Archive of the former Yahoo!Groups mailing list: Homebrew PCBs

previous by date index next by date
previous in topic topic list next in topic

Subject: Re: [Homebrew_PCBs] Re: plating holes with an exploding wire

From: Charles R Patton <charles.r.patton@...>
Date: 2013-01-03

My take is that is for both.
A bit of history that weighs on my opinion. Some 40 or 50 years ago,
one method of doing two layer boards was the use of brass rivets that
were then solder re-flowed. Sounds great and reliable --right? As
happens, this turned out to be an incredibly bad technique. The thermal
expansion/contraction of the board fractured the rivet/solder bond to
the traces and the boards would go intermittent. I was in the two-way
business then and remember resoldering all the connections on boards
done with this technique. I read of other similar stories. Furthermore
the technique really didn't work on more than two layers. The cure was
the move to plated through holes. Both problems were solved.

I have used the method previously mentioned on this list of inserting a
wire through the via and soldering to both sides. I feel this deviates
considerably from the brass rivet problem. The wire is copper and
consequently can stretch with thermal expansion, not breaking the solder
bond, but it adds severe constraints on the layout as vias really need
to be extra to the through-hole parts and mandatory for SMT parts.
Although it's generally considered bad practice to put the via in an SMT
pad, anyway.

So my take on this is the use of copper metallurgically bonded to the
foil traces by use of plating yields robust connections that can stand
the thermal cycling. The concept of exploding wire is appealing as the
idea of hot copper vapor moving at supersonic speeds will surely bond to
the foil edges it encounters (two or more) and thereby yield a reliable
"plated" through hole. But I've been a electronics designer long enough
to be skeptical of simple statements. Simple designs could have amazing
ways of failing. So one wire explodes OK, but will the next 1000 wires
do so? I know how to find 1 via that's poor in a sea of a 1000, but how
many people know the techniques? Bad vias will lead to incredible
frustrations. The method developed must be very reliable and repeatable.

Regards,
Charles R. Patton


On 1/2/2013 2:38 PM, Bob Macklin wrote:
>
> What is the purpose of this discussion? Is this for two layer boards
> or boards with more than two layers?
>
> Bob Macklin
> K5MYJ
> Seattle, Wa.
> "Real Radios Glow In The Dark"
>



[Non-text portions of this message have been removed]