[SI-LIST] Re: Traces don't cause EMI - really?

  • From: "Lee Ritchey" <leeritchey@xxxxxxxxxxxxx>
  • To: "Chris Cheng" <chris.cheng@xxxxxxxxxxxx>,"bdewitt@xxxxxxxxxxxxx" <bdewitt@xxxxxxxxxxxxx>,si-list@xxxxxxxxxxxxx
  • Date: Wed, 22 Oct 2003 13:48:53 -0700

Chris,

It is true that it is possible to infer from the statement that because PCs
pass EMI and have traces on outer layers that traces on outer layers don't 
cause EMI.  That was an error on my part.  It would have been better for me
to state that I had been involved in a number of disc drive designs that
had traces on the outer layers and passed EMI without benefit of a Faraday
cage.  That's not so hard to visualize.

As to my claim that I route members of differential pairs on different
layers successfully, meaning that they function properly, I've got a number
of PCBs in my lab done this way.  I've also got a number of clients with
designs done this way that are meeting all their margins.

Do I deliberately route all members of diff pairs on different layers?  No.
Only when it is necessary to escape a connector or part.  What I stated was
that I have had to route members of the same pair on differnet layers many
times without adverse effects.  It is true that if you route one of them
over a Vdd plane and the other over a ground plane, the noise on Vdd will
couple onto one line and not the other, resulting in differential noise
coupling.   That has to be guarded against.

As to my challenge that traces on outer layers casue EMI, I have seen two
papers cited, one of which does not claim this, the other of which I have
yet to receive a copy.

This is a good dialogue, if for no other reason than that proponents of
particular design rules are being challenged to show their validity.

Chris, I'm not hiding behind anything.  Don't have any reason to.  I'm also
not attacking any of the respondents.

Lee


> [Original Message]
> From: Chris Cheng <chris.cheng@xxxxxxxxxxxx>
> To: bdewitt@xxxxxxxxxxxxx <bdewitt@xxxxxxxxxxxxx>;
<leeritchey@xxxxxxxxxxxxx>; <si-list@xxxxxxxxxxxxx>
> Date: 10/20/2003 5:52:31 PM
> Subject: [SI-LIST] Re: Traces don't cause EMI - really?
>
> Brent & Lee, 
>
> Nice spin on the issue but unfortunately you both tried to avoid the
> original problem that Lee stated :
>
> >As far as EMI is concerned, it has been demonstrated many times, once in
> >the paper done by Doug Brooks with the staff at UMR, that traces
traveling
> >over planes are not a detectable source of EMI.  Therefore, constraining
> >the routing of differential pairs to prevent them from creating EMI is
not
> >appropriate or necessary.
>
> I have seen and explained how surface trace can both fail EMI and signal
> integrity even if it is referencing to a solid plane and perfectly
> terminated. If the edges are fast enough and the reference plane is
> unrelated to the I/O power, the image current will be denied a low
impedance
> return path and will exhibit strong EMI and power/ground bounce. This is
the
> exactly reason why tight coupling differential traces can help. This is
also
> the original claimed by proponents of coupled differential signals and Lee
> has chosen not to response to me. Whether the image current is flowing
> between the differential traces or on the reference plane, they cancel
each
> other out at both the driving and receiving end and thus minimize both EMI
> and power/ground bounce. Whether the reference plane is related to the I/O
> power or not does not affect the outcome in differential case but make a
big
> difference in the single end signal case. 
>
> To hide behind claims that since PC's or workstations have surface trace
> passing EMI and magically deduced that traces does not cause EMI problem
> makes as much sense as driving in the highway seeing no cops pulling
people
> over for speeding and extending that to no one is speeding in the highway.
> Let me ask you this way, have you seen a highspeed system that has surface
> trace referencing to the wrong voltage plane pass EMI with an open
enclosure
> without those crazy highspeed decoupling caps or thin core planes to
return
> the image current ? I have seen plenty of them fail that way. And yet I
have
> seen plenty of coupled differential signals route that way and pass EMI or
> signal quality. This is the fundamental advantage of coupled differential
> signals that you are so ignorant about.
>
> Lee,
>
> You seem to like to make claims that I can easily counter but when face
with
> the rebuff you like to ignore it and continue to repeat your flawed claim
as
> if no one can give you a counter example. 
>
> In this forum, you have been asked many times the following question :
>
> "If tomorrow you are going into a client's office to consult on designing
a
> 2.4GB/s differential signal system. Will you recommend them to "routed
> thousands of differential signal where each member of the pair is on a
> different layer". Do you think that is good engineering practice ? Do you
> think you can still keep your job as a consultant after making that
> statement ?"
>
> And you seem to ignore the fact that this is how all these differential
> traces vs. single un-coupled lines discussion starts.
>
> This is a simple yes or no question. I don't even want to throw in any
> technical point or science into it. You either do it or don't. If you do,
I
> would propose you put that in your lecture notes and whatever book you are
> writing and call it "Lee Ritchey technique" or may be even patent it as I
> sure haven't seem anyone designing >GHz signals doing that. As an inventor
> of that technique, you deserve to publize it and make sure people follow
it.
>
>
> On the other hand, you silence suggests you may have something to hide in
> you claim. It is plain wrong and even you yourself won't dare to do that.
>
> Which way is it ?
>
> -----Original Message-----
> From: Brent DeWitt [mailto:bdewitt@xxxxxxxxxxxxx]
> Sent: Sunday, October 19, 2003 8:28 PM
> To: leeritchey@xxxxxxxxxxxxx; si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Traces don't cause EMI - really?
>
>
> It seems this has become something of a tempest in a teacup.  As I
mentioned
> in an earlier post, I believe much of the energy of the dispute has
> developed over differences in terminology.
>
> First, in the vast majority of cases, I agree with Mr Ritchey.  In the
> twenty five years or so I've been involved in EMC, I've never seen the
> radiation from  surface trace fail a product.  That said, I've only worked
> on old/slow boards with fundamental clocks less than 1.5 GHz, although
I've
> worked with 900 MHz intentional radiators to their 10th harmonic.  Rules
of
> thumb only work until the thumb is too fat to see the problem underneath
it.
> As I mentioned in an earlier post, resonant patch antennas and other
> intentional radiating pcb structures are nothing more than fat, well
> designed traces.  Somewhere between them and our ideal EMC designs lies
> practicality.  As frequency increases, I find myself needing increasingly
> closer inspection of detail.
>
> In somewhat oblique support of Mr. Ritchey, most failing EMC issues I've
> observed have been associated with L di/dt induced voltages on the
reference
> plane caused by surface traces.  I/O cables, using said poorly controlled
> reference planes out to the world, are often a major emissions issue, but
> that is entirely another subject for discussion.
>
> Finally,  I believe Mr. Ritchey is correct, but using the following
> assumptions:
>
> - The trace structure and geometry does not approach a resonant structure
at
> the fundamental or appreciable harmonics of the operating frequency.
> - The reference plane structure supporting the return currents of the
> surface trace does not significantly contribute to reference plane
resonance
> and induced voltage on attached cables.
>
> Respectfully,
>
> Brent DeWitt
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
> List archives are viewable at:     
>               //www.freelists.org/archives/si-list
> or at our remote archives:
>               http://groups.yahoo.com/group/si-list/messages 
> Old (prior to June 6, 2001) list archives are viewable at:
>               http://www.qsl.net/wb6tpu
>   



------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                //www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
  

Other related posts: