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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Poor soldering on fine pitch?

Views: 1808

Shawn Vike

#42039

Poor soldering on fine pitch? | 8 June, 2006

We are a small company, reletivly new to doing our own SMT (1 year) and we are having an issue, I think.

We manufacture many board with no problems, but we have one that constently defeats us. This particular PCB has ENIG finish, and we have had the same problem with both tin/lead (no-clean, and SAC305 (no-clean and aqueous).

The problem is with QFP100 & TSSOP48 packages. They just don't seem to want to solder their leads down correctly. We have leads that seem to be soldered, but with a little poking come right off the pads. SOICs, 0805s, 1206s, crystals all seem fine.

Now I have a couple of questions.

1) How much presure should a lead be able to take in a "push-pull" type of test? I read on article on this forum that implies that this kind of testing is rediculous, but how else do you verify that you have good joints?

2) What can cause this kind of problem? We have had it on 2, no 3 revs of the PCB, from different suppliers. We have pre-baked any "open" components. the problem spans over most of the past year.

I have tried profiling the oven, but have had limited sucess. any suggestions??

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »


Rob

#42042

Poor soldering on fine pitch? | 9 June, 2006

Hi Shawn,

that you're baking the components makes me think that they are either old, or have been sitting around for a while (or none of the above & you are really dilligent!) Whilst baking will sort out the MSD issues (such as popcorning) it will not help you if the solderability coating has actually oxidised or wasn't put on there in the first place.

You could try a number of things:

1) Profile your oven & see if the temperature at the pins is ok. 2) Ask you solder supplier for a free sample of a paste with a more agressive flux. 3) Manually tin the leads on one side of one device then place & reflow as normal - see if that side solders. 4) Dip an IC in flux directly before placement. 5) If available try an alternative manufacturer for this part.

Good luck

reply »

#42046

Poor soldering on fine pitch? | 9 June, 2006

did you look at the pad vs the stencil design? are you putting down enough solder paste on the pads? what are your reduction ratios for these packages?

reply »

#42050

Poor soldering on fine pitch? | 9 June, 2006

The leads may be that Alloy 42. I would look to be sure your peak temp at reflow is over 220C, like 222C is nice. Also check that time-over-liquidous is meeting solderpaste mfrs. specification.

reply »

Chunks

#42057

Poor soldering on fine pitch? | 9 June, 2006

Rob and Pete are both right. "We" had the same problem, with just one manufacturer of the same style QFP (don't want to name names here). We had the same style of QFP from a different manufacture right next to the problematic one on about 3 product we build too. Same issue, easy to pop the leads with little or no effort. Paste was reflowing, but it almost seemed like the lead would not take the solder. This QFP was no-lead and we are using a leaded paste (yup we are loop-holed still).

We had a variety of issues. Co-planarity was one. Got that straightened out at placement. Next was our profile. It was OK, but we ramped it up to the higher limits of our paste. But we really noticed a problem at wave too. The wave was re-reflowing the QFP. This caused some co-planarity issues too since the board bows down whilst over wave.

The bottom line is we have a part that is plated like no other. We also could probably do better if we had a transitional paste too. But the biggest bang for our buck was at wave. We increased the conveyor speed a bit and the problems went away. I'd bet we're right near the edge, but after a dozen different people, prying a dozen different crow bars - no failures yet. Just something to think about.

reply »

#42064

Poor soldering on fine pitch? | 9 June, 2006

That is the problem I am having! Except with 3 separate manufacturers.

To answer some of the other questions...

Baking - diligence

Stencil - as far as we can tell it is good

Placement is good

We are using Pb-free parts and solder, but I have seen the same problem with Pb part & solder.

The only thing different about this board vs. most of our others is this is 4 layer (our standard is 2).

I don't have the option of waveing these, I am kind of stuck with manual rework at the moment. This is frustrating, I went to aqueous flux to try and help solve the problem, but it seems worse than no-clean..

