Fluke 5500A Startup-Error

[Echo88]

At the moment im trying to repair a Fluke 5500A Calibrator and cant find the fault.
When powered up,the two Displays say "FLUKE Calibrator Starting up", after 2s "Starting up" vanishes and the switch-LEDs on the keyboard "Operate, Earth, etc." stay lit indefinitely.
So it seems the calibrator is stuck in the boot-loop. I optically checked every board and deconnected/connected every plug on the boards to clean them a little bit.
The 5V/+-12V-test points on the motherboard/power supply are okay and i also could measure those voltages on the cpu-board at the caps.
Before this startup-error the calibrator also couldnt finish its Zero-Calibration at the TC-Terminals, so i suspect that this error led to the startup-error. The calibrator also doesnt answer via RS232 and isnt visible on GPIB.

At the moment im thinking its either a defective cpu-board (maybe corrupted data because of the unfinished Zero-Calibration) or a component (opto-coupler for example) which doesnt answer to the cpu, so it does get stuck in the startup.

Maybe someone knows this problem or has access to the schematics of the calibrator?

Edit: After the inspection by Fluke they said the CPU-board had an error and built in a new one. After that the problem was gone.

[TiN]

I'd start with taking photos of all stuff and posting . We missing F5500A teardown thread in metrology section 

Outguard section troubleshooting perhaps good start. If power is all good (there are +12, -12 and +5V), check resets and check that there is life between outguard CPU and inguard CPU on optical interface.

For schematics perhaps you can reference to 5700-5500A zombie, 57LFC manual, which have schematic diagrams and resemble 5500A for most of the parts. Also check if there is life between front panel and outguard CPU. I have 5500A A2 board and front panel LCDs, which I do not need.

Be careful not to get zapped, there are high voltages inside the inguard box.

[Echo88]

I measured all voltage test points, checked caps, fuses, etc. and still couldnt find the error. Theres no opical communication in the 5500A. The 57LFC indeed was useful and the mentioned cpu-board is very similar to the one in the 5500A. 
 
I still suspect the cpu-board, maybe an error in the RAM. My supervisor has now decided to get it repaired by Fluke. Will be interesting to see what they locate as the error-source.

Anyway, it was an interesting teardown and ive collected photos for the Metrology-guys: https://www.eevblog.com/forum/metrology/fluke-5500a/

@TiN: Id like to send you the uncompressed photos for your archives, can you tell me how i can upload them?

[TiN]

Usually Fluke does board level repair for these calibrators.

You can upload pics to my FTP by following these instruction.

[Kenny Powers]

Hello,

Did you ever find a solution to this? I have 2 with similar symptoms if left long enough they eventually start and work as normal?

[Echo88]

Nope, mentioned it in the first post:

CPU-board was faulty and Fluke changed the CPU-board for the repair-fee, after that it worked flawless.