HP 34970A Ser.Nr. US37007297 repair help needed -FIXED -

[Spectrum]

Hi, I am writing from Italy and my English is not so accurate. I have this measuring instrument which needs repair.
At the start there are 5 errors, in particular: 
1) 742 CAL DATA LOST: DCV CORRECTIONS
2) 743 CAL DATA LOST: DCI CORRECTIONS
3) 744 CAL DATA LOST: RES CORRECTIONS
4) 745 CAL DATA LOST: FRES CORRECTIONS
5) 746 CAL DATA LOST: AC CORRECTIONS

I read that these errors refer to the EEPROM memory which, in this unit is marked 24C04-S and is close to the back-up lithium battery that (battery) I have already replaced.

Furthermore, if I perform the extended test, I find over 10 errors: the same as the start twice repeated and a new one marked with the code 621 SELT TEST:AC RMS FULL SCALE

Almost all the components are manufactured in 1997, browsing through the various menus I could read:
REV 06-01-02
PWR ON LAST
DMM ENABLED
CAL SECURED

HOW CAN I attempt the repair? Does the 24C04-S need to be programmed? Where can I find this complete firmware component? What is the error 621 due to?
Could someone who knows this measuring tool well help me?
Thanks for the answers. Stefano (Steve)

[pico61]

You have to download the HP34970A Service Guide from Keysight website. The link:
https://www.keysight.com/main/techSupport.jspx?cc=IT&lc=ita&nid=-33261.536881544&pid=1000001313%3Aepsg%3Apro&pageMode=PL

I  looked at the manual but the errors you indicate are not reported. There is only the 621
Fortunately in the service guide there are all the wiring diagrams to try a repair.

[Shiv]

Hi, I am writing from Italy and my English is not so accurate. I have this measuring instrument which needs repair.
At the start there are 5 errors, in particular: 
1) 742 CAL DATA LOST: DCV CORRECTIONS
2) 743 CAL DATA LOST: DCI CORRECTIONS
3) 744 CAL DATA LOST: RES CORRECTIONS
4) 745 CAL DATA LOST: FRES CORRECTIONS
5) 746 CAL DATA LOST: AC CORRECTIONS

I read that these errors refer to the EEPROM memory which, in this unit is marked 24C04-S and is close to the back-up lithium battery that (battery) I have already replaced.

Furthermore, if I perform the extended test, I find over 10 errors: the same as the start twice repeated and a new one marked with the code 621 SELT TEST:AC RMS FULL SCALE

Almost all the components are manufactured in 1997, browsing through the various menus I could read:
REV 06-01-02
PWR ON LAST
DMM ENABLED
CAL SECURED

HOW CAN I attempt the repair? Does the 24C04-S need to be programmed? Where can I find this complete firmware component? What is the error 621 due to?
Could someone who knows this measuring tool well help me?
Thanks for the answers. Stefano (Steve)

Don´t hestiate too much with the cal errors, if the memory was erased or factor lost, you will keep this error until unit is calibrated and adjusted.

The fisical damage could be the failed self test...

621:  AC rms full scale failed This test configures for the 100 mV ac
range. The 1 mA ohms current source is switched on the charge the
ac input capacitor A4C301. This produces a pulse on the output of the
rms-to-dc converter which is sampled 100 ms after the current is
applied. A 20 ms A/D measurement is performed and checked against
a limit of 10 V ± 8.5 V into the ADC.



Looking at SM schematics, I would check +-15V (wich powers the OP apms) and try to follow back the signal from C301 to the amplifiers...


If that looks ok, replace nvram, initialize and calibrate it.

Look at this:
https://www.eevblog.com/forum/testgear/hp-34970a-error-741-fix/

[Spectrum]

Thanks so much for answer. The voltages are OK, for the rest I will test in labo next time.
For first I discovered there are two FRAM, one in the main and one in the DMM, is it right ?
Is the DMM module the one positioned above the main next to the slots and above the back up battery?
Above my IC there is written 24C04-S, is it a FRAM ?

I just bought online nr. 10 24C04-S, I think the package will be here in 10 days

[Shiv]

Oh...

I have read the first post again and I see you replaced the battery, some of this old boys have a special procedure to do it (power inputs, jumpers or so...)
Before doing more, read this:


https://community.keysight.com/servlet/JiveServlet/download/63858-1-5176/Error%20740%20Series%20for%20fix%2034970A.doc

I don't know if was before the fails or the battery change, it may be just power loss and the memories are ok.

I would try it without replace it.
But be sure you have the proper equipment to do it.

