Searching \ for '[EE]: Grinning, Macs, and the FCC' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=grinning+macs+fcc
Search entire site for: 'Grinning, Macs, and the FCC'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Grinning, Macs, and the FCC'
2000\07\31@094550 by Dan Michaels

flavicon
face
Anybody who has seen Steve Jobs lately, of course, notes his
trademark ear-to-ear grin. Apple is doing well, and he keeps
birthing sleek new Mac designs. Getting to the point:

Apple's new marketing point is "transparent" iMacs and iCubes.

Question is, how does a 450 Mhz machine with a transparent
plastic case pass FCC EMC requirements? I may be wrong, but my
impression is that you could not embed enough conductive material
into something that "appears" to be transparent to be an effective
shield. So, possibly, internal shielding in exactly the critical
places?

Also, on a more philosophical level, is it ever possible for a
"bare" unshielded pcb [eg, multilayer or not], with chips
running at say 20+ Mhz, to pass FCC reqs? Or does it almost
always require some local shielding or a shielded case? For
that matter, will "local" shielding ever actually do it?

best regards,
- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==========================

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2000\07\31@120641 by Phillip Vogel

flavicon
face
Dan Michaels wrote:

> Also, on a more philosophical level, is it ever possible for a
> "bare" unshielded pcb [eg, multilayer or not], with chips
> running at say 20+ Mhz, to pass FCC reqs? Or does it almost
> always require some local shielding or a shielded case? For
> that matter, will "local" shielding ever actually do it?

Several years ago, I did a product in a plexiglass box. It contained an 8051
running at 12MHz on a double sided board. No problem getting through FCC. I
had ground poured everywhere that wasn't a signal, and that helped. I'm sure
that a multilayer board with ground & power planes near the outside would have
done even better.

Come to think of it, does anyone ever make multilayer boards with ground
and/or pwer planes as the outside layers? It would alo be nice to have chips
in shielded packages, but I can imagine what would happen to the price...

And then, I could speculate on why jobs (and Mac users in general) all have
that stupid grin, but I'm not going there today :-/

--
Phillip M. Vogel, President   | "It's not what you've been taught,
Bartal Design Group, Inc.     |  it's what you've learned." (me)
318 Marlboro Road             | +1-201-567-1343 FAX:+1-201-568-2891
Englewood, NJ 07631  USA      | spam_OUTphillipTakeThisOuTspambartal.com

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2000\07\31@131300 by David VanHorn

flavicon
face
>
>Also, on a more philosophical level, is it ever possible for a
>"bare" unshielded pcb [eg, multilayer or not], with chips
>running at say 20+ Mhz, to pass FCC reqs? Or does it almost
>always require some local shielding or a shielded case? For
>that matter, will "local" shielding ever actually do it?


Sure!  Though I've not gone into the 200-300 MHz arena, I use two layer
boards in plastic enclosures routinely.
You have to pay attention to what you're doing.

--
http://www.SpamWhack.com  A pre-emptive strike against spam
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

My transistor sings
with unintended parasitic
the smoke escapes

By Jeff Stout

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2000\07\31@133154 by mike

flavicon
face
On Mon, 31 Jul 2000 12:06:44 -0400, you wrote:

>Dan Michaels wrote:
>
>> Also, on a more philosophical level, is it ever possible for a
>> "bare" unshielded pcb [eg, multilayer or not], with chips
>> running at say 20+ Mhz, to pass FCC reqs? Or does it almost
>> always require some local shielding or a shielded case? For
>> that matter, will "local" shielding ever actually do it?
>
>Several years ago, I did a product in a plexiglass box. It contained an 8051
>running at 12MHz on a double sided board. No problem getting through FCC. I
>had ground poured everywhere that wasn't a signal, and that helped. I'm sure
>that a multilayer board with ground & power planes near the outside would have
>done even better.
>
>Come to think of it, does anyone ever make multilayer boards with ground
>and/or pwer planes as the outside layers? It would alo be nice to have chips
>in shielded packages, but I can imagine what would happen to the price...
I have seen one PCB with outer planes, but I was told this was as much
to improve track density as EMC (I think the argument was that
through-hole component pads took more space on surface layers, so
there is more rouring area on inner layers)

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


'[EE]: Grinning, Macs, and the FCC'
2000\08\01@022451 by rubenj
flavicon
face
> Also, on a more philosophical level, is it ever possible for a
> "bare" unshielded pcb [eg, multilayer or not], with chips
> running at say 20+ Mhz, to pass FCC reqs? Or does it almost
> always require some local shielding or a shielded case? For
> that matter, will "local" shielding ever actually do it?
>
> best regards,
> - Dan Michaels
> Oricom Technologies
> http://www.sni.net/~oricom
> ==========================
>

I did a project using a Scenix SX52 running at 10MHz in a plastic
housing that at the first testing had way too high levels in the
emission tests (at 150 - 250 MHz). The two sided board had mainly
tracks on the bottomside and a filled ground plane on the topside. It
turned out that it was the powersupply to the SX chip that was bad.
When the chip drew current (at every clock pulse) the powersupply
voltage to the chip fluctuated (rippled with overtones up in the 100-
300MHz range). All IO outputs that was either high or clamped to the
internal pull up resistor (transistor) carried this ripple, all the
way out to the connecting cables, which worked fine as antennas (most
cables where unshielded).

