Electronics > Repair

HP 8116A Error 42 (E42) Repair

(1/6) > >>

henmill:
Hi y'all,

I recently bought this 8116A function generator off eBay hoping to 1) fix it 2) learn something in the process and 3) gain a function generator for my bench!

The listing was scant on details but said that it powers on but little detail beyond that. On first inspection of the outside, there is a concerning amount of what looks like corrosion and corrosion-esque residues all along the bottom 1/2 inch of the enclosure, front panel and all over the connectors. Not looking so hot so far. (you'll see in the pics)

But, I was delighted to find it more-or-less working! I don't get a whole lot of time to work on my projects during the week so didn't get to let it really warm up until a couple days later, when I decided to crank it up into higher frequencies, ended up letting it run at 100kHz 75% duty 1V amplitude and would periodically check on it.

Lo and behold after a couple hours maybe, the negative half of the waveform had gone all wonky. Positive side still doing fine. Okay, now we're getting somewhere. Upon reboot it greeted me with E42 but would still output, just nothing below 0V pretty much.

I should mention, soon after those early tests I cracked it open to look around, but not much stood out from the top side. Except it does look like somebody has worked it over a time or two, probably even replacing the pesky HP ASICs I've heard about (evidenced by discolored flux residue underneath). (yay! maybe?)

Upon getting the error code 42 and a quick google, this does seem to indicate a problem with the negative half of the output amp, nice!

Last night I went ahead and paid somebody for a scan of the manual, and happy to have it now. Today I've taken a closer look at the output area and oh boy there's lots of corrosion. My mind jumps to leaking electrolytics ...but maybe not so, as you'll see later. *suspenseful music plays*

I cleaned it up and to me it looks as though the Unknown Past Servicer may have already replaced the output diodes and big electrolytic. I'm judging this a couple ways: 1) the amber flux residue on particular components 2) the slight difference in color of solder joints compared to rest of board and trimmed leads that gives it away

I think there's a story here. Imagine, if you will... brilliant engineer working into the night finally gets the thingamabob to work, and in a moment of ecstacy smacks his beer over, spilling right onto his trusty 8116A (which he just had repaired). He quickly turns off the unit and does a hack job cleaning it. It sits unused while engineer achieves great success and ultimately sells off his old gear for new kit and he lives happily ever after.

But seriously, could the leftover flux and proximity to the chassis in a corrosive environment (humid + degrading battery on top PCB) contribute to this type of failure?

At any rate, next I need to actually go through checking voltages at all the nodes in the negative half of the output amp I reckon. All the power rails checked out good.

I really appreciate any advice you folks can give!

I'll be adding pics here, trying to keep it chronological:


https://imgur.com/a/Tkf4w8F


stevopedia:
Given how far apart the battery and the corrosion near the front of the outputs are, I'm doubtful... but you never know. I've heard that long-term exposure of rosin flux residue to a humid environment, such as storage in a non-climate-controlled area, can encourage corrosion as rosin flux residue attracts moisture, but usually rosin flux residue is Mostly Harmless™ when left in place.

Humid storage conditions do seem to be a factor generally, especially given the appearance of the instrument's frame and front panel--all that crusty stuff screams "moisture exposure" to me. Maybe even "covered in condensation." The white deposits on the board in picture #9 might just be a result of that and not corrosion per se. If you clean it off and the board is undamaged underneath, then it's just the salts/whatever left behind by the moisture exposure.

Either way, get rid of that crusty old battery ASAP and clean off the corrosion as well and thoroughly as you can. Be sure to check whether or how much the traces the battery connects to have been eaten away. You may need to carefully remove some solder mask to get a good look or if an ohmmeter doesn't give good results. Battery corrosion has a nasty habit of spreading if unchecked, but it's usually only areas that directly contact the corrosion or which the leaking battery juice falls onto.

henmill:
Yeah, I think the corrosion down underneath the output is unrelated to the battery. But my thinking is it must have been stored in a not so great environment, maybe after breaking just sat for a really long time. The way the salts or "corrosion" as I'm calling it travelled all around the frame makes me think that. My first thought when I saw it was I hope it wasn't in a flood or something because of the gunk all along the bottom. But I don't think that's the case.

Btw the battery is still holding a voltage. Front panel will recall last used settings. Although it came to with defaults when I turned it on first.

But I've removed it anyways to clean and inspect the PCB. Will replace eventually.

I've read water + baking soda and/or vinegar can neutralize the battery corrosion residue. Any other tips welcome!

stevopedia:
It almost certainly got very wet at some point. The fact that the white stuff seems to have cleaned off very easily points to waterborne salts or whatever. That's the only explanation for all the residue I can think of. But for all that it's in really fantastic shape! I think you lucked out on this one  :)

Regardless of whether that old battery is holding a charge, the fact that only the battery leads seem to have actually corroded is a sure sign that it's leaking electrolyte. If you leave it in, a good portion of the whole board will look like its leads do. Old rechargeable batteries just like that one are notorious for absolutely wrecking boards beyond repair.

Edit: I had another look at your pictures, and the second to last shows some clear evidence of battery leak damage. See how some of the pins on R11 have a dull appearance, the same as the two vias next to the battery's negative terminal? That's early-stage corrosion. You can see some marking on the traces from those vias too. But the biggest warning sign of all is the green corrosion on the one leg of the resistor pack next to the little blue capacitor on the bottom left.

You caught this one early. The only place that old battery belongs is in the trash.

The good news is that NiMH cells/batteries are still readily available and are usually drop-in replacements for the original (probably NiCad) batteries.

MarkL:

--- Quote from: stevopedia on October 21, 2024, 02:32:58 pm ---...
The good news is that NiMH cells/batteries are still readily available and are usually drop-in replacements for the original (probably NiCad) batteries.

--- End quote ---
The old battery is NiCad and is 2 cells, HP part #1420-0251.   Possibly 2x1/3AA.   What battery did you find to replace this?

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod