SMT Components Popping Off During Reflow

Phil
Welcome to Board Talk. Yes it's Jim Hall and Phil Zarrow, the Assembly Brothers here to attempt to tackle your SMT process problems. Jim, what do we have today?

Jim
A thing dear to our heart as we're old oven guys. The question relates to SMT components popping off during reflow.

One particular SMT component is popping off during the reflow process leaving indentations in the solder. Could the fans in our convection reflow oven be blowing the part of the PCB, or is it more likely due to a component moisture problem? This comes from D.L.

In this case we're assuming that if you're talking about moisture damage, you're probably talking about a plastic IC package. And the answer to your specific question is it could be either.

It could be a fan problem, or more likely a diffuser or nozzle plate problem. Most convection ovens are designed to give you pretty uniform gas flow and will not blow of components, particularly larger components. So the question, has somebody been tinkering with the oven internal setups?

Or have you gotten some flux or other contaminant material into the oven such that it's plugged up some of the holes and caused a high velocity of gas to come out of one hole because adjacent holes are plugged up.

Certainly we have seen moisture cause parts to blow of boards, to be lifted off their pads causing indentations. This is also likely. If that's true, you should be able to look at the part after it comes out of oven. You may have to open the oven after it's cooled off and find the component inside.

You should be able to see permanent damage, popcorning type cracking on the bottom side of QFP, SOIC or BGA components. You'll find that the bottom has bulged, usually with permanent cracks, and that's what's typically called popcorning because you usually can hear a pop inside the oven. You can also stand near your oven when the parts are going through and listen to hear the popping.

Phil
That would be a challenge to hear the popping above the sound of the fans and everything else. But if you hear popping, that means you have a moisture sensitivity control program that has gone astray or is nonexistent and that's another whole topic area.

Jim
If parts are blowing off I would suspect moisture sensitivity because it's one of the most universal problems that we see in our consulting business. We have been in only one facility in our many year of combined consulting experiences where we have observed a fool-proof MSD program.

Many people have good MSD programs and some of them are pretty good, but we've always been able to find holes in them where something can slip through the cracks.

With more parts being thinner, plastic, larger dies, and higher temperatures for lead-free - all of those have made the potential for moisture damage greater. So look for those popped off parts, see if they aren't bulging.

If not, go back to your oven, turn it on and try to locate physically where the condition is happening in the oven. Check the flow uniformity out of the diffuser and nozzle plate on the top side of the board.

Moisture and bad solder paste might also be the cause. But if the paste was generating the explosive force I would anticipate you wouldn't see indentations. You'd see solder balls.

Phil
And whatever you do in the meantime ...

Jim
Don't solder like my brother.

Phil
And don't solder like my brother and watch out for those flying parts.

We have seen assemblies with kapton tape masking that went through a wash cycle after primary exhibit this. If it's close to the component, the moisture under the tape will outgas at the easiest escape route, a wrinkle in the tape is a good place for failure. But, you should see some solder balling as well. All of the other suggestions are valid; vias with moisture, moisture in paste, moisture in package (probably the root cause if it's one per). I have not personally seen clogged vents cause this, but of course it's possible.

Here in KOMEX we have experienced phenomena like the mentioned in the paper. After several head aches we found that the main root cause is moisture trapped in the components. And components that are marked as MSL components.

A special case was originated by a tantalium capacitor. This component was pushing out a resistor away. Firstly we looked for something mechanical touching the component, deformation in the resistor itself, vias with moisture trapped.

Finally we baked some capacitors before placing and problem disappeared. This capacitor was exactly next to the resistor but, not marked as sensitive component.

What kind of part? What is the part termination finish? You say "one particular SMT component" but we need to know if it is the same part used in many locations, but only coming off in one location? Without that information we can speculate all day, but that does not get you anywhere.

Hopefully Hector's comments were helpfull. If not, the first thing that occured to my mind, was: "what about trapped air in plugged via's located below the components, that are popping off? Of course in that case we are dealing with a raw print malfunction.

When I started at our current employer we had similar issues and thought the same thing, our oven did not have the variable fan speed settings option. With the help of an oven profiler (thermocouples) that we later acquired we found via the profiler temperature graphs that we where ramping up to temperature too quickly and this rapid temperature increase created excessive out gassing and therefore a popcorn effect. Parts where being blown off during this rapid out gassing.

With the help of the oven profiler we where able to fine tune profiles to our solder, board density and thickness. The problem went away. I'd recommend you start with verifying that your oven parameters coincide with the recommended temperature graph as suggested by the manufacturer of solder paste and attempt to mimic those settings. Process of elimination start there. Hopefully handling issues associated with moisture control are taking place, if not, correct handling issues.