Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Aluminum Capacitor disassociated from sodler joint

Views: 3182

#81999

Aluminum Capacitor disassociated from sodler joint | 15 February, 2019

Hello,

I have a capacitor that doesn't form intermetallic joint with SAC 305 solder paste.The board is ENIG. Solder melts and forms good intermetallic joint with the PCB finish. However, capacitor can be removed with physical force - it looks like a cold solder joint. We did recheck profile and it looks that we have a 70seconds in TAL and 245C peak. What am I missing here?

Attachments:

reply »

#82000

Aluminum Capacitor disassociated from sodler joint | 15 February, 2019

Pd in the termination plating, maybe?

That stuff has reeked havoc on wetting for me in the past, but that was in leaded assemblies at leaded temperatures.

Maybe now with Pb free temps it works better?

reply »

#82028

Aluminum Capacitor disassociated from sodler joint | 19 February, 2019

Thank you Steve! I added couple of pictures from experimental board.

reply »

#82029

Aluminum Capacitor disassociated from sodler joint | 19 February, 2019

I'm with Steve. Check the datasheet, and make sure that the terminations are meant for solder application, and not conductive epoxy.

I had that happen a year or so ago with an MLCC. Drove us nuts for two days (not to mention the significant rework my floor tried to do) before we reviewed the datasheet. I'm not sure that electrolytics are made that way, but, worth the check.

For electrolytics like this, I've also had issues with thermal shadowing, and/or poorly thermally relieved pads. When you're running tests, try to get thermocouples on the pads to see what temperature they're getting to during reflow. It's possible that you're not getting enough heat at the solder joint to flow the component.

And, of course, the usual checks for solderability/contamination. Try to tin the leads on the bench with an iron and some SAC305 wire solder, and make sure that it's taking solder.

cheers, ..rob

reply »

#82032

Aluminum Capacitor disassociated from sodler joint | 19 February, 2019

Thanks Rob,

actually the thermal relieve is very poor. We did attach the thermo couples on the pad so that is checked. I will now check solderability.

reply »

#82035

Aluminum Capacitor disassociated from sodler joint | 19 February, 2019

Everything rob said makes perfect sense.

That said, do you have a robust, infallible paste management system? ;)

Have you tried multiple tubes and multiple date codes of paste and had the same results?

Is it stored properly (on the tip, assuming it's in tubes)?

I only ask because once in a while we get surprised when for some unknown reason, replenishing the paste on the stencil from a fresh tube solves the problem even if our paste management procedures have (allegedly) been followed.

reply »

#82038

Aluminum Capacitor disassociated from sodler joint | 19 February, 2019

Steve,

we have great system in place to trace the paste and it's compliance. This board is actually very busy and all is soldered great besides this capacitor. I will keep investigating.

reply »

#82060

Aluminum Capacitor disassociated from sodler joint | 24 February, 2019

Guys, I have been down this road also spent a lot of time working on this. Electrolytics often are difficult to solder like this. For the sake of testing, try a water soluble more active flux as an initial test. IF you compare hand solder wetting the lead between fresh and old parts you can see a difference in the wetting time a bad part takes. **** look at your profile and see if adding duration in the "soak/flux activation" range MIGHT help giving the flux longer to do its job. Another experiment would be to hand solder/tin the caps 1st and see if thay improves things. The joint you are seeing is called "foot in mud" result beautifully wetted to the pad, APPEARS nicely wetted to the part but given shock/force the joint breaks just as you have shown.

reply »

#82064

Aluminum Capacitor disassociated from sodler joint | 25 February, 2019

Solved it last week! Thank you all for your help. PRoy I wish you posted that earlier :) it would have saved us time and effort. We did extend the soak time(from 65s to 80seconds) in the reflow profile which resulted in a better wetting and stronger joint. Regards, Emil

reply »

Fluid Dispense Pump Integration

SMT spare parts - Qinyi Electronics