Help with 3458A board (repaired)

[PartialDischarge]

As some of you may know I got a 3458A some months ago. It's 18 yo but probably totally unused by the looks of it. Well it has a fault at bootup and I've been spending a lot of time trying to debug all the signals within. I've replaced many obtainable devices in A5 just in case and tested anything i can think of.
It may seem that the fault lies on A5 but the truth it that it seems fully functional, so I had my suspicions on A3. However, TiN even sent me the MCU on A3 but no luck, that wasn't it.

So in sort, this fault beats me. TiN offered to test my A5 board but I'm afraid I can't ship it to the other side of the world since I'll have to pay import taxes (as I had with the MCU he sent me for free, go figure).

Before buying another A5 board on ebay, which I will if there is no other solution, I'd like to ask if someone in Europe would be willing to test mine. There's virtually no risk since A5 communicates via fiber optic with A3 so this board is not going to damage any equipment

Here's a video of the current state of affairs. If I disconnect one of the fiber optics it will show "Isolator fault", ie the optic link is working and transmitting/receiving

[PartialDischarge]

Hello MasterTech - are you sure that all CAL data on board A5 is still present ? (no data corruption or data loss of the CAL parameters ?).

If I restore the original memory contents (which still work and are writable/eraseable) those CAL messages disappear.
I've tested both old memories and new NVrams sourced from mouser, I've tried with ROMs V8 and V9 too, but nothing.

[Kleinstein]

The video mainly shows missing calibration. So a first step would be to give it some kind of valid calibration data (old memory or just a dummy calibration even if not accurate). To me it is not clear what the problem is beyond that.

Is the unit able to run self tests and the ACAL ?

[TiN]

It doesn't matter about calibration data. Meter would still work normally even with default calibration values.
The issue is that board self-resets before entering main operation loop. Which leads me to issue with power supply (A6 board) or A5 itself.

As I mentioned before, this is a step before actual A5 talking to A3 MCU (which we confirmed at this point).

If that would be me, I'd hook A5 standalone to +5V bench PSU and try to test if it's stable (monitor reset, TX/RX, etc).
I don't remember if there is watchdog for reset, so take a look on that as well.

In worst case getting A5 from ebay could be the way, as it's not hard to resell the board if it happen not to be the issue. However seller who have A5 currently listed is bit shoddy, so beware.

[PartialDischarge]

I'll leave you scratching your heads a bit.
Same meter with original NVRAMS, does not display CAL errors, but blocks instead of rebooting 

Only hope I have to blame A5 for the fault

[engiadina]

I'd check the firmware roms on A5 ...

[PartialDischarge]

I'd check the firmware roms on A5 ...

Checked. The original ROM is ok and I also tried with V9. PSUs are ok. *everything* seems to be ok. Its like a deep down hidden logical issue.
Anyway someone offered to help and I hope this mystery is solved soon 

[PartialDischarge]

Good news, 3458A is fixed and calibrated  . ADC drift checked, input filter changed just in case, upgraded to option 001 which it didn't have. Will post pics when it arrives from Germany

Eventually I sent it to Adrian (ap) @ ab-precision.de who took good care of it. He was very professional and a pleasure to work with, I'll post the cal results later.

Also thanks to TiN for al the good help, rumor has it that he never sleeps as he will reply messages at any time 

So the problem was in the Gate Array from A3, which was replaced, and this is what was happening:

When A5 does not receive any bits from A3 at startup it will throw an ISOLATOR FAILURE error. However in this case it did receive some lonely bits. This fooled A5 CPU which did wait for more data, until a timer restarted the communication process again. This continued for 5 retries after which another timer did cause the reset of the equipment with no errors displayed.

[IanJ]

Eventually I sent it to Adrian (ap) @ ab-precision.de who took good care of it. He was very professional and a pleasure to work with

I can back that up!........highly recommended.

Ian.