Re: EMC compliance question

On Sat, 12 Oct 2024 12:25:27 -0400, bitrex <user@example.net> wrote:

On 10/12/2024 11:22 AM, john larkin wrote:
>

48 is super common now. All our phones are PoE powered, which is
typically about 54 volts. Digikey sells warts up to 65.
 
The phones are cool. I can take one to Hawaii and plug it in and it
works just like it does here.
 
I imagine that europe has tens, maybe hundreds of millions of PoE
devices with the chinese version of the CE mark molded into the case.
 
So if european manufacturers realy have to do all the CE certs and
testing, they have one more reason that they can't compete with
imports.
 

>
So as of 2019 it looks like the US rules are similar to the European
"can't enforce" rules in that the manufacturer takes responsibility for
everything and it's up to the mfgr how and what tests they perform to
determine compliance:
>
<https://www.electronicdesign.com/technologies/test-measurement/article/21209868/recent-developments-in-emc-legislation>
>
This makes it sound like it's not too expensive to do some basic
compliance tests on a small-volume product in house:
>
<https://incompliancemag.com/emc-bench-notes-how-to-use-spectrum-analyzers-for-emc/>
>
Need a 1 GHz-ish spectrum analyzer at least as the main tool which
aren't exorbitantly expensive nowadays.

I can buy a spectrum analyzer and a surfboard antenna for under $1000,
and can take a product out in the country and do an open-field test
and crudely ballpark its EMI signature.

CE requires screen room testing and more quantitative measurement.

The reality today is that few products are honestly certified for EMI
or safety, and life goes on pretty well. If a product causes massive
EMI problems or hurts people, civil and criminal liabilities apply.

It's impressive how few EMI problems there are in real life.