Electronics > Repair
HP 8116A Error 42 (E42) Repair
henmill:
Hello again,
I had hoped to make more progress by now and/or update sooner, but alas.
Anyhow, see attached spice file. Since last posting I've added the offset control amp and honestly the whole thing kinda works. If you have time maybe you (and the greater "y'all") can dig up models for the original components rather than the ones I used included with ltspice. Could have a digital twin, so to speak.
Thanks for attempting to explain the feedback path. That makes a lot more sense although I don't quite get how to understand the gain network with the dual polarity feedback(?). I have a lot to learn about how the different stages of this amp work, which boils down to my understanding of bjts..... which I haven't thought about in ages.
Thanks also for showing me how schematics work *facepalm*
On Q507, by all nodes I meant all the transistor pins: emitter, base, collector.
To your point of comparing the voltages, I once again had to photoshop myself a better guide because it is so tedious trying to probe each pin from underneath and know what you're poking. See attached pic of underside of amp with Q503-513 pins labeled. I also made an excel sheet to record my measurements, see screencap of said sheet. I would totally take your source document so I can type in the numbers directly! I don't have a printer at home so don't have an easy way to record values except by the ol' pen and paper.
Now, on to another thought that's been creeping into my mind... if it's difficult or expensive to find replacement transistors, what if we bypassed the output power stage with a modern high speed op amp? Call me crazy but there is probably a part or two out there that could do what this unit does, maybe slightly less amplitude into 50 ohms. I also barely know what I'm talking about, if at all ;)
Thanks for any more feedback on this resurrection project! From my measurements, the only thing that's really jumping out at me is Q507 being equipotential at all pins... love to know your thoughts.
-H
MarkL:
Here's what I see from your latest readings:
Recall that the output is idling at about +250mV and not reaching 0V, where it should be.
Q506 is not pulling down, which in turn explains why emitter followers Q509 and Q511/Q513 are not pulling down on the output.
Q506 appears to not be on by any significant degree, given the Vbe of only 0.5V.
Q506's base is being driven by Q504. Q504 appears to be in saturation with a Vbe of 0.8, and a Vce of 0.5V.
Q504's base is sitting at the negative rail of -24.1V, which would say Q502 is not pulling up on R527 (100R) at all. So why isn't Q502 on?
From some of your earlier measurements (reply #20), the junction of CR501/CR502 is very negative at -8.7V when the offset is set to cause a failure. (Unfortunately, the table doesn't have 0V offset filled in, so I'm going to assume the unit is in non-operational territory with voltages similar to setting a negative offset.)
R517 is pulling down on the other side of CR502 (the cathode), so cathode of CR502 should be one diode drop less, or approx. -8.7V - 0.7V = -9.4V. This is also Q502's base.
But Q502's base is not at -9.4V, it's at +1.05V.
Ok, so is Q502's E-B diode pulling the base up to +1.05V? No, because the emitter is at -9.4V.
I didn't notice this discrepancy when you posted it. I don't know where a positive voltage on Q502's base is coming from when it's surrounded by negative voltages. Maybe the wrong points were probed? I think those numbers need verification. You could add Q502 to your spreadsheet.
Maybe R517 is open and Q502's base is floating. But I would expect the readings to be more random. Your earlier reading in reply #16 would imply R517 might is ok. But then again, that diagram shows (after correction) that Vbe is 1.0V, which is not right.
I've never looked around for a replacement power opamp. I suppose there could be one out there somewhere. But I wouldn't give up on a repair yet. This amp worked once upon a time, and it can be made to work again. It already works half way.
Attached is the source document for the annotated schematic.
MarkL:
Also... That's a nice photo overlay you made there!
On Q507, you have E and C reversed. A couple of clues: Variable resistor RV535 is between the base and collector and, the tab on metal-cased transistors (such as TO-5, TO-18, etc) is next to the emitter which in this case is pointing to the back.
Navigation
[0] Message Index
[*] Previous page
Go to full version