Back to the drawing board on Monday!

Shawn

reply »

#42157

Poor soldering on fine pitch? | 13 June, 2006

So many processes outof control all at the same time. Do you gentlemen get paid to do this? Shawn, do you have any updates?

reply »

adlsmt

#42163

Poor soldering on fine pitch? | 13 June, 2006

If the only difference is the layer count, check the profile as stated above. If you dont have a profiler you can probably get the local Kic or Mole rep to do a demo on that board. I believe I was the one who stated in an old post that prying the leads is rediculous. If the joint is bad you should be able to tell by looking at the joint under magnification. If the fillet looks really good with good weting you should be ok. Prying on the leads would be totally subjective and is an unacceptable way to check the joint. How did you find the problem? Were the parts failing electrically? Or did someone start prying them and decide they were too weak?

reply »

#42164

Poor soldering on fine pitch? | 13 June, 2006

"I believe I was the one who stated in an old post that prying the leads is rediculous."

I have been down this road as well. The prez of my previous employer decided that because he could pry .012" leads off the pads with a screwdriver that we had a process run amok. I looked at the board and found gouges and an 0805 cap that had been crushed while being asked to perform as a fulcrum.

Turns out the failures were software related and it cost us several thin sections to get them to even look at their end of the @#$%&*% product.

I'd be looking at paste volume and registration, paste condition, reflow profile, pad layout, and lead and pad finish, pretty much in that order.

reply »


RDR

#42165

Poor soldering on fine pitch? | 13 June, 2006

Which processes are out of control here?

Do you ever have any input? YOU asking for updates? what the hell for? You don't even know what we are talking about!

Man, thing, lady, whatever, you are getting annoying! And yes I know I just encouraged your childish little game!

Russ

reply »

#42166

Poor soldering on fine pitch? | 13 June, 2006

"Which processes are out of control here?"

Forum moderation?

reply »

#42167

Poor soldering on fine pitch? | 13 June, 2006

What???? Look here buddy, What are you a Phd in? I have a few thoughts but I'de like to hear it from you. Just an observation, IF you posted solutions or other useful information you MIGHT get a better response. Jay

reply »

#42188

Poor soldering on fine pitch? | 14 June, 2006

How did you > find the problem? Were the parts failing > electrically? Or did someone start prying them > and decide they were too weak?

Originally found the problem going through some older assemblies that had failed testing. Poked at leads and found them not soldered (at least we could move the lead off the pad with not too much force). After soldering those components down found that the board worked (sometimes). So when I went to do another build last week, decided to look for this problem and would you believe it was still there. Couldn't seem to overcome it and was wondering if I am going insane (BTW, I am).

Grabbed a board that we had a CM manufacture in 2002 and found same issue. Really started questioning my sanity. Took an ethernet switch PCB with a popular brand name on it and found the same issue.

So now I am starting to agree with you that this is a poor test method, but is there a good method out there?

I mean with Pb solder it is easier to check for fillets and vissually inspect. Not quite so easy with SAC.

Pretty sure we are putting enough paste on the board, and I think I have a good profile, so we will see what happens with testing today/tomorrow.

Shawn

reply »


Rob

#42191

Poor soldering on fine pitch? | 14 June, 2006

Hi Shawn,

See if you can get hold of a free tub of Koki S3X58-M406, it claims to be good with Alloy 42.

http://www.ko-ki.co.jp/product/new_product.html#new03

Cheers,

Rob.

reply »

#42193

Poor soldering on fine pitch? | 14 June, 2006

Hi,

We have just started running this Koki paste and it's working really well. Funny, the joints are a lot more shiny than I expected.

Grant

reply »

Chunks

#42195

Poor soldering on fine pitch? | 14 June, 2006

We too have run different pastes and found similar results. I do not repsond to this post to help the Doctor out.

reply »

Cogiscan Track Trace Control

Hioki Flying Probe Tester