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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Solder joint issue

Views: 1540

Mark

#41589

Solder joint issue | 17 May, 2006

Hello,

I have recived from customer PCB where he found that big SMD electrolytic capacitor just fell off.

I checked this boards (FR4 , Au/Ni coating). I noticed that all solder remained on the termination of capacitors and looks ok but on the board pads are like someone just cut it - very flat surface. What can be a resone PCB coating?. Wetting ? Process is lead free by the way. Alloy SAC305.

Regards

Mark

reply »


RDR

#41591

Solder joint issue | 17 May, 2006

What color is the pad? coated with solder?, black?, still gold?

Russ

reply »

Mark

#41593

Solder joint issue | 17 May, 2006

Hi , It looks like there is solder (shiny like mirror surface). Defently not black and not gold.

Regards

Mark

reply »

Chunks

#41595

Solder joint issue | 17 May, 2006

I would suspect touch-up/repair, unless all your caps are falling off. I know our operators use two irons to solder both sides at the same time. If it's not flay and square, yo may get the defect you are talking about.

Another area is co-planarity. I have some SMD cpas like this and the leads are simply bent into a plastic carrier. You may want to double check for this as well. Again, I'm not sure if you have one or a bunch of failures.

reply »

#41613

Solder joint issue | 17 May, 2006

Chunks learn to spell first, then you can place blame on the operators. That's all yu do - blame operators. Get the facts first, even we know how to do that.

reply »


RDR

#41629

Solder joint issue | 18 May, 2006

Good post WML! you helped a lot!,

Chunks brings up good point. Are there vias that can be shorted by this component being slightly misaligned? When reworking there could be residual solder or even flux that has shorted the pads or component, etc...

Are you using immersion or plated gold?

Russ

reply »

#41636

Solder joint issue | 18 May, 2006

Perhaps a PhD should be able to recognize the difference between a typographical error and a spelling error. Even WE can see that one.

reply »

#41638

Solder joint issue | 18 May, 2006

Are you seeing added solder or is it possible you're looking at nickel?

reply »

Chunks

#41645

Solder joint issue | 18 May, 2006

Thanks Doctor - as always, you're right. I must've been typing with my mittens on. It's starting to get cold on this part of the planet these days. I wasn't blaming anyone, just trying to take a few questions marks away so "we're" left with the answer. Prescribe yourself a Chill Pill and call me in the morning.

One other thing to look at is if this part has a homeplate style aperture in the stencil. I have a few like this, and they cause some skewing. Just another thought to take a question mark away.

reply »

#41648

Solder joint issue | 18 May, 2006

I'm not on the production floor on a daily basis like the other guys here but here's my 2 cents: With a Ni/Au plating the inter-metallic that needs to be formed is Ni/Sn. Ni/Sn inter-metallic requires higher peak assembly temperature and a longer dwell at peak temperature compared to fex. hasl (where the IMC is already made). What I suspect happening is that due to a lack of heat to the "big" capacitor an IMC is made between the Ni and Au only, leaving the solder strength depending on just the Ni/Au interface.

I would check with a thermocouple and make sure you reach the peak temp as required for the type of paste you use

reply »

Chunks

#41666

Solder joint issue | 18 May, 2006

Hi Mark,

I feel like a broken record since I keep replying to this thread. BUT, I remembered a large lytic cap that we had a couple years ago, doing the same thing. The base of the cap would expand in reflow and the bottom would bulge out and only allow on leg to solder. Just another thing to look for (i.e. Good Doctor need not reply).

reply »

#41682

Solder joint issue | 18 May, 2006

Hypothesis #1: We agree with Steve. You are probably seeing the nickel underplate. The gold was absorbed into the solder in portions of a second, so that you don't see it is not surprising.

We're guessing that the gold was porous and allowed the nickel to corrode, making it difficult to solder. Prove this is correct by soldering the pad with a hand soldering iron and the paste that you use routinely.

When corroded nickel problems occur, they are usually more wide spread that an isolated pad; which is the criticism of this hypothesis.

Hypothesis #2: That the solder climbed the lead and departed the pad is curious. When this happens, it usually means that the lead was much warmer than the pad.

We're guessing the pad is heavily heat-sinked, preventing proper reflow of the solder. Prove this is correct by profing the board with thermocouples on this pad and other pads as a reference.

reply »

 Reflow System