[SI-LIST] Re: Radiated Emissions continued
- From: Larry Smith <Larry.Smith@xxxxxxx>
- To: Ken.Cantrell@xxxxxxxxxxxxxxxx
- Date: Wed, 24 Aug 2005 10:16:11 -0700
Well, it was something like that...
Actually, I was in the lab "helping" the EMI guy because he was
stealing all of my PDS caps in order to fix HIS EMI problem! But we
worked together well and I learned a lot from him as I trust he did
from me. We were using all available capacitor sites to see if we
could reduce the EMI. After carefully selecting capacitors that
resonated at the problem frequency and mounting them in the various
locations, we found that sometimes the EMI got better and sometimes it
got worse. We hypothesized that locating a low impedance 1/4
wavelength away from a source actually applied a high impedance to the
source, just like a 1/4 wavelength transmission line resonator.
During that period of time, I was developing power plane analysis a
tool based on transmission line matrices and we used that tool to
confirm that this was indeed what was happening. This underscored the
importance of using power plane analysis tools to place high frequency
decoupling capacitors. This was back in the days of 14 mils
separation between power and ground planes and 56pF capacitors
actually did something. Now we place power and ground planes next to
each other with 4 mils or less separation. We have fewer EMI problems
and small value decoupling capacitors have very little affect compared
to the capacitance of the planes.
regards,
Larry Smith
Sun Microsystems
PS - sorry for the tardy response, I have been on a nice 2 week
vacation and am just sorting through all my email!
Ken Cantrell wrote:
>
> Robert -
> I remember something that Larry Smith said once. He was failing EMI at some
> frequency, and called in the EMC guy. The EMC guy looked for lambda/4
> distances on the PC board, and found some bypass caps at that distance from
> one of the chips. The transfer impedance at lambda/4 is infinite, so the
> bypass caps weren't bypassing. You're lambda/4 (air) is 4.292 inches. Trap
> cap value is ~ 56pF.
> Larry - feel free to correct anything that I missinterpreted.
>
> Ken
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Americomrh@xxxxxxx
> Sent: Wednesday, August 17, 2005 1:27 PM
> To: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Radiated Emissions continued
>
> Content-Type: text/plain; charset="US-ASCII"
> Content-Transfer-Encoding: 7bit
>
> Thanks for the informative replies with regard to the radiation emissions
> question. Here is more info detailing the problem.
>
> The chassis, backplane, and front panel are all grounded. The PCB has
> frame
> ground floods on both the upper and lower layers. The front to back
> distance of the cavity is about 5.5 inches. The left to right width of the
> cavity
> is about 15 inches, but there is some intervening metal for card guides.
> It
> is mostly open. The top to bottom size is 1.3125 inches.
> The other anomoly is where it comes from. I believe the frequency is
> actually 687.5MHz. We have 125MHz on the board. At one time, I divided it
> by 2 to
> get 62.5MHz to drive some logic so I assumed it was the 11th harmonic. I
> changed the 62.5MHz to 41.67MHz, but the 687.5MHz did not move. Several
> other
> lower level spikes did move, but they are not a concern. The 687.5MHz
> appears to be the 5 1/2 harmonic. Fourier says that can't be. Perhaps a
> chip
> does a divide by 2 internally which I don't know about. We tried bypass
> caps on
> the power supplies as small as 100pf without effect.
>
> Robert Hanson
>
> -- Attached file included as plaintext by Ecartis --
>
> Return-Path: <Gary.Miller@xxxxxxxxx>
> Received: from rly-xa04.mx.aol.com (rly-xa04.mail.aol.com [172.20.64.40])
> by air-xa03.mail.aol.com (v107.10) with ESMTP id MAILINXA31-67430337d224a;
> Wed, 17 Aug 2005 09:12:54 -0400
> Received: from mimesweeper2.hubbell.com (mimesweeper2.hubbell.com
> [63.111.141.185]) by rly-xa04.mx.aol.com (v107.10) with ESMTP id
> MAILRELAYINXA43-67430337d224a; Wed, 17 Aug 2005 09:12:50 -0400
> Received: from CRPORGCTNW03-IA.HUBBELL.COM (gwise.hubbell.com
> [63.111.141.139])
> by mimesweeper2.hubbell.com (8.12.11/8.12.11) with ESMTP id
> j7HDvhe8009428
> for <Americomrh@xxxxxxx>; Wed, 17 Aug 2005 09:57:43 -0400
> (envelope-from Gary.Miller@xxxxxxxxx)
> Received: from HUBBELL_GWGATE-MTA by CRPORGCTNW03-IA.HUBBELL.COM
> with Novell_GroupWise; Wed, 17 Aug 2005 09:12:49 -0400
> Message-Id: <s302ff91.086@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
> X-Mailer: Novell GroupWise Internet Agent 6.0.4
> Date: Wed, 17 Aug 2005 09:12:40 -0400
> From: "Gary Miller" <Gary.Miller@xxxxxxxxx>
> To: <Americomrh@xxxxxxx>
> Subject: Re: Response to EMI Questions
> Mime-Version: 1.0
> Content-Type: text/plain; charset=US-ASCII
> Content-Transfer-Encoding: quoted-printable
> Content-Disposition: inline
> X-AOL-IP: 63.111.141.185
>
> Robert,
> The resonant cavity is an interesting thought. What is the Q of a
> reson=
> ant cavity?
> The chassis, backplane, and front panel are all grounded. The PCB has
> fr=
> ame ground floods on both the upper and lower layers. The front to back
> dis=
> tance of the cavity is about 5.5 inches. The left to right width of the
> cav=
> ity is about 15 inches, but there is some intervening metal for card
> guides.=
> It is mostly open. The top to bottom size is 1.3125 inches.=20
> The other anomoly is where it comes from. I believe the frequency is
> ac=
> tually 687.5MHz. We have 125MHz on the board. At one time, I divided it
> by=
> 2 to get 62.5MHz to drive some logic so I assumed it was the 11th
> harmonic.=
> I changed the 62.5MHz to 41.67MHz, but the 687.5MHz did not move.
> Several=
> other lower level spikes did move, but they are not a concern. The
> 687.5MH=
> z appears to be the 5 1/2 harmonic. Fourier says that can't be. Perhaps
> a=20=
> chip does a divide by 2 internally which I don't know about. We tried
> bypas=
> s caps on the power supplies as small as 100pf without effect.
> Perhaps we can discuss this when you arrive. For now, we are moving
> tow=
> ard a plastic front panel as a solution.
>
> regards
> Gary Miller
> Pulsecom
>
> >>> <Americomrh@xxxxxxx> 08/16/05 07:42PM >>>
> =20
> Gary,
> =20
> My first thought is that a resonance frequency(s) may be inherent in
> your=20
> chassis. A quarter wave length at 688 MHz is about 4.36 in. which can
> exis=
> t=20
> within your structure. Any harmonics of 688 MHz (assuming they exist)
> woul=
> d=20
> have wave lengths of decreasing distance. Your front panel is grounded,=20
> how-wire, Cu, braid, other and how long? I assume that the enclosure is
> al=
> so=20
> metal, but it is floating. Therefore can I assume it is isolated from
> the=20=
> front=20
> panel? Also, what is "ground," i.e. chassis, earth, green wire, etc.?
> =20
> When the front panel is removed the cavity is eliminated which stops the=20
> build up of resonant modes. Now this is all conjecture on my part in
> that=20=
> I=20
> have not experienced any of your development knowledge; therefore, it is
> ju=
> st an=20
> educated guess.
> =20
> Robert
> =20
> message dated 8/16/2005 8:01:22 A.M. Pacific Daylight Time, =20
> Gary.Miller@xxxxxxxxx writes:
>
> Robert,
> We have just experienced an unexpected EMI phenomena, and I would like
> to=20
> add a question to our list of questions.
>
> One of our units had an emission problem at 688MHz. The device is
> roughly=20
> 5" x 5" x 1.5" with a grounded, metal front panel. When we removed the
> fro=
> nt=20
> panel (leaving a gaping hole) the overall emission reduced by 20dB. How
> is=
> =20
> this possible? We are looking at isolating the panel from ground as our=20
> current solution.
>
> Gary Miller
> Pulsecom
>
> *******************************************************************
> This email and any files transmitted with it are confidential and
> intended solely for the use of the individual or entity to whom they
> are addressed. If you have received this email in error please notify
> the system manager.
>
> This footnote also confirms that this email message has been swept
> for the presence of computer viruses.
>
> www.hubbell.com - Hubbell Incorporated
>
> ------------------------------------------------------------------
> 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 FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
> List technical documents are available at:
> http://www.si-list.org
>
> 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
>
> ______________________________________________________________________
> All email being sent to or from SRC Computers, Inc. will be scanned by a
> third party scanning service.
> ______________________________________________________________________
>
> ------------------------------------------------------------------
> 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 FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
> List technical documents are available at:
> http://www.si-list.org
>
> 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 FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
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:
