BGA soldering debate

If a BGA has lead-free balls, is it OK to solder it with leaded solder, using a temperature profile that melts the solder on the board, but doesn't melt the ball?

Seems OK to me; we're soldering the ball to the board. There seem to be references that say it's OK, and others that say the balls must melt.

The high-temp profile affects some other parts, and tends to bake flux ehough that it's harder to clean the board.

My production manager says that he's done it a lot, and hasn't had problems.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin
Loading thread data ...

Den tirsdag den 21. november 2017 kl. 00.15.34 UTC+1 skrev John Larkin:

formatting link
formatting link

Reply to
Lasse Langwadt Christensen

I went through this a few weeks back. the bottom line is that the lead free balls are probably coated in a tin-silver-copper (SAC) alloy, and a typical 60/40 profile will not be hot enough to make a reliable connection.

This was a customer who uses the largest Xilinx parts available, and who Xilinx bend backwards to support. The above advice was from Xilinx.

Reply to
JM

Am I reading Figure 4 correctly (SAC balls + leaded paste at a low temperature is the most reliable)?

Reply to
JM

The red graph is apparently tin-lead solder and lead-free balls, and has the lowest failure rate. A 220C, the balls probably didn't collapse. That contradicts most of what everyone else says.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

OK? No, it's definitely immoral. The priests of all that's green will curse you.

He's right. It's done a lot but it's not recommended. For military/aerospace applications, they have (or at least did have) the chips re-balled with leaded balls. In my last job, we used leaded solder on RoHS parts all the time. We had a lot of problem with the high-temperature profile of socialist solder (shorted transformer windings, popped AL caps, and gooey LEDs), so used leaded solder until they figured out the problem (needed a new oven).

Reply to
krw

Some of our aerospace customers tell us to use tin-lead solder. But the BGAs that we buy have lead-free balls, and we don't want to have them re-balled. So we use tin-lead solder paste on the BGAs. Then the question is, what temperature profile? We prefer the tin-lead profile, because it damages parts less, and is easier to clean. But it doesn't collapse the balls.

Probably doesn't matter.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin
[about tin-lead solder paste and SAC (Pb-free) solder BGA]

If it doesn't collapse the balls, the meniscus might be wrong, too. That implies that the mechanical strength differs from what your paste metering was intended to accomplish.

And, the 'collapse' refers to thinning of the solder plating, which (if it doesn't happen) means solder is in a thick layer in the support. That changes the THERMAL connection, solder is poorer conductivity than the (nickel/copper) ball inside.

Scary. Before Challenger exploded, the shuttles all seemed flightworthy by functional test...

Reply to
whit3rd

Bad news.The balls won't collapse properly. Correct procedure is to re ball. The only other sensible option is to advise the customer and get a disclaimer from them that they accept responsibility for any issues.Might be ok for a 'retail' product IPC class 1. MIL/ Aerospace is a no-no.

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
TTman

It' not really soldering the balls to the board as if they were copper balls. The molten 60/40 will become >60/

Reply to
Clive Arthur

What about using ordinary tin/lead solder on lead-free SMD components?

Reply to
Steve Wilson

They will cease to be lead-free and the unleashed karmic currents will make the Universe a better place. I do it all the time. :-)

Best regards, Piotr

Reply to
Piotr Wyderski

They must have messed up the colours. The graphs don't tie up with the document text.

Is this a high volume design? If not you could just solder on the BGA part with an IR tool after the rest of the board has been assembled.

Do you always use solder paste with BGA's? My preferred assembly house just uses flux, but they use vapour phase reflow machines which will control the temperature of the inner balls much better than convection ovens. AFAIK Motorola never intended paste to be used with BGA's.

Reply to
JM

They were invented by IBM--Controlled-Collapse Chip Connect (C4) and first used (iirc) in the Thermal Conduction Module of the 3081. (Might have been the 3033 but I don't think so.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
https://hobbs-eo.com
Reply to
Phil Hobbs

High for us. Maybe a few thousand BGAs per year. Some individual products could be close to a thousand.

If not you could just solder on the BGA

Sounds messy.

Yes. I think that is common. All the appnotes and such show paste being used.

Motorola is ancient history!

We have a big BTU International nitrogen-purged reflow oven.

formatting link

Can vapor-phase control temp profiles? The oven has 7 zones and lots of control.

We have actually had very little problems with BGAs, but someone recently questioned whether it was safe to use leaded paste and a low-temp profile on a lead-free BGA.

Here are some horrors, since corrected:

formatting link

formatting link

formatting link

I'll ask my folks to do some boards with the leaded paste and the low-temp profile, and take some more pics.

--

John Larkin   Highland Technology, Inc   trk 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

I think General Electric developed the flip chip (in 1963) before IBM started using it. Anyway, Mullen's 1993 patent seems to be considered as "the" BGA patent.

Reply to
JM

We do a lot of ROHS assemb ly, and it's ugly but seems to work OK. The high temp profile bakes the flux hard, so it's a little harder to clean.

All the components are ROHS now, but solder fine with either kind of solder.

--

John Larkin   Highland Technology, Inc   trk 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

That's what I want to do. How do you get around the ROHS requirements in Europe and I think California?

Reply to
Steve Wilson

Long before even the 3033. C4 was used long before TCMs. System 360s used C4 SLT modules in the early 60s.

Reply to
krw

You don't, that's why everyone uses RoHS techniques and generally don't bother with non-RoHS in the few locations where you might be able to use it.

--

Rick C 

Viewed the eclipse at Wintercrest Farms, 
on the centerline of totality since 1998
Reply to
rickman

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.