CE compliance testing

Wed Aug 18, 2010 5:11 pm   

I have a small electronic unit which needs some CE compliance testing.

Cost is a major issue and was wondering if anyone here can offer advice on
an affordable solution.

Wed Aug 18, 2010 7:19 pm   

On Wed, 18 Aug 2010 17:11:21 +0100, Fredxx wrote:


You should contact a certification lab in your area. CE and UL are mostly
a lot of paperwork with a small amount of actual testing. They are
designed to be fee generators, and in CEs case trade barriers, do there
isn't going to be a cheap way to do it. FCC is easier because it's based
on physics not on the whims of bureaucrats, you can do FCC certification
in a few hours because it's just a matter of measuring the emissions of
the system, you are either within bounds or not.

Wed Aug 18, 2010 9:10 pm   

<fredxx_at_spam.com> wrote:


If you are located within Europe CE conformation is just a matter of
putting a CE sticker on your product. It really is simple as that. The
sticker says you take responsibility for the product.

If you really want compliance testing then it is advisable to find a
company with an EMC test setup to do some pre-compliance testing. This
will tell you where the problem areas are. Next you'll need a spectrum
analyzer to find the problem areas in your product.

This is a very cheap one but it does the job:
http://www.door2doorshop.com/wholesale/Atten-Spectrum-Analyzer-AT6011-213.html

If you shop on Ebay you can get these for half the price (new
ofcourse).

The last step is shopping around for a cheap EMC lab.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico_at_nctdevpuntnl (punt=.)
--------------------------------------------------------------

Thu Aug 19, 2010 8:21 am   

"Nico Coesel" <nico_at_puntnl.niks> wrote in message
news:4c6c3aef.2204064734_at_news.planet.nl...

The two replies above do not accurately reflect the practical and legal
situation in the UK with regard to CE marking.
You can just put a sticker on your product and a declaration (mandatory) in
the manual and you may get away with it. If your product is investigated and
does not comply with the regulations then when they discover that you have
done no testing at all Trading Standards (the UK authority which enforces CE
marking of electrical and electronic devices) will throw the book at you.
If you have done proper testing and have some evidence (in which case it is
unlikely that you will have any problems anyway) they will discuss the issue
with you in a sensible way and give you time to fix things.

You need to test for radiated RF emmissions, susceptibility, electrostatic
discharge susceptibility (8kV sparks to any exposed part), power surges from
anins or power supply. If you are connected to the mains you must also test
for mains pollution as well.
You must also comply with ROHS and WEEE regulations.

A lot of small organisations don't bother with some or all of this - for
obvious reasons. If you are going to spend money its worth going to a decent
lab who will advise you about what you need to do. If you are short of cash
then you should understand the rules yourself.

My experience is that it is unusual for a device to pass all the tests
without some tweaks to the design.

I don't know your circumstances but my advice is that you are taking a big
risk with customers and enforcers if you are going te sell a lot of stuff
and don't test.
If you are selling a few parts you have a good chance of getting away with
it - but you are screwing your customers by claiming things which you
haven't tested and they will never forgive you if they find out.

I'ts all horrible - good luck.

Michael Kellett

Thu Aug 19, 2010 2:11 pm   

"General Schvantzkoph" <schvantzkoph_at_yahoo.com> wrote in message
news:8d2mieFco2U4_at_mid.individual.net...

I have, and we're talking about £1,500 just for a pre-test, and £3,000 for
final testing and documents. £300 just for a plots is OTT.

Thu Aug 19, 2010 2:16 pm   

"Nico Coesel" <nico_at_puntnl.niks> wrote in message
news:4c6c3aef.2204064734_at_news.planet.nl...

I am aware that you can self certify, but I wanted to collate some evidence
that it would actually pass the EMC requirements, especially for radiated
emissions. I'm very confident about the rest.

I can see the cost of a spectrum analyser is a fraction of the cost of any
assessment or testing by an EMC lab. Many thanks for the pointer and am
looking on eBay for suitable equipment.

One thing that does bother me still, is sourcing wideband aerials suitable
for measure field strength.

Many thanks.

