[sci.electronics] FCC approval needed for MPU projects?

peg@psuecl.bitnet (PAUL E. GANTER) (11/27/89)

If you want to manufacture a microprocessor-based device on a small scale,
what do you do about FCC testing, approval, registration?

Is there a special class of FCC approval for small production quantities
(say under 100 units/year)?

Can you use the cop-out clause:

  This device must accept any interference.....
  The user of this device must correct any interference.... etc.

that you see so often on cordless phones, garage door openers, etc. and
avoid getting FCC approval at all?

If you must have FCC certification no matter what, does anyone have
an example of the costs involved?

I've heard a lot about manufacturers of MIDI interfaces getting busted
for failure to meet FCC regs.  I'm just wondering if they could get me
for custom-building and/or producing a few MPU devices for sale.

Thanks for any input!

Paul

elliott@optilink.UUCP (Paul Elliott x225) (12/01/89)

In article <67885@psuecl.bitnet>, peg@psuecl.bitnet (PAUL E. GANTER) writes:
> If you want to manufacture a microprocessor-based device on a small scale,
> what do you do about FCC testing, approval, registration?

You need to perform compliance testing on most any "Computing Device".  Only
a few catagories are exempted from this; these excluded catagories are:

1) A computing device used in any transportation vehicle.

2) An electronic control or power system used by a public utility or in
   an industrial plant.

3) Industrial, commercial, and medical test equipment.

4) Computing device utilized in an appliance, e.g. microwave oven, dishwasher,
   clothes dryer, etc.

5) Specialized medical devices ...[prescribed devices, not marketed through 
   retail stores]...

I smell some money changing hands in the construction of this list, don't
you?  Especially #4.


> Is there a special class of FCC approval for small production quantities
> (say under 100 units/year)?

Unfortunately, no.


> Can you use the cop-out clause:
> 
>   This device must accept any interference.....
>   The user of this device must correct any interference.... etc.
> 
> that you see so often on cordless phones, garage door openers, etc. and
> avoid getting FCC approval at all?

These labels are required even though the equipment has been certified.  The
above devices fall under part 15, subpart E : "Low Power Communications 
Devices", and generally must be type approved.

> 
> If you must have FCC certification no matter what, does anyone have
> an example of the costs involved?

For a "Computing Device", there are two catagories: Class A, which is
a "commercial / industrial / business" catagory, and Class B, which is the 
"residential area" catagory.  Class A is less stringent, and no formal
filing with the FCC is required.  These devices are "verified" by the
manufacturer, more or less on the honor system.  The FCC reserves the
right to review the measurement records, however.  The verification
proceedure is pretty simple, requiring a site (a grassy field works well),
antenna(s) covering 30 MHz to 1500 MHz (most people use biconical dipoles
which are broadband, but the specs ultimately use a tuned dipole as the
reference), and a spectrum analyzer or equivalent calibrated receiver.

Class B is tougher to comply with, and some types of equipment have to
be "certificated", which requires that the measurement data be filed with
the FCC.  Other Class B equipment only requires verification.

There are many consultants who have the equipment and will make the 
measurements for a about $1000 (typical as of a few years ago).  Many
electronics manufacturing organizations have set up a measuring facility
and if you can use the equipment, you might be able to barter for some time 
on the site.

> I've heard a lot about manufacturers of MIDI interfaces getting busted
> for failure to meet FCC regs.  I'm just wondering if they could get me
> for custom-building and/or producing a few MPU devices for sale.

I've done verification measurements on a MIDI adaptor, and even a device
as simple as this needs care in design and layout to meet either Class A or B
requirements.  My advice is to think of the FCC as the IRS: Your chances
of being audited on a tax return are low, but the penalties for screwing up
are awesome to contemplate.

Regards, Paul


-- 
Paul M. Elliott      Optilink Corporation     (707) 795-9444
         {pyramid,pixar,tekbspa}!optilink!elliott
"I used to think I was indecisive, but now I'm not so sure."

elliott@optilink.UUCP (Paul Elliott x225) (12/02/89)

In article <2695@optilink.UUCP>, elliott@optilink.UUCP (Paul Elliott x225) writes:
 
> > I've heard a lot about manufacturers of MIDI interfaces getting busted
> > for failure to meet FCC regs.  I'm just wondering if they could get me
> > for custom-building and/or producing a few MPU devices for sale.
> 
> I've done verification measurements on a MIDI adaptor, and even a device
> as simple as this needs care in design and layout to meet either Class A or B
> requirements.  My advice is to think of the FCC as the IRS: Your chances
> of being audited on a tax return are low, but the penalties for screwing up
> are awesome to contemplate.

I've had a chance to regret making the above comparison.  My dealings with
the FCC have actually been very pleasant.  The Laboratory division has been
helpful and reasonable when I was designing "low power communications devices",
and needed interpretations of the regulations when the device didn't neatly
fit into one of the pre-assigned pigeonholes.  I just haven't had the 
opportunity to experience first-hand what can happen if you don't touch all
the bases, and don't intend to let it happen.

Can anyone out there provide information on what penalties or fines the FCC
has imposed for non-complying "computing device" manufacturers?  I have heard
of some companies coldly taking the calculated risk of not being caught, and
shipping obviously out-of-spec equipment, but haven't seen the results.

Any comments?
-- 
Paul M. Elliott      Optilink Corporation     (707) 795-9444
         {pyramid,pixar,tekbspa}!optilink!elliott
"I used to think I was indecisive, but now I'm not so sure."