(Manual requires 5700A, but maybe a 3458a and various stable sources will make an approach)

[Spectrum]

Hi,
before battery change the fails were the same...and the old battery (SANYO) was 2,4 or 2,7V not so bad. With new battery nothing has changed.
Could it mean anything?

This afternoon, if I have time, I try to make some measure about C301 and amplifiers

thanks for the availability shown

[coromonadalix]

the battery did not leak on the board ?  no corroded traces ?  are all chemicals okay,  leaks corrosion ?

[Spectrum]

No leak or acid traces. It was the original one (1997) but still intact

[Shiv]

Hi,
before battery change the fails were the same...and the old battery (SANYO) was 2,4 or 2,7V not so bad. With new battery nothing has changed.
Could it mean anything?

This afternoon, if I have time, I try to make some measure about C301 and amplifiers

thanks for the availability shown

Don´t worry, I don´t mean the cause was the battery change, but if was it, there is no need to change memories, only to clean it and write again.

If the failures was before the battery change and was 2.x V, is probable that was bad memory instead of memory out of power.

You would probably need a new ones. (follow the entire procedure instead of start from cleaning)

I would start ignoring ERR 621, is possible that get this error from going out of adjustment range (fine adjust at selftest), with memory lost, it can happen.


NOTE: I would do that just because I have equipment at work to factory adjust it, I prefer adjust without change nothing and check if works.

At home without equipment, I´ll do it otherwise, check for faults to try to get the memory intact until no more options available.


NOTE2: My english is more horrible than yours and here we are... understanding us....

[Spectrum]

OK thanks Shiv. I am waiting for new 24C04-S

This afternoon, if I have time, I try to make some measure about C301 and amplifiers

Sorry, no time, I hope tomorrow

Fortunately in the service guide there are all the wiring diagrams to try a repair.

it would be nice to have the schematics, but neither in the service manual nor in the user are there. Do you know where is possible to find them ?

[coromonadalix]

the 34970 service manual is available on the web      check at KO4BB  website too

[Shiv]

OK thanks Shiv. I am waiting for new 24C04-S

This afternoon, if I have time, I try to make some measure about C301 and amplifiers

Sorry, no time, I hope tomorrow

Fortunately in the service guide there are all the wiring diagrams to try a repair.

it would be nice to have the schematics, but neither in the service manual nor in the user are there. Do you know where is possible to find them ?

Here it is:

https://drive.google.com/file/d/1zbxCo8pLu3gEZDXRaVp-ABRcauAjqjLf/view?usp=sharing

There are the adjusment procedures too, check it.

[Spectrum]

Wonder, thanks a lot

[jogri]

I have a 34970A that experienced the same failure, so i am rather certain that the second FRAM is dead. Not the one near the battery, but the one on the DMM module (that's the thing you have to remove to get to the battery). Start the 34970 without this module (perfectly okay to do this), it shouldn't give you the 740-errors.

The FRAM (yes, those are FRAMs) work without a battery, so this isn't the problem here. You can try to programm the new FRAM with the content of the old, but sinces this FRAM is dead it is probably corrupted (happened to mine). So your only realistic option is to bite the bullet and install a clean FRAM and initialize it via the GPIB commands. There's only one small problem: The instrument is proper f***ed without calibration, mine was around 2-3% off on nearly everything.

(And you should replace the battery while you're at it)

Btw, the 24C04 FRAMs can be an absolute nightmare to program with a programmer, so just don't bother with that, install clean ones and run the two initialization commands.

Protip: Make absolutely sure that the new battery can't get lose (by cracking the solder joint) as this will result in a "dead" 34970: Won't power on and troubleshooting this can be quite time consuming as nothing appears to be broken (i'm speaking from experience...).

[Spectrum]

Thanks jogri is all quite clear, only the last part I did not understand completely."" Protip: Make absolutely sure that the new battery can't get lose (by cracking the solder joint) as this will result in a "dead" 34970: Won't power on and troubleshooting this can be quite time consuming as nothing appears to be broken (i'm speaking from experience...).""

The new battery has soldered in his site, definitely.

I will do a new calibration, I have some reference voltages and resistances. A problem will be initialize it via the GPIB commands, never done before...

[jogri]

The new battery has soldered in his site, definitely.

I will do a new calibration, I have some reference voltages and resistances. A problem will be initialize it via the GPIB commands, never done before...

That's what i thought when i installed a new battery in my unit... Only to have it shaken loose by who knows what after a few months of constant use.

The initialization is the easy part, just get the TekVISA software and use the Instrument manager to send a GPIB command to it. Btw, this thing requires an AC signal in the kHz range for the calibration, so you'll need some sort of AC calibrator (mains + Variac won't cut it, tried that).

