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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Soldermask Defined BGA Pads

#30975

Soldermask Defined BGA Pads | 14 October, 2004

We are experiencing a problem with soldermask defined BGA pads not accepting solder during the reflow process. The paste is there when the boards come off the printer but after reflow, no paste. Some other pads in the same pattern are fine. Here is the really weird part, we only saw it on 3 of 6 boards. The other 3 were fine, 100% coverage.

Opening sizes are .012 round.

Our process reflows the boards without components so we are expecting a meniscus of paste on the BGA lands but there is none, only the HASL finish.

The pads that wont accept solder, for the most part, are not well defined or as round as the pads that do accept solder. Underdeveloped? Contamination? Sabotage?????

Any feedback is appreciated. mk

reply »

#30979

Soldermask Defined BGA Pads | 15 October, 2004

Hi Matt, You've got an interesting problem. Sounds like something from an X-Files episode but it's still interesting. In your posting, you say that the paste is "there" after screen printing. Release isn't the greatest with a 0.012" round aperture, even with a .005" foil. You might want to try a square aperture for land geometries like this. They tend to release better when you get to lower area ratios. Are you confirming that the paste is there with a visual verification under power, naked eye, or are you doing a volumetric measurement? A quick test for you would be to see if your screen is plugged after printing. That will give you an idea of how well the paste is releasing. You may just be transferring very little solder paste to your substrate through some apertures and that's why your not seeing the same geometry on some pads vs others. As for the contamination concerns or sabotage, I would think sabotage would be more likely than a solderability problem on a HASL board. I've never had an issue with soldering to solder.

Regards, Justin

reply »

#30980

Soldermask Defined BGA Pads | 15 October, 2004

Hi Justin,

Thanks for the feedback. We are inspecting with the naked eye. Most of what we do are prototype quantities, in this case 6 pcs, so AOI or measurement equipment has never seemed justified.

The release looked uniform during the printing although we have no way to actually measure it. Even if the volume was not consistant, wouldn't some of the paste have fused and stuck to the pad during reflow?

We even went back and tried to appply the paste by hand and heat it with an iron and it balled up and rolled off the pad.

I dont get it??

mk'

reply »

#30981

Soldermask Defined BGA Pads | 15 October, 2004

Matt, are there any discoloration of the pads on the raw boards. i have seen what we refer to as gray pads. Ran into this in the past and it turned out to be a problem at the board house with one of their dipping tanks. The proper levels for the process were out of the prefered range.

MrMaint

reply »


RDR

#30983

Soldermask Defined BGA Pads | 15 October, 2004

sounds like you have a very thin layer of mask on some pads to me.

Russ

reply »

#30984

Soldermask Defined BGA Pads | 15 October, 2004

Thats what I was thinking especially because the pads are not round, they have jagged edges which to me indicates some kind of development issue.

mk

reply »

HOSS

#30985

Soldermask Defined BGA Pads | 15 October, 2004

Matt,

Had a similar problem with HASL that wasn't fluxed properly during fab. The appearance of the HASL looked dull and/or mottled (sp?). Solder was curled under on the edges where it should have been smoothly joined to the pad. Almost like a puddle.

The HASL plating was not my first, second or third consideration when troubleshooting. Never had it happen before. Lesson learned. If your board house is capable, a plating analysis may be done to determine plating thickness and composition.

Good luck.

reply »

C Lampron

#31024

Soldermask Defined BGA Pads | 20 October, 2004

Hi Matt,

Sounds like the HASL may be to thin on some pads and you're getting copper migration. If this is the case, and the solder paste is printing, I would asume you would see some solder balls under the BGA at X-Ray or incosistancies in ball size. Have you seen anything like this?

Chris

reply »

#31028

Soldermask Defined BGA Pads | 20 October, 2004

Not really. Our process is a solid solder deposition application so we are not placing components into the wet paste prior to reflow.

We print the paste then reflow is without components, trying to create a meniscus which is then washed, flattened, coated with adhesive flux, and covered with a release paper to protect the tacky surface until it is recieved at the end user. During the final assembly, components are then placed into the adhesive flux on the solid solder pads and reflowed to attach them without printing paste.

That is why it is so unusual for us to see missing solder because nothing happens to the wet paste prior to reflow. Normally, if the paste is there after print, the fused meniscus is there after reflow. That is why we suspect that there was contaminatoin or oxidation preventing the fused solderpaste from sticking.

The fact that it was only on 3 of 6 boards was a bit puzzling as well. The small aperture size, .012 round, is prompting us to consider going to a type 5 solderpaste when processing small openings in the future to give us the best shot at consistant deposition at the printer.

Thanks to all for their valuable feedback.

mkehoe@sipad.net

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

PCB Soldering Tools

Manufacturing Software