Lead free solder joint rework

[IconicPCB]

A client has a problem with lead free solder joints.
The joints are cracking open. The assemblies are failing.
I have been asked to reflow the joints with the view to repairing the assemblies.Now that is not a problem. I can certainly reflow the assembly.

My question is .... will reflowing the joints provide a reasonable resolution to this problem or will they fail in a similar fashion in not too distant a future.

[Stray Electron]

  Lead free solder is absolute shit in my opinion.  A friend of mine has a job at KSC resoldering circuit boards for satellites with leaded solder after they had countless failures with lead free solder.  He's been at it for about four years now and there is no end in sight. I worked for a large aerospace company that built systems for the .MIL and we never stopped using leaded solder. That should tell you something.

[Kasper]

[...] I worked for a large aerospace company that built systems for the .MIL and we never stopped using leaded solder. That should tell you something.

All that says is they don't like change.  Changes require documentation and testing.  In aerospace for .mil, that's a huge hurdle.

I used to work in avionics and we used all sorts of old components and did lots of mods.  New compoments and new design would have been much better but it would have required years of work to re-certify it.

[Whales]

There are lots of variables that control solder joint strength and resilience to cracking. 

• Solder alloy (initial used)
• Solder alloy & gradients (final product, affected by working time, pad plating, tip material, etc)
• Forces being applied (large/heavy cantilevered parts or human-handled parts, vibration, multiple mounting surfaces, etc)
• Geometry of solder joint: gap between pad and leg
• Wetting of solder (unclean joints only partly attach)
• Heating & cooling of solder (cold joints, grain structure, etc)
• Probably more

If the client says they are seeing this particular problem often: I would not rely on just reflowing the solder, it might not be enough or it might be irrelevant and the problem will just re-occur again. If there are problems occurring reliably on the same parts across many boards (but not other parts!) then it smells like it might be a non-alloy related issue.  You need to get the boards in your hands and find out everything.

If you choose to try removing and replacing the solder: don't just base your decision on leaded vs lead free.  Go for a solder that you know works.  There are many more things to go wrong with solders.

[SMTech]

  Lead free solder is absolute shit in my opinion.  A friend of mine has a job at KSC resoldering circuit boards for satellites with leaded solder after they had countless failures with lead free solder.  He's been at it for about four years now and there is no end in sight. I worked for a large aerospace company that built systems for the .MIL and we never stopped using leaded solder. That should tell you something.

We've been using exclusively lead free ever since the legislation introduced it in 2006. We have never seen a failure we can attribute to the alloy. Our assemblies go into all sorts of industries and all sorts of environments, but probably never see severe temperature extremes which some alloys(low temp ones in particular) certainly don't like. We also do some prototyping/low volume work for people in the military space and not once have they specified lead-free be used for their assemblies.

The OP however has a bigger problem, identifying what joints are failing and why, physical or thermal stress would be the obvious ones and are not something lead is magically immune to.

[jmelson]

A client has a problem with lead free solder joints.
The joints are cracking open. The assemblies are failing.
I have been asked to reflow the joints with the view to repairing the assemblies.Now that is not a problem. I can certainly reflow the assembly.

My question is .... will reflowing the joints provide a reasonable resolution to this problem or will they fail in a similar fashion in not too distant a future.

What environment are these parts in?  Large temperature changes?  High vibration?  As in cars parked outdoors and driven on rough roads?
I've been making gear with lead-free solder for many years and have not had these sorts of issues.  BUT, I have to say my gear is in a much less demanding environment, as far as I know.  I do use SAC305 solder, which is less prone to this.  I tested some SAC100 solder (pure Tin) and thought it was quite awful and would never work.

Jon

[KaneTW]

I've only ever used lead-free. Never had issues, but none of it is in high-vibration applications.

[thm_w]

Where are the joints?

If they are on a component thats a different story than if they are on a connector that is constantly being inserted and flexed. Different solutions may apply.

[KaneTW]

That's true. I usually avoid solder connections in connectors where I can help it.

[abquke]

I discovered bismuth alloy solder and got rid of anything lead.

[IconicPCB]

Thank You all for your comments. They are much appreciated.
I do not know the provenance or age of the assembly.
From a few hints it is a JLCPCB board made and assembled by jlcpcb.

Inspection of the PCB assembly suggests a batch of PCBs going back a few years.
The faults are breaks in 0503 resistor joints. Some cracks are close to PCB pad others are half way along a45 degree plane between pcb and component intermetalic bonds.

On the question of bismuth not as good as tin lead alloy.
Bismuth typically added to create grainy appearance to the solder joint.
This grainy texture facilitates less strain under unaided visual inspection regime.

[Whales]

Are the 0503 cracks always in approximately the same parts/locations, or are they randomly distributed throughout the boards?

Localised might suggest physical stress (after soldering, eg flex during scored board snapping or final install).  Random might suggest alloys, thermal profiles & processes (during soldering).  Not 100% certain.

[onsenwombat]

Reflowing cracked joints just like that? Yea, depending on your luck, you might get some life to the gizmo for a while. Expected lifetime? Days, weeks, months, possibly years if you're lucky indeed. For anything important/critical/worth your money, such is just a bad bad way to "repair" stuff. The joints probably failed due to bad intermetallic compound formation in the first place, just melting the same poop again won't do miracles.
This post was not written by a chemist nor a metallurgist, so take it with a grain of salt.

[IconicPCB]

Update...

Boards reflown with liberal lashings of flux come to life.

Majority of problems in area dominated by a USB C connector.

Will see how the product survives in the months to come.

[Whales]

Hope things go well

Sounds like it could be localised flex related.  I wonder if it's possible to elasticise solder with additives.

[jmelson]

Update...

Boards reflown with liberal lashings of flux come to life.

Majority of problems in area dominated by a USB C connector.

Will see how the product survives in the months to come.

Hmmm, that points to mechanical stress on the USB connector.  Is there a way to glue the connector to the case so it can't flex the PC board?  That would be my first thing to try.
Jon

[Kasper]

I'd be plugging / unplugging that connector a bunch of times on one board to try to re-create the problem.