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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

What is the typical BGA substrate pad alloy

Dave Cunningham


What is the typical BGA substrate pad alloy | 2 December, 1999

Is there a typical alloy used by vendors on the substrate pad? What is their process used to attach a eutectic ball to that pad? Thanks.

reply »


Re: What is the typical BGA substrate pad alloy | 3 December, 1999

Dave: There's lots of stuff on the web describing BGAs. Examples are:

From top to bottom, a plastic BGA might have the following layers, if you sliced one in half (top to bottom) and looked at a side view:

* Plastic overmold * Semiconductor die & wire bonds & wires * Die attach epoxy * Gold plated die attach & wire bond pads & wires * BT resin/glass epoxy * Copper foil pads and interconnect * Solder mask * Solder ball

The balls are eutectic for PBBAs and 10/90 for most other BGAs. They are attached with 63/37 solder.

Good luck

Dave F

reply »

Dave C.


Re: What is the typical BGA substrate pad alloy | 7 December, 1999

Thanks for all the info Dave.

It's the interconnect between the copper foil pad and the eutectic solder ball that I'm most interested in. The BGA in question is very thin gold plated over nickel over the copper. When we are reworking the BGA, it looks as though the initial reflow by the manufacturer may be inadequate in that the balls come right off the pad without leaving evidence of having reflowed. However, I wonder if the rework process may be causing this. Do you think the rework process can actually pull all the solder back into the ball and make it spherical again?

Thanks again for all the good information via the links.

reply »


Re: What is the typical BGA substrate pad alloy | 8 December, 1999

Hi Dave I think you just provided the key to understanding the problem. By your description you are talking about a CBGA (ceramic). These typically have balls composed of 10/90 Sn/Pb which melts at a higher temperature than the 62/36/2 Sn/Pb/Ag (or 63/37 eutectic) you are used to working with. These hi temp balls are soldered to the Ni/Au pad of the CBGA with conventional eutectic solder. This is why the ball seems to come off without melting. PBGA's normally have a lo temp ball that melts and collapses during reflow. And, no, surface tension to the BGA pad would prevent truly molten solder from reforming a sphere. John Thorup

reply »


Re: What is the typical BGA substrate pad alloy | 8 December, 1999

Dave: You should see a solder smudge on the gold plating where the ball was attached, if there is (was) solder reflow during either the (1) BGA fabrication, (2) your BGA attach, and / or (3) your BGA removal.

You rework will not remove the solder smudge from the gold.

Every liquid solder ball would prefer to be round, because that shape minimizes the forces of suface tension. If the solder is touching something when it cools, the surface tension forces will not be symetrical and the ball will not be round. No, I would be amazed if your rework process will cause previously reflowed balls to be round again.


Dave F

reply »

PCB Manufacturing

Voidless Reflow Soldering