I'm a novice electronics engineer and I am currently trying to fix an old linear power supply of a video door phone system.

This power supply works as intended with some of its features (12 VDC and 12 VAC outputs) but on the 18 VDC timed output it has a problem.

When the control signal (a button is pressed) from the door station then a relay is energized and 18 VDC output is fed to the monitors and camera. You can set the time with a potentiometer from 20 seconds to 90 seconds. The problem is that when some seconds have passed after the button is pressed, the relay starts to bounce on and off for some seconds and then to off. This happens only when the time is about to run out. I have checked all the components with my multimeter and they seem to work fine.

I attached an image of the circuit. If you need some more I will provide them.

Do you have any idea what causes this relay bouncing effect?

Edit: There are no components beneath the PCB.

It seems I have resolved the problem. As the circuit was active I noticed that the right transistor of the darlington pair was overheating as the voltage was dropping (time out). So I changed it and the supply works as it should. The transistor was a BC 337-25. The new one does not seem to overheat. As I was checking the "bad" transistor with my LCR-TC1 tester, I couldn't see what the problem was, and also the multimeter diode test was OK. Is this possible?

Power supply pcb

  • \$\begingroup\$ If you have access to the schematic could you share it with us, please? \$\endgroup\$
    – Electrocol
    Commented Jul 24, 2023 at 21:02
  • \$\begingroup\$ Please update your question with a photo of the underside of the board. Often there are surface-mount components on the bottom side. Also, are there any markings on the relay? Also, Can you post a scope shot of the coil voltage when chattering occurs? \$\endgroup\$
    – AnalogKid
    Commented Jul 24, 2023 at 23:40
  • \$\begingroup\$ Check my edit please :) \$\endgroup\$ Commented Aug 28, 2023 at 19:51

2 Answers 2


That it works to some degree tells me it's not likely to be something like a bad transistor.

In older equipment as this appears to be a common cause of problems is the electrolytic capacitors. There may be some AC ripple in the relay circuit voltage due to reduced filtering, or a bad cap in the timing circuit may be making it go wonky (that's a technical term).

If you have an oscilloscope you can check for excessive ripple on the DC supplies, if not you can use a multi-meter on the AC volts setting to get some idea of how much ripple there is.

Another thing that happens in older equipment is the resistor values drift, that might cause a circuit to be working right on the ragged edge.

Capacitors can be checked out of circuit with an ESR meter, simply testing for capacitance will often miss bad ones. And there's always the old 'just replace them all' approach.

  • 1
    \$\begingroup\$ I'd add that potentiometer might be a culprit too since they often oxidize. \$\endgroup\$
    – Julien
    Commented Jul 24, 2023 at 21:42
  • \$\begingroup\$ Check my edit to the question please :) \$\endgroup\$ Commented Aug 28, 2023 at 19:51
  • \$\begingroup\$ @j_pentester Well that's an odd one. Sometimes a transistor can degrade over time or from something like a transient, it doesn't stop working completely but it's characteristics change. This could make it heat up and when it gets hot enough it starts working erratically. At least it's working now. \$\endgroup\$
    – GodJihyo
    Commented Aug 29, 2023 at 18:16

If the voltage across the relay coil is decreasing slowly, it passes through a point where the relay armature (the part with the moving contacts) is making very light contact with, or is suspended above the stator (the structure with the fixed contacts). There is a burst of output changes until the coil voltage is low enough that the armature is completely disengaged from the stator. This is called relay chatter.

If the unit was not doing this before, it is possible that a component failure has changed the rate at which coil voltage is decreased.

  • 1
    \$\begingroup\$ Fixed. 5 4 3 2 1 \$\endgroup\$
    – AnalogKid
    Commented Jul 24, 2023 at 23:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.