TECHNET Archives

1996

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Date:
Thu, 24 Oct 1996 08:47:42 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (78 lines)
     Paul,
     
     For most of our board vendors, hole count doesn't seem to be as much 
     of a cost issue as layer count and trace/space and minimum hole sizes. 
     The vias should only reduce stack height if you used a smaller size 
     than were used elsewhere in the board.  
     
     Yes, the vias will likely cause impedance bumps for traces running 
     near them. Does it matter?  I don't know. If the impedance needs to be 
     controlled, you'll need to allow sufficient clearance from them. The 
     hardest part for me is negotiating enough board area from the 
     mechanical people for things like guard rings and striplines.
     
     I agree, edge plating is electrically probably much better than vias 
     but I think boards would have to be singulated before plating.  If 
     your boards are going to be assembled as multi-up panels you will 
     probably be stuck with using vias.
     
     [log in to unmask]


______________________________ Reply Separator _________________________________
Subject: Re: Internal and external ground planes
Author:  [log in to unmask] at Internet2
Date:    10-23-96 21:46


In your message dated Wednesday 23, October 1996 you wrote :  
>      I recently attended the University of Wisconsin's "System EMC 
>      Compliance through Printed Circuit Design" seminar and was taught that 
>      if the external planes are for shielding purposes, it is a good idea 
>      to run a guard ring of ground around the periphery of the board on all 
>      layers and tie the external planes and the internal ground plane 
>      together through closely spaced (.25 inch) vias only at the guard 
>      ring. Tying the IC and Cap ground pins to the external plane only puts 
>      noise onto the shield.  The course contained other subtle tricks. Take 
>      it if you can get the chance.
>
There are serious cost implications here with a significant increase in hole 
count together with reduced stack height for small holes. This could up the 
drilling cost by a factor of 3 or more. On one design we had to manufacture, it 
added 5000 holes to the panel.

I was also wondering what the impedance characteristics would be like for 
inner layer signal tracks running close to the vias. Would it be spiky with the 
fluctuating dielectric? Does it matter?

If there is no other way of shielding within the casing of the finished unit, 
then perhaps edge plating could be a lower cost method and certainly more 
uniform. This would mean routing the profile, leaving attachment points, at the 
first drilling stage. The attachment points are then routed through at the final

profile stage. If the screening is really critical, vias could be placed at the 
points where there will be no plating.

We have done this on a number of designs over the years with no manufacturing 
problems.
-- 
Paul Gould
Teknacron Circuits Ltd
EMail [log in to unmask]

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************



ATOM RSS1 RSS2