All Testing passed

I must applaud the inclusion of the multiple certifications for Europe and North America.  At least this will alleviate any possible delays based on testing.

I am curious as to what the "Canadian thing" was that had to be tested?

Kevin,

I'm not as confident as you seem to be that the testing done so far will

"alleviate any possible delays based on testing".

The FCC rules, Title 47, section 15.3 "Definitions", paragraph i, say:

"Class B digital device.   A digital device that is marketed for use in a

residential environment, notwithstanding use in commercial, business

and industrial environments.  Examples of such devices include, but

are not limited to, personal computers, calculators, and similar

electronic devices that are marketed for use by the general public."

But apparently, Class B testing has not been done, even though the

RPi is being marketed as a PC for use in residential environments.

There is another section, 15.32 "Test procedures for CPU boards

and computer power supplies",  but it doesn't seem to have been

followed, from what we know.  It refers to testing within an enclosure

with the cover removed.

As I understood it, the initial release was to be to allow enthusiasts to develop software and accessories that would be useful to the educational world when the full release takes place in September.

Ironically, Schools are the same as offices and only need to be class A.  While enthusiasts developing software and hardware may wekk be doing so in the home, they should be suitably equipped to resolve any interference issues themselves (i.e. mobve the radio away from the workbench).

Someone mentioned the ESD packaging earlier, this is likely to be one reason Class B needs a case - it was testeed with one ESD test, but direct contact to pins is another matter

It would be nice to see the results for ourselves, to see how close it was to Class B in each area.

Ellis,

   I think the purpose of residential Class B requirements is not so much to prevent

interference within one residence, (that is easily resolved as you say), but to

prevent interference with a neighbor who may have no idea which direction

the interference is coming from.  So even enthusiasts may be ill prepared to

resolve such issues.

Probably all that you need to do if you are getting complaints, is to clamp some ferrite clamps on the cables connected to the R-PI. Without these "antenna's" it's very unlikely such a small board will still radiate.

Martin,

  From the pictures, it looks like the guys in the test chamber did in fact use some ferrite clamps,

but apparently it still didn't meet Class B requirements.  Awaiting details.

Anyway, the FCC rules are designed to prevent a device from being used in a residence

without Class B certification.  As opposed to preventing unsophisticated users from using

such a device.  The intent of the rules is to avoid having to ask your neighbors to add clamps

to whatever devices they may be using that could be causing interference.

Just picking up on one point from one of the above posts, I do not believe that ".... the RPi is being marketed as a PC for use in residential environments."

Although the bare board uncased 'version', which is currently the only one available, is being ordered by 'home users', I would not say that it is being marketed or advertised as being particularly suitable (yet) for that particular section of the market.

We must remember also that the compliance testing was carried out under carefully regulated conditions by people who are experts in their fields.

It is a shame that some posters on here (and one in particular) seem to have nothing better to do than to continually criticise the foundation, and to try and pick holes in, and to undermine the excellent work they are doing.

Do I detect the presence of a troll perhaps?

David,

  You are trying to make a distinction between "being ordered by home users" and "being marketed

to home users".  But the FCC rules define "marketing" as selling or offering to sell, so I don't think

this is a valid distinction.  Anyway, the RPi promotional videos, such as Quake3 and XBMC are

clearly aimed at home users rather than industrial. 

If there are any home users, the device is considered Class B.  It doesn't need to be the majority

or anything like that, although in the case of RPi, I have no doubt that home users will in fact be

the majority.

You say that compliance testing was carried out by experts, and no doubt that is true.

I look forward to hearing how they justify Class A testing for the RPi.

For details on the FCC rules, see Bulletin 62 here: http://transition.fcc.gov/Bureaus/Engineering_Technology/Documents/bulletins/oet62/oet62rev.pdf

I just scanned the document mentioned, and it lists as exempt: things like digital watches that use less than 6 nanowatts. I find that 6 nanowatts very very little. I measured my 1980ies watch once, and got about 0.3 micro amperes. So about 450 nW. I'm pretty sure that that watch is pretty harmless at that powerlevel even if it would TRY to interfere with radio communications.

The Foundation certainly promoted the board as an effective media computer, running XBMC and OpenELEC very nicely.  They even had prominent articles about that on their blog, and they also showed it hosting  normal Linux desktop distros and have many times said "It's just standard Linux".  This ensured that the board would attract interest from ordinary consumers worldwide, and hence end up going into normal domestic consumer situations worldwide too, not just into the hands of electrical engineers and radio amateurs who know how to deal with EMC problems.

There can't be the slightest doubt that this was intentional.  They didn't try to restrict the audience for batch #1 to techies only, quite the opposite, so feeling aggrieved when the need for full domestic certification was explained to them was fairly comical.  They ensured that it would have to be so by their promotion.

Using ferrite clamps during testing is a common way to indentify which of the cables is radiating. It says absolutely nothing about whether the device passes any tests with or without these clamps. Using them is simply a testing tool.

@Martin,

Quite  a lot of devices are shipped with ferrite lumps of different sorts on the cables (I've hade a Cannon camera with one on the USB lead and a network switch with one on the power cable as well as being invloved in product developments where the ferrites were on the cable but inside the box). It's an expensive solution but may the best bet if the cost of a re-design is significant. Of course the device is no longer compliant if a different cable is used and the documentation should reflect this.

Michael Kellett

@Martin,

Quite  a lot of devices are shipped with ferrite lumps of different sorts on the cables (I've hade a Cannon camera with one on the USB lead and a network switch with one on the power cable as well as being invloved in product developments where the ferrites were on the cable but inside the box). It's an expensive solution but may the best bet if the cost of a re-design is significant. Of course the device is no longer compliant if a different cable is used and the documentation should reflect this.

Michael Kellett