The sollution was to improve the powersupply circuit and decoupling
capacitors to the chip and I also placed a (3 legged)
capacitor/inductor filter on every IO and powerline directly at the
connection to the PCB. This dropped the emission levels by 20dB, so I
could still use my plastic housing and unshielded cables.

At the first look at this problem I was thinking of isolating the
powersupply to the microcontroller (impedance wise) by putting a
filter (inductor and capacitors) in series with the powersupply trace
to the chip (as I also have seen mentioned on this list) because I
thougth that the emission was spread too the rest of the board via
the power supply lines. This would only have made my problems worse
since it turned out that the emission was instead spread by the IO
lines of the microcontroller and the filter would have made the
ripple larger.


==============================
Ruben Jvnsson
AB Liros Elektronik
Box 9124, 200 39 Malmv, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
.....rubenKILLspamspam@spam@pp.sbbs.se
==============================

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamKILLspammitvma.mit.edu>

2000\08\01@150857 by kirmse

flavicon
face
Phillip Vogel wrote:
>
> Dan Michaels wrote:
>
> > Also, on a more philosophical level, is it ever possible for a
> > "bare" unshielded pcb [eg, multilayer or not], with chips
> > running at say 20+ Mhz, to pass FCC reqs? Or does it almost
> > always require some local shielding or a shielded case? For
> > that matter, will "local" shielding ever actually do it?

Given the construction of many PC cases they might as well be open.

{Quote hidden}

Going from a 2 sided board to a board with a proper ground plane has a
significant effect on EMI. There is some benefit from having high frequency
signals sandwiched between ground planes but the effect is not huge.

> And then, I could speculate on why jobs (and Mac users in general) all have
> that stupid grin, but I'm not going there today :-/
>


--

---------------------------------------------------------------------
| Dr. Kevin Dale Kirmse, PhD EE
| Portable System Design, High Speed Serial Links
| FPGA Design, Video Hardware, Graphics Hardware
|
| King of Prussia, PA 19406
| .....kirmseKILLspamspam.....netaxs.com
---------------------------------------------------------------------

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspam_OUTspamTakeThisOuTmitvma.mit.edu>

2000\08\01@153844 by James Paul

flavicon
face
All,

Just as an FYI, if you have a power plane on one side of a two sided
PCB, and a ground plance on the other, and some of the components on
the power plane side need to be bypassed, bypassing them to the
power plane is almost as good as bypassing them to the ground plane.
Also, one point grounding to prevent ground loops is desirable.  It's
not always achievable though.

                                             Regards,

                                                Jim


On Tue, 01 August 2000, Kevin Dale Kirmse wrote:

{Quote hidden}

KILLspamjimKILLspamspamjpes.com

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu>

2000\08\01@160753 by Mitchell D. Miller

picon face
>  Just as an FYI, if you have a power plane on one side of a two sided
>  PCB, and a ground plance on the other

Uh ... doesn't that mean you have no signal layers? <g>

-- Mitch

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestspamBeGonespammitvma.mit.edu>

2000\08\01@162900 by James Paul

flavicon
face
No, that does not mean you have no signal layers.  The signal layers
are between the 2 outer layers. I guess I should have stated that
explicitly, but I thought everyone would understand.

One thing the internet and email lacks that would be beneficial to
groups like this is body language and facial expressions.  You know
how when you talk to someone face to face, they see your body
language and gestures, and your facial expressions.  These parameters
add significantly to ones understanding what someone else is saying,
especially if the receiving person is unfamiliar with the topic that
is being discussed.  If anyone can think of a way to add these
parameters to an email, you might have a niche here.  Could make a
million.  Food for thought.      :')

                                               Regards,

                                                 Jim




On Tue, 01 August 2000, "Mitchell D. Miller" wrote:

{Quote hidden}

RemoveMEjimspamTakeThisOuTjpes.com

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestEraseMEspam.....mitvma.mit.edu>

2000\08\01@164143 by David VanHorn

flavicon
face
>  If anyone can think of a way to add these
>  parameters to an email, you might have a niche here.  Could make a
>  million.  Food for thought.      :')

My favorite ID book is titled "Turn signals are the facial expressions of
Automobiles".

:)


--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspammitvma.mit.edu>

2000\08\01@164559 by Dan Michaels

flavicon
face
James Paul wrote:
>Mitch Miller wrote:
>>  Just as an FYI, if you have a power plane on one side of a two sided
>>  PCB, and a ground plance on the other
>>
>>Uh ... doesn't that mean you have no signal layers? <g>
>
> No, that does not mean you have no signal layers.  The signal layers
> are between the 2 outer layers. I guess I should have stated that
> explicitly, but I thought everyone would understand.


