Agilent E3633A / E3634A Power supply - current calibration (13mA offset)

[michal82]

Hello

    I already have a working E3633A. He had a damaged transistor Q5.
The power supply is functional but has an offset in the measurement of current, which I cannot calibrate.
After switching the output on, the display shows 13mA.
However, the output current is 0mA because there is no load.
I suspected that this was an error generated by the ampmeter in the power supply.
To remove this incorrect indication I even gave a 20k potentiometer to compensate input offset on the U44 opamp. I connected 20k pot between between pins 1 and 8 (Vos trim) and 7 – it is responsible for measuring the current on the display.
Thanks to this, I could set 0mA on the display when measuring current.
However, CC settings were still incorrect.
To align the CC settings and current measurement, I performed the calibration procedure again.
To my surprise, after calibration the previous 13mA result returned on the display, the potentiometer settings were leveled.
So calibration “works” because it removes the offset caused by the potentiometer, which appears on U44.
Potentiometer re-adjustments 0mA on display and calibrations procedure caused the same behavior, so I removed the potentiometer.
The -13mA current is generated from the current source R147 and Q7.
When the CC setting is below 13mA, the output current begins to flow in the opposite direction from the Q7 current source and the voltage is negative -0.4V .
When setting CC is 0.001mA the current flowing to the output is -12mA.
CC= 0mA output curent is -13mA 
CC=13mA output curent is 0mA   
CC= 20mA output curent is 8mA

Maybe someone knows where the problem is.
Michal

[gerrysweeney]

Hi Michael,

Looking at the video, I don't think you have an offset or cal issue, there is definitely a fault. I would suggest you are going to need to look at each stage in isolation and determine if it's working as expected - the is no easy fix for these, its just an (isolated if possible) scope, DMM and lots of time sadly. Good luck with the repair.

Gerry

[CM]

I have a E3633A that fails OCP calibration.  I use an external shunt resistor to measure current, so it's not noise from an Electronic Load. 

The actual operation of the OCP is within spec.  When OCP is set to 20A, it trips at 19.8V, which is within the specification of +/- 0.5% 0.5A (which is +/- .5A at 20A).  Since the OCP works as expected, what can I do to address the calibration issue?  What does this unit do when it internally calibrates OCP and OVP?   

All other functions work fine.  The unit passes all verification tests when the sense inputs are not shorted to the main + and - inputs.  When they are, it fails the Voltage programming accuracy and readback tests at 20V.  But I think that's related to Op Amp U36. I don't think that's in play for OCP calibration.

Thanks in advance for any guidance.