Dave has been licking the beryllium again.
Doug Pauls
From: "David D. Hillman" <[log in to unmask]>
To: <[log in to unmask]>
Date: 05/09/2013 11:39 AM
Subject: Re: [TN] Gold Removal
Sent by: TechNet <[log in to unmask]>
Hi Dean - yes, the 1% concern is when you have a "segregated zone" where
the gold is not uniformly distributed within the solder joint. The JSTD
001 Handbook goes into fairly good depth on soldering process constraints
that deal with the gold distribution. Using your 1% rule would keep you
out of trouble without question.
Good discussion - better than that "goops and glues" stuff Doug was asking
about. Metals rule (Ok, please don't shoot me for that, its directed to
Doug and Dewey).
Dave
From: "Stadem, Richard D." <[log in to unmask]>
To: "[log in to unmask]" <[log in to unmask]>,
"TechNet E-Mail Forum" <[log in to unmask]>
Date: 05/09/2013 09:56 AM
Subject: RE: [TN] Gold Removal
Thank you, Dave. That is greatly appreciated! And now:
The 3-5% gold content is an industry-accepted ballpark figure. Gold
embrittlement can also be seen with gold percentage as little as 1%. Much
also depends on the base metal that the SJ is formed with, and the
geometrical shape of the solder joint can also contribute heavily to the
embrittlement level. For example, gold plated wire cups are notorious for
embrittlement if not pre-tinned properly. So are gold plated terminals,
certain types of IC and hybrid leads, and a few other configurations. For
very high reliability applications, it is very very important to monitor
the gold (and other impurity levels).
From my personal experience, a limit of 1% gold works 99.99999999999% of
the time. Several times in my career I have seen termination finishes that
met the 3-5% limit fail in the field later. Believe me, you do not want
that to happen to you!
And when working with gold-plated parts, especially SMT connectors, it is
always a good idea to perform XrF periodically at receiving inspection to
make sure that the gold thickness is actually remaining within the
component SCD limits.
From: [log in to unmask] [mailto:[log in to unmask]]
Sent: Thursday, May 09, 2013 9:13 AM
To: TechNet E-Mail Forum; Stadem, Richard D.
Subject: Re: [TN] Gold Removal
Hi Dean - you may not be a metallurgist but you must have stayed at a
Holiday Inn recently. Very very good process details. Let me change your
last paragraph just a bit.
Gold and tin will form a intermetallic compound (IMC) - AuSn4 - when the
solder joint composition contains 3-5 weight % gold. This IMC is very
brittle and will crack during vibration, drop shock or thermal cycling
causing the solder joint to degrade and fail (not a porosity issue). The
good thing is that gold diffuses into solder at a rate of 100 microinches
per second so when the procedures Dean listed are followed, there is very
little chance of having a gold embrittlement issue. The process goal is to
not allow the gold content in the solder to exceed the 3-5% range. The
IPC JSTD 001 Handbook and the IPC AJ 820A Handbook have a very good
section on this issue.
Dave
From: "Stadem, Richard D." <[log in to unmask]>
To: <[log in to unmask]>
Date: 05/09/2013 08:30 AM
Subject: Re: [TN] Gold Removal
Sent by: TechNet <[log in to unmask]>
Good question, Bob.
Gold does not "melt" at soldering temperatures. A dissolution process
takes place, and because gold plating on component leads is so thin, the
dissolution is typically enough to absorb all of the gold into the solder
bath upon contact with the molten solder. This is also how a solder joint
is formed with copper and certain other metals. Copper, gold, silver, and
nickel do not melt at normal solder temperatures, a small amount is
dissolved to form the bond with the solder, called the intermetallic
formation.
A double tinning method utilizes the first tin pot to dissolve and absorb
the gold into the molten solder, and the second pot is required to ensure
the remaining solder is relatively free of gold (the first pot has gold
levels increasing over time as the tinning is performed, so the second pot
is used to make sure no gold remains). The dual pot method is used where
only a small amount of solder is used for tinning, and is thus more easily
filled with unwanted "impurities" (other alloys and elements other than
Sn63Pb37 or whatever the desired alloy is). The dynamic wave is simply
another option for obtaining the same results, where a larger volume of
solder is pumped to provide a laminar flow that will ensure the gold is
removed and replaced with the target alloy, Sn63Pb37 as an example.
Because of the larger solder volume in the dynamic wave method, the gold
content is diluted such that it does not go above the limits listed in
J-STD-006.
When using a double pot method, the gold content must be monitored in both
pots. This is done with a regular pot analysis, and the data is used to
determine the frequency of solder replenishment to dilute the gold and
other alloys to an acceptable level. The solder test is relatively
inexpensive and can be performed by solder companies such as Alpha,
Kester, etc. With the laminar flow method, you still need to test at
regular intervals, but because of the larger volume of solder it takes
longer for the gold and other impurities to go above the specified limits.
Tinning data should be kept to determine the frequency of sampling for
test, as well as how often the solder should be refreshed with new solder
to dilute the impurities to an acceptable level. A simple tinning log is
valuable in determining the history (number of components tinned, what
type, and when) to quantify the amount of tinning over time in order to
help determine the safe frequency for adding solder that will ensure the
alloy will remain within spec, the spec being J-STD-006, including
Appendix A and B, which list the allowable levels of "impurities" (it's
hard to consider how gold and silver can be considered "impurities", but
it's all relevant to the goal of Sn63Pb37 for leaded solder, and nothing
else).
J-STD-006 also lists the test methods used to determine the impurities.
The concern with having gold levels above the limits in the J-Standard is
that gold in the solder joint will nucleate through the hardened solder
joint towards the junction of the component lead and the solder, and/or
the pad and the solder (the intermetallic junction). Gold wants to return
to this intermetallic boundary, like many other elements seeking
equilibrium. So the gold molecules will actually travel through the solder
joint over time (nucleation), leaving behind a more porous solder joint,
which is called embrittlement, which leads to solder joint fractures and
potentially catastrophic electrical failure.
I am not a metallurgist by training, so this explanation may be somewhat
crude to those who are, but I think it will serve the purpose of
explanation. For all of you metallurgists out there, feel free to expand
on this or correct me.
Hope this helps you,
R. Dean Stadem
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Robert DeQuattro
Sent: Thursday, May 09, 2013 7:28 AM
To: [log in to unmask]
Subject: [TN] Gold Removal
Hello Fellow Technetter's,
I have a question on J-Std-001E 4.5.1 Gold removal which states a double
tinning process or dynamic solder wave may be used. Can any of you
describe this process. How is the gold actually removed? Or is it just
being tinned? Gold melts at over 1000 C so I am a little confused.
Thanks,
Bob
______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
|