[Spectrum]

I have an Agilent 82357B USB with software, but never used...

I have a Bradley 192 (Oscilloscope Calibrator), it is a stable voltage source, DC (from 5 mV to 120 V), AC (only 1KHz square) and frequency (impulse). If it is not enough, I have some function generator (HP33120 ecc..) so somehow I'll do

What do you think about Error 621 AC rms full scale failed ?

[Shiv]

I have an Agilent 82357B USB with software, but never used...

I have a Bradley 192 (Oscilloscope Calibrator), it is a stable voltage source, DC (from 5 mV to 120 V), AC (only 1KHz square) and frequency (impulse). If it is not enough, I have some function generator (HP33120 ecc..) so somehow I'll do

What do you think about Error 621 AC rms full scale failed ?

That isnt enought for that calibration.. .

Bradley has 0.25% in DC mode, cool for 2% scopes, but not for this.

DC can be done with a clean ps and a well calibrated DMM (8.5 digits... Aka 3458)

AC is difficult to get it without an ac calibrator or a lot of things.

Tipically: 5700A... Maybe 5500A.

I doubt if a 33120 with a 3458a could do the job...
Maybe, at least for home use.


Resistance: maybe with a decade and a well calibrated DMM..

Same with current.

Too difficult on AC mode without a nice standard

[coromonadalix]

should be barely enough to check if the 34970  respond and get rid of the errors   and if its is,   send it to a cal lab

[Spectrum]

Let's start with fixing the device, which is the important thing, then I'm sure to do a decent calibration ... probably not at 2ppm, but to be used accurately enough at 0.01%

I can also have for DMM a 3458A, 34401, Kethley 2001 and Prema 5017 with short and good cables, it is not a problem for me to find the correct value, but to have a stable, non-variable standard. Obtained this, I can measure it exactly. The Bradley, after the correct warm-up, is sufficiently stable

However, nothing prevents me from having it calibrated, but I cannot bring to calibrate a faulty or out of specification device

[jogri]

I doubt if a 33120 with a 3458a could do the job...
Maybe, at least for home use.

No, it needs something like 300 V @ 10 kHz for the full-scale cal. You could maybe use it for the 10 V range, but that's it. And just forget the AC current cal, it needs 1 kHz.

In regards to the 621 fault: Haven't had it personally, but i'd check the responsible capacitor as well as the formed pulse.

621 AC rms full scale failed This test configures for the 100 mV ac range. The 1 mA ohms current source is switched on the charge the ac input capacitor A4C301. This produces a pulse on the output of the rms-to-dc converter which is sampled 100 ms after the current is applied. A 20 ms A/D measurement is performed and checked against a limit of 10 V ± 8.5 V into the ADC.

[Shiv]

I doubt if a 33120 with a 3458a could do the job...
Maybe, at least for home use.

No, it needs something like 300 V @ 10 kHz for the full-scale cal. You could maybe use it for the 10 V range, but that's it. And just forget the AC current cal, it needs 1 kHz.

In regards to the 621 fault: Haven't had it personally, but i'd check the responsible capacitor as well as the formed pulse.

621 AC rms full scale failed This test configures for the 100 mV ac range. The 1 mA ohms current source is switched on the charge the ac input capacitor A4C301. This produces a pulse on the output of the rms-to-dc converter which is sampled 100 ms after the current is applied. A 20 ms A/D measurement is performed and checked against a limit of 10 V ± 8.5 V into the ADC.

For "home use" I mean "barely clean errors"...

Just check nothing is broken and can be adjusted by comparing with a 3458a...

For hi V AC... A lot of things...


For ACI... I don't know how...

[Spectrum]

I have a 34970A that experienced the same failure, so i am rather certain that the second FRAM is dead. Not the one near the battery, but the one on the DMM module (that's the thing you have to remove to get to the battery). Start the 34970 without this module (perfectly okay to do this), it shouldn't give you the 740-errors.

Correct, no errors wtihout DMM, when I will have the 24C04-S I try to change this for first.

The voltage are OK only the +18 V is about +19

For error 621, by a quick check, the C301 is OK and the amplifier 301 is supplied (+/-15V) and seems work. However, it is better to check more accurately

[Spectrum]

Hi,
replaced the new 24C04-S in the DMM but the errors are always the same.  The 24C04-S is supplied. I need to initialize the unit or something is wrong anyway ?

[jogri]

Yes, you have to initialize the new FRAMs by sending the two commands via GPIB. Do that, then restart the unit and the errors should be gone (you'll probably still get a date/time lost error on the first start, but that's okay).