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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


BGA voids

Dennis _F

#7811

BGA voids | 17 January, 2000

We seem to have voids show up in one out of 5 BGAs during x-ray inspection. The x-ray is set up to accept the joint with anything less then 30% void area. I need anyones input of what during the process we can do to decrease the voiding and/or a good source of the accdeptance level for BGA voids. The products are all following the same process using no-clean paste. Also, it seems that the voiding is alwas at the top of the solder ball.

reply »

William

#7812

Re: BGA voids | 17 January, 2000

Voids can be acceptable @ 24% per 5 balls area, and 8% per single ball area. In fact, Proceeding Book 1996; Vol. I, Page #126 conclusions saids: that solder joint voiding at the maximun levels pbserved in this study (were voided area was up to 24 percent of pad area) caused no negative effect on PBGA board level relaiability. PBGA solder joints with voids had 16 percent better reliability than those without voids. Also N2 will reduce voids in BGA process as long as baking process takes place prior to do this process.

reply »

#7813

Re: BGA voids | 17 January, 2000

Dennis: On your BGAs:

* Never seen voids any place other than the top ... lower density stuff rises. * Our spec on voids says:

1 Voids 1a Voids in soldered balls shall be less than 20% of the diameter of the solder ball. 1b No void shall show on the exterior of a solder connection. 1c. Multiple voids may sum to less than 20% of the diameter of the solder ball.

* Beyond the issue of a spec for voids, I appreciate your point: voids in only 1 of 5 BGAs, same process, same paste etc ... what's going on?? Something is not the same. Something is changing your profile. The no. 1 suspect is the board design ... the way the vias sink heat to internal ground plains. * Check the archives. We've talked about voids in BGAs before.

Good luck

Dave F

reply »


RDR

#7814

Re: BGA voids | 18 January, 2000

I have found that reducing the amount of metal in the joint makes a difference, we have reduced the aperture sizes to provide a minimal amount of paste. just enough to hold the part during handling. You may also want to try and slow the initial ramp up during the profile and hold the soak a little longer. I have also seen voids in balls directly from the manufacturer!

Russ

reply »

park kyung sam

#7815

Re: BGA voids | 19 January, 2000

In my case, when i solderd the bga in normal reflow(convection hot air). There is a lot of boid. i had tried to reduce void long time. so i reduced just a little. It did't satisfied me but now i cannot looking for the void in soldered bga in the N2 reflow.i had got the HP x-ray machine. I had 1 n2 reflow and 3 normal reflow. I always put the bga b'd at n2 reflow. all my concern is cleared.I did do nothing except takeing profile. sorry. i know the money for n2 .

reply »

Mark Williams

#7816

Re: BGA voids | 9 February, 2000

Hi

We use 2 criteria... maximum void size is 10% of the solderable surface and a maximum area of voids of 10% of the maximum cross-sectional area of the ball.

Any views

Mark

reply »

Software for SMT

Reflow Oven