Re: [SI-LIST] : Different board vendors different impedances

Ron Miller (rmiller@Brocade.COM)
Mon, 29 Mar 1999 13:00:49 -0800

--------------A37F1873AFB8EE99F9017014
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> Hi Chris

While the idea of specifying 50 ohms sounds simple, when you are using
differential traces and have capacitive pads for tuning and inductors it makes
more sense to put a calculated /simulated 50 ohm test trace on the board for
the board house to shoot at and assume that other structures will be correct
if the 50 ohm is correct. Then the 50 ohm test trace can be trimmed off the edge.

ron miller

> Specifying Er of FR4 is not an easy task. It depends on resin content,
> which changes with dielectric thickness and it also changes with
> frequency.
> Should we specify the fundamental bandwidth of the TDR risetime that is
> used to measure Zo? (Example, Tr of TDR = 35ps, which slows to around
> 140ps the the test trace, so 0.35/140ps = 2.5GHz). I think not.
>
> Another problem is that FR4 is not uniform. Digging into how many Er's
> exist in a stackup will make your head spin. C-stage has one resin
> content
> and therefore an Er. B-stage (partially cured material) has a different
> resin content and therefore another Er. Both B-stage and C-stage are
> comprised of glass bundles (another Er) that criss-cross through the
> material with resin (another Er) in between.
>
> Take all these lovely variables into play and you may conclude that
> specifying a single value of Er is quite hopeless.
>
> Further, the traces we are so careful about are not rectangles, they are
> trapezoids with "teeth" on the c-stage side of the trace that can
> account
> for as much as 30% of the thickness of the line. The b-stage side of
> the
> trace is "roughed up" during the etching process to help adhere to the
> b-stage.
>
> Solution?: Let the vendor be responsible for Zo with a tolerance.
> Suggest
> a line width and overall board thickness and let the vendor do the rest.
>
> Chris Heard
> 3Com Corp
> Marlboro Ma
> 508-323-2115
>
> **** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****

--
Ronald B. Miller  _\\|//_  Signal Integrity Engineer
(408)487-8017    (' 0-0 ') fax(408)487-8017
     ==========0000-(_)0000===========
Brocade Communications Systems, 1901 Guadalupe Parkway, San Jose, CA  95131
rmiller@brocade.com,  rbmiller@sjm.infi.net

--------------A37F1873AFB8EE99F9017014 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">

Hi Chris


While the idea of specifying 50 ohms sounds simple, when you are using
differential traces and have capacitive pads for tuning and inductors it makes
more sense to put a calculated /simulated 50 ohm test trace on the board for
the board house to shoot at and assume that other structures will be correct
if the 50 ohm is correct.  Then the 50 ohm test trace can be trimmed off the edge.

ron miller
 
 

Specifying Er of FR4 is not an easy task.  It depends on resin content,
which changes with dielectric thickness and it also changes with
frequency.
Should we specify the fundamental bandwidth of the TDR risetime that is
used to measure Zo?  (Example, Tr of TDR = 35ps, which slows to around
140ps the the test trace, so 0.35/140ps = 2.5GHz).  I think not.

Another problem is that FR4 is not uniform.  Digging into how many Er's
exist in a stackup will make your head spin.  C-stage has one resin
content
and therefore an Er.  B-stage (partially cured material) has a different
resin content and therefore another Er.  Both B-stage and C-stage are
comprised of glass bundles (another Er) that criss-cross through the
material with resin (another Er) in between.

Take all these lovely variables into play and you may conclude that
specifying a single value of Er is quite hopeless.

Further, the traces we are so careful about are not rectangles, they are
trapezoids with "teeth" on the c-stage side of the trace that can
account
for as much as 30% of the thickness of the line.  The b-stage side of
the
trace is "roughed up" during the etching process to help adhere to the
b-stage.

Solution?:  Let the vendor be responsible for Zo with a tolerance.
Suggest
a line width and overall board thickness and let the vendor do the rest.

Chris Heard
3Com Corp
Marlboro Ma
508-323-2115

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP.  si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****

-- 
Ronald B. Miller  _\\|//_  Signal Integrity Engineer
(408)487-8017    (' 0-0 ') fax(408)487-8017                 
     ==========0000-(_)0000=========== 
Brocade Communications Systems, 1901 Guadalupe Parkway, San Jose, CA  95131
rmiller@brocade.com,  rbmiller@sjm.infi.net
  --------------A37F1873AFB8EE99F9017014-- **** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****