Tektronix DPO3000 SPC Fail Error

[analogRF]

I had not run SPC on my DPO3034 for a very long time and today i realized that it fails SPC 
It fails at almost the end of the SPC run. But self tests always pass even in loops of 10 it never fails. also the scope shows signals correctly on all channels.
I ran the SPC in manufacturing mode with no difference but I can now get a slightly more meaningful error message in the error log
which unfortunately goes beyond the window  the morons who wrote the FW didn't wrap the text so that it continues on the second line instead of disappearing out of the window....

anyways, using :ERR:FIR? command from the scope's web page I can finally extract the error message as this:

Code: [Select]

846: 3 0 0 Calibrate 22:52:11 03/03/2021 CH4 NREJ comparator A 2mV/div -FAILED- 2mV Trig Hyst Rslt too hi

any ideas? any suggestions as to where I should be looking at? I believe NREJ means Noise Reject mode in Trigger setup
and the rest are pretty much self explanatory except the last two words "too hi"? what is "too high"?

I noticed that if I change down to 2mv/div or 1mv/div on all channels, CH4 has more noise (some spikes actually) than the other 3 channels
Those other 3 channels dont show spike noise. Could that be the cause?

any suggestion is appreciated

[analogRF]

I cannot figure out how the trigger works on this scope
I mean which chip is responsible for trigger? In other Tek scopes you can always find a custom ASIC that is clearly doing the trigger
but in this scope I suspect it is done by the Altera FPGA maybe? I dont see any 'comparator' chip which is referred to in the error message

[YetAnotherTechie]

I would try running the spc several times from cold to warm, maybe a paramenter is outside a tolerance-from-last window.

[analogRF]

I would try running the spc several times from cold to warm, maybe a paramenter is outside a tolerance-from-last window.

I have run it maybe 10 to 12 times by now both cold and warm and always the same error. I have not changed location though. I did turn off
everything else in the room even the lights.
I wish I knew this is happening in the front end or further down the line.
I think those small spiky noises in CH4 at <=2mV/div, must have something to do with this...

[YetAnotherTechie]

this may sound silly, but what about putting a BNC cap on the port?

[analogRF]

this may sound silly, but what about putting a BNC cap on the port?

yeah I had done that too 

[analogRF]

I did a test yesterday that I wish someone else on the forum can repeat and let me know the results.
I applied a small 8mVpp to all channels (50 ohm) one by one and tried to have stable edge trigger
only on channel 4 trigger is sort of not as stable as the others. occasional ghosting occurs
also there is extra noise on the signal. at higher amplitudes like 100mVpp for example, this never occurs
changing the Noise Reject trigger does not do anything on any of the channels

surprisingly I did the same test on my Agilent DSO-X3054A and guess what!?
exactly channel 4 behaves exactly like the DPO3054. Trigger is not completely stable like the other channels when amplitude is very low!
but that scope never complained when I ran the self calibration!

can someone try this experiment?

I cannot figure out how the trigger circuitry works in this scope. In most other scopes it is easily figured where the trigger signal goes
with all the complex triggering options that it has, I suppose there must be a dedicated ASIC for that but I cannot figure where/what that is

[analogRF]

does anybody have an older firmware for DPO3000? I mean older than 2.31?
the last time I had run SPC on this unit and it passed was long time ago and at that time it had FW 2.25
later on I updated to 2.40 but never ran SPC...I would like to try that old firmware again but I cannot find it

[analogRF]

I also changed the room and ran the spc again, same fail 
still I would like to try FW 2.25 if someone can post a link to it.

can it be a capacitor failing somewhere? It must be a small problem
Bizarrely my Agilent 3000a behaves exactly like this on ch4 only but didnt complain when I ran calibration

[SMdude]

I would try heating around the board(hot air solder station @ 140c or so) and see if that has any effect.
If there is a bad cap or something like that it might start to work when it gets hot. And then it probably will work for a while just to make it hard to track down!

[analogRF]

I would try heating around the board(hot air solder station @ 140c or so) and see if that has any effect.
If there is a bad cap or something like that it might start to work when it gets hot. And then it probably will work for a while just to make it hard to track down!

done that already, no change. the thing is the front end is a separate board underneath the main board so
it wont see the heat directly.
if anyone has FW rev 2.25 or older please let me know