Recently we encountered defects on some bare boards from our supplier. The defect is a small dimple. What might be the cause of this defect? Board Talk
Board Talk is presented by Phil Zarrow and Jim Hall of ITM Consulting.
Process Troubleshooting, Failure Analysis, Process Audits, Process Set-up CEM Selection/Qualification, SMT Training/Seminars, Legal Disputes
Phil Zarrow
With over 35 years experience in PCB assembly, Phil is one of the leading experts in SMT process failure analysis.
He has vast experience in SMT equipment, materials and processes.
Jim Hall
A Lean Six-Sigma Master Blackbelt, Jim has a wealth of knowledge in soldering, thermal technology, equipment and process basics.
He is a pioneer in the science of reflow.
Transcript
Phil
And welcome to Board Talk with Phil Zarrow and Jim Hall, the Assembly Brothers pick and place coming to you from high atop Mount Rialto.
What is the question today Jim?
Jim
The question comes from J.J.
Recently we encountered defects on some bare boards we received from our supplier. The defect is a small dimple on some of the VGA pads.
These pads have blind vias. What might be the cause of this defect?
Phil
Blind vias are definitely a handicap.
Jim
Particularly via in pad. You have got to deal with them, before you try to solder them because if you just leave a blind via, you've got this little cylinder of gas, you put solder paste on top of it, the gas is going to do you no good.
It's going to cause voids, bubbles, some of it can get out and some of the solder can wick down the hole giving you an insufficient.
When you're dealing via pad, you have to fill them, so I'm assuming that these are being filled and capped in some way and that it's just not being done properly.
Think about a PCB Fab trying to fill these and cap them so that you have a nice, flat surface an un-dimpled surface to solder on. They've got the same problem with whatever they used to fill it, they've got to get all the air out as they put the filler in.
If you just try to tent it, you've now got this little expansion chamber of trapped gas in there that's going to expand or contract depending on temperature.
So that's not a very good plan either.
You've just got a tough thing designing boards with blind via -- via in a pad is bad enough, even when it is through, so that you have some hope of and confirmation that you filled it and you get all the gas out by filling from one side, the gas goes out the other end. But, blind via, to deal with it in pad is going to be problematic.
I think you need to have a long talk with your PCB supplier about can they really repeatably deal with these and maybe you have to change your design to go to through vias will allow them to be filled properly and reliably.
Phil
Whatever kind of flux is entrapment you may or may not have in your vias whether they can see you're not. Whatever you do...
Jim
Don't solder like my brother.
Phil
Please don't solder like my brother.
Comments
For blind vias in pads, we require that the dimple has to less than 80 microns. Check IPC 6012 and IPC 600 for requirements too.
Andrew J Levesque, Gentex Corporation
If you have a blind via that has a foil cap on the internal side of the hole, is it a defect if that foil has a void or is missing?
Judy Milin, Valley Labs, Inc.
There are electroplated copper baths available that address this problem specifically for laser in outerlayers of PWB's. The preferentially plate in the via, rather than on the surface of the part, and slowly fill the laser vias. There will always be a small imperfection, but plenty of planarity for BGA assembly.
There is also a novel concept used by some fabricators where instead of lazing from layer 1 to layer 2, the fabricator lazes the vias while the layers are still in layer form, and lazes from layer 2 to layer 1. The fabricator then laminates/assembles the final PWB with the laser vias buried within the PWB. The advantage is that the outerlayer pads are completely planar, with not surface imperfections at all, it doesn't require special plating process, and the prepreg fills the vias, so no air entrapment. The only disadvantage is that it doesn't work on foil lam constructions - only cap lams. So, it only works on 6, 10, 14, 18, etc. layer constructions.
If you are interested, I have a diagram illustrating the construction.
Bob Burns, Printed Circuits Inc.
I designed a twelve layer .062 thick FR4 board with blind vias from layers 1-6 and 7-12 in 0402 pads. The blind via drill diameter specified was 0.006 with a final board finish of ENIG (3-5mu AU/100mu NI). The final plating filled the blind vias.
