2

I thought, I will not bother any more with this project of mine, but I really run out of ideas trying to get it to work.

When using breadboard, the design works as expected (did it couple of times already), however, when I put it onto universal (3-track) PCB, it stopped to work.

I divided the testing into two parts: disconnected R8 to see if U1b generates sound. It does not. The breadboard setup was quickly done using the same pins and yes, it worked right away.

What I have done:

  • List item
  • measured "voltages" on 8, 9, 14 and compared (I am aware this is oscillating, so values are averages, I guess). Values were very similar.
  • touching pin 8 by finger actually produced repetitive sound on the PCB solution, while on the breadboard counterpart it added noises, otherwise reaction was similar.
  • checked if C1 is not shortcut, and trying another capacitor
  • U1a part also did not work, but I have not researched it extensively.
  • of course, checked continuity of all junctions and checked the schematics, more than once

The question is, what else can I try to pinpoint the problem or is it time to rework, for example, get IC out and check if it's still good? (I think, I was very careful and quick soldering IC with 350 deg.C iron, but...). I do not have an oscilloscope yet.

The picture of the board and breadboard is more an illustration, as connections aren't seen. The breadboard implements only oscillator part, but at prototyping stage I had both.

schematics pcb vs breadboard

UPDATE: Solved by adding more resistance to Rload, thanks to Rafal Powierski.

UPDATE2: Datasheet for the comparator also advices to connect unused inputs to ground, as Spoon suggested.

Roman Susi
  • 985
  • 3
  • 13
  • 30
  • First are all your unused inputs pulled up/ down to 0V or 5V? These can oscillate and influence other inputs (the proto board may not have experienced this as it has a small amount of extra capacitance.) Do you have decoupling capacitor for the IC supply ? Finally when prototyping I use sockets for the IC s just in case I have to debug this sort of issue. – Spoon Jul 21 '13 at 15:37
  • I have tried 10uF capacitor - no effect. Unused inputs aren't connected. I can try it, thanks. – Roman Susi Jul 21 '13 at 16:03

1 Answers1

1

Output stage might have insufficient current to drive low voltages, because of low resistance of speaker. For testing add a serial resistor 470 ohm ... 1k in series with speaker. Output stage drive capability is varying from chip to chip, depending on output transistor beta. (assume 6mA as minimum)

Rafal Powierski
  • 54
  • 4
  • Same speaker works perfectly on breadboard with 100 ohm resistor. I will try your advice however. Thanks – Roman Susi Jul 21 '13 at 16:05
  • Thank you! It's a power issue indeed. When I added another 100 ohm resistor in series to the speaker, I got it to work. I've always prototyped with TWO separate ICs, assuming, there is no much difference. – Roman Susi Jul 21 '13 at 16:15