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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Oop's Pb BGA's in a RoHS process?

Views: 4806

Hi, We ran some boards and the girl on the line put in so... - Jun 16, 2006 by Grant  

Under fill them. ... - Jun 20, 2006 by

#42248

Oop's Pb BGA's in a RoHS process? | 16 June, 2006

Hi,

We ran some boards and the girl on the line put in some Pb BGA's instead of using RoHS parts. This should not have happened as we have almost no Pb parts left, and we spray painted Pb parts fluorescent pink so it was obvious.

The boards seem to work, and passed testing.

Does anyone think there is going to be any issues with this as far as reliability. I know the boards are not RoHS, but I am worried about other issues that might arise if the boards were sold as non RoHS. The issue is one BGA that's Pb, and it was soldered with lead free paste, on a board that's otherwise all lead free. We have a small bunch of units like this that would be a shame to waste.

Are these going to be reliable? Does anyone have any info on that?

As far as I had heard it was a big no-no, but the units work, so I would like to sell them if they would be ok for sale in non RoHS regions.

Regards,

Grant

reply »

#42250

Oop's Pb BGA's in a RoHS process? | 16 June, 2006

Hi Grant, I can't speak for reliability, I don't know how your boards are going to be used or what alloys you are using. As for RoHS compliancy as far as I understand the cut off if July 1. So any boards sold before that in the EU are OK. Best regards, Jay Brower

reply »

GS

#42252

Oop's Pb BGA's in a RoHS process? | 16 June, 2006

Hi Grant,

just my opinion, I think the concern could be related to the extra temperature the BGAs have been exposed during the Lead Free Reflow Temperature. So you need to check ( data sheet) what is the max temperature and how long the Pb/BGA can withstand under such temperature without get internal delaminated. Not so easy predict the BGA reliability soldered in a wrong way like that.

You should perform few SAM (Scanning Acoustic Microscope) on samples of suspected BGAs in order to check if there are symptoms of delaminations.

Regards....GS

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

reply »

#42253

Oop's Pb BGA's in a RoHS process? | 16 June, 2006

Hi,

Reliability was what I was worried about.

We have not shipped these yet, as we are not sure about reliability. We can ship them in regions outside of the EU anytime so we are ok on that, and because of supply chain delays, we are already shipping fully RoHS into Europe now, and mostly worldwide as well.

But we could ship these into other locations if there is no reliability issues with soldering a Pb BGA onto a board using a Non Pb paste. Does anyone know about reliability in this situation?

They seem to work ok!

Grant

reply »

#42254

Oop's Pb BGA's in a RoHS process? | 16 June, 2006

Hi Grant, I understand your concerns about the reliability. It depends what the boards are use for. For ex. if it is in a telecom system market where lifetime and reliability is a big issue 10-20 years lifetime (our customer reqs. that at the system level and this is very hard to fulfill with RoHS), then there could be a big issue on this matter.

On the other hand if this is for the consumer market (short product life time), like Pc:s, microwaves, cell phones or x-rated tools, etc. where lifetime is not that critical, then I don't think there should be any worries for you. We have together with our customer made some tests in a lab on this scenario, and the answer is not conclusive. It will work with some components and sometime the same component from different vendors will not. Our company and customers will just go on the safe side.

Questions: 1: Are these pcba:s expensive? 2. Do you need to take back the faulty pcba:s for repair or will you just discard them and send out new ones? 3: If some of these boards fails in the field, will that seriously damage your company�s reputation?

/Sincerly

reply »

Loco

#42259

Oop's Pb BGA's in a RoHS process? | 19 June, 2006

There are plenty of studies on this subject. If the BGA can handle the extra temperature and they are soldered on a leadfree profile, the general result seems to be a better reliability than Pb :)

Just out of the top of my head, there was a link to a motorola BGA study on this site, there's another one called 'NPL REPORT DEPC MPR 030 Measuring the Reliability of Electronics Assemblies During the Transition Period to Lead-Free Soldering' which i quiete like and should also be found somewhere on the web.

reply »

WEEEsted

#42261

Oop's Pb BGA's in a RoHS process? | 19 June, 2006

Hi, In our setup what we usually avoid is to use RoHS BGA on Pb process. But we use sometimes Pb transistor(4~6pins) on RoHS process, no BGA yet have been reported.

Its really difficult to say about reability unless you start running some functionality test on the boards.

You can also try to check the thermal profile of the BGA if still marginal.

