[SI-LIST] Re: Effect of Stub-Length on Max. Length Calculation
- From: chris.mcgrath@xxxxxxxx
- To: si-list@xxxxxxxxxxxxx
- Date: Fri, 15 Nov 2002 05:48:52 -0800
You are correct in your assertion that in your situation (PCI?) your max bus
length is governed by the load farthest from the driver, although since test
points are generally not used except in the lab, I prefer to put test points
at some point within the bus and not on its own stub. This can be a little
trickier during routing and layout, but I think it is a better approach to
the problem if constraints don't restrict this approach
As always, stubs or other topologies that induce similar effects in
reflective switching environments should be simulated (most applicably,
post-route) extensively to insure that stray reflections induced by stubs
don't adversely affect other loads on the bus during times when the bus is
active. In other words, care should be taken in looking at the signals
being received at all points in the load individually and not just at the
Driver and Receiver1. I apologize if this seems intuitive.
-Chris
> -----Original Message-----
> From: Adeel Malik [mailto:adeelm@xxxxxxxxxxx]
> Sent: Friday, November 15, 2002 6:08 AM
> To: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Effect of Stub-Length on Max. Length Calculation
>
>
>
> Hi Experts,
> In order to calculate the maximum length of the
> bus signals
> (address, control and data) to meet the set-up and hold times, the
> calculation usually starts with a formula such as :
>
> tflight_time < tperiod - tco - treceiver_setup - tcrosstalk - tjitter
>
> where tco is clock to output delay.
>
> If we consider a following bus with a single driver and
> several receivers
> and assuming that the bus employs 'Reflective Switcing' mechanism,
>
> Driver---Receiver1---Receiver2--------------Receiver3
>
> One needs to find the round-trip delay for a signal to travel
> from driver to
> Receiver3 and back all the way and then calculate the max.
> allowable trace
> length by allowing 5 to 10 round trips for a bus to quiet down.
> So let say if we change the scenario a little bid and add
> some fixed length
> stub close to Receiver 3 (for the purpose of adding the Test
> Point). So the
> modified figure will now be:
>
>
> < A >
> Driver---Receiver1---Receiver2-------------Receiver3
> |
> |
> |
> | --> a stub of length L
> |
> |
> TEST POINT CONNECTOR
>
> My question is that how our original calculation to determine the max.
> bus-length will be affected by the addition of this stub of
> length L. If the
> length L is less than or equal to distance 'A' as depicted in
> the figure,
> shouldn't our original estimation of the max. bus length
> remain valid ?.
>
> Regards,
> Adeel
>
>
>
> ------------------------------------------------------------------
> 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:
