Broken 34401a - where to begin

[JJalling]

Hi,

I have a broken 34401a. When the power button is pressed, all the display segments light up, and then turn off.
A long beep comes from the speaker and thats about it.

The shift button on the front panel can toggle the "SHIFT" segment of the display but none of the other keys seems to do anything.

How would you proceed from here?

Thanks, Jonas

[vinicius.jlantunes]

Have you tried running the self test? I think most HP/Agilent/Keysight equipment allow you to run some sort of diagnostics by holding a front pannel key (not quite sure if any key or a specific one) and then turning it on - release the key when it starts the testing routines.

I don't have one of these but it's a pretty popular meter and I think I've seen other repair threads in the repair subsection, in case you haven't checked there yet.

[radioFlash]

I think that's the behavior when the front panel display can't communicate with the mainboard. You should download the service manual which as a schematic and check the cable commendations between the display panel and the mainboard and the power supply voltages.

[JJalling]

Ok, the reset pin is oscillating...

[JJalling]

I tried to hook up the display to another meter, and it is working fine. So the fault must be on the main board.
As I wrote earlier, the reset line is oscillating- could U553 (lt2925) be broken? I can measure 5v at the output so at least the regulation part is working.

Br Jonas

Edit: Minor corrections

[HighVoltage]

Did you test all the rail volatges?
I had several 34401A that showed really weird problems and at the end it was a voltage regulator or a bad tatalum cap.
So, I always start there, including ripple and noise on the rails.

[JJalling]

Did you test all the rail volatges?
I had several 34401A that showed really weird problems and at the end it was a voltage regulator or a bad tatalum cap.
So, I always start there, including ripple and noise on the rails.

No, I haven't tested them yet. My plan was to do it tonight, but now my Rigol DS1054z is stuck during boot - just my luck.

BR Jonas

[JJalling]

Did you test all the rail volatges?
I had several 34401A that showed really weird problems and at the end it was a voltage regulator or a bad tatalum cap.
So, I always start there, including ripple and noise on the rails.

No, I haven't tested them yet. My plan was to do it tonight, but now my Rigol DS1054z is stuck during boot - just my luck.

BR Jonas

Fixed my scope - back to the 34401a.

All the voltages are ripple free and the voltages are within spec.

/Jonas

[JJalling]

Did you test all the rail volatges?
I had several 34401A that showed really weird problems and at the end it was a voltage regulator or a bad tatalum cap.
So, I always start there, including ripple and noise on the rails.

No, I haven't tested them yet. My plan was to do it tonight, but now my Rigol DS1054z is stuck during boot - just my luck.

BR Jonas

Fixed my scope - back to the 34401a.

All the voltages are ripple free and the voltages are within spec.

/Jonas

Ok, maybe the comm between the earth-referenced circuit and the floating circuit is bad.
If I, in my good meter, disconnect power to the floating circuit, I get the same long beep from the speaker, as in the broken meter.

The 5V in the floating circuit is good.

BR Jonas

[JJalling]

I feel like I'm running out of options :-( I'm not sure what to look for.

BR Jonas

[JJalling]

The LM2925 (actually its a CS8126-1 in this worn - its a LM2925 in the working meter) seems to be working, so what is pulling the reset line low.
There seems to be communication on the RAM and ROM data lines, but not on the CAL eeprom.

/Jonas

[free_electron]

check the tantalum caps on the power rails. they may cause the 5 volt to collapse.

[JJalling]

check the tantalum caps on the power rails. they may cause the 5 volt to collapse.

Hi free_electron,
Thank you for replying.
Well, the tantalums looks ok - at least visually. There is no ripple on the rails (or it is at least comparable to the ripple in the working meter). should I unsolder and replace the tantalums?

/Jonas

[JJalling]

check the tantalum caps on the power rails. they may cause the 5 volt to collapse.

Hi free_electron,
Thank you for replying.
Well, the tantalums looks ok - at least visually. There is no ripple on the rails (or it is at least comparable to the ripple in the working meter). should I unsolder and replace the tantalums?

/Jonas

The highest ripple I can measure is about 2mV ac. The measurement is done with the working 34401a.

/Jonas

[HighVoltage]

If you have a working 34401A, you can compare voltages and signals step by step.  In one other 34401A repair thread, someone posted a nice page of the schematics with voltages as they are suppose to be. May be you can start there.

[JJalling]

Ok, I measure about 14v between the two LO terminals on the front. The left one is 14V with respect to ground.

/Jonas

[tautech]

Use the search button next to help button, not the search box and you will find much information including schematics in various threads:

https://www.eevblog.com/forum/repair/repairing-an-hp-34401a-dmm-51092/
https://www.eevblog.com/forum/repair/help-me-repair-agilent-34401a/
https://www.eevblog.com/forum/repair/34401a-test-failures-(many)/
https://www.eevblog.com/forum/repair/please-help-to-fix-my-agilent-34401a/
https://www.eevblog.com/forum/testgear/34401a-service-notes/

[JJalling]

Use the search button next to help button, not the search box and you will find much information including schematics in various threads:

Actually, a better way to search the forum is to use Google. Write "site:eevblog.com/forum" plus your search term in the google search box. This way will also give a better overview than the forum search function.

https://www.eevblog.com/forum/repair/repairing-an-hp-34401a-dmm-51092/
https://www.eevblog.com/forum/repair/help-me-repair-agilent-34401a/
https://www.eevblog.com/forum/repair/34401a-test-failures-(many)/
https://www.eevblog.com/forum/repair/please-help-to-fix-my-agilent-34401a/
https://www.eevblog.com/forum/testgear/34401a-service-notes/

Thank you for the links, but none of them has the same problems as me. The reason I keep updating the thread is (mostly) for my own future reference. When the DMM is fixed, the plan is to update the first post of the thread with a step-by-step guide of what I did and what I observed during the repair.

BR Jonas

[tautech]

Thank you for the links, but none of them has the same problems as me. The reason I keep updating the thread is (mostly) for my own future reference. When the DMM is fixed, the plan is to update the first post of the thread with a step-by-step guide of what I did and what I observed during the repair.

Don't forget to add (Fixed) in the thread title when finished.

[JJalling]

Thank you for the links, but none of them has the same problems as me. The reason I keep updating the thread is (mostly) for my own future reference. When the DMM is fixed, the plan is to update the first post of the thread with a step-by-step guide of what I did and what I observed during the repair.

Don't forget to add (Fixed) in the thread title when finished.

Good tip!! Will do that.

/Jonas

[JJalling]

Does anybody know how the main program of the processor functions? Is it worth mucking with the analog circuits, before I have sorted out the reset problem of the main processor?

BR Jonas

[JJalling]

This is a scope plot of the 5V (CH1) and the reset line (CH2).

Working meter:


Broken meter:

[JJalling]

I don't think I'm able to repair this thing unfortunately, so the plan is to use it for spare parts.

/Jonas

[jitter]

Not so fast...

You've only looked at voltages and the reset circuit, as far as I can tell.

According to this service guide the symptoms seem to indicate a failure in serial communication between main board CPU and front panel CPU, see page 127 (as indicated by the browser/pdf-reader, not the number on the page) item 601. Item 601 is also part of the power on self-test (page 126).

Next step would be to check if there's a two way communication between both processors.

Edit: also look at page 113. The front panel processor can force the main processor to reset. Also interesting: for the floating front panel, 13 V is a logic 0 and 18 V a logic 1.

[JJalling]

Not so fast...

You've only looked at voltages and the reset circuit, as far as I can tell.

According to this service guide the symptoms seem to indicate a failure in serial communication between main board CPU and front panel CPU, see page 127 (as indicated by the browser/pdf-reader, not the number on the page) item 601. Item 601 is also part of the power on self-test (page 126).

Next step would be to check if there's a two way communication between both processors.

Edit: also look at page 113. The front panel processor can force the main processor to reset. Also interesting: for the floating front panel, 13 V is a logic 0 and 18 V a logic 1.

I have another (working) meter to compare with. If I disconnect the front panel from the working meter, it still resets the way it is supposed to, so I don't think that the front panel (or the connection to it) is the problem.

/Jonas