Regards, WEEEsted

reply »

#42268

Oop's Pb BGA's in a RoHS process? | 19 June, 2006

whats an x-rated tool?? sounds like something you can buy in the amsterdam redlight district.

reply »

#42287

Oop's Pb BGA's in a RoHS process? | 19 June, 2006

Loco, I will look for that Motorola study. Personally, we had an issue and hired a lab (Cookson or UIC) a few years back. The results for our large board (14x18) with 16 large BGA (Xilinx >300 balls) was very negative. The lab gave us an estimate of 24~48 hours of run time before failure. I have had a number of lab experiences but never with as definate a report.

reply »

#42301

Oop's Pb BGA's in a RoHS process? | 20 June, 2006

Again, the verdict is mixed here as well, but from the seminars, white papers, etc. the consensus seems to be that Pb BGA's are not compatible in a Pb-Free RoHS Process...but... Pb-Free SAC BGA's are compatible in a Sn-Pb process as long as your profile is hot enough. Make sense?

From a materials science standpoint, here is a good paper from AIM regarding this scenario: http://www.aimsolder.com/techarticles/Lead%20Contamination%20in%20Lead-Free%20Electronics%20Assembly.pdf

To be on the safe-side and to eliminate the "reliability" question mark, I would block out stencil apertures on the Pb BGA device, and manually apply tacky flux before pick-and place. The hotter Pb-Free temps. should be fine to collapse your Sn-Pb ball. The main concern, as the paper from AIM states, is intermixing SAC alloy with a Sn-Pb joint. For a BGA, the bulk of your solder joint is going to be Sn-Pb.

reply »

Chunks

#42304

Oop's Pb BGA's in a RoHS process? | 20 June, 2006

Under fill them.

reply »

#42305

Oop's Pb BGA's in a RoHS process? | 20 June, 2006

This has been debated several times on this forum, but I have to say the opposite of Samir. I would rather see a SnPb BGA in a Pbfree process(as long as the component can take the heat) than a Pbfree BGA in a SnPb process. There has been some mention of extra voiding with a SnPb ball, and SAC paste, but at least you know that that SnPb ball has fully liquified, and should mix relatively evenly with the SAC paste. With a SAC ball, and SnPb paste, the ball liquifies at a much higher heat, and if you don't get there, chances are you will get layers of different intermetallics throughout the ball, with some very lead-rich zones, that will destroy your longterm reliability. Yeah, it works now, but for how long?

reply »

#42306

Oop's Pb BGA's in a RoHS process? | 20 June, 2006

Hi,

That's a really interesting article, however one thing I am a little confused on. They seem to say a Pb-Pb solder joint is almost as bad as a Pb-lead free joint in their results. Are they saying the testing which looks like it cycles up to 125 Dec C with an all lead process is also not reliable?

Apart from that it's a good article, but just confusing on the test results.

Grant

reply »

#42316

Oop's Pb BGA's in a RoHS process? | 20 June, 2006

Hi,

What are your thoughts about the article saying that the lead will leach out of the solder joint to the last place to cool, creating a weak point?

I have two worries about the report.

First, is that we have always used tin lead solder, and lead free solder is mostly tin, so adding some lead should just make it a kind of wacky lead based solder? Why does lead not come out of a tin-lead joint like it says in the report?

Secondly, the report takes the joints up to 125 Dec C, and our product has a bit heat sink, and good air flow, so we don't expect it to get above 50 Dec C often, but never over 70 Dec C in normal operation. So we won't be getting the temp cycling that they seemed to use in their tests in the report.

So I am not sure what to think of the report. Does anyone have any thoughts on it?

Regards,

Grant

reply »

#42320

Oop's Pb BGA's in a RoHS process? | 20 June, 2006

We all know those academic 'reliability studies' are very focused and only give a snapshot of real life. The real reliability studies of leadfree are starting as we scramble to meet our numbers.

We have no idea about your customers' expectations about reliability. If you absolutely do not want to comprimise, * Remove the BGA * Reball the BGA with the proper ball type * Dress the BGA pads on the board with the proper solder * Replace the BGA

Light the underware of the person that gave 'the girl on the line' these BGA on fire [if they should have known better]. Otherwise light his/her boss' shorts up.

reply »

Loco

#42322

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

Finally found a link to the NPL report, good reading.

http://publications.npl.co.uk/npl_web/pdf/depc_mpr30.pdf

reply »

#42328

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

Hi Muse, what do you mean by "with a SAC Ball and SnPb Paste the ball liquifies at a much higher heat"? Actually depending upon the ratio of solder contributed by the pad (SnPb) to the ratio of the SAC solder contributed by the BGA Ball, the SAC solder starts to liquefy anywhere from 210 deg C onwards. also, more the Ag content, the lower the temp for liquefaction. when the ratio is 75-25 approx, at 217 deg C, 99% of the SAC is in liquid state. the key to this "hybrid" profile is to increase the time above liquidus (90-120 sec) and not go a degree higher than you absolutely have to.

