Kevin,
What type of solder joints are you interested in?
For surface mount applications, in cases where sufficient solder
joint strength can be provided from/along the edges of the SMD
land, there may not be a pull strength/reliability issue when
using an electroless/flash gold over nickel finish. In other
surface mount applications, such as soldering to a land/perimeter
grid array with large/round surface mount pads, our experience
would support the findings of K.Banerji & E. Bradley.
Regards,
Karl Sauter
Sun Microsystems, Inc.
[log in to unmask]
----- Begin Included Message -----
>From [log in to unmask] Fri Dec 8 09:59 PST 1995
Resent-Date: Fri, 8 Dec 1995 10:02:20 -0800
Old-Return-Path: <miso!radstone.co.uk!berry>
To: [log in to unmask]
Mime-Version: 1.0
From: Kevin Berry <[log in to unmask]>
Subject: Reliability of Ni/Au, PCB Finishes
Date: Fri, 08 Dec 95 13:11:20 PST
Content-Transfer-Encoding: 7bit
Resent-Message-Id: <"ciJFG1.0.6ZJ.tV6om"@ipc>
Resent-From: [log in to unmask]
X-Mailing-List: <[log in to unmask]> archive/latest/2016
X-Loop: [log in to unmask]
Resent-Sender: [log in to unmask]
I am looking for references,relevant to the reliability of
sn63/pb37 solder joints, when used in conjunction with an electroless
nickel / immersion gold finish, applied (over copper) to PCB's. Pull
strength of the solder joints, is also of interest.
Note: Electroless nickel thickness :- 4 to 8 um
immersion gold thickness :- 0.03 to 0.07um
My particular interest, is the solder joint reliability, when
subjected to thermal shock, thermal cycling, (-50c to 100c)
vibration & high humidity levels.
Most of the nickel/gold process suppliers are listed below :-
Lea Renol
Shipley
Elf Atotech (or Atochem)
Mac Dermid
Technique
We have information from most of these companies, but they
tend to concentrate on solderability. The only relevant article I
have managed to find, is by K.Banerji & E. Bradley of Motorola,
titled, "Manufacturability & Reliability of products assembled with
new PCB finishes".(SMI proceedings '94)
The article mentions crack propagation at the nickel / tin
intermetallic layer, when assemblies are subjected to mechanical &
thermal shock. The Authors appear to have concluded, that the nickel
/ tin intermetallic is weaker, or more brittle, than a copper / tin,
intermetallic. The phosphorous content of the nickel is also thought
to cause problems, as it is "displaced ahead of the growing
nickel/tin intermetallic & forms a layer of a complex compound, that
causes degredation of interface strength"!
Is there any informaion that can substantiate (or deny) this
report?
Regards
Kevin Berry
Radstone Technology Plc
Waterlane
Towcester
Northants
NN12 6JN
Tel: 01327 359444
Fax: 01327 358100
Email: [log in to unmask]
----- End Included Message -----
|