[SI-LIST] Re: Long T reflection problems

  • From: "Andrew Ingraham" <a.ingraham@xxxxxxxx>
  • To: <si-list@xxxxxxxxxxxxx>
  • Date: Thu, 3 Mar 2005 11:20:08 -0500

I'm noticing that the signals are really "slow" (change infrequently), but
that you expect them to have rather fast, sub-ns edges.

Also that you expect very long lengths when the logic analyzer is attached.

I don't have much experience with logic analyzers, but I would have thought
that any modern one has some sort of isolation, probably in the form of
buffer amps, in the test head that attaches to your board.  Expecting any
system to work with an extra meter or so of T-line attached, is NOT normal,
and I doubt that a logic analyzer today would do that sort of thing.  Logic
analyzer vendors know about the SI problems that would be introduced by
attaching a one or two meter line to a board trace, so they probably design
them to present no more than a few centimeters at most of additional length,
followed by a buffer, before the cable back to their main unit.

That being said, make sure that your model does that.

It sounds like you could isolate your 150mm "stub" with a big resistor, and
be done with it.  With 500 ns signals, unless the logic analyzer needs to
capture the signals in a lot less time than that, let it eat a few dozen
nanoseconds by using big (>>50 ohm) resistors between the "T" and your stub.

With only 1.5cm between devices, getting two sets of series terminating
resistors (one at each driver) wouldn't make sense.

Or maybe you can live with your 150mm unterminated stubs and accept the
reflections.  If you've really got 500ns for signals to settle, crappy
waveforms for the first dozen - even hundred - nanoseconds are OK as long as
they don't overshoot severely, or radiate.

5V signals with 0.3ns edges are really screamin'.  Maybe they aren't quite
that fast.


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:

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List FAQ wiki page is located at:

List technical documents are available at:

List archives are viewable at:     
or at our remote archives:
Old (prior to June 6, 2001) list archives are viewable at:

Other related posts: