From: Dennis Schmitz ([email protected])
Date: Wed Jan 10 2001 - 09:28:00 PST
As it happens, I've done a fair amount of work on backplane LVDS signals.
The stub terminator resistors are critical -- they must be as close as physically possible to the connector. He may complain about tooling clearances, so beware. Similarly, the length of the conductors in the connectors (both together) is also critical - this might mean using fewer rows than you'd like.
Also get LVDS transceivers closer to the connector -- insist that he not exceed a stub length greater than the board pitch and the effects of adjacent boards aren't additive in effective impedance of the stub field.
Your problems will be significantly reduced if the longest propagation time is less than half a bit-time.
If you can't get those design rules, then you'll need to simulate combinations of adjacent boards.
----- Original Message -----
From: John Lin (ªL´Â·×)
To: [email protected] (E-mail)
Sent: Wednesday, January 10, 2001 4:00 AM
Subject: [SI-LIST] : How to simulate SCSI backplane with 15 slots?
Dear SI gurus,
One question regarding SI simulation for SCSI backplane. How do I define the worst case of SCSI backplane with 15 slots for 15 SCSI drives?
Shall I do the SI simulation, theoretically, for all combinations of 15 SCSI drives?
There are huge combinations -- the number of drives mixed with the number of slots (location on the backplane). Based on my knowledge, I would like to put drives on even spaces for the best case of SI simulation. for example, three drives on slot 1st,8th, and 15th. For the worst case, intuitively, I would like to put them on slot 1st,2nd, and 3rd because of far away from end terminators.
The question is that how to assure the worst case for simulation? Or, doing all combinations is only way to do even it is exhaustive.
Your comments are highly appreciated.
Best Regards,
John Lin
Senior SI Engineer, Server Team, ARD4
Quanta Computer Inc.,Taiwan, R.O.C.
Email: [email protected]
Tel: 886+3+3979000 ext. 5183
**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected]. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****
This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:30:36 PDT