OK, I'll bite. It's a "two sided" pcb with power plane on one side
and gnd plane on the other side, and 2 signal layers in between.

[Oops, I just accidentally superglued my forefinger to my middle
finger, and my little finger to my ring finger]. Do I have it
staight now?

- danM

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestEraseMEspamEraseMEmitvma.mit.edu>

2000\08\01@165014 by Dan Michaels

flavicon
face
Dave VanHorn wrote:
>>  If anyone can think of a way to add these
>>  parameters to an email, you might have a niche here.  Could make a
>>  million.  Food for thought.      :')
>
>My favorite ID book is titled "Turn signals are the facial expressions of
>Automobiles".
>
>:)
>

OK, I'll bite - I have the book and just looked in it, but couldn't
find any funny faces. [hmmm, guess I bought the wrong edition].

- danM

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestspam_OUTspamKILLspammitvma.mit.edu>

2000\08\01@165629 by David VanHorn

flavicon
face
>
>
>OK, I'll bite - I have the book and just looked in it, but couldn't
>find any funny faces. [hmmm, guess I bought the wrong edition].

No, just the concept that facial expressions convey a lot of subtext, "what
I mean" and "what I'm going to do".

While I've been told emoticons are "unprofessional" I use them a lot,
because of many experiences with misinterpreted email.. I suppose
"professionals" can only use :-|

--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestTakeThisOuTspamspammitvma.mit.edu>

2000\08\01@181350 by Bob Ammerman

picon face
Um, is a one-sided PC board a mobius (sp?) strip?

Bob Ammerman

----- Original Message -----
From: Dan Michaels <EraseMEoricomspamspamspamBeGoneLYNX.SNI.NET>
To: <RemoveMEPICLISTKILLspamspamMITVMA.MIT.EDU>
Sent: Tuesday, August 01, 2000 4:45 PM
Subject: Re: [EE]: Grinning, Macs, and the FCC


{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestSTOPspamspamEraseMEmitvma.mit.edu>

2000\08\02@072548 by W. K. Brown

flavicon
face
IMHO, an 'n-layer board' has 'n-layers of copper'. Am I wrong?

wkb

They're just Jealous
That the Voices
Don't talk to Them.


       {Original Message removed}

2000\08\02@072948 by W. K. Brown

flavicon
face
Love it!
wkb
They're just Jealous
That the Voices
Don't talk to Them.


       -----Original Message-----
       From:   Bob Ammerman [SMTP:KILLspamRAMMERMANspamBeGonespamPRODIGY.NET]
       Sent:   Tuesday, August 01, 2000 6:12 PM
       To:     EraseMEPICLISTspamEraseMEMITVMA.MIT.EDU
       Subject:        Re: [EE]: Grinning, Macs, and the FCC

       Um, is a one-sided PC board a mobius (sp?) strip?

       Bob Ammerman

       ----- Original Message -----
       From: Dan Michaels <@spam@oricom@spam@spamspam_OUTLYNX.SNI.NET>
       To: <spamBeGonePICLISTspamKILLspamMITVMA.MIT.EDU>
       Sent: Tuesday, August 01, 2000 4:45 PM
       Subject: Re: [EE]: Grinning, Macs, and the FCC


       > James Paul wrote:
       > >Mitch Miller wrote:
       > >>  Just as an FYI, if you have a power plane on one side of a two
sided
       > >>  PCB, and a ground plance on the other
       > >>
       > >>Uh ... doesn't that mean you have no signal layers? <g>
       > >
       > > No, that does not mean you have no signal layers.  The signal
layers
       > > are between the 2 outer layers. I guess I should have stated
that
       > > explicitly, but I thought everyone would understand.
       >
       >
       > OK, I'll bite. It's a "two sided" pcb with power plane on one side
       > and gnd plane on the other side, and 2 signal layers in between.
       >
       > [Oops, I just accidentally superglued my forefinger to my middle
       > finger, and my little finger to my ring finger]. Do I have it
       > staight now?
       >
       > - danM
       >
       > --
       > http://www.piclist.com hint: To leave the PICList
       > .....piclist-unsubscribe-requestspam_OUTspammitvma.mit.edu

       --
       http://www.piclist.com hint: To leave the PICList
       TakeThisOuTpiclist-unsubscribe-request.....spamTakeThisOuTmitvma.mit.edu

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\02@092756 by James Paul

flavicon
face
Only if it's twisted.  A Mobius strip has only one surface, but a
one sided PC Board has two surfaces.  One copper, and one not copper,
correct?

