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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


BGA REFLOW PROBLEMS

Robert Hutton-Squire

#4572

BGA REFLOW PROBLEMS | 25 March, 2000

I am assembling a board with 2 x ALTERA FineLine PBGA devices. (256-pin and 100-pin) These devices have:

surface pad size = 0.41mm (16) ball size = 0.48mm (18.9) ball pitch = 1.0mm (40)

For my FIRST attempt I screen printed a fine pitch water soluble solder cream onto the board (admittedly with an etched stencil) which resulted in a rather irregular solder deposition. After reflow I had 3 inter-ball shorts - the balls also appear rather "fat".

For my SECOND attempt I applied a thin layer of water soluble flux cream, the same as used in the paste. The result, more shorts than before.

I have put the X-ray images online at www.elprom.co.za/tech1.htm

I am baffled as to why these shorts still occur - could it be the surface tension of the flux pulling the balls together? Could it be the surface pads are too small to accomodate the re-flowed solder ball?

Any help would be appreciated.

- Robert

reply »


JAX

#4573

Re: BGA REFLOW PROBLEMS | 25 March, 2000

Robert, Have you checked for any form of contamination that might cause the solder not to adhere to the balls? From the pictures you have on the net it appears that you have too much solder on the pads. What thickness is your stencil? If it is to specs have you measured the thickness of your board masking?

That's just my..............

reply »

Mark Alder

#4574

Re: BGA REFLOW PROBLEMS | 27 March, 2000

Dear Robert

The reason for the bridging problem you are experiencing on your BGA's is due because of too much solder. The PCB pads need to be at least 15% larger than the sphere size of the BGA balls. If you are not using C4 (63/37) BGA spheres and therefore have to print solderpaste, I would also strongly recommend using a laser cut stencil with appropriate aperture reductions.

Regards Mark

reply »

#4575

Re: BGA REFLOW PROBLEMS | 27 March, 2000

Robert: Howya doin' bud?

You say "BGA Shorting!!!" I say "Too Much Solder!!!" So, earlier comments about aperture design are appropriate, make good points, and I think we both agree that improvements could be made in your stencil design, but I'm not inclined to steer you in that direction, because your bridging occurs both with: � Solder paste (including flux). � Flux only.

OK. Let�s try it again: You say "BGA Shorting!!!" I say "Vias!!!"

Just drifting-off (sorry, Nancy), but if that view that you posted is the flux only, there seems to be a pant load of paste there (I know, I know, that�s what�s causing the shorting, Dave), but most connections seem fine. It�s that the shorted areas seem to be getting extra material from someplace. Where the blank is the excess paste on the shorted leads coming from? What�s different about those pads?

OK, I�m back, I wonder if we can narrow our focus to either BGA or board design by: 1 Reflowing the component, bug-down only with flux on a ceramic or oven-proof glass plate. 2 Similarly, reflowing the board only with flux only and balls. I reckon you wouldn't need to "ball" all the pads, just those that are bridging.

Continuing to be contrary, but taking a different angle(s): � Moisture entrapment in array devices causes solder shorts 80% of the time. Cracks in array devices from moisture entrapment occur the bottom side of components. Use an acoustic microscope to find cracks. � From the looks of things, I�d guess that you�re using Liquid Photo Imageable (LPI), but do you have "solder-dams" between the pads and vias?

Good luck Dave F

reply »

Ravi Bhat

#4576

Re: BGA REFLOW PROBLEMS | 2 June, 2000

Robert,

I think I am late, but better be late than never. Did you get solution to your problem. If not, Can you give me the stencil aperture dimensions and aperture type ? I think I can give you some insight into this.

reply »

ICT Total SMT line Provider

Reflow Oven