[SI-LIST] Re: Developing High Speed Routing Rules
- From: "Salkow, Steven" <steven.salkow@xxxxxxxx>
- To: "'jonpowell@xxxxxxxxxxxx'" <jonpowell@xxxxxxxxxxxx>,Wilbur.Harvey@xxxxxxxxxxxxxx, si-list@xxxxxxxxxxxxx
- Date: Tue, 17 Jun 2003 19:43:29 -0700
There are always more issue about cost than you can shake a stick at. (Model
library integrity is one such issue.) Doing a "what if" by simulation will
get you to design goals. I agree with Jon Powell, nothing is as good as SI
on the completed board. That said, distilling rules for the router is
probably, by itself, not definitive enough. The PCB designer needs feedback
to get it right. Hence, the PCB designer gets this from you or has SI tools
for this purpose. SI is a long term investment. If doing a lot of it and
your to become an expert, then its worth the time effort and $.
Alternatively, a good idea of the layout arrived at by simulations to derive
board requirements followed by one or two prototype build may be quicker for
a company that does not have the dollars and the time to invest in SI
methods.
Technologically, those that do not have SI tools may never know the extent
of their design margins and possibly will not be able to arrive at complex
high-speed designs on their own. A recent example was a design where the
FPGA had an unexpectedly fast rise/fall times hence should have had some
form of output termination but this was overlooked in the actual first
article because traces were less that 4.5 inches. With spins of the artwork
maybe approaching $4000 to 10,000 depending on complexity, SI can pay for
itself within 4 projects. When you consider , however, a mistake found later
in production after even a 100 units are already manufactured, the cost of
SI looks very small.
Steven Salkow
Lockheed Martin
3200 Zanker Rd
San Jose, CA 95134
(408) 473-4058
steven.salkow@xxxxxxxx
Fax (408) 473-3044
> -----Original Message-----
> From: Jon Powell [SMTP:jonpowell@xxxxxxxxxxxx]
> Sent: Tuesday, June 17, 2003 6:34 PM
> To: Wilbur.Harvey@xxxxxxxxxxxxxx; si-list@xxxxxxxxxxxxx
> Cc: Yao, Jimmy; Au, Gabriel; Yuen, Patrick
> Subject: [SI-LIST] Re: Developing High Speed Routing Rules
>
> OK,
> Some quick generalizations and then specifics.
>
> I believe that routing rules are secound best approximations that should
> be
> used in situation where you can't simulate. For instance, you can simulate
> a
> complete board SI (to acceptable accuracy for most people) but you can't
> do
> EMI simulation for a full board (to acceptable accuracy for most people).
> So
> you may have to use Rules for EMI.
>
> Now, I know that some simple rules for routine (CLOCK to CLOCK spacing
> etc.)
> are good things to start with, but if you have really comprehensive rules
> (If you find a series termination on a CLK net make sure the device is
> place
> within .4 inches of the driver) I think you will find that you quickly
> overwhelm any auto-routing engine.
>
> So, use simple rules for routing, and check everything aftwards with SI
> simulation.
>
> As to whether you should buy are rent. I would say it depend on how many
> designs you are going to do using a particular stackup and layout strategy
> (cause the rules change as you change techs).
>
> that is all
> jon
>
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Harvey, Wilbur
> Sent: Tuesday, June 17, 2003 1:02 PM
> To: si-list@xxxxxxxxxxxxx
> Cc: Yao, Jimmy; Au, Gabriel; Yuen, Patrick
> Subject: [SI-LIST] Developing High Speed Routing Rules
>
>
> We are trying to make the tradeoff of purchasing tools such as HyperLynx
> or
> Specctraquest.
>
> The most important use of these tools that we see, is to develop the
> Specctra routing rules.
>
> Specifically we are working in the domain of 600Mhz LVDS and Rocket I/O,
> as
> well as the normal STTLII for the DDR memory interfaces.
>
> The question is, over a several year period, would we be better off to
> hire
> outside companies to develop these rules for us as needed, or to have the
> tools in house and develop them ourselves. Any comments would be
> appreciated.
>
> If anyone has any leads on who we could contact to get prices on
> developing
> these rules it would be appreciated as well.
>
> Wilbur Harvey - Engineer
> Spirent Communications Inc., www.adtech-inc.com
> First Hawaiian Center, 999 Bishop St.. Suite 1700
> Honolulu, HI 96813
> Tel: +1 (808) 440 3494, +1 (808) 440 3363 (Direct)
> Mobile: +1 (808) 221 0395
> email: wilbur.harvey@xxxxxxxxxxxxxx
>
>
> ------------------------------------------------------------------
> 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
>
------------------------------------------------------------------
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:
