[SI-LIST] Re: nonlinear in power ground noise of clock driver
- From: "Brent DeWitt" <bdewitt@xxxxxxxxxxxxx>
- To: <gene.glick@xxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Thu, 4 Dec 2003 22:12:03 -0700
They are not non-linear with respect to amplitude below saturation, but the
ratio of L to R is highly non-linear with frequency. The vector sum of the
two components is generally published by the manufacturer as a curve of loss
tangent. Most filter beads are designed to have the lossy component equal
to dominant in the recommended frequency range. The description of
"frequency dependent resistor", while not entirely accurate, is much better
than considering them as inductors.
Hope that helps.
Brent DeWitt
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Glick, Gene (IndSys,
Interlogix)
Sent: Thursday, December 04, 2003 11:40 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: nonlinear in power ground noise of clock driver
Can someone enlighten me on ferrite bead behaviour? My rudimentary =
understanding of these non-linear devices (and I've never persued it any =
further) is that they act as frequency dependent resistors. Do they =
have inductance as well? What other non-linear stuff is going on there?
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Raymond Anderson
Sent: Thursday, December 04, 2003 1:23 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: nonlinear in power ground noise of clock driver
In this particular case with the info given I can't be sure if the =
difference
seen is caused by just the difference in the decap ESR or by the =
difference
in ESR combined with the LC filter design. I'd kind of guess it is a =
function
of the ceramic cap having a much lower ESR.
As Dav0 wrote, LC power filters can be very problematic is not designed
correctly. The problem stems from the fact that the filter (be it pi, t, =
or L
section) is being driven from a very low impedance source (say 10's of
milliohms or less) and is loaded by a moderate (10's or 100's of ohms) =
load.
The large ratio of load to source impedance can cause huge (10 or 20 dB =
or
more) peaking in the filter's response. A filter that you might hope has =
a
nice low pass roll-off at about 10 kHz may have 20 dB peaking very near =
to
that frequency. A person might think they could synthesize a proper =
low-pass
filter using classical synthesis methods to produce a nice Butterworth =
(or
whatever) LPF. On paper you can, but you will find the element values =
are
essentially unrealizable. A filter that is driven from near zero =
impedance
into say a 100 ohm load with a cutoff freq of around 10 kHz can require
Henries of inductance and perhaps Farads of capacitance (not the sort of
thing you want on your PCB. You can actually measure more noise on the =
filter
output than on the input (due in part to the different impedance =
levels). The
typical cure is to place a small amount of resistance in series with the
inductor to effectively de-Q it and limit the circulating current or as =
Dave
mentioned, go to a RC filter instead of an LC filter. Build yourself a =
good=20
spice model of the filter and play around with it to understand what is =
going=20
on. BTW, you can generate a pretty good model for the ferrite or =
powdered iron=20
inductive element, but you won't find it in the vendor's data sheets or =
web site.
-Ray Anderson
Staff SI Engineer
Sun Microsystems Inc.
Menlo Park, CA
Dav0 Lieby wrote:
>I had this experience a while ago.
>It boils down to the bead and the
>capacitance. We found a resonant
>circuit. Various capacitors gave
>huge noise results, another lot of
>them was OK. The final solution
>we did was to replace the bead with
>a resistor, the clock chip did not
>eat too much current so it did not
>go out operational range. =20
>
>dav0
>David Lieby
>Siemens Medical Systems, Ultrasound Division
> =20
>
------------------------------------------------------------------
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 technical documents are available at:
http://www.si-list.org
List archives are viewable at: =20
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages=20
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
=20
------------------------------------------------------------------
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 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 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:
