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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


problem in solderability

Views: 7066

we have solderability problem(some pins aren`t soldered) wit... - Aug 26, 2008 by omid  

... - Sep 04, 2008 by WaveMasterLarry  

#56184

problem in solderability | 26 August, 2008

we have solderability problem(some pins aren`t soldered) with two devices in our board with the below detail EP1K30QI208-2 brand:ALTERA with the package PQFP-208 TMS320VC5416PGE160 brand:TI with the package TQFP144 these two devices are pb free but the solder paste alloy is sn63 pb37 please tell me what is the best oven profile we can use ? also pay attention to this topic that other devices in the board doesn`t have any solderability problems.

reply »

#56198

problem in solderability | 27 August, 2008

Hi Omid,

Before tweaking the profile, it's a good idea to figure out why the devices were not soldered. It well could be that there is a problem with their finish, so changing the profile might not help. That is what our cutomers ask us to do if they face a similar problem.

Regards,

Vlad

www.sentec.ca

reply »

#56199

problem in solderability | 27 August, 2008

Hi,

also you can check the board design. If always the sme pins are not soldered, a via could be close to the pin and take part of the solder in.

Regards, Emil

reply »

#56212

problem in solderability | 27 August, 2008

nothing relates to ROHS components and tin lead process?

reply »

#56218

problem in solderability | 27 August, 2008

Soory, but I don't understand what you are saying/asking. Parts with Pb-free finish existed in the Pb-Sn world, so what is so special about your QFPs?

Vlad

www.sentec.ca

reply »

#56227

problem in solderability | 27 August, 2008

i mean that the parts that we have here is ROHS compliant(pb free)but the process that we have(the type of solder paste & the surface finish of the board has pb in its mixture) this is not the cause of problem ?

reply »

#56230

problem in solderability | 28 August, 2008

I've been there. Soldering no lead parts in a leaded world can be difficult. Sounds like you are not wetting to your parts becasue you ar enot getting hot enough. Here's the trick - you can turn up your profile to look like a no-lead profile, and the parts will probably solder, but he joints may be very weak. This is due to the fact that your leaded solder paste flux will burn off way too early. Many solde rpaste companies produce a paste that is designed just for this. The paste is leaded, has 2% silver for better wetting and has a higher temp flux that won't burn off.

Try increasing your profile and see if your parts solder. If so, I would look into the different solder paste as well. Like I said earlier, you solder joints will be weak.

reply »

#56234

problem in solderability | 28 August, 2008

Well, as I've said before, it's much better to figure out why they don't solder and then take a corrective action. As long as we don't know the root cause, it is difficult to find the cure :-)

One of our customers had a similar case ricently and as it turned out the component finish was at fault. To make it even more interesting, we had another case with a leaded component (a huge IC) and again, the root cause was the lead finish.

From my experience, it's always better to establish the root cause for a problem and get rid of it permanently, rather than fixing it most probably just temporally

reply »

#56241

problem in solderability | 28 August, 2008

> I've been there. Soldering no lead parts in a > leaded world can be difficult. Sounds like you > are not wetting to your parts becasue you ar enot > getting hot enough. Here's the trick - you can > turn up your profile to look like a no-lead > profile, and the parts will probably solder, but > he joints may be very weak. This is due to the > fact that your leaded solder paste flux will burn > off way too early. Many solde rpaste companies > produce a paste that is designed just for this. > The paste is leaded, has 2% silver for better > wetting and has a higher temp flux that won't > burn off. > > Try increasing your profile and see > if your parts solder. If so, I would look into > the different solder paste as well. Like I said > earlier, you solder joints will be weak.

i should say that we have another parts (SO packages)in our board that they are ROHS compliant too but they dont have any solderability problem

reply »

#56246

problem in solderability | 28 August, 2008

how can i be sure that the device isn`t affected by moisture ? i mean that if the pins are oxidized due to the moisture how can i remove this oxidization ?

reply »

#56248

problem in solderability | 28 August, 2008

That is why you need to do analysis of the failed components first!!!

Regards,

Vlad

www.sentec.ca

reply »

#56251

problem in solderability | 28 August, 2008

Vlad, Where might one do this analysis???

reply »

#56252

problem in solderability | 28 August, 2008

We do it on a regular base. You can check out the website below (we are still working on it, though).

Regards,

Vlad

www.sentec.ca

reply »

#56254

problem in solderability | 28 August, 2008

So just send a board to your company and that's root cause analysis? LOL... Ok sure. I guess I'm too much of an engineer and like to get to root cause by eliminating all the question marks and double checking my findings. Part of THAT kind engineering is TRYING things and then determining what to try next. Just simply sending boards to a company is not engineering. Most companies that pay us engineers to DO engineering are not too happy with us simpley sending out all our problems and waiting for the analysis to come back a couple weeks later. Is that why internet was invented? To waste time whilst we wait?

Then what do we do if the gurus at your place aren't right? Send more boards?

reply »

#56261

problem in solderability | 29 August, 2008

i read in smt soldering handbook that by placing the device in a 125`c within 48 hours the moisture will be removed what do you think about this method ?

reply »

#56262

problem in solderability | 29 August, 2008

Omid,

Try changing your profile. I had the exact same thing happen to me in 2006. Eventually I had to change my paste. No lead parts can and DO have a different finish that Pb parts. Why do you think the rest of the world increased their oven temps when going to no lead?

reply »

#56265

problem in solderability | 29 August, 2008

Omid It depends. Follow the guidelines of ANSI/J-STD-033, Standard for Handling, Packing, Shipping, and Use of Moisture/Reflow Sensitive Surface Mount Devices. It's available on-line.

reply »

#56269

problem in solderability | 29 August, 2008

Everyone has been playing a guess game so far :-) but we don't even know what the lead finish is:-) Even old (Sn-Pb finish) days, the ammount of Pb in the plating would be on a single digit level (just to mitigate Sn-whiskers growth), meaning that the melting temperature would be close to the one for pure Sn. So, a reflow profile for Sn finished leads wouldn't be that much different from a leaded one.

However, if the part isn't Sn finished, then it would be Pb-free by default. :-) That is why i keep saying: find the root cause and then you will know the cure :-)

Regards,

Vlad www.sentec.ca

reply »

#56273

problem in solderability | 29 August, 2008

Vlad; you should pay for advertising like everyone else on this site.

reply »

#56278

problem in solderability | 30 August, 2008

we are doing many experiment on this problem 1)sn63 pb37 solder paste +normal profile 2)sn62 pb36 ag2 solder paste +normal profile 3)pb free solder paste + pb free profile 4)sn63 pb37 solder paste + pb free profile but still the problem exist so i think we are dealing with oxidized pin . please give me a solution for removing oxidization from the pins .

reply »

#56280

problem in solderability | 31 August, 2008

All that they are telling is important (variants in the process), what you can do in a fast way is: change the component for a new one (date of manufacture / other supply - if you have); change the component (test - same package but other specification); if you don't have other component with the same package, try to assembly any other component in that pad.

Question: at reworking it's getting solder?

reply »

#56282

problem in solderability | 1 September, 2008

we use another device with the same package but it doesn`t have any solderability problem . the things that we are hesitate on is the difference between the pin alloy in these two different packages . also i should say that in rework station it will be soldered without any problem .

reply »

#56287

problem in solderability | 1 September, 2008

I told a friend this problem, and he also thinks the problem is the component, and suggested a different solder paste (more acid), maybe you can ask your supplier for a sample and orientation. Until there I think you have to rework if its only one component per board.

Good lucky!

reply »

#56288

problem in solderability | 1 September, 2008

I would do just that (and eventually will), but I've been just trying to give "free advices" so to say. What I've been suggesting is a typical way to solve a manufacturing problem.

Please let me knw if you think I'm confused.

Regards,

Vlad

reply »

#56295

problem in solderability | 2 September, 2008

That could lead to oxidation of the leads and worsen solderability. It's much better to make sure the components are handled properly and not exposed to moisture.

The reason for baking is moisture in the body not in the leads.

Why is no one suggesting that he use Siplace boards? *ducks and runs*

reply »

#56322

problem in solderability | 3 September, 2008

> That could lead to oxidation of the leads and > worsen solderability. It's much better to make > sure the components are handled properly and not > exposed to moisture. > > The reason for baking is > moisture in the body not in the leads. > > Why is > no one suggesting that he use Siplace > boards? *ducks and runs*

what do you mean by siplace board *ducks and runs* ?

reply »

#56323

problem in solderability | 3 September, 2008

I've seen where a new person comes on with some strange problem. Then someone else new comes on and says that the perfect solution is siplace boards. And for a while it seemed like Siplace boards were the solution to every problem. Or maybe they were basically advertising.

Vlad reminds me of them.

reply »

#56326

problem in solderability | 3 September, 2008

Wow, what a bender. too many Miller Lites! Anyway, Vlad is a guy working for Suntech or some company out there. I looked it up and they are some kind of Trace Labs wanna be. Not sure what duck and runs are, but I got the drinkin' runs.

reply »

#56330

problem in solderability | 3 September, 2008

Omid,

In some cases we have been able to restore acceptable solderability by dipping the leads (one side at a time) in an active flux and then re-tinning. Usually a good visual inspection after tinning would reveal whether this was going to work. This is tedious, but probably worth a try. Be sure to wash the parts thoroughly after the re-tinnning.

Glenn

reply »

#56331

problem in solderability | 3 September, 2008

No, I don't work for Suntech (me company name is different but close, it much have been misspelled:-)) and the company isn't exactly a trace lab, However, that wasn't the point why I �jumped in�. Defining a problem as �solderability� is too generic. It's been almost two weeks and around 30 posts, but the problem is still there. Isn't it? Some questions can be asked and successfully answered on the technical forums like that, but some can not be answered. I got an idea. I'm posting a picture showing a case of poor solderability. The board was ENIG, the process � Pb-free, the re-flow profile was a typical one for the process. Can someone tell me how to fix the problem (better yet what the root cause was)? I'm almost tempted to put a reward for the correct answer :-).

Regards,

Vlad

Attachments:

reply »

#56339

problem in solderability | 4 September, 2008

I happen to agree with Vlad and I'm not in the same business as him. Randomly turning knobs isn't the way to solve a problem. Do a proper analysis and let the parts tell you what's happening.

Seems like a lot of problems people post on this forum get replies to change the reflow profile without really knowing the root cause. That's not problem solving - that's shotgunning.

reply »

#56340

problem in solderability | 4 September, 2008

In the REAL world most people don't have time to send out problem boards and wait for someone else to do their work. In the real world most of these problems on this forum are caused by someone doing something dumb, or a change was made that the poster was unaware of. That's why most posts here are never ended. The original poster finds that someone has moved, tweaked, changed a setting and never report that fact back here.

Randomly changing several knobs at one and looking for results is stupid. But so is just stopping your production line and sending out problem boards every time you have a problem. Believe it or not, companies do not pay engineers to sit back and analyze problems. They are there to overcome problems and get product out without sacrificing quality in a timely fashion. The few times when we had sent stuff sent our, brought people in, they got it wrong. Once I told to solve a similar problem like this by �Wear smocks and finger cots�. Come to find out it was improper plating from the board house. I could go on and on and on�.

But my real question to you is; what do you call it when someone looks at a problem, makes a change, looks at the result, makes some other adjustments and fixes the problem?

reply »

#56343

problem in solderability | 4 September, 2008

You are right. That is what process engineers are for. HOWEVER, in many cases they just can not fix a problem because it either in the part, or board or paste and so on. I used ot work for a large company in QA doing lab work for production. In many cases, our QA engineers could handle various process related problems but problems like "poor solderability" or functinal faiilures after assembly - forgetr about it. We'd have to do appropriate analyses to find the root cause and once it's done we'd go after the supplier at fault (if it was a supplier related problem), as they are VERY relactant to admit it.

Regards,

Vlad

reply »

#56349

problem in solderability | 4 September, 2008

#56355

problem in solderability | 4 September, 2008

This is fine from a quality engineers point of veiw, but as you said the vendor never admits to nothing. So the company that has the problem has to resolve the problem for themselves. Generally this turns into a finger being pointed at the process and the Process Engineer has to try and resolve the problem (Yes turning knobs IS a part of this process). And most times they can over come the major part of the problem but the ripple effects such as solder balls, solder shorts etc... may occur until weeks go by and the study shows the vendor is at fault.

reply »

#56357

problem in solderability | 4 September, 2008

Well, it depends on what kind of evidence you'd have. I dealt with quite a few of them including a VERY big one (a chip maker) and all of them would try to prove you wrong, until they realize that the data you have speak for themselve. Saying data I mean results of, sometimes, very sophisticated analysis, which might involve SEM/EDS, FTIR and other "toys".

Regards,

Vlad

reply »

#56368

problem in solderability | 5 September, 2008

What was the prize for guessing the solderability problem Vlad

reply »

#56372

problem in solderability | 5 September, 2008

Well, I've never "set it up" (wasn't sure whether it would be appropriate or not:-)), but considering we are all rather far away form each other, I would be happy to send the best pizza from the area where the winner lives :-)

How does it sound? :-)

Vlad

reply »

#56378

problem in solderability | 7 September, 2008

OK Pizza it is then, here goes. Dewetting of Ni/Au pads, paste favouring the Tinned component lead hence excess paste wicking on top of the components. Reason is EITHER Nickel migration into the gold due to high levels of Ionics from poor rinse OR more likely residues of Solder Resist deposited back on the pads due to poor extraction during cure of the resist/misregistration of resist Cheers Greg

reply »

#56380

problem in solderability | 7 September, 2008

Hi Greg,

It's an interesting scenario, but not even close to the root cause. :-) Let's wait a day or two, may be someone else will come forward with a guess and then I will poste three more images with a VERY explicit clue. :-)

How does it sound?

Cheers,

Vlad

reply »

#56415

problem in solderability | 9 September, 2008

Hey Greg,

Do you still want to see the answer? :-)

Vlad

reply »

#56423

problem in solderability | 9 September, 2008

yes please

reply »

#56424

problem in solderability | 9 September, 2008

OK. Here it goes. It's a series images taken after cross-sectionning one of the components.

I hope that it's obvious now that teh "knob turning" technique wouldn't work that time around :-)

I have one more image in my "back pocket", which explains everything. Let me know if you want to see it right away or need some time :-)

Attachments:

reply »

#56426

problem in solderability | 10 September, 2008

Wheres the Gold layer on the non wetted part of the pad

reply »

#56430

problem in solderability | 10 September, 2008

It's never been there and it wasn't supposed to be there, as it is not the original surface.

Should I post the last image? :-)

Regards,

Vlad

reply »

#56432

problem in solderability | 10 September, 2008

If its Ni/Au and there is no wetting then it should be there you have everything else shown.

reply »

#56439

problem in solderability | 10 September, 2008

Well, it wasn't as easy to figure out the root cause. The last image is below. It shows how the board finish looks like (the trace was under solder mask). AS you can see, the layer of E-Ni is spliting (the board shop admitted the problem afterwards). So, once a solder pad with such a bad layer of E-Ni gets wetted with solder, top portion of teh layer (E-Ni) "floats in" the bulk of solder, causing "dewetted" appearance of the surface and weirdly looking (thinning) layer of Ni.

It's not the only case I have and all of them are different :-)

Do you have any other comments?

Regards,

Vlad

Attachments:

reply »

#56441

problem in solderability | 10 September, 2008

Yes why did the Electroless Nickel layer split this is the ROOT cause. My guess is still that the resist was slightly miss registered over the copper and therefore poor adhesion of the Nickel ensued OR like I said originally the resist had contaminated the pad. Did you complete any chemical extraction of the Circuit followed by IR analysis to determine if there was any Organics present. Saying the Nickel layer had split is not the ROOT cause only a problem caused by it. Sectioning the joint is not a fool proof answer. The floated part looks as though it may have some organic contamination in and Plasticizers from resist will not volatise under normal reflow conditions To be honest a simple swab test that can be done on the production line could determine if this was the case as I have been seeing for 16 years now and even more so with Lead Free This board in question was sectioned and then scanned as an etch depth and silicon was the main item found on the surface. Just a hunch Cheers Greg

Attachments:

reply »

#56442

problem in solderability | 10 September, 2008

Root cause analysis is simply looking for what caused a problem, then looking for what caused that, and so on, until you reach someone you don't like, or can pass the buck onto. I don't see why Vlad's method is better than tweaking your knobs until you can show they are not the problem then go after the board shop or switch to one that is reasonably honest with you. Vlad's analysis only makes it easier to go after board shops that I wouldn't want to be dealing with anyways.

reply »

#56443

problem in solderability | 10 September, 2008

VERY TRUE, well said Cheers Greg

reply »

#56444

problem in solderability | 10 September, 2008

No, we didn't do the test you mentioned, but did EDS on the joints and didn't see carbon their (which would indicate the presence of any organic-based contaminant).

For the customer the issue was closed and passed to the board shop. We did work with them and they COULD NOT reproduce the problem or they didn't tell us what they did wrong (the shop is a very good one, though).

From our original customer point of view, the probelm with E-Ni was the ROOT cause for their solderability problem and they couldn't have it figured out without the analysis we had done. It gave them enough ammunition to go after the board supplier with a smoking gun and they even got them pay for the analysis.

I'd also have to mention that the customer (a very good CM company) usually tries to exhaust all the options turning knobs, before asking us for help.

However, from the board shop point of view, that was the aftermath of what they had done and you may be right with your theory.

Do you still think that that kind of analysis is almost useless? :-)

Regards,

Vlad

reply »

#56457

problem in solderability | 10 September, 2008

I dont know about analysis on boards it is OK as long as the people carrying it out know how to interogate the answers.I worked on a problem just last week where an analysis house in China tested the resist then the pads and found on both surfaces Carbon and silicon and then blamed the customers solder was contaminated with Aluminum instead of solder resist this was also Ni/Au finish PCB's. So I think it has its place but turn the knobs and wipe the PCB first then think about 'expert advice' when that fails and you want to apportion blame in a direction. Cheers Greg

reply »

#56458

problem in solderability | 10 September, 2008

does that mean your buying me half a pizza then Vlad

reply »

#56461

problem in solderability | 10 September, 2008

Why just a half, I'd be glad to send a whole one. Just let me knwo what you want and the phone number of the one you want to get it from. :-) The discussion was well worth it :-)

On your China board "experience" I completely agree with you. It is also very true taht it's extremely difficult to find a lab you can trust the knowledgebase of, but once a mutial trsut has been proved, it really helps the both parties. :-)

So, please send me the "delivery and order details".

Cheers,

Vlad

reply »

BGA Rework Services

SMT Equipment Online Auctions