Stolz ICD2 problem - seeking help

[philba]

Hi, I am banging my head against the wall over this problem. I built a version of stolz's design - serial interface but used an OpAmp to generate target mclr (as discussed on his site). I simulated the opamp in SPICE and bread-boarded just the opamp. I get good values and the target_mclr voltage is adjustable.

However, on the full ICD2 PCB that I built, I am unable to adjust the output voltage (target mclr) via the trimmer (R10). The voltage reads (via DVM) 12.6 to 12.9 which is within the range but should be adjustable and I believe the PIC is not able to control target_mclr. I tested the PCB for shorts and it is ok. I also measured the resistors and they are correct. The supply voltage for the opamp is 15.6V. I am noticing that the 15V supply is getting hot (though not excessively so) - I assume it is drawing excess current. A short makes sense but I can't find it.

The ICD2 connects to MPLAP (6.6, I believe) and I can download the os. self test passes and voltages read within 50 mV of my DVM measurements. It can erase chips but not read or write. I've confirmed the erase with another programmer. I assume this is due to the opamp problem.

Am I missing something obvious? I am stuck at this point. I would be very gratefull for any help you can give me.

I have included just the opamp section and also the full schematic.

 

[narccizzo]

you should test it by hand but with a pic that already has one pin in a logical 0 and other with a logical 1, then you make all the possible combinations with the circuit inputs, in three of the four possibles 00, 01, 10, 11, you have to measure 0, 5, and 12.5 to 13.5, if this is correct the circuit its ok, and the problem can be the pic firmware.