My 2002 Kitchen Aid built in fails self tests #3 (evaporator fan not coming up to speed) and self test #5 (compressor).

All other self-tests are OK, but #3 and #5 both return "02".

The refrigerator works - but I have to set the freezer section to "0 F" and the refrigerator section to 32 F in order to keep the refrigerator cooler than 40F. And it seems to run all the time. There is no ice buildup (I have had the evaporator "shield" off and checked that out.)

I have had two repairmen check it out. The first didn't even try the self-test and told me that it needed $700 worth of refrigerant. I didn't believe him since the whole evaporator was very cold.

The second told me he thought it was the control board, but he didn't know for certain. He told me that it definitely did NOT need refrigerant. Since he knew no more than that, I paid him for the service call minimum and sent him home.

So I took it apart myself. I checked the evaporator fan and motor. Everything spins freely. No binding whatsoever. When I go into self test, the motor spins slowly (maybe a steady 60RPM or so). It spins very smoothly, no jerks or hiccups. It doesn't seem to be labored. Unfortunately, I didn't put a voltmeter across the leads at that time to see if it was 12V or so.

But self-test #5 (compressor) failed as well. During the self-test, it doesn't seem to run at all. BUT IT DOES RUN IN NORMAL OPERATION. When running, I find that it is quite hot to the touch (probably about 170F or more), and the room was no more than 65F at the time. I have a small, almost unreadable schematic. It doesn't look like the compressor has a start or run cap. The control board must create 3-phase with some "H" bridges and and power it with that.

I put a Wattmeter in line with it and a short time after startup it was using about 360W, but it went down to about 210W after awhile.

I'll probably put a voltmeter across the evaporator fan motor fan tomorrow to see what the voltage is. Since it appears like I have 2 problems, I am reluctant to simply replace anything without knowing more.

The control board looks perfect when I inspect it. It doesn't appear that any part has been overheating. It is conformal-coated, so any hot part should have at least discolored the conformal coating.