              Regards,

                Jim



On Wed, 02 August 2000, "W. K. Brown" wrote:

>
> Love it!
> wkb
> They're just Jealous
> That the Voices
> Don't talk to Them.
>
>
>         {Original Message removed}

2000\08\02@101906 by Bob Ammerman

picon face
Sorry, it was just an attempt at humor, but the HML was omitted <G>

Bob Ammerman

----- Original Message -----
From: James Paul <TakeThisOuTjimKILLspamspamspamJPES.COM>
To: <.....PICLISTspamRemoveMEMITVMA.MIT.EDU>
Sent: Wednesday, August 02, 2000 9:26 AM
Subject: Re: [EE]: Grinning, Macs, and the FCC


{Quote hidden}

> >         {Original Message removed}

2000\08\02@153058 by Peter L. Peres

picon face
>translucent RFI stop

I don't know about Macs, but anyone familiar with medium power RF homebrew
will tell you that there are scores of 'household' plastics that are
translucent (if not transparent) yet melt and stink when exposed to
medium strong RF fields (as in coil formers for RF amps and drivers). So I
would not be surprised if some plastic or other would be rather effective
at stopping higher frequency RF (>10MHz or so, maybe even 1MHz) given
some thickness. Plastic cases are always much thicker than metal anyway.
Straight conductive plastic that is transparent has been available for
years now afaik.

There are ways to design a board such that most of the energy stays
trapped inside it (between the outer ground planes). The signal ground
returns are NOT on the outer ground planes for this (they are buried). BGA
and PGA packages also help with this. Afaik at the present state of
technology (in consumer etc area) a correctly designed board radiates 6 dB
or less, more than the same board cased in a commercial case (not RF proof
special case). (Reading taken at some distance).

However, I do not really want to know about the magnetic mode (H) noise of
such a machine, as that is probably not shielded at all by the transparent
case. It would be nice to ask whether a transparent Mac owner has used a
small AM radio near a running Mac and a running PC to compare "results".

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\02@153101 by Peter L. Peres

picon face
>and/or pwer planes as the outside layers? It would alo be nice to have chips
>in shielded packages, but I can imagine what would happen to the price...

BGA chips and PGA chips are naturally shielded by their own substrate and
mounting plate. What is really bad is crosstalk 'inside' the planes.
Because of the outer shield the traces (buses) become very closely coupled
and extra grounds are required to separate them and provide return at HF.
A 10 cm portion of 4-bit data bus between ground planes makes a very
passable 1:1:1:1 transmission line transformer at 100 MHz and higher f.ex.
Also all the lines are terminated as transmission lines at both ends
inside the chips or outside using SMD resistor arrays etc.

MILSPEC chips in ceramic packages with metal tab covers are also shielded.

For normal production consumer equipment a adhesive copper (or cheaper,
tin) foil can work wonders, when stuck over noisy parts of the circuit.
Some people use metalized paint spray or even graphite spray to make the
inside of their plastic boxes conductive. One of the latest tricks is the
RF absorber sponge pad (which is an insulator but absorbs RF) glued
directly over the noisy parts.

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\02@155806 by Dan Michaels

flavicon
face
Peter Peres wrote:
>>translucent RFI stop
........
>
>However, I do not really want to know about the magnetic mode (H) noise of
>such a machine, as that is probably not shielded at all by the transparent
>case. It would be nice to ask whether a transparent Mac owner has used a
>small AM radio near a running Mac and a running PC to compare "results".
>

Good info - first about plastics end of question.

However, I am not too sure the AM radio is a Mac good test, as I
routinely use one to monitor do-loops on my PC, tune my PIC boards,
and test my CRT monitor. [despite those little FCC stickers on everything,
my little AM radio is the best test instrument in my building - even
predicts thunderstorms].

Thanks,
- Dan Michaels
Oricom Technologies
===================

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\02@160502 by Dan Michaels

flavicon
face
Peter Peres wrote:
>>and/or pwer planes as the outside layers? It would alo be nice to have chips
>>in shielded packages, but I can imagine what would happen to the price...
>
>BGA chips and PGA chips are naturally shielded by their own substrate and
>mounting plate. What is really bad is crosstalk 'inside' the planes.
>Because of the outer shield the traces (buses) become very closely coupled
>and extra grounds are required to separate them and provide return at HF.
>A 10 cm portion of 4-bit data bus between ground planes makes a very
>passable 1:1:1:1 transmission line transformer at 100 MHz and higher f.ex.
>Also all the lines are terminated as transmission lines at both ends
>inside the chips or outside using SMD resistor arrays etc.
..........


More good info. From your experience with multilayer boards and
BGA/PGA chips, do you think it's better to have the power and gnd
planes as the *innermost* planes, or as the *outermost* planes?

Thanks,
- Dan Michaels
Oricom Technologies
===================

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\03@163644 by Peter L. Peres

picon face
>inside or outside ground/power planes

The best seems to be power/ground planes on both sides as far as possible,
but NONE of them used as signal grounds. They act as shields and supply
most of the time and they use oodles of through-holes wherever possible,
and distributed decoupling. Take a good look at a recent PC motherboard
esp. around the CPU/local memory module area for a good example. In a
"quiet" board you have mostly the same setup buried between layers.
However this costs power. The drive power required to run the lower
impedance 'buried' lines is higher than that required to run the surface
'transmission lines'. Sometimes giving up a lower RFI design is required
because you exceed dissipation in certain driver chips... Anyway an open
line over ground plane is 50-120 ohms I think and a buried one can be 2 to
3 times less. Most fast chips for commercial use cannot drive such lines
afaik.

Each of the RAM bus line drivers of a modern CPU must output more power
than a ham QRP transmitter (and there are hundreds of them working in
parallel) just to drive the short lines to the RAM.

Note that I do not design boards, but I use a lot of them and I get to see
them 'perform' from time to time.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\05@104024 by picxpert

picon face
I know for certain that the analog board is shielded. It's located
underneath.

Also, the bus isn't running at 450 MHz - it's running at 100MHz, as I
recall. The chip (and, with the G3 processor) cache are running at 450MHz
and 225MHz, respectively (a G3's cache can run at the speed of the chip, but
you'd have to pay for memory that fast)

Those factors mean that the problem of shielding probably isn't as bad as it
would be otherwise, but yes, it still does need some shielding. My theory is
that the carbon in the polycarbonate shell is helping with that - carbon is
conductive, right?

-Randy Glenn
RemoveMEPICxpertANTISPAMspamspamBeGonepicxpert.com (remove ANTISPAM)
http://www.picxpert.com/ << WORKING NOW!

Those packing a big grudge, usually pack a big mouth along with it.

{Original Message removed}

2000\08\05@121204 by Don Hyde

flavicon
face
People doing stuff like cell phones and wireless internet stuff are learning
a lot about how to control RF on PC boards.  They have to so that their
processors won't kill their own radio receivers, and their radio
transmitters won't kill their own processors.

The basic trick is to keep the high frequencies where they are needed and
never let them out any further than necessary.  I'll bet that if you look at
the boards in the iMac, you won't see very many traces because the outer
layers of the multilayer boards are all power and ground planes, which
effectively shields all the interconnects.  That alone, plus a little care
around the wires going in and out is often enough to meet the FCC emission
requirements, which are, after all, not really all that stringent.

Another trick that seems like cheating but is legal is "spread spectrum"
clocks.  Here the clock frequency is intentionally dithered a little (i.e.
frequency modulated with noise).  That way the emissions show up on a
spectrum analyzer as a broad hump rather than a narrow line.  Since the same
energy is spread over a wider bandwidth, its peak intensity is lower, and
that's what is spec'ed by the FCC.  Supposedly that is OK because it is the
peak intensity that matters most as far as interference with intentional
signals is concerned.

> {Original Message removed}

2000\08\05@205124 by Russell McMahon

picon face
>Another trick that seems like cheating but is legal is "spread spectrum"
>clocks.  Here the clock frequency is intentionally dithered a little (i.e.
>frequency modulated with noise).  That way the emissions show up on a
>spectrum analyzer as a broad hump rather than a narrow line.  Since the
same
>energy is spread over a wider bandwidth, its peak intensity is lower, and
>that's what is spec'ed by the FCC.  Supposedly that is OK because it is the
>peak intensity that matters most as far as interference with intentional
>signals is concerned.


Which can sort-of :-) be achieved with a PIC by returning the crystal
capacitors to Vcc rather than to ground.

RM

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-request@spam@spamspam_OUTmitvma.mit.edu>

2000\08\06@023734 by Sean Breheny

face picon face
I have been listening to this thread and wondering about the whole
conductive plastic thing. How conductive does plastic really get? Most
conductive plastics I have seen have a few k ohms of resistance from a
typical point to another point, say, a centimeter away, only good for
bleading away static charges. I have a hard time believing that this would
affect RF fields much at all.

I don't think regular polycarbonate is conductive. It contains carbon, but
conductivity is totally dependent on the configuration of the electrons in
the molecules, which depends on how the atoms are bonded, etc., not just
what element you are talking about.

Sean

At 10:39 AM 8/5/00 -0400, Randy Glenn wrote:
{Quote hidden}

| Sean Breheny
| Amateur Radio Callsign: KA3YXM
| Electrical Engineering Student
\--------------=----------------
Save lives, please look at http://www.all.org
Personal page: http://www.people.cornell.edu/pages/shb7
shb7EraseMEspamcornell.edu ICQ #: 3329174

____________NetZero Free Internet Access and Email_________
Download Now     http://www.netzero.net/download/index.html
Request a CDROM  1-800-333-3633
___________________________________________________________

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\06@125146 by Robert Francisco

flavicon
face
I have always thought polycarbonate is an insulator, don't they use this as
a dielectric in capacitors?

robertf
{Original Message removed}

2000\08\06@203037 by Dan Michaels

flavicon
face
Sean Breheny wrote:
>> I have been listening to this thread and wondering about the whole
>> conductive plastic thing. How conductive does plastic really get? Most
>> conductive plastics I have seen have a few k ohms of resistance from a
>> typical point to another point, say, a centimeter away, only good for
>> bleading away static charges. I have a hard time believing that this would
>> affect RF fields much at all.
>>
>> I don't think regular polycarbonate is conductive. It contains carbon, but
>> conductivity is totally dependent on the configuration of the electrons in
>> the molecules, which depends on how the atoms are bonded, etc., not just
>> what element you are talking about.
>>

Robert Francisco wrote:
>I have always thought polycarbonate is an insulator, don't they use this as
>a dielectric in capacitors?
>

I started this thread specifically to get info about possible
conductive RFI-shield plastics - from wondering about the new iMac
cases. Have no idea about polycarbonate - but we now have a couple
of differing opinions. Possibly like silicon, where conductivity is
related to impurity doping - [dang, where are all the chemists when
you need them].

