[SI-LIST] Re: Why 220ohms at driver end in PECL Terminations
- From: "Suresh Sivasubramaniam" <Suresh.Subramaniam@xxxxxxxxxx>
- To: Andrew.Ingraham@xxxxxx
- Date: Mon, 07 Oct 2002 10:10:53 -0700
Apart from the excellent responses already received I would recommend app
note # AND8020-D by Paul Shockman of OnSemi. It is a tutorial on ECL
terminations and is available from www.onsemi.com under the technical
documentation tab.
Suresh
"Ingraham, Andrew" wrote:
> > The recommended Termination for PECL has a 220ohms to ground on each
> > leg of
> > the differential pair at the driver end. Can anyone explain what is
> > the
> > purpose of 220ohms. If it is to damp the reflections from the receiver
> > why
> > 220ohms ?. Is it good to design without the 220ohms.
>
> ECL/PECL are similar to open-drain or open-collector outputs, in that
> the driver only pulls one way. Open-drain outputs only pull down, and
> you must have an external pull-up resistor. ECL/PECL outputs only pull
> up, and you must have an external pull-down resistor. Otherwise, you
> don't get any signal.
>
> With ECL (and PECL is essentially the same), there were two standard
> recommended "terminations".
>
> (1) Wires longer than "several" inches (depending on rise/fall times)
> require true terminations to control overshoot and ringing, typically
> about 50-75 ohms (to match the impedance of your wires). The resistor
> should be at the receiver end, and goes to Vtt. It serves the dual
> purpose of terminating the transmission line, and providing the current
> path to keep the output transistor conducting. (One could use a
> two-resistor combination that was electrically identical to 50 ohms to
> Vtt, without requiring a Vtt supply.)
>
> (2) Shorter wires can have just a pulldown resistor to Vee (=ground when
> using PECL). The resistor could be anywhere, driver or receiver end.
> Typically 220 or 330 ohms to Vee works pretty well. The advantage is no
> need for a Vtt supply, or less power dissipation than the two-resistor
> Thevenin equivalent.
>
> Other variations are possible, including pulldowns plus series
> termination resistors. No matter what you do, you need to provide a DC
> path from the output pin to a more negative voltage (Vtt or Vee), or it
> doesn't work.
>
> If you have AC (capacitor) coupling between one stage and the next, you
> need to put a pulldown resistor before the capacitor (i.e., somewhere
> near the driver), even if the line is long and there is also a matched
> termination at the receiver end. The pulldown does not terminate or
> damp reflections. However, in this case there is an extra thing to
> consider. The pulldown resistor is needed not only to keep the
> transistor conducting, but also because the current to the AC coupled
> load, flows both ways. So the pulldown resistor has to be small enough
> to keep the transistor conducting a little, even when the output
> switches from high to low and the resistor sinks the transient load
> current too.
>
> Andy
>
> ------------------------------------------------------------------
> 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:
