Conducted emissions on the AC line side can be handled two ways. If you’re
using a Part 15 qualified AC adapter, you should be fine. Otherwise, you need a
LISN and a spectrum analyzer to do a pre-qualification test. You can buy or
rent a LISN. My RFI web page has info on how to build one. As for radiated
emissions, W0LEV has already given an outline of that. It’s challenging to
build your own test facility for that, but as he said, you can probably find
any really bad problems with some simple tests.
As for the Raspberry Pi, I have several deployed in ham stations and they are
fairly noisy. ALL connections to the Pi require filtering in order to avoid RFI
in the HF bands, at least if your antennas are fairly close. The (wired)
ethernet line is surprisingly bad and requires a CM choke; it’s not just
ethernet-related racket that is carried on that cable. I can’t speak for the
HDMI port but I trust it not. For 5V power, I use either a carefully-filtered
dc-dc converter off of 12V, or a genuine Apple USB power adapter, which is the
quietest switching supply that I have ever measured.
-Gary NA6O
> From: Charles Gallo <charlie@thegallos.com>
> To: rfi@contesting.com
> Subject: [RFI] Not QUITE OT
>
> Hey Gang,
>
> It is starting to look like the company I'm working for MAY (note the
> MAY) start selling complete devices, vs components, which of course
> means Part 15 etc.
> Now, as far as we know, simply because of what we measure, out existing
> stuff is low noise (I2C bus on one side, all uV and mV level near DC
> (basically DC, but the levels tend to fluctuate over many minutes -
> think temperature in a fish tank, or pH - makes for the worlds most
> boring graph why you look at a number that is either 7.1 or 7.2 all day,
> every day, and the value is a DC mV signal. We had to do a LOT of work
> to make the circuits EMI clean on the INBOUND signal, which tends to
> clean up the outbound
>
> Anyway, I'm thinking "How do we informally test our stuff BEFORE we go
> out to the Part 15 lab?"
>
> My actual biggest worries are the USB power supply (I posted one here,
> someone asked RE RFI - quick answer - the Made in the USA version SEEMS
> ok, the one that looks the same, weights 1/2 as much and costs 1/2 as
> much from China is a POS, and we're never ordering them again), and
> believe it or not, the Raspberry Pi we plug our probes into.
>
> Anyone know of a good informal way? I assume a spectrum analyzer, with
> some probes...
>
> Also anyone know if the RPi is clean, assuming clean power in?
>
> --
> 73 de KG2V
> Charles Gallo
> http://www.thegallos.com
>
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|