Hello David,

This is a serious defect; the main issue in your case is the quality of the 
plated copper since it should not have failed even after more than 5 cycles. I 
do not know if interconnects at each layer would reduce the likelihood of 
cracking but would doubt it if the copper is that brittle. If it would, someone 
would have to explain why all those double sided pth boards don't need nfp's 
inside them to hold the plating together.

Ductility in plated copper is affected by the amount and type of additive used, 
and the time between carbon treatments. Normally, solutions should be treated 
every 6 - 12 months, depending on throughput, to remove all organic material and 
return them to a healthy blue colour. Organics come from the breakdown of the 
additives, and leaching of the dry film resist. I'm sure there are many experts 
out there who can add other factors which produce less ductile copper deposits.

It is better to design your pcb's with all pads present as this helps to ensure 
proper clearances between traces and holes. Allow the fabricator to remove 
nfp's if they believe it will improve reliability. This is easy and automatic on 
CAM software.

Paul Gould
EMail [log in to unmask]

In your message dated Thursday 26, September 1996 you wrote :
> I am currently debating the pro's & con's of suppressing  non-functional
> inner layer pads and have found our PCB fabricators have differing views
> on which is best. Here are the 2 views.
> 
> 1.  Suppressing inner layer pads will give a better drilled hole, and so
>     will produce a more consistant & cleaner plated hole.
> 
> 2.  NOT suppressing inner layer pads will produce a stronger & more 
>     reliable PTH less prone to failiure.
> 
> We first changed to using inner layer pads after a new design, which
> required 2 reflow soldering processes & 1 wave solder process, failed
> ICT afer assembly.  The cause was a process problem at our supplier which
> resulted in via's cracking in the barrel. Having this problem made us
> consider what we could do with the design to reduce the risk of similar
> problems happening again.
> 
> I would appreciate any views you may have on this topic.

***************************************************************************
* 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.        *
***************************************************************************