Help! BGA Socket / BGA Direct Placement

We have a tidy clause in our contract where as we are not responsible for functionality of prototypes. With that in mind we do what ever the customer is willing to pay for! Experiment away my boy!

Now for my two centavos on your question. Without some way of keeping the balls from sucking up into the plated through holes this won't work. Although I have not tried this I think that when it went liquidous that the weight of the BGA falling would cause some of the solder to blow out the back. Not to mention the fact that the hole volume to ball mass is probably 2 to 1. Plugging the holes so it doesn't blow out like Mt. St. Helens. Lets say you could accomplish filling the back of the holes with solder mask, or similar, and lets assume that it doesn't run up to the top and get onto your annular ring, (the only place left to make contact) now how do you calculate how much paste to apply because I think if you put your usual amount you will end up with too much due to the greatly reduced pad area and you will end up with shorts. Not to mention if you don�t get shorts you will have a very small contact area for soldering and reliability will suffer. Of course, that's just my opinion, I could be wrong.

reply »

Kevin, I have done the following process to a similar problem:

1. Plug holes (even to both sides of the board) 2. Clean any flux left on the surface, especially on bottom side. 3. Apply Kapton tape to the bottom side, you mite want to apply 3 layers. 4. Apply a small layer of paste flux to the top side. 5. Apply BGA and Reflow.

Note: Applying paste flux will only work if attaching a plastic BGA with SN63/37. If you are applying a Ceramic BGA with SN 90/10 than you have to apply solder paste and than apply the BGA.

Also step 5 might also require shims under BGA to keep BGA from collapsing more than it should.

Good luck

reply »