Hi Blue Smurf -
G. Franck Jr.'s answer cover's 99% of the basics (a wonderfully
complete and detailed answer too!). As he said, copper on the surface
is BAD. Both the Cu3Sn and Cu6Sn5 intermetallic oxide species require
very active fluxes to permit soldering. Test data has shown that the
intermetallics will oxidize within ten minutes of reexposure to the
atmosphere if they start out oxide free. With that in mind, any amount
of either Cu3Sn or Cu6Sn5 intermetallic is going to cause problems. A
secondary issue is if the intermetallic gets too thick it can be a
source for crack initiation leading to cracked solder joints. How do
you not get copper/tin intermetallics? - keep you solder finish thick
enough that the copper can not diffuse to the surface. I have seen
some data that suggests 60 microinches minimum. For more intermetallic
info take a look at figure 6-10, page 178, "The Mechanics of Solder
Alloy Wetting and Spreading", ISBN 0-442-01752-9 (the whole book is
pretty good).One additional chunk of data - the tin preferentially
oxides over the lead in a tin lead alloy thus making soldering even
tougher for surfaces that are old (SnO2 is the main culprit- I have
some SERA curves that show that point).
Dave Hillman
Rockwell Collins
[log in to unmask]
____________________________ Reply Separator _________________________________
Subject: Re: ASSY/FAB: SnPb Oxidation
Author: [log in to unmask] at ccmgw1
Date: 11/18/96 2:50 PM
I still have some questions about this response:
1. What if the Cu3Sn is too thick?
2. How thick is too thick? (2 microns??)
3. How can you get rid of it?
To my knowledge, the thickness of Intermetallic Compound (IMC) formation is
poportional to the square root of time. (Considering constant temperature)
Blue Smurf
----------
From: TechNet-request
To: John_Gulley; technet
Subject: Re: ASSY/FAB: SnPb Oxidation
Date: Sunday, November 17, 1996 2:12PM
John,
A quick response to oxidation/solderability
The Lead oxide is not a problem, the tin oxides are the ones we are
dealing with. Now, simple tin oxide is not a problem for most fluxes,
if the oxide is not TOO thick. But you probably do not have simple tin
oxide.
While the solder is sitting on the surface of the copper, copper atoms
will migrate/diffuse from the copper layer into the solder, and tin will
migrate the other way into the copper. The lead, being physically much
larger, just kinda gets into the way.
What happens when copper starts showing up on/near the surface? Glad
you asked. It oxidizes, just like the tin. HOWEVER, the tin and copper
start making interesting alloys. And it is the oxide of the alloy which
needs to be stripped. Without getting into metallurgy (which I cant
even spell correctly) and all the different alloys, just remember that
the higher the copper content at the surface, the harder it is to remove
the oxide.
OK John, on your boards, enough copper has made its way to the surface to
create a copper/tin alloy which you can not solder to. What happens if
you remove the top layer of oxide and metal? Answer: You expose a layer
of tin/copper/lead which is slightly higher in copper than the one on the
surface. In other words, a layer slightly harder to solder to.
So, to your question, how do I remove Oxide from solder? Well, you can
use Thiourea Hydrochloride, often found in Fab shop solder brightners,
(and I think Wal-mart sells it as Tarn-ex, next to Cascade in the PWB
section) but what have you accomplished? You will very quickly form
another even harder to solder to oxide coating.
Summary: Copper is a major contributor to solderability problems.
Copper is at its lowest concentration on the surface of the solder.
Therefore, the solder which is easiest to solder is on the surface of
the solder. If the surface solder is not solderable, none of the solder
is solderable. Then your only hope it to a) re-HASL or b) scrap and
reorder.
There is a relationship which goes something like this:
PWB solderability shelf life (Time)
=3D solder thickness (L) / Copper diffusion rate (L/Time)
(The diffusion rate in a function of Temperature, and alloy.) There are
other factors in the equation, but this is first order.
In my experience, this is the major problem of HASL boards that are 9+
months old.
There are other causes for solderability problems. May none of them be
as bad or prevalent as this one.
George Franck Jr
Raytheon E-Systems
Product Assurance Engineer
=46alls Church Va
***************************************************************************
* 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. *
***************************************************************************
***************************************************************************
* 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. *
***************************************************************************
***************************************************************************
* 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. *
***************************************************************************
|
