[SI-LIST] Re: Ferrite Beads on Differential Signal Lines?
- From: "billg12@xxxxxxxxxxxxx" <billg12@xxxxxxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Fri, 12 Sep 2014 15:21:02 -0400 (EDT)
Hey Guys and Gals, A twisted pair keeps the signal a closely held secret
between the originator and receiver. IF you have a lot of common mode noise or
other AC trash, put a bead over the two wires. Put a bead at both ends, but be
careful to include the signal return path or the complementary signal.
A bead on a pair tends to make the pair a differential pair, which is good. A
bead at the transmitter (including the return path in the bead, and a bead over
both conductors at the far end. If you are sending a single ended signal, use
twisted pair with a bead over both conductors at both ends.
It has been 40 years ago that a young man complained about the signal from my
box. I went in the lab and looked at his setup. He had connected the 16 data
lines and the 4 strobe lines, and connected his scope ground to the receiver in
his box... When he looked at the signal at my box, there was one clean pulse.
At his box there were three pulses. I had to tell him to hook up all of the
grounds and twist the ground with the signal. And terminate the grounds in his
box on the matching transceiver chips.
After much arguing and pawing the ground and stamping the feet, he twisted
his ground wires between the cable connectors, and connected the grounds in his
box. Suddenly (as soon as he turned the power back on) the twisted pairs were
balanced and the signals didn't bounce. This was the same guy who hooked a
piece of wirewrap wire between a 50 MHz source and his counter chip. Again, his
ground was through the scope probe and the counter chip was defective. He went
through several chips before he twisted a return wire with his signal wire and
terminated the grounds properly. No more defective chips.
This was in the early days of TTL and people had problems with the speed. I
go back to RF and I know "Ground is Ground the World Around!" is false. Ground
is where you put it. And if you don't 'Put It', expect a lot of radiation.
Always put the ground where you want it. Or where it needs to be...
When you route a signal, look at where the return is. If the signal is
crucial, make sure the signal return goes hand in hand with the signal. Think
about a transmission line with big hunks of sheet metal attached to one of the
conductors. That is what it looks like when you go single ended over a ground
plane. The impedance of your t-line changes when the return goes across a
ground plane. And if you cross a split plane... too awful to describe. Put on
your RF hat and look at what you are doing. If you are going to route a 500 MHz
broadband signal, Consider the reflections if your return is not a constant
size and constant distance from the signal trace(wire, ...).
If you were the chief engineer for a broadcast station, think about how you
would route the energy from the transmitter to the antenna. Would you drive a
ground stake at the door to the transmitter building and ground your
transmission line there, then run single ended to the base of the tower,
connect the single ended T-Line to the matching network and use a ground stake
for the return, and then attempt to match the antenna. You can save a lot of
copper that way... Probably kill some earth worms, and heat the soil. And get
fired!
OK, you are routing signals with edges in the GHz range. Be very paranoid
about who gets to provide the return. If you have a differential signal, keep
the pair close, keep the impedance constant, and watch out for lines that run
parallel for a ways and then go elsewhere. They may have some of your signal.
Be scared, very scared! ;)
Best Regards
BillG
-----Original Message-----
>From: Orin Laney <olaney@xxxxxxxxx>
>Sent: Sep 12, 2014 11:47 AM
>To: leeritchey@xxxxxxxxxxxxx, clfranci@xxxxxxxxxxx, si-list@xxxxxxxxxxxxx
>Subject: [SI-LIST] Re: Ferrite Beads on Differential Signal Lines?
>
>All generalizations are false.
>
>-----Original Message-----
>From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
>Behalf Of Lee
>Sent: Friday, September 12, 2014 8:15 AM
>To: clfranci@xxxxxxxxxxx; si-list@xxxxxxxxxxxxx
>Subject: [SI-LIST] Re: Ferrite Beads on Differential Signal Lines?
>
>It makes no sense to put ferrite beads in any signal line. How did this get
>started?
>
>-----Original Message-----
>From: Craig Francis
>Sent: Thursday, September 11, 2014 4:11 PM
>To: si-list@xxxxxxxxxxxxx
>Subject: [SI-LIST] Ferrite Beads on Differential Signal Lines?
>
>SI-List,
>I haven't seen any literature concerning using ferrite beads on each signal
>line of a differential pair.
>Usually only a common-mode choke is used to reduce common-mode EMI.
>
>Can ferrite beads be used on a differential pair to increase the the rise
>time of the signal and reduce unnecessary high-frequency content that may
>radiate?
>
>The two ferrite beads would be sized to allow the bandwidth necessary for
>the differential signal's highest data rate.
>
>
>Thanks,
>Craig Francis
>
>
>------------------------------------------------------------------
>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 forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
>List archives are viewable at:
>//www.freelists.org/archives/si-list
>
>Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
>
>
>-----
>No virus found in this message.
>Checked by AVG - www.avg.com
>Version: 2014.0.4765 / Virus Database: 4015/8184 - Release Date: 09/09/14
>
>------------------------------------------------------------------
>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 forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
>List archives are viewable at:
> //www.freelists.org/archives/si-list
>
>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 forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
>List archives are viewable at:
> //www.freelists.org/archives/si-list
>
>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 forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
