EMI/EMC Tests on a Simple Power Supply Panel

What does the end Client's standard have to say?

Would I be asking here, had the end client said anything?

Is any reader here at all interested in rhetorical questions (rhetorical question - NNTR)?

If the end Client's standard is absent, and there is no call for it on any other document, then the concept of a <...warrant...> is a non-starter, in which case so is <...asking here...>.

"Would I be asking here, had the end client said anything?"
Probably "yes", as the question implies you don't have much experience of the subject .
Ignoring your sarcasm and trying to be helpful:-
You could do a paperwork exercise showing that any reasonable or likely failure doesn't cause a dangerous outcome.
The problem with most EMC testing is that you define your own fail criteria.
E.G If it's a pump, "stops running" could be considered safe, whereas "continues running indefinitely" could be considered a failure.
However I'm sure we can all invent scenarios where the opposite could be true.
Del

https://en.wikipedia.org/wiki/Electromagnetic_compatibility

Probably none.

As stated previously by PWSlack, does your end user/customer require this to meet any standards such as MIL-STD-461, CISPR, FCC, CEN, BSI, VDE?

https://en.wikipedia.org/wiki/Electromagnetic_compatibility#Laws_and_regulators

Is the concern that this power supply panel will affect other nearby equipment or is the concern that this power supply would be susceptible to other nearby equipment?

Is this equipment for your own in-plant use or for sale to others? That usually makes a big difference in your decision making process.

Some years since I was directly involved in that stuff, but basically in Aus

• If you are selling to a client on their engagement, then they specify the expectations.
• If you are intending to sell it retail/wholesale for use by others then you need to satisfy whatever the local/federal expectations are for EMI/EMC and get the appropriate type testing done to get your compliance certificates.
• If you are intending to use it yourself, then the local power authorities will have expectations on the effect on metering devices and such in their supply system.
• If there is evidence of interference with other appropriately designed and manufactured items owned/operated by neighbours, then you would need to rectify the situation.

For you to decide whether testing is warranted depends on the level of risk you are prepared to take.

Note that most testing authorities would conduct "indicative" tests using all the necessary equipment and such without certification for significantly reduced costs to those requiring certification. They are mitigating their own risk.

If you decide to test, then you still need to determine the appropriate testing. Is this going to be a consumer device, or a simple item used within a HV switchyard for relay operation?

It depends on where you are selling the panel. In the EU it is mandatory that the panel complies with the 2004 EMC Directive regardless of the client specification. Up until the Directive came into force you could effectively self certify through the Technical Construction File, but that route is no longer available. I am not sure about other parts of the world but in Europe you have to test.

I sounds like you don't have anything that will oscillate so for most of us the default answer would be "no". Now, if your application is something special or you have customer or legal requirements you didn't disclose then the answer might be "yes".

If you're testing something it has to be for a reason, either to certify for some defined use, or you are experiencing interference from an unknown source in some nearby equipment, or if you suspect that it might pose a problem in certain locations and you wish to mitigate any problems going forward, or you are required to by law or customer request.....Does your panel fall into any of these categories?....then yes...if not, then no....

I used to deal with transformer - rectifier - diode - choke - capacitor supplies at a telecoms company. Output 24 volt 15 amp to 1000 telephone channel microwave radio equipment. There was a VDR on transfo input to stop a serious on-off spike but I cannot remember even snubbers on the diodes, certainly no input - output EMC filters & they were emission tested

Nothing you describe in the panel is likely to be susceptible to interference from anything else. I suppose this is not a high power item capable of affecting the supply, something that could plug-in? Any transformers? I take it "rectifiers" are just diodes & not SCR or a switch-mode power supply?

The MC...Bs operate so rarely they are outside the scope of emission regulations & most installations have them anyhow - if they are to a standard like CE, they will already meet local rules for EMC.

Only the rectifiers can generate noise [by charge storage, they make a spike every time they go from conduction into reverse bias]. This can be reduced by a snubber (R-C network) across each diode.

It matters a lot if you are making a single panel for use on your own site & if it is close to other buildings owned by others. If it is your own site & build & not for sale you are only likely to be affecting your own equipment or susceptible to own installation. The regulations allow this without testing.

If you are a telecom installation you might do testing for your own security.

I do not know your situation or power rating , but I would probably go for snubbers and a line filter without testing. Also R-C/VDR suppression of any transformer primary/contactor/relay coils (helps control contact/switch life anyhow). The other question is whether the output goes to much cabling which could radiate or conduct to loads. Unsuppressed relays or coils connected to controls/instrumentation were the only things I found causing trouble.