granted, you would need a DOE to do this and x-sections to determine the degree of mixing. but when once those parameters are down, one should have repeatable results everytime.....Just my 2 cents worth.

reply »

#42337

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

Muse, I agree. There has been much debate but the previous scenarios discussed in the forum have been regarding the SAC/RoHS BGA and Sn-Pb paste. I have successfully collapsed a SAC BGA with a Sn-Pb paste using the "hybrid" profile that Amol refers to...I had the joints pull tested (but not thermal cyled) and it was just as strong as SAC on SAC.

As far as the Sn-Pb BGA and SAC paste scenario, various white papers, PhD materials scientists, etc. have advised against this. I looked on my seminar notes and there was one MEPTEC Lead-Free symposium study that shows that a Sn-Pb ball soldered with SAC alloy will start to fail shear testing exponentially after 400 thermal cycles, whereas the regular controlled SAC BGA's (also soldered with SAC alloy) stay pretty consistent after 1,000 thermal cycles.

Bottom line...most of us on this forum are not Materials Scientists, (except maybe the Guru and his Behodium solder)but you as the Engineer needs to take into account the final application of the product (ie will this product see harsh environments, and potentially fail in the next 5-10 years)?

Personally, I wouldn't take that risk. I would do DaveF's suggestion and rework these puppies.

reply »

#42338

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

Burn from dave on again Chunks!

I agree with dave - get rid of the women. Less problems.

reply »

#42343

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

Hi,

That was another very interesting article posted.

This article seems to point to the same conclusion, which is that lead free soldering is more reliable than Pb components with Pb paste, and Lead Free components with Pb paste?

From what I see, they are saying that there is not much difference to soldering lead free parts with lead paste, and lead parts and lead paste?

It's looking like we just have a non RoHS product, but there is not much other issues?

Perhaps it's late and I am misreading these reports, but it seems to be saying that? Does anyone else read this conclusion from these reports?

Regards,

Grant

reply »


RDR

#42347

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

That is about what I gathered as well grant, if the pb part can withstand the elevated temp of pbfree profile there are no issues

Russ

reply »

#42349

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

I see what you're saying - I'm not a process guru by any means. I was going on the premise that SAC melts at a higher temperature than SnPb. Basically I think of it as a BGA is joining solder to solder, unlike an SOIC where it is tinned metal lead to solder. The SnPb solder will melt sooner than the SAC solder. As I've seen, most profiles are geared to getting just the paste to its critical temperature to make a bond. However, with a BGA, you need to get ALL of the solder (both the ball and the paste) to the critical temperature. If they are different and their melting points don't match, you have to gear to the higher one. Just achieving wetting with the paste, but not fully collapsing/etc the solder ball will definitely affect reliability.

reply »

Chunks

#42353

Oop's Pb BGA's in a RoHS process? | 21 June, 2006

Unless it was an x-rated tool built by a woman.

reply »

#42365

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

Grant,

The best rule is- "If it works don't touch it". What I mean is check your component datasheet. Max allowed temperature should be there. You know the profile of your reflow oven and the temperature reached during reflow. So if there is no conflict, I believe the boards are reliable. If you change the BGA's you can have more problems in and after rework.

reply »

#42376

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

Ha ha, very funny. Not as funny as our burn on YOU!

reply »

#42377

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

Yesa yes Samir - you are stupid to form any other opinion but davefs! You should always follow his advice. Maybe you Samir, need to talk to your engineers for further advise.

reply »

Chunks

#42381

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

Oh yeah, I forgot about the great big burn of 2006 you and davef laid on me. Wow, good one fellas. Now high five each other. Don't forget the butt slap. Now belly slam! (wow 578 lbs of flab just met in a cube somewhere!) Rub each others bald spots for luck and wink at each other like "you are SO good looking brutha!".....

reply »

#42385

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

Hi,

Yes, I agree, and it seems like these studies were designed to show that lead free soldering was better than lead based soldering, and they threw in some results of lead components soldered in a lead free process, which seemed about the same reliability as a normal lead based solder process.

Funny after changing to RoHS, it seems so old fashioned to be talking about lead soldering now. It seems had to remember back two weeks ago to those heady days of lead based soldering, where I would not dream of licking the product as much as it's possible to do today. We certainly live in a better world now.

Regards,

Grant

reply »

#42388

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

Shocker, You truly live up to your name with those "sexist" remarks! Are you Islamic? "