[SI-LIST] Re: Differential signals - skew - and EMC
- From: Ralph Morrison <ralphmorrisonee@xxxxxxxxx>
- To: Charles.Grasso@xxxxxxxxxxxx
- Date: Sat, 2 Nov 2013 13:34:46 -0700
Hi Charles,
I read your question on differential signals and I would like to offer
some commentary. I feel the only way to appreciate the issues
involved is to look at the fields produced by the signals of interest
and the fields that you desire to reject. The signal of interest must
be carried from one circuit to another across a space usually in a
cable. This space contains fields that can be labeled
interference. These fields can be from your own hardware or from
other transmitters. This interference is already radiation from many
sources and produces surface currents on all conductors that reflect
the interfering fields. By attaching any cable to a piece of
hardware, these external fields will cause current flow on the cable
shield or on bare conductors.. This current flow will then flow on
the outer conducting planes of a circuit board. This is expected.
It is normal. This is not all bad and here is the reason.
A logic signal of 5 volts moving on a 5 mil spaced transmission line
has an E field of about 200,000 volts per meter. Interference fields
are rarely more than 2 volts per meter. So the coupling is the ratio
of E fields or about one part in 100,000. Interference fields
(television, radio, radar etc.) are not an issue on the board where
transmission lines are well controlled. They will not add significant
signal to normal transmission paths. An entering cable is another
issue. If there is a shield there is a phenomena know as transfer
impedance. On a cable, interfering surface current produce fields
that couple to conductors. This field usually couples to all
conductor pairs and specifically between every conductor and the
shield. It adds a signal that is given the name common-mode
interference. This is the true culprit. The fields that couple are
large because the conductor spacings and cable lengths are not
small. The cable thus couples to the interfering field. This
coupling is called common-mode interference. This signal can be
rejected by the circuitry in the receiving components.
Signal lines on a board have well controlled transmission paths and do
not couple to interference. The best way to carry balanced signals on
a board is to treat them as independent signals and carry them to the
components. There will be no added interference and the transmission
lines will be easy to terminate.
I hope this helps.
Ralph Morrison
On Nov 1, 2013, at 12:31 PM, Grasso, Charles wrote:
> Hello,
>
> In an ideal situation differential signals will have no skew through
> the transmission path and (as I understand it)
> the common-signal (emi) will be very low as a result. Given that
> EMC is very system dependent - does any one have
> a rule of thumb or anxiety(!) factor for how much is skew is
> tolerable before becoming an emissions (regulatory)
> Issue?
>
> Thanks in advance.
> Charles Grasso
> ------------------------------------------------------------------
> 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
>
>
------------------------------------------------------------------
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:
