I recently acquired an Agilent 53131A for free which when first powered on seemed to work fine. I tried adjusting the internal clock around the back but it didn't change the reading and then i noticed the High stability option on the back was marked as installed (option 010). So I opened it up to adjust it and thought "well I'm in here might as well clean out the fan (mad noisy)" took out the power supply to get at the fan and cleaned it out and added rubber grommets to the mounts to get it to be quieter. Then i put the supply back in and powered it up...... "FAIL: GPIB" and then nothing none of the buttons would work at all or do anything. I restarted it into the utility menu (POWER+RECALL buttons) and it got into the menu fine moved to test all and started it instantly failed gpib but then skipped to testing FPGA but got stuck. So i was wondering what could I have done to fuck this up? I unplugged the option to see if the counter worked without it and found it worked fine without it self test and all.
So if the option was causing problems with the self test maybe something I screwed up so i took it out, found the pinout for the TCXO itself and took the TCXO off. I first applied power to the heater (rated 20V to 30V) to see if it was fine and it was constant power input (current lowered with increased voltage) and it heated up happily no issues here. Then I powered the OSC supply with 12V and on the 10MHz output i read a happy (roughly) 10MHz on the very 53131A i pulled it out of. So since the TCXO module itself worked i suspected the board itself with the DAC and signal conditioning. Checked for shorts and dodgy caps nothing wrong found. After a while of having the TCXO powered it plugged it into the board and installed it into the 53131A to measure the operation in circuit..... "Self Test Passed" so now it worked fine...
Now i tried to see if calibrating it was possible and the on module tuning adjustment was fine and worked did some tuning fuckery but couldnt get it inline against my reference (which was the REF out of a rigol DG812 function gen). Initially it was inline but as it heated up it moved well out of range of the rigol reaching a delta of 610.4Hz by the time i turned it off for the night to continue the following morning. This drift was constant and seemed to be linear.
So in short, I have a strange issue that seemed to have fixed itself and then broke overnight by some angry elves overnight. Does anyone have any suggestions on the possible issues here?