Copper dissolving in tin?

[nsv]

We have a proces where we dip solder a 0.1" lead wire made as a braid of 0.003" (approx. AWG 40) copper strands. Apparently the copper shrinks or disappears in the process.

Could it be that the very thin strands simply alloy with the tin to bronze and are dissolved in the tin bath even though the temperature 700°F is way below the melting point for copper?

NSV

 

[EdStainless]

Have you looked at a phase diagram? You can disolve a few percent Cu into tin at 227C (441F).
What is your solder composition?

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[nsv]

Ehrr - phase diagram - hrrm, afraid I'm more electrical than material so I can't answer that.
According to the manufacturers declaration for the automotive industry the solder is composed of 99.28% Sn, 0.7% Cu and 0.02% Si. There could be trace amounts of other materials in concentrations too small to be mandatory to declare.
In metric: The temperature is approx. 350°C, the braid is 0.25mm and the strands 0,01mm.

 

[redpicker]

Well, phase diagrams are easy to read and understand, if you have the training. Without the training, they can be very confusing and just plain confounding.

The solder you have mentioned is essentially pure tin. At 350 C, it will dissolve copper until the concentration in the bath is approximately 4%. So, the answer to your question is, yes, you can dissolve copper with the tin and in all likelyhood, that is what is happening.

What can you do about it? Limit the exposure time so that you still have some copper left.

rp

 

[EdStainless]

Sorry about the snide comment.
This is one that your solder supplier should be willing to help you address.
A lower temp would also help slow down the rate of attack.
There may also be acceptable solders that have other alloy elements in them that will slow the rate of Cu attack.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[kenvlach]

From the Cu-Sn phase diagram in ASM Handbook, vol. 3 Alloy Phase Diagrams.
The solubility of Cu in Sn:
0.7 wt% at 227 ^oC (eutectic point)
1.5 wt% at 250 ^oC
2.7 wt% at 300 ^oC
4.5 wt% at 350 ^oC
6.8 wt% at 400 ^oC
Two intermetallics may also form: [η] of about 60.7 wt% Sn and [ε] at 38 wt% Sn.

The solder is at the eutectic composition.
Your temperature is way too high.

 

[nsv]

Thnx for the help.

I had a typing error - the strands are 0.08mm and not 0.01mm, but still they are very fine and I suppose the finer the more readily they dissolve.

The temperature is indeed very high. The process solders a 0.12mm insulated aluminium wire to the braid and the high temperature strips off the insulation. Not a very good solution, but the wire is so fragile that mechanical stripping is impossible. At the moment we carry out some experiments with ultrasound to see if this can remove insulation without breaking the wire. If we succeed we can lower the temperature.

By the way, is it OK in these fora to ask question in metric units? I easily make mistakes when translating to inches and feet and pounds etc.

NSV

 

[kenvlach]

Perhaps pass the insulated wire over a torch to burn off the insulation & preheat the aluminum. Or, use a CO₂ laser to flash off the insulation; I think it won't hurt the aluminum due to reflectivity.

mm units are fine for small items like wire diameters, microns or micrometers ([μ]m) for things like plating thickness, and nm for nano. Metric units are preferred for chemistry, too, but perhaps most American learned cgs system in school, e.g., g/L, not Kg/m³, for concentrations. Also, mm units are good for measuring a value off a print or on-screen graph; it's how I got the Cu concentrations above (other than the eutectic value).

 

[CoryPad]

Of course it is acceptable (even preferred) to use SI.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[nsv]

It's a bit complicated to remove the enamel. A torch does not remove it before the aluminium melts, even though we use thermoplastic enamel and we would prefer to use thermosetting for better stability.

But that bit about the laser was interesting. Where do you buy lasers?

 

[kenvlach]

I'm not sure laser idea will work; remembered something of similar principle in a machining magazine:
Prior to chemical milling of aluminum, it is coated with a maskant (resin that cures to a peelable plastic film), the maskant is scribed per blueprint & then removed from areas to be chemically milled. Once, scribing was done manually with a knife, but it can be now done by a laser on an programmable movable carriage. Also, aluminum reflects ~90% of the beam from ordinary CO₂ lasers, which made some modification necessary for cutting aluminum.

of course, the wire geometry may cause some difficulty, maybe require 2 lasers, one on each side.

There are laser vendors everywhere. Just get a salesman to bring a small one to test on your line.

 

[nsv]

Well here's one more on the same subject.

This copper braid is 100 mm long and pretinned for about 10 mm in each end. When I put it into an oven at 220°C for 30 minutes it turns golden.
Again, being more electrical than material, I have to guess wildly that tin is migrating over the surface and alloying with the copper to form bronze - hence the color.
Is this possible, and if not can someone give an explanation to the color change?

The braid when installed at the end user is subject to acoustic frequency vibrations at 2-20kHz, so it must be very flexible. Will this conversion of the material affect the brittleness?


NSV

PS. I haven't found a laser vendor yet, but I still find it interesting