Agilent 34401A, SelfTest OK, but 20 Errors (CAL Checksum Failed) [FIXED]

[HighVoltage]

This 34401A was working perfectly until yesterday.
Turning it on today, gave an "ERROR" message in the front display after bootup.

Interestingly, the self test is passing without any problems.

Since the 34401A stores only up to 20 Errors, there are probably more, but may be they era just stored new, every time the unit is turned on.

Following Errors are listed in the Error section of the Menu:
740, 747, 748, 741, 742, 743, 744, 745, 746, 740, 747, 748, 741, 742, 743, 744, 745, 746, 740, -350
The last (-350) one also indicates that there are more than 20 errors.

 The user guide is listing these errors as:

- 740 Cal checksum failed, secure state
- 741 Cal checksum failed, string data
- 742 Cal checksum failed, DCV corrections
- 743 Cal checksum failed, DCI corrections
- 744 Cal checksum failed, RES corrections
- 745 Cal checksum failed, FRES corrections
- 746 Cal checksum failed, AC corrections
- 747 Cal checksum failed, GPIB address
- 748 Cal checksum failed, internal data

Does anyone here have an idea where to start to look for the problem?

Thanks for any advise.

[acbern]

it seems that the cal data, stored in eeprom, are corrupted somehow. try to calibrate it. that may not fix the problem but give you more insights into what happened.

[HighVoltage]

Thank you acbern,
I tried the calibration procedure.
The 0 Offset is ok
The DC gain is off and after calibrating, it does not store the new calibration data

For instance, if I calibrate the 10 VDC range:
Applied voltage from a standard: 10.0000 V
Shown value before calibration: 10.08479 V
Then follow calibration procedure
Shown value after calibration: 10.08479 V
So, essentially nothing changes

The same is true for the other ranges of AC and OHM calibration

The service guide says following:
This error may be generated if you turn off the power during a calibration.
Recalibration is required to clear the error. If the error persists,
a hardware failure of U505 may have occurred.


Does that mean I have to exchange the U505 EEPROM?
If so, can someone tell me if I have to watch out for anything special?

Thanks for any advise.

[acbern]

I would think you have nothing to loose. sounds like the eeprom is dead. this could be the device itself, but is could also be its conrol logic. so befor desoldering, you should check if read/write access does occure. if so, the data probably is corrupted. if not, there seems to be an issue with accessing the device and you should trace that.

[HighVoltage]

Thanks for your reply acbern,
The 5V power to U505-A is perfect, and so is the GND connection.

I measured the signal at pin 3 (IN) of U505-A and at the end of the calibration cycle, the uController U500 is sending some data.
It really seems that the EEPROM is defect.

I will order a new part and replace the old one and then test again.

Enclosed is the circuit around the EEPROM U505-A
May be there is something else to look out for?

Thanks for any advise and help.

[HighVoltage]

I have ordered the EEPROM from Keysight
The part number is: 1818-5236
(IC 4K-bit EEPROM 500ns CMOS 8- SOIC)
Cost is: EUR 4.74

May be I am lucky and it is only the RAM

[acbern]

Keep us posted

[HighVoltage]

Just to follow up on this repair, it was a success.

The EEPROM from Keysight arrived today and I soldered it on to the PCB.
(See pictures)

The errors persisted until the instrument was calibrated.
After a successful calibration, the errors went away after a power cycle.
The instrument holds the calibration data perfectly and works really well on all settings.

Thanks for all your input in helping with this repair.

[y2khris]

How much did the 1818-5236 module cost?  I am looking at a similar issue at a 34401A I might be picking up.

[HighVoltage]

How much did the 1818-5236 module cost?  I am looking at a similar issue at a 34401A I might be picking up.

See above...

Cost is: EUR 4.74

This instrument still works perfectly after this repair.
Good luck with yours!