more BGA pix

Now all you need is that nice universal cutting laser to blow the little ball into atoms. ;)

Cheers

Phil Hobbs

Please John,

What is that exactly ? A short circuit between BGA balls being melted or what ?

Habib.

Larkin has only one ball ?>:-} ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

I'm looking for work... see my website.

You *are* one asshole.

Looks like a short that didn't melt into the main balls; maybe a temperature profile problem. I think they ran that one through the oven again and fixed it.

Since that shot, we got a 7-zone oven with nitrogen, and that process works a lot better than the old gear.

No Xrays?

Cheers

oh ! you did it at home ! impressive ! Here in France there are many very skilled EMS subcontractors doing that job and i'm very satisfied with them.

Best regards, H.

Jim,

This ng deserves more from you. Really. And you're capable of that. Don't be shy :-)

Best regards, H.

No, we did it with an older production line, simpler oven and no nitrogen. That was an especially dramatic photo. On both the old and the new lines, BGA assembly is very reliable, better than leaded parts.

I'm not a specialist on that area but why a nitrogen atmosphere is a plus for reliability when melting BGA balls ? I thought that the temperature profile along the oven was the main criterion.

No. I understand that xrays are seldom used in production inspection. We have sent some BGA boards out for xray, at a customer request, and frankly I found the images fuzzy and not very useful. The optical thing is great for process control, but it can't see balls deep in the array. It can look down the rows of balls, backlit, all the way to the other end, so it can spot things like that mini-ball, or crud, between balls.

When I was at IBM, the server folks in Poughkeepsie had a big hard-X-ray machine for C4 inspection. (C4 means "controlled-collapse chip connect", i.e. solder balls.)

Cheers

Phil Hobbs

Low oxygen reduces oxidation and makes things wet and flow better. It's especially advantageous with lead-free solder, where temperatures are higher and there's more oxidation of the solder and the flux. The profile is very important, too.

We use gold-plated boards, ENIG, which helps a lot too. It's planar, doesn't oxidize, solders beautifully. Looks cool too.

I hear they tend to fall off during a shake test.

Cheers

We haven't done any shake testing, but we've had zero field failures out of tens of millions of balls. One of our customers recently shook some of our VME modules, no problems. When hundreds of balls connect a chip to a board, there is lots of ability to withstand force. The parts that fail vibration tend to be tall stuff with few leads, things with bendy resonances. You can anticipate failure of those parts, the ones that are visual blurs on the shake table.

Small BGAs are great. Big ones need underfill, or they rip out the pad metal on the corners under thermal cycling. That's one of the two main reasons chips haven't got any bigger on the last 15 years. (Yield is the other.)

Cheers

Phil Hobbs

The BGAs that we use are really small PC boards.

I imagine that the TCEs match the PCB pretty well. We don't underfill them, but we don't do hard thermal cycling either.

We do some pretty serious shake testing. Never had a problem with BGAs coming off, either in testing or in the field.