However, I do use ABS cases from Serpac in one of my projects that
are rated [ie, what the sticker says] as 55-60 dB attenuation at
30-2000 Mhz. They are black and somewhat conductive - I cannot
measure any conductance with my DMM, but they'll kill an input
signal if you touch the signal lead to the box.

I haven't actually tried measuring the RFI qualities of the boxes
[maybe Dave VanHorn has], but I can tell you they quite effectively
shield the high-gain analog section from pickup of digital noise
generated elsewhere on the pcb inside. Leave off the box - lot of
low-level random noise, put on the box, noise level drops 2-3x.

cheers,
- Dan Michaels
Oricom Technologies
===================

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\06@205946 by David VanHorn

flavicon
face
>
>I haven't actually tried measuring the RFI qualities of the boxes
>[maybe Dave VanHorn has], but I can tell you they quite effectively
>shield the high-gain analog section from pickup of digital noise
>generated elsewhere on the pcb inside. Leave off the box - lot of
>low-level random noise, put on the box, noise level drops 2-3x.


I've not messed with them, never needed to.
I have had good luck with using carbon loaded foam as an absorber here in
the shack, to absorb computer hash.

I'd expect the conductive cases to have a metalized layer, but I've never
seen one.
I know they offer various conductive sprays, but we always rejected it
because of the cost delta.

--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements

2000\08\07@124146 by Dan Michaels

flavicon
face
Dave VanHorn wrote:
>>
>>I haven't actually tried measuring the RFI qualities of the boxes
>>[maybe Dave VanHorn has], but I can tell you they quite effectively
>>shield the high-gain analog section from pickup of digital noise
>>generated elsewhere on the pcb inside. Leave off the box - lot of
>>low-level random noise, put on the box, noise level drops 2-3x.
>
>
>I've not messed with them, never needed to.
>I have had good luck with using carbon loaded foam as an absorber here in
>the shack, to absorb computer hash.
>
>I'd expect the conductive cases to have a metalized layer, but I've never
>seen one.
>I know they offer various conductive sprays, but we always rejected it
>because of the cost delta.
>

I tried cutting and sticking aluminized duct tape to the inside of
regular ABS cases, but it was too much trouble. Never tried spray paint.

The info on the Serpac EMI cases says material equals "ABS stainless
steel fiber" and "higher degree of protection than painting the
surface...". SIgnificant cost increase, however, compared to regular
ABS cases. Money vs time - always a tradoff.

- danM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@125740 by Dan Michaels

flavicon
face
I wrote:

>Sean Breheny wrote:
>> I have been listening to this thread and wondering about the whole
>> conductive plastic thing. How conductive does plastic really get? Most
>> conductive plastics I have seen have a few k ohms of resistance from a
>> typical point to another point, say, a centimeter away, only good for
>> bleading away static charges. I have a hard time believing that this would
>> affect RF fields much at all.
>>
...........
They are black and somewhat conductive - I cannot
>measure any conductance with my DMM, but they'll kill an input
>signal if you touch the signal lead to the box.
.........


Oops, have to revise this. Last time I measured on the surface of
the case, and no conductivity, ie >10Mohms. But went back and measured
at the "edges" of cut-thru holes in the case - readings are in the
range of 20-50 ohms, depending.... So, the surfaces appear to be
fairly insulative, highly conductive layers inside.

cheers,
- Dan Michaels
Oricom Technologies
===================

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@130408 by Peter L. Peres

picon face
>that's what is spec'ed by the FCC.  Supposedly that is OK because it is the
>peak intensity that matters most as far as interference with intentional
>signals is concerned.

The peak intensity is the most important because intermodulation is a
multiplicative process (as in mixing). However the total intensity can get
you too. For example in some applications the noise is deliberately kept
on certain frequencies to avoid raising the noise floor all over the
place. In other words a spread spectrum supply will likely pass FCC easier
AND disturb a DX receiver MORE than a fixed frequency unit.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@190801 by Peter L. Peres

picon face
>How conductive does plastic really get? Most
>conductive plastics I have seen have a few k ohms of resistance from a
>typical point to another point, say, a centimeter away, only good for
>bleading away static charges. I have a hard time believing that this
>would affect RF fields much at all.

>I don't think regular polycarbonate is conductive. It contains carbon, but
>conductivity is totally dependent on the configuration of the electrons in
>the molecules, which depends on how the atoms are bonded, etc., not just
>what element you are talking about.

