repair work on my board is more extensive than I first realized--and all of it sloppy. Excess solder all over the place, damaged traces, and evidence of overheating the epoxy.

If you still have that board available, contact me off-list (dan at shragmir point com).

[rastro]

Getting a replacement board is your best bet, because the original problem(s) where probably never resolved by the previous troubleshooter.

[monkeysuncle]

Just got the replacement board from MadTux--thanks!!!

Probably won't get to putting it together again until the holidays are past.

--Dan