Electronics > Projects, Designs, and Technical Stuff

EMC Pre-Compliance Visit: What to Expect and Bring?

<< < (5/6) > >>

techy101:

--- Quote from: Gyro on April 18, 2019, 07:18:22 pm ---20dB below the line on emissions is very respectable, especially in an ABS case.

--- End quote ---

I actually asked the guy to show me a run with nothing in the chamber because I didn't believe him. The board is probably going overkill on ground floods and ground transition vias, but it worked so I'm not going to try and get fancy with it. It's not a product that needs to be value engineered to the point of reducing drill hits.


--- Quote from: Gyro on April 18, 2019, 07:18:22 pm ---Floating the shells of connectors is always a bad one - you can get solder tags for those connectors to allow them to be connected together and to the right grounding point.

--- End quote ---

That was going to be the starting point, but I'm struggling with the "right" grounding point since there are four "grounds" on the device. We'll see how things shake out next week when the tools come in and I might end up making a dedicated post on the topic.

As for the 870MHz immunity, am I correct that there's no good and cheap way to test this other than going back to the test house? I also forgot to mention that it happened only in the horizontal antenna orientation, not vertical.

Gyro:
Yes, the "right" ground is tricky - but you've taken the right action, hiring an ESD gun.

Well 870MHz is around the UHF TV transmission band (don't know if you have that in the US), you're unlikely to have a TV transmitter but an antenna might be handy. Useful to know the horizontal polarization, I'm assuming that's the plane of the PCB too. In that case you're probably looking for a feature on the PCB that acts as a dipole at that specific frequency. For the fix, you may have to rely on going back to the re-test with a bunch of ideas (possible suspects), some low value caps and copper foil.

jmelson:

--- Quote from: techy101 on April 18, 2019, 07:39:57 pm ---As for the 870MHz immunity, am I correct that there's no good and cheap way to test this other than going back to the test house? I also forgot to mention that it happened only in the horizontal antenna orientation, not vertical.

--- End quote ---
You could get an oscillator, likely need a small power amp, and a hand-made dipole around that frequency.  Probably you can rent this gear, too.
I'd strongly recommend getting or renting the gear and setting up the immunity test in YOUR shop.  emissions testing is difficult due to the noise everywhere, but immunity should be simple to do in-house.  As long as you aren't doing it all day, nobody is likely to complain, unless it takes KW of RF to cause the effect.

Jon

Berni:
Yeah it should only take a few watts of RF power to reproduce any of those immunity issues if you stick the antenna close enough. Not too difficult to find a RF amplifier for that power and make your own antenna. Tho some gear for tuning the antenna to the exact freqency is helpful (Like an antenna tuner or network analyzer)

But i do recommend getting a small EMC chamber for your diy pre compliance testing. You won't get busted for blasting out RF a few times here or there, but you should generally try to avoid doing this as you have no idea what RF communications you could jam with your signal.

The EMI tents are the best way to go. They are simply a tent made out of conductive stainless steel fabric. They don't cost too much, are easy to ship when packed up and provide a surprisingly good amount of RF shielding. You can get them in pretty much any size you want. They make EMI emissions testing much much easier because there is way too much RF in urban enivorments to do a proper test and for EMI immunity it allows you to pump in as much RF as you want without jamming anyone and getting a visit from the authorities.

Solving EMI problems can be a very iterative process where you make fixes and test them. Some fixes do nothing even tho one would think they should, while other fixes that you think wont help at all actually help a lot. So having a way to do a crude test in your lab helps a lot and gives you confidence for when you go the real tests at the certification facility.

Oh and a poor mans ESD gun is a piezo igniter from a gas grill, they produce a pretty consistent high voltage pulse. If you can't kill it with one of those then it will likely also survive a ESD gun.

Gribo:
870 MHz is within the Cellular bands in north America. You can get a cellphone as an interference source.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod