For the second time in two years the switching power supply to my 14 year old microphone preamp is not passing power.

No rail voltage is present. Thinking the bridge rectifier may be bad, I replaced it. Still no rail voltage. I've searched for faulty diodes and found none to be bad. Today I decided to check primary voltage at the transformer. I powered up the board and, to my surprise, I now have power at the output pins.

Is there a particular component that would allow a power supply to work well for a time period, completely shut down for a period and then begin working properly?

I've attached a schematic for a similar IPD power supply. Mine was custom built by IPD for the application and has +29V, -29V and +52V outputs.

Thanks for you help,
mrvo

 

Hello,

It can be that a capacitor gets worn out.

I have circled the part that I suspect to be faulty.

Greetings,
Bertus

 

Bad solder connection.

First check the heaviest components that are mounted by their leads . The transformer is the most likely suspect if it is pc mounted.

 

Thanks so much for your input. I think I'll recap the 14 year old power supply. Who knows? I may find a bad solder joint while I'm doing the work.

mrvo

 

The art of systematic trouble shooting seems to be a lost art. Arbitrarily replacing components is amateurish. The reply from Flat5 has some logic associated with it, as PC mount transformers sometimes develop a 360° solder fault halo that can only be seen under a magnifying glass. This can happen over years of heating and cooling which cause expansion and contraction of the board.

 

Hello,

As I wrote, the capacitor gets worn out afther some time.
(specialy in a very warm surrounding as in the case at my work).
This will lead to this hicking responce of the powersupply as I have had this with several powersupplies at my work.
Due to the bad capacitor the resistor also got burned afther some time.
Thats why I noted them as "probably defective".

Greetings,
Bertus

 

Hello,

As I wrote, the capacitor gets worn out afther some time.
(specialy in a very warm surrounding as in the case at my work).
This will lead to this hicking responce of the powersupply as I have had this with several powersupplies at my work.
Due to the bad capacitor the resistor also got burned afther some time.
Thats why I noted them as "probably defective".

Greetings,
Bertus

I have no argument with pointing out probable defective components (known from experience with the item) but that's what the DMM and Scope are for. They should be tested before replacing. Sometimes trouble shooting can be a greater art than design.

 

Caps do go bad, but seldom go good again. Also, when bad, usually the voltage doesn't go away, it just goes out of spec.

However, bad solder joints act exactly like this, and it doesn't necessarily take hot-cold cycling to make it happen. These could have either started out bad, or gotten bad over time due to mechanical strain (heavy components, especially the physical mounting points that double as an electrical connection), or the previously mentioned temperature cycling (usually almost anything but a capacitor). Some things mounted with screws can also get loose and have intermittent contact on the case.

If you can shake it, bend it or otherwise mechanically contort it, and it makes it work again, then a mechanical connection is almost certainly the source of the problem. If you can't find it visually, sometimes you can find this sort of problem by pressing on different areas of the board (with a wooden prod). When you find the point where the least pressure "fixes" it, you found the physical location of the problem. Search nearby.

Failing this, you may have to resort to actually using a meter on it.