HP 34401A random function selection / reset problem

[wbooth.clearscene]

Hi all,
Great joy, I have just laid my hands apon an HP 34401A. However it has a very odd fault.
After running for a few minuets it will randomly select a different function, eg continuity. After doing this It will flick around more, beeping as it goes, until settling on ohms. Once there the unit will, every few seconds, beep & display the lamp test, returning to the ohms setting.
I've checked the voltage at u553 (in 10v, out 5v) and all the electrolytic look OK. A full self test passes each time I do it.
Unfortunately the service manual does not give any assistance for this type of trouble, so any insights you may be able to provide will be greatly appreciated.
Cheers.

[Dr. Frank]

Check the front keyboard.
Sounds as if there's a problem inside that circuitry, or somebody has poured some liquid over it.
What's the history of this instrument?

Frank

[wbooth.clearscene]

Cheers for the reply. A dirty front panel was my first though before receiving the unit (I knew it was faulty when I purchased it). But the unit inside and out is in immaculate condition. However, I did remove the front panel last night to check, and everything does look good.
After replacing the front panel, the problem persisted until I decided to test the output of u553 over a "flipping about" cycle. Strange, while I was doing that the unit did not misbehave! Further, it looks like to  problem has now gone away completely. I really don't understand what could have caused/solved this, so am not classifying this one as"fixed" just yet.

[SeanB]

Tin whiskers on the front panel causing intermittent switch activation? Clean al the boards and the front panel with IPA and dry thoroughly.

[Dr. Frank]

Cheers for the reply. A dirty front panel was my first though before receiving the unit (I knew it was faulty when I purchased it). But the unit inside and out is in immaculate condition. However, I did remove the front panel last night to check, and everything does look good.
After replacing the front panel, the problem persisted until I decided to test the output of u553 over a "flipping about" cycle. Strange, while I was doing that the unit did not misbehave! Further, it looks like to  problem has now gone away completely. I really don't understand what could have caused/solved this, so am not classifying this one as"fixed" just yet.

U553 is also the main RESET / brownout circuitry.
If you have measured on its pins with a nedle, you might have bended the IC legs a bit, and perhaps virtually repaired a bad solder junction.

If the problem pops up again, I would visually inspect the junctions of U553, if enough solder is available on each pin, and all components around, esp. output cap and reset cap.

Maybe the IC itself is not healthy anymore, depending on temperature, or so..

Frank

[VK5RC]

I checked out the hp_agilent_equipment yahoo group, a quick look there seem to have quite a few problems with the display panel of the 34401, if you do solve it I am sure the group would be interested to hear
Regards Rob

[Barry Ward]

I also have the same issue with 1 of 2 of my meters.   

I'm an experienced component level troubleshooter and I've looked in detail at my meter that suffers the exact same symptoms.   When laid apart on the bench it does not seem to happen.   I also suspected dendritic or conductive issues, which still might be the case if they are under the VFD (as it is yet to be removed). 

I've heated all areas of the front panel board and the main board as well as pushing-tapping-flexing the boards.  Nothing seems to aggravate it.
I've tested the series termination resistors on the FP signal lines.   
The supplies are good.
The meter passes the self test when it's in a working mood.
It's been in the same moderately dry air conditioned environment for a few years. 

The one interesting issue is the display interrupt signal from the main CPU is terribly non-digital.   It seems to be almost a float that slowly gets to a threshold then flips to a high state. See attached.
It goes to the expected rapid rise time when a 10K pullup is superimposed.

I've not been able to catch it in crazy-town mode with the scope nearby.

[DMME-NL]

Sorry for reviving a very old thread... However, I might have found a (very dirty) solution!
At least, it works for my unit that shows prety much the exact same behavior...

In my case, after the DMM is powered on for a while; it will beep, light up all display segments, and reset to 2W resistance mode.
I could not get a consistant occurence, or capture any interesting waveforms. Checking the supplies and brownout circuit also showed no issues.
However, I noticed there is a reset signal in the schematic between the main board and the front panel (FPRST / IGFPRES via R514).
This signal is used to reset the micro controller of the front panel.

Long story short; I disconnected the ribbon cable and lifted this reset pin so it does not connect the two boards anymore.
Result: The unit has been running for a few hours now and it is still perfelctly set to DCV, without any random swithing of modes!

My guess is that this reset signal is used during power-up or perhaps when the watchdog is unhappy?
Anyways, I am sure this not the brightest or most sophisticated idea, but everything boots and seems fine + I can use my DMM as normal again!