audio-talk

Ok, so is it a bug that's causing the processor to reset?

Nick wrote: ↑Wed Jan 04, 2023 3:07 pm Ok, so is it a bug that's causing the processor to reset?

The processor is absolutely fine - the issue I mentioned happened several months back when I had it in that build with the switched mode supply. In this build, the software is running perfectly, no apparent glitches.

My mistake, I misread the word reset in the following:

I then removed the darlington chips for the volume control side and retested. The panel lights no longer dimmed, and this was an interesting data point - when the source is switched, the volume is reset as well (it stores the last volume setting for each source) so you get activation of the volume relays - and it was this that was causing the slight dimming.

Well, the issue is proved - seems the PCF8574 can't deliver enough current in this circuit to trigger two darlingtons at once - it has to be one per port.

I added some test diagnostics to the software which enabled me to probe around with a multimeter, and I got to the point where I could see the trigger voltages appearing at the input pins of the Darlingtons on the source select side, and the volume control side. Then I measured a corresponding output from the darlingtons on the latter, but not the former where the inputs had been paralleled.

As a final proof, straightening out pin 1 of one of the source select ULN2803's so that it was disconnected from the circuit (meaning P0 of the corresponding PCF8574 only had to trigger pin 2, rather than pin 1 and 2 together) then showed the darlington working on pin 2, but not the others.

So yep, problem solved, but it means a PCB redo.

I think we can close this one off now - thanks for the help everyone.

That's odd, looking at the circuit of the 2803 I would have expected at most 2ma per input. Ok, so now looking at the output stage of the 8574 it looks like it can at most source 100uA, but it can sink much more so I would think some 1k pull up resistors on the output would do the job.

Edit: TBH, I would have said its more luck that anything else that they drive one input without pullups.

Just realized there are pull up, it may just be that 47k is to high for the two inputs. So the 2803 specifies 0.93ma input current, that would suggest 5k3 so I would have expected the single inputs to be 4k7 not 47k. Wonder if it was a typo somewhere.