>Sean

Conductivity is not the key here, but a certain bulk resistivity at AC. If
that is equivalent to the characteristic impedance of the plastic (i.e.
the complex components of Z at F0 in that material) then the material is
the best possible absorbent. Much better than metal which reflects
everything back in to bounce around and exit through the fan gills and
power cords. This is why absorbent sponges are often used inside RF proof
cases. They attenuate the energy bouncing around inside and dissipate it.
I suppose that you could make the case out of 'sponge' and live to see a
FCC test.

Since the characterisitc impedance of the plastic is fairly high (and
depends on frequency) the required resistivity for perfect match is also
relatively high. F.ex. certain PVCs exhibit very high DC resistance and
absorb RF above ~30MHz like a sponge (and get hot and melt in higher
fields).

RF (300kHz-10MHz) melting of large plastic (or what passed for plastic
then) parts for machining/forming purposes has been known since the 1920s.
Look up dielectric heating.

In practice for cases and shields the resistivity is lower than 'best' and
some energy is reflected back in in an effort to minimize transmission.

Think adapted transmission line (or blackbody), not wall/mirror.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@192450 by Dan Michaels

flavicon
face
Peter Peres wrote:
........
>Conductivity is not the key here, but a certain bulk resistivity at AC. If
>that is equivalent to the characteristic impedance of the plastic (i.e.
>the complex components of Z at F0 in that material) then the material is
>the best possible absorbent. Much better than metal which reflects
>everything back in to bounce around and exit through the fan gills and
>power cords.


So what does this say about my Serpac ABS box, whose material is
touted as "ABS Stainless Steel Fiber"?
=========

This is why absorbent sponges are often used inside RF proof
>cases. They attenuate the energy bouncing around inside and dissipate it.
>I suppose that you could make the case out of 'sponge' and live to see a
>FCC test.
>

The sticker says 55-60 dB attenuation at 30-2000 Mhz. Is the energy
absorbed or reflected back inside? The case is certainly not spongy.

- danM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@194758 by David VanHorn

flavicon
face
>
>This is why absorbent sponges are often used inside RF proof
> >cases. They attenuate the energy bouncing around inside and dissipate it.
> >I suppose that you could make the case out of 'sponge' and live to see a
> >FCC test.
> >
>
>The sticker says 55-60 dB attenuation at 30-2000 Mhz. Is the energy
>absorbed or reflected back inside? The case is certainly not spongy.


Both. Nothing is prefectly reflecting, or perfectly absorbing, but I
suspect it's more reflective than it is absorptive.
High resistances would indicate absorptive to me.
The ideal material would have a gradient of impedances arriving at near
zero for the outer wall.
All the RF would then (in theory) be converted to heat, which is still
legal to radiate.

Somehow, it seems like cheating to upconvert all the noise beyond the FCC's
reach in the spectrum though :)


--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@200453 by Dan Michaels

flavicon
face
Peter Peres wrote:
>>
........
>>The sticker says 55-60 dB attenuation at 30-2000 Mhz. Is the energy
>>absorbed or reflected back inside? The case is certainly not spongy.
>
>
>Both. Nothing is prefectly reflecting, or perfectly absorbing, but I
>suspect it's more reflective than it is absorptive.
>High resistances would indicate absorptive to me.
>The ideal material would have a gradient of impedances arriving at near
>zero for the outer wall.


Actually, as I loosely determined by poking with my DMM, the
outer surface of the case has very low conductance, >10 Mohm.
However, poking the edges of holes cut thru the case, you get
readings around 20+ ohms. So, would seem this is like
1/[your ideal material]. Gradient is infinite on outer/inner
"surfaces" - with low Z inbetween.
============


>All the RF would then (in theory) be converted to heat, which is still
>legal to radiate.
>
>Somehow, it seems like cheating to upconvert all the noise beyond the FCC's
>reach in the spectrum though :)

Guess they trap the heat inside the box after all, rather than radiate
it. Hmmm, box is black - must mean something - heat eventually
"conducted" to outside surface, then radiated.

- danM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@202000 by David VanHorn

flavicon
face
>
>Actually, as I loosely determined by poking with my DMM, the
>outer surface of the case has very low conductance, >10 Mohm.
>However, poking the edges of holes cut thru the case, you get
>readings around 20+ ohms. So, would seem this is like
>1/[your ideal material]. Gradient is infinite on outer/inner
>"surfaces" - with low Z inbetween.
>============

Well, I'm not sure what you have there is a gradient material. That would
probably be very expensive to produce.
One step, does not a gradient make.

--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@204101 by Dan Michaels

flavicon
face
Dave VanHorn wrote:
>>
>>Actually, as I loosely determined by poking with my DMM, the
>>outer surface of the case has very low conductance, >10 Mohm.
>>However, poking the edges of holes cut thru the case, you get
>>readings around 20+ ohms. So, would seem this is like
>>1/[your ideal material]. Gradient is infinite on outer/inner
>>"surfaces" - with low Z inbetween.
>>============
>
>Well, I'm not sure what you have there is a gradient material. That would
>probably be very expensive to produce.
>One step, does not a gradient make.


