Advice on repairing a TTI CPX200

[AudioFile]

I have an oldish TTI CPX200 dual PSU I’m trying to fix, and not having worked on one of these before I’m learning as I go along. I wonder if any of the experts here could give me their thoughts.

The PSU is working properly in that the voltage and current can be set and current is well controlled. However, it has an issue in that on one of the two channels, when power is enabled, the reading for actual current is all over the place. The display starts at around ‘20A’ and goes up to ‘40A’, even though the PSU is delivering the ‘correct’ current (as set by the front panel controls).

I have tracked the problem down a bit to either the 7109 ADC or one of the CD4051 multiplex ICs, but want to be sure of the likely cause before I put the soldering iron on it.

At the input to the A/D converter, pin 35, there is a 4Hz squarish wave of +/- 50mV on the good channel (pic 9), but this is <10/0mV with a 450mV negative going spike on the bad channel (pic 2).

At pins 30, the buffer stage, pic 3 shows the good channel and pic 4 the bad one.

After the buffer and integrator stages of the ADC, there are huge spikes and waveforms up to 3V at pin 32 (see pic 7 (bad) and pic 5 (good)), and I’m assuming that these are the result of the spikes at the input being amplified and integrated within the ADC.

The reference voltage on pins 36 and 39 is 187mV on both good and bad channels.

Working backwards from the ADC input, there is the spike on the output of the ‘high’ MUX IC 13 pin 3, but I’m not sure if that is causing the issues with the ADC or whether the ADC is producing the spikes. Everything before the MUXs seems to be fine. The Output Current Sense signal is clean and linear, giving 9mV/A into the MUXs. The digital control signals to pins 9, 10 and 11 of the MUXs are also fine (pic 8).

I know the microcontroller (IC11) is fine as I’ve swapped them on good and bad channels.

All suggestions gratefully received before I take the invasive step of removing the ADC.

[fcb]

TTi are good at releasing manuals wiuth schematics - drop their tech help line an email.  There is often a troubleshooting section.

It does sound like you've got rubbish being multplexed into the ICL7109, you can check the inputs to the 4051? And all the rails?

[AudioFile]

Thanks for the super fast response fcb.  I have the service manual and it has a good circuit description but no troubleshooting as such.

Inputs to the 4051 are fine, so I'm suspecting either that (IC13 - the high MUX on the bad channel) or the ADC.

Rails are not too bad, no obvious mains ripple, although there is 'fine grass' of about 50mV on all supply lines good and bad.

[fcb]

The fine grass probably just pickup on your scope gnd. Red-herring.

Can you lift the MUX output leg and examine the switched waveform before it hits the 7109?

[AudioFile]

Just lifted the MUX output and glad to see the spikes are there on the chip! And the large excursions on the 7109 are gone.

Replacement 4051 now on order.

Many thanks fcb - don't know why i didn't think of that. Wood...trees I suppose!