EEVblog 1418 - The Most EMBARRASSING Repair

[EEVblog]

This "Repair" of a Keysight DSOX1102G oscilloscope was a tad embarrassing...

[TheSteve]

The scope clearly doesn't like it there, even though it now appears to be working again it should be sent to my lab for disposal. Keysight products are always happy in my lab.
 

[skander36]

At that time also my "hacked" scope (DSOX1002A) has beginning to behave like Dave scope. But after reverting the hack, which many times has been succesfully, last time has not been return to normal. Instead it has failed calibration on channel 2 and deepened this simpotm with every failed calibration, until both channel have completely get out of scale. No reset, downgrade/upgrade fw, have been able to restore it. After sending to a local dealer they replaced it with a new one. 

[Kleinstein]

Different hardware configurations can have a different format for the cal data (e.g. 2 channel version needs less than 4 channel). An incompatible data format coud srew up the calibration.  Having a different HW than the SW expects can definitely make the internal CAL fail. The resistors are not just to sell crippled versions. Chances are they have the same SW for different versions like a 2 and 4 channel verion and they may use those resistors to detect those diferences too, though some may be detectable in a different way too.

[Bud]

At that time also my "hacked" scope (DSOX1002A) has beginning to behave like Dave scope. But after reverting the hack, which many times has been succesfully, last time has not been return to normal. Instead it has failed calibration on channel 2 and deepened this simpotm with every failed calibration, until both channel have completely get out of scale. No reset, downgrade/upgrade fw, have been able to restore it. After sending to a local dealer they replaced it with a new one. 

Seems you made an improper selection of the strapping resistors, especially if you did not apply the channels and trigger hardware mods. Or had a poor soldering job.

[skander36]

Poor soldering job can be excluded, but I have done succesfully multiple times, mod and revert. There were just two resistors, so as long as the mod was succesfully more times, it is not possible to put other resistors.     
But at one point in time reverting was no more posible.

[amyk]

I suspect this could be something the AgilentKeysight people put into the firmware to mess around with those trying to hack their way to a free upgrade...