WEll, actually 2 steps, one infinite [outer surface] to zero
[inside], two zero [inside] back to infinite [inner surface].
But this is just quibbling. There probably is "some" reflection
going on inside the box, but I am more than happy that the box helps
shield the analog section from the digital - as I said, low-level
analog noise drops 2-3X when the box is installed - and I seem
to recall having done some tests months ago to determine this
*is* coming from the digital insdie the box, and not from
outside sources. Unplugging chips/etc.

- DanM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\07@204730 by David VanHorn

flavicon
face
>
>WEll, actually 2 steps, one infinite [outer surface] to zero
>[inside], two zero [inside] back to infinite [inner surface].
>But this is just quibbling.


Second one dosen't count, as a source inside the box can't see it.


I do a similar thing here with PCs (and several have called me less than
sane for it)
I glue a sheet of mosfoam to the cover, such that it can't contact the
boards. This reduces emissions quite a bit, and the model I've used to
explain it is like a sheet of black (or grey) paper in a box made of mirrors.

I also noticed that my "quiet box" for prescanning PCBs was a lot better
behaved when I lined it's insides with foam.
It was bare copper, and measurements were very unrepeatable without the foam.

--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2000\08\08@161339 by Peter L. Peres

picon face
>The ideal material would have a gradient of impedances arriving at near
>zero for the outer wall.

I think that that would be bad. You don't wnat a conductive case on an IP2
class case. Anyway the inside should match the Z0 of air and progress in
depth to a minimum Z0 and then an outer layer of insulator would be
applied. This looks like diffusion-doped plastic (one side) with a second
layer undoped molded onto it from the outside. I am quite certain anyway
that the injection molding for those cases is not a one-step process. The
problem is that IP2 dielectric testing must be satisfied and this can be
very hard for a case with only 2 mm thickness (at most) that has a
conductive layer in it. So most things probably use constant Z0 and high
DC insulation (like the steel filled ABS).

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2000\08\08@161343 by Peter L. Peres

picon face
>So what does this say about my Serpac ABS box, whose material is
>touted as "ABS Stainless Steel Fiber"?

I suspect that it is ABS loaded with chopped stainless steel wire. Which
would make it an insulator at DC and a maze of coupled dipoles at higher
frequencies. Whether it attenuates or reflects depends on the plastic not
on the steel filling. If the plastic is loaded with some carbon component
(like carbon fiber) then it dissipates. If the people who made this know
what they are doing then it almost certainly dissipates.

If you do not know whether a material absorbs or not you can apply the
microwave test. Cook a piece of it in a microvawe oven with the obligatory
cup of cold water for 20 seconds and CAREFULLY touch it when done. If hot,
then it dissipates ;). The 'test frequency' is about 2450 MHz (2.5 GHz)
for legal microwave ovens in most cuontries. It pays to put the test piece
on an ashtray (not metal or plastic) you don't need anymore, in case it
melts or something, and do it on a day when your SO is not around so you
have time to clean the mess ;) Note that glass (ashtray) gets hot itself
in the intense field. You need to test it 'dry' (with water cup) before
the live run. Now, why would you want to apply this empirical test on a
material that was certified by others using much better instruments ?

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2000\08\08@162802 by David VanHorn

flavicon
face
At 11:02 PM 8/8/00 +0300, you wrote:
> >The ideal material would have a gradient of impedances arriving at near
> >zero for the outer wall.
>
>I think that that would be bad. You don't wnat a conductive case on an IP2
>class case.


I was merely looking at it from an EMI viewpoint.

Once I get to the near zero Z layer, I can't "see" anything beyond that in
terms of EMI, so feel free to insulate, paint, color, texture, and flavour,
whatever floats your boat.



--
Where's dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2000\08\09@184412 by Peter L. Peres

picon face
>I was merely looking at it from an EMI viewpoint.

>Once I get to the near zero Z layer, I can't "see" anything beyond that in
>terms of EMI, so feel free to insulate, paint, color, texture, and flavour,
>whatever floats your boat.

If it is battery operated stand-alone you are right. If there is even one
single wire connected to something else coming out of it (and that is not
a car) then it needs to pass IP2 testing unless you want to use a grounded
power cord and a matching case. This can increase (double) the cost of a
project. So it's rather important pricewise I think. The grounded case is
usually RF shielding too so sometimes you just use the metal case and
grounded power cord, and othertimes you use a cheap plastic case, and
spray or foil the inside ... the one goes with the other ;)

This is significant even for one-off projects. F.ex. a plastic case can be
had for ~$8 and a same size metal case will not be had under $22, needs
machining to make control elements available, a grounded lug power cord
($3-5) and a matching receptacle ($5). So the grounded case costs 4 times
more than the plastic one. If the spray costs less than $20 then its
investment returns after one single plastic box treated like this... and I
was able to buy Graphite spray for ~$15 (by CRC Industries).

It is not a coincidence that most portable RF handheld unit manufacturers
use custom injection molded metal outer cases (some are plastic or rubber
clad on the outside). No machining, perfect form, best shielding... and
sky high tooling costs for injection forms. ;(

Peter

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestEraseMEspamspam_OUTmitvma.mit.edu>

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...