In a message dated 04/27/98 4:01:12 PM Pacific Daylight Time,
[log in to unmask] writes:

>
>  We are evaluating a fully gold plated (30uin) connector for a new product.
>  I have heard that soldering a gold plated connector to a PCB can cause a
>  brittle solder joint because of the intermetallic bonds made between the
>  gold and solder.  The assembly is through-hole and wave soldered.  Gold
>  plating is required of the contact portion of the connector because of
>  environment and signal levels.  I am trying to weigh the gold contact pro
>  against the gold solder joint con.
>
>  Can anyone verify this or point me to evidence that this is not a problem?
>
>  Thank You
>
>  Cory Steeby
>  [log in to unmask]
>

Hi Cory!

      Well, you dropped in at just about the right time...there's been a lot
of discussion going on about gold plating. Thirty micro-inches shouldn't hurt
anything, especially since it's going thru wave solder. While the board is
passing over the wave, a good portion of the gold will go into the solder pot
along with the solder that's washing over the bottom of the board.

      There was a good post a last week from gentleman named Gerry Gagnon that
went over a little history about soldering and gold plating thickness...I'll
paste it at the end of this email.


Have a great day!


-Steve Gregory-

*******************************************************************
Hello All

I have seen the various posts on this topic with amusement and my
opinion will tend to be different and slightly cynical. I too have a
very long history with electroplate Gold and Nickel/Gold systems.

Initially we used gold as an etch resist. The same Gold was used for
edge contact fingers. The finger spec was 100 microinches of gold over
copper and it worked very well. Gold was well under $50 an ounce.

Gold went up to $100 an ounce. Because of the cost (and because of gold
embrittlement of solder joints!) we converted to tin/lead etch resist.
We also converted to "vertical tab plating" for fingers using hard gold,
which was a brand new process at that time. Also at that time we added
Nickel underplate. The Nickel allowed a reduction in gold thickness
while maintaining the ability to pass our tests for porosity and
mate/unmate durability.  The parameters that met our specs at that time
were 70 microinches gold over 300 microinches nickel. The reduction in
gold and the added plating precision were a real cost saver. This system
also worked very well.

Gold prices skyrocketed and consistent supply was uncertain. I can
recall long term futures contracts at $400 and $800 an ounce. Security
guards supervised all gold adds. A weird time indeed. Also vertical tab
plating machines had a hard time meeting the 300 microinch min Nickel
req. unless the nickel cell was extremely long. So, the new specs became
70 microinches gold over 150 microinches nickel and these also worked
very well.

Next came the "commoditization" of  PC style memory (a.k.a. the SIMM).
Due to cost pressures and the redundant SIMM contact design, the
"acceptable" level of gold dropped to 30 microinches and we began seeing
even lower cost HASL coated fingers and solder mating connectors. How
ell these systems work depends on who you ask. I'll stay quiet on this
one.....

Today we see a variety of metallurgies that basically serve as gold
replacements or extenders.......improvements.

The moral of my story is: Technical issues aside (and all opinions I've
seen on this subject are basically TRUE), for PWB edge card contacts,
what has become "acceptable" over time has largely depended on the price
of gold, the and the level of reliability of the board-to-connector
contact "system",  relative to the level of reliability needed for the
end product.

Phew......enough history for today.

Regards

Gerry

################################################################
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 web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################