My 34970A keeps rebooting as soon as i power it on

[Skyllet20]

Hello there,

i've been using the data acquisitor 34970A for years, and today when i powered it on, it just kept rebooting without allowing me to acces any of its functions.

I've tried to do a factory reboot and a self test run, but it just beeps and keeps rebooting after that.

Has anyone a clue about what is happening?

Here's a video of my problem

[HighVoltage]

Remove all plugin cards to make sure it is not based on a faulty card?

The service manual is available for download from Keysight for these instruments.
Open it up and check all power supply rails first, may be you are lucky and its only a faulty tantalum.

[nfmax]

Also check the cable between the main board and the front panel - the soft power switch runs through this and can give rise to intermittents. IIRC, it is a transition on the main PCB, not a connector, but if you can push the top part of the assembly down by hand, the IDC joints may have worked loose.

[TizianoHV]

A bit late but it might be useful to someone else.
My unit had the same behaviour (continuously resetting), after some panic I noticed that the mains plug had a bad earth connection, changing it solved the problem. 34970A modules (*plugin cards) are susceptible to mains transients. Without earth shielding all the mains noise reach the modules (*plugin cards) causing it to reset.

[Dr. Frank]

Very good, so I also have the opportunity to add my repair case.

I bought my unit 1 year ago as defect.
The Ohm range was not working, repair description here: https://www.eevblog.com/forum/repair/hp-34401a-error-612-613-615-617-618-619-621/msg3571485/#msg3571485

The unit was sensitive to tapping on its case; it made a reset as if I would change its cards.
I also found out, that all of the clamps on the mains plug were quite loose. I assumed brief disruption of power, or that the instrument is sensitive to arcing on these contacts, or similar EMC noise.
I pulled all clamps and tightened them carefully with pliers. Still the clamps were not very well seated, as they seem not to fit correctly to the mains jacks contacts. Anyhow, all connections felt more rigid now, and the tapping sensitivity seemed to be much better, although it still happened from time to time.

The battery was depleted, so I cut it off and soldered another one on the remaining pins. Clock and memory were fine again.




I also replaced the firmware from faulty V9 to V13, and nearly forgot the problem.



Last week, when I used the instrument in my hot room upstairs and not in my cool basement lab, this problem re-occured in a very frequent manner, so I first disconnected the DMM, the backplane, disassembled the instrument completely and by tapping on the boards in the end localized the fault to be somewhere on A1, the main controller board.
Power supplies, regulators and capacitors were ok, and by chance I found the culprit. The solder joint of the batteries minus pole made a loose connection.
The reset and battery supervisor IC, ADM691, also might create resets, when the battery is changed.. that's a vague hint from the data sheet.



The assembly problem is that the battery weighs several grams, but the diameter of the via and the solder pad for this single minus pole is very tiny, and the battery probably has to be hand-soldered. As one does not want to heat up the battery, the soldering process might have been too short, so that not enough solder was sucked into the via to the top side, for making a good mechanical connection.
After 20 years, and by the higher temperatures and vibrations, this created a solder joint crack.   

So I made a proper solder joint by feeding some excess solder, maybe over heating the battery a bit.
I also changed R183, from 8k25 to 300k, as the battery type I'm using is not intended for recharging.



The instrument works rock stable now.

Frank

[TizianoHV]

Note!
The power cord fixed random shutdowns but after a day (obviusly just after posting) it started rebooting (without recovering)!
I found that the battery had a strange behavior: without mains power voltage was low but still enought to keep memory but after applying power it was acting like an open circuit and after a few hours 34970A (detecting "no battery") started resetting.
Now I put a 100kohm resistor and it's doing fine...waiting new battery...

[TizianoHV]

Thanks Dr. Frank for the article. I'll keep the rechargeable / non-rechargeable battery issue in mind when choosing a new one.
Not so sure about soldering directly on the battery. Once I did the same and it started leaking like crazy!
I love the 34970A but it can be quite fragile: /

[Dr. Frank]

Your (first) name really is Tiziano? Bellissimo!

I did not solder directly on the battery, as it's got short leads.

Next time, after depletion, I will assemble a battery holder for CR 1/2 AA Li batteries, 3V.
3.6V does the job as well, as you can see.

There are several threads which propose other solutions, like CR2016 / 2032 cells and holders.

R183 should be changed, in any case.

Frank
 

[TizianoHV]

Yes, grazie!

Interested in the battery holder mod