TECHNET Archives

February 1999

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:
Dan Brandler <[log in to unmask]>
Reply To:
Date:
Tue, 23 Feb 1999 09:04:49 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (76 lines)
John,

An empirical solution to the problem, is to accurately test, cross section and measure a coupon (trace
width and thickness and dielectric thickness). Use the results to "backcalculate" the effective dielectric
constant. Then use that constant in future calculations. Of course, if you change the material or
construction, you have to repeat the procedure. Also, send some coupons to your customer to correlate your
test measurements.

Dan Brandler.

John Bownass wrote:

> Colleagues,
>
> I wonder if you can throw any light onto a problem I have with a
> specific impedance controlled board where the prediction does not match
> the actual measured results.
>
> The board is 12 layer FR4 with controlled impedance on layers 3, 4, 6,
> 7, 9 and 10.  Layers 2, 5, 8 and 11 being the planes.  So there are two
> signal layers between each pair of planes, all impedances should be 48
> ohms (+/-10%).
>
> The problem is that the measured impedances from a finished board do not
> agree with the prediction based on the actual measurements from a cross
> section of the measured board.  I give 2 examples:
>
> All dielectric spacings are copper to copper, all figures in microns:
>
> Layer 3: track width avg 126, spacing to layer 2 = 135, to layer 5 =
> 260. Thickness = 16.
> Layer 6: track width avg 121, spacing to layer 5 = 100, to layer 8 =
> 270. Thickness = 16.
>
> L3 measured impedance = 57.3 avg, and calculated from above figures =
> 50.6
> L6 measured impedance = 54.3 avg, and calculated from above figures =
> 46.2
>
> Dielectric constant of 4.6 used.
>
> Of course the Er can be out but it could not explain such a large
> difference.
>
> The offset stripline model used works within normal tolerances for all
> our other work but for some reason not this particular job.  Any ideas?
>
> Has anyone a model that predicts nearer to the actual results?
>
> Regards
>
> John Bownass
>
> ################################################################
> TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
> ################################################################
> To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
> To subscribe:   SUBSCRIBE TechNet <your full name>
> To unsubscribe:   SIGNOFF TechNet
> ################################################################
> Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
> For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
> ################################################################

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################


ATOM RSS1 RSS2