Hi,

I'm just wondering how I could test a mains noise/suppression filter. The same as this:



Something in my synthesizer emitted smoke, but I only recently got around to checking it. There are just two things in the area from which the smoke came that could have emitted smoke: this or the transformer. This smells quite funky which leads me to suspect it, coupled with the fact that the transformer is outputting the correct voltages and thus seems to be operating normally.

So how can I test if the noise filter is busted, besides taking it out and splitting it open to check the components inside?

 

This is tongue-in-cheek because it is a lot of work to perform correctly, but it will give you an idea of how it is done formally and you can improvise from there.

Put your filter in place of the filter in a switching power supply, the build or make one of these:

Line Impedance Stabilization Network

Switch on your power supply and look at signals on the output ports of the network with a spectrum analyzer equipped with a quasi-peak detector. You can compare your results with the table below.

 

Thanks for the reply.

I don't have a switching power supply (or a spectrum analyser for that matter) to test. I suppose it'd be easier for me to just replace the noise filter if I suspect it's busted?

 

If it is suspected of smoking you can do a quick and dirty test. Assuming the parts inside are laid out according to the diagram on the side:
Check resistance between AC3 and 5 - should be <5Ω
Check resistance between AC4 and 7 - should be <5Ω
Check resistance between 5 and 7 - should be infinite
Check resistance between 6 and 7 - should be infinite

 

Thank you! Yes, quick and dirty is what I'm after.

 

Sorry, I missed a test:
Check resistance between 5 and 6 - should be infinite.

 

This is my particular one. In any case, it is the same.

So the readings I get are as follows.

AC3 - 5: .5 ohms
AC4 - 7: O.L (infinite?)
5 - 7: O.L
6 - 7: O.L
5 - 6: 49.4 ohms

So... Not as expected! Which means it's busted as I thought?

 

Yes, 5 to 6 at 50Ω is definitely bad.

 

It's replacing time! Thanks a million for the help, very much appreciated.

 

Did you disconnect the wires from the filter to prevent you from getting misleading results due to other items it is connected to ? I would have also suggested taking capacitance readings between 5 and 6 (I would expect a little over 0.1 uF ) and also between earth and 5 and earth and 6. (I would expect these two readings to be a little less than 6600 pF.)

Les.

 

Hi. No, I didn't disconnect it. Of course I should have. Oops. I'll do so when I get some time during the week. Even still, if 5 to 6 should be reading infinite, that I'm getting any resistance at all seems odd, does it not?

I'll remove it and check capacitance too. Thanks.