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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

BGA Rework - Pads being lifted

Views: 4470


BGA Rework - Pads being lifted | 3 April, 2009

I have a VJE Summit 750 Rework machine... I don't think that matters much, but I put it out there anyway.

I am removing a BGA (29mm x 29mm) off a board, but some of the pads on the board are being lifted during the process.

Does someone have any suggestions on what maybe causing this and what I can do to fix it?

reply »


BGA Rework - Pads being lifted | 4 April, 2009

Cause: Solder between the pad and the BGA on the pads being lifted has not melted. So, when you lift the BGA, you also lift those pads with the unmelted solder with the BGA. These pads are probably sinking the heat to a large copper plane in a inner layer of the board.

Solution: You may want to add more heat to the board prior to lifting the component. Do this through: * Longer preheat * Increased bottom side heating * Longer soak

Each component on each board may require a different thermal recipe to remove that part. The good news is: you know which pads require more heat and can monitor the temperature at those pads.

There's lots of good articles on determing thermal recipes for removing BGA. Here's one:⊂section=Display&ARTICLE_ID=112089&KEYWORD=Rework

reply »


BGA Rework - Pads being lifted | 6 April, 2009

I agree the longer soak will help provided you are not applying to much heat/time to the board now.

-are the pads that lift connected to anything? This has happened to me in the past where un-connected pads would come of far easier than those connected to trace or ground. This was actually when an operator used a ROHS profile instead of a leaded profile.

Are you actually profiling the component or just using a genaric profile? I dont recomend winging it, as Dave mention each part/board combo require their own settings

reply »


BGA Rework - Pads being lifted | 6 April, 2009

Here are some pictures of the part.

I created a profile for this part. I've been in touch with the vendor and he seems to think the profile is too hot. I am waiting for another board from R&D so I can work on it again.

If you look at the picture, you can see that the pads are still on the BGA.

When creating a profile for rework. Do you still keep the same reflow specifications as if I were doing it through an oven? Meaning, should I reflow for 50 seconds with a peek temperature of 245 deg C? So, should I just pull it off as soon as I hit Liquidious? Do I need to follow TAL when removing a part?

This is a lead-free part.


reply »


BGA Rework - Pads being lifted | 8 April, 2009

I would use almost the same profile you use when putting the component down on the board. Your pre-heat is essential in rework. Also even though your machine tells you your at a specific temperature I would add a thermocouple next to the component your removing attached to a seperate temp meter. This will ensure to you that the temperature of the area has reached the solder liquidus range, and then wait another ten seconds to go above the liquidus before removing.

reply »


BGA Rework - Pads being lifted | 9 April, 2009

Another suggestion for it. Baking the whole boards before rework the BGA. It is not only benifit to the BGA but also to the pad.

This message was posted via the Electronics Forum @

reply »


BGA Rework - Pads being lifted | 9 April, 2009

Well, I did manage to solve this problem. I lowered the max boards temp from 245 to 235. Then I decreased my TAL. This seemed to work perfectly.

Thanks everyone for your suggestions! They got me in the right direction.

reply »

Dual Lane Reflow Oven