HP 34401A Error 606 on Self Test (Solved?)

[_rushin_or_draggin_]

Hey all,

I ran a self test on my 34401A and it has the error 606. This error says that it "Cannot calibrate rundown gain: This test checks the nominal gain between the integrating ADC and the U500 onchip ADC. This error is reported if the procedure can not run to completion due to a hardware failure. I saw (I believe on this forum actually for a 34970A) that someone replaced U420 and U402 to fix this issue. I am not entirely sure how to test these opamps without taking them out of the circuit so I went around different components in the ADC schematic page. I noticed that +5VB going into U411 is actually around 9.7V and seems kind of unsteady. What I'm confused about is where the +5VB is coming from as I am struggling to find anything in the schematics that show it. Is it actually supposed to be somewhere around 5V or is this 5V referenced to somewhere else? I checked the 5V rail and it is rock solid at 5V.

Thanks,
Josh

[Kleinstein]

The 5 VB should be very stable. It is a voltage generated via U401B (pin7) from the main 7 V reference, not the normal 5 V supply. It is also related to the 5 V that goes to U500 for it's internal ADC.  The actual voltage can be a little off, like 4.9 V.

U401B is wired as a simple buffer, so one could check the input and output of that relatively easy.
A dropping voltage could also be cause from excessive load.

[_rushin_or_draggin_]

I’ll have to check that. I’m just confused why I’d be getting almost double the voltage on +5VB. Also, do you know the difference between +5VB and +5BV? I see multiple voltages like this on the schematic.

[Kleinstein]

There seem to be several parts of the 5 V supply, for more analog and more digital part.

Having really 9.7 V for 5VB is bad and very means a bad U401. Or a serious problem with the -15 V.
The high voltage may have damaged U411 and worst case also U500, but there is a series resistor that could have saved the day. The 74HC4053 is usually relatively robust, but if one orders a replacement for U401, one could as well get a HC4053. In my tests with a similar ADC the HC4053 from Ti was slightly better than the chips from NXP and ST, though for the 34401 the difference likely won't matter.

[_rushin_or_draggin_]

I see on the theory of operation part for the ADC that the FLASH input is clamped to either 0 or 5 VDC as a protection, plus the series resistor you mention. The funny thing is, the meter (atleast voltage, frequency, and resistance) measures spot on. I haven’t checked current yet. It’s just when I do the self calibration it gives me the error 606. I’ll check the voltage across R442 to see if it’s feeding a good 5V into pin 5 of U401-B. If it isn’t I’ll have to backtrack to the U403 voltage reference. I’ll be sure to check the -15 volt line that you mentioned. I don’t think the filter cap or regulator were bad when I checked, but then again I didn’t look too closely. I’ll update you when I get to check it later tonight.

[_rushin_or_draggin_]

I can’t believe this. I get home, self test, it fails. I turn it off and back on, check the voltage going into U401-B, solid 5V. I check the voltage going into U411 (pretty sure that’s the right number, going off memory right now) and it’s a perfect 5V. Turned it off, self test, and I haven’t been able to get the darn thing to fail since. Everything is spot on too. AC/DC voltage, resistance, frequency, continuity, everything that I’ve checked. I have no idea what changed, but something did. I’m wondering if maybe the input voltage was low? Maybe I could hook it up to a variac and sweep from 100-120 VAC? It doesn’t make much sense to me but that’s about all I can think of (besides that X-Class solar flare that just hit, but that’s a stretch I think). I don’t know, I’m at a loss…but atleast it’s working (knock on wood) for now.

[Stray Electron]

  FYI, I've found that when things suddenly work again after you probe them is often because there usually is a hairline crack in a solder joint. And at least once I found the bad connection was actually inside the IC and that I could make or break the connection depending on how I pushed on one particular pin on the IC.

[_rushin_or_draggin_]

I was beginning to think that. I had the board out with the plan to recap it and the entire time it was out I was worried it might get stressed. After getting it all together is when the error started. The only chip I probed was the MUX and R442 before it started working again, so it should’ve narrowed it down to that area. I did press all around the area I was working and slightly flexing the PCB to try and recreate it but it resulted in nothing. I may just leave the caps in it, I’m sure they’re fine. Thanks for the info!

[_rushin_or_draggin_]

Yeah, I knew I was going to replace them while I was typing that last reply. New caps in, turns on fine, passes self test. I guess I’ll just watch it and see if it ever happens again.

[_rushin_or_draggin_]

Just testing these caps. The 2200 uF shows 0.112 ohms on ESR testing. I’ll have to look at the schematic to see what they’re using this 2200 uF for, it was on the opposite side of the PCB.

[Kleinstein]

A cracked / cold solder joint is a real possibilty. To get a higher voltage for the 5VB the suspect points would be around U401, especially the feedback path and decoupling capacitors (could make U401 oscillate).

I doubt tha a low mains voltage would cause such an issue. The HP meters are usually made to work also with a rather low voltage. The symptoms from a low mains would be more additional noise from ripple coming through the regulator.