Thu Aug 19, 2010 2:21 pm   

"Michael Kellett" <nospam_at_nospam.com> wrote in message
news:VtWdnRCx4Ob0RvHRnZ2dnUVZ8lqdnZ2d_at_bt.com...

That is perhaps the primary reason I want some evidence of compliance
testing.


ROHS is easy to comply with if sourcing known components from reputable
companies.


That is the concern!!


Many thanks.

Thu Aug 19, 2010 2:36 pm   

On 8/18/2010 5:11 PM, Fredxx wrote:

I have no idea if this is appropriate to your situation, but you might
avoid a lot of the CE compliance issues by using a wall-wart or somesuch
which is already CE marked. Then all the mains powered stuff is
pre-certified by someone else. Get one with a ferrite thingy near the DC
jack!
HTH., Syms.

Thu Aug 19, 2010 2:42 pm   

"Fredxx" <fredxx_at_spam.com> wrote in message
news:i4jb3m$5nq$2_at_news.eternal-september.org...


There is a few hands-on test that I always do to get an idea if my design
has weaknesses:
-Does it work when I call someone with my cellphone and put the phone over
my electronics (remember to make noise to get max radiation).
-Does it block my cellphone? You may try a FM radio closeby to see how bad
it is.. Compare with similar products.


If your project cant afford a test, then its unlikely that you will ever get
sued for not being within specs. You are just "noise" in the marked.


On the other hand, be sure that nobody can get killed by touching your
product. And that it doesnt burn down the building if something fails.
Then you can put the CE mark on it and sleep well at nights.

Thu Aug 19, 2010 3:54 pm   

"Symon" <symon_brewer_at_hotmail.com> wrote in message
news:i4jc0c$e1a$1_at_news.eternal-september.org...

Many thanks. I was already intending to use an external wall-wart. As you
say this circumvents much testing. The total power dissipation is likely to
be around 3W in any case.

Thu Aug 19, 2010 4:00 pm   

"Morten Leikvoll" <mleikvol_at_yahoo.nospam> wrote in message
news:dPqdnSoZacs3qfDRnZ2dnUVZ8iSdnZ2d_at_lyse.net...

This is for a niche market, though any market has competitors who can play
informer to the authorities.

It's difficult to justify a cost of several £000s on a relatively small unit
where the maket may be just 100 per year.


I would put my hand up and say that good safe working practices have been
followed. Also the power supply will be a wall-wart, which will have its
own certification.

Thu Aug 19, 2010 7:13 pm   

"Fredxx" <fredxx_at_spam.com> wrote:


Don't bother. They will be more expensive than renting a lab to do
pre-compliance testing. Without a properly shielded cage (which is
difficult to construct anyway) it is difficult to test. For bench
testing a small pick-up loop is much more usefull to identify the
exact source of the excess radiation.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico_at_nctdevpuntnl (punt=.)
--------------------------------------------------------------

Thu Aug 19, 2010 11:36 pm   

On Thu, 19 Aug 2010 08:21:41 +0100, "Michael Kellett" <nospam_at_nospam.com> wrote:


However this is extremely unlikely - it is a complaint-driven system, and Trading Standards have
better things to do.
Germany can be a bit riskier as I believe they do random test purchases, but in a niche market this
is unlikely.


Has anybody ever been done for non-safety critical products failing immunity?
In practice it's a product quality issue. Waving a cellphone near it is as much of a test as needed
to cover 99% of what it will see in the field.
Immunity testing on something that presents no hazard is a ridiculous and unnceessary expense.


Be aware that most approved PSUs are tested with resistive loads.
In practice it is easy to make a supposedly compliant PSU kick out noise above limits with a
moderately noisy load.
And of course many cheap supposedly approved PSUs aren't.

Thu Aug 19, 2010 11:38 pm   

On 18 Aug 2010 18:19:58 GMT, General Schvantzkoph <schvantzkoph_at_yahoo.com> wrote:


CE is no different. Once you've figured out what standards apply.

Thu Aug 19, 2010 11:40 pm   

On Thu, 19 Aug 2010 14:11:58 +0100, "Fredxx" <fredxx_at_spam.com> wrote:


the only difference between 'pre-test' and 'final test' is the test house's libility,
A pragmatic approach is to book a day or half-day at a test house, test as much as you can & use the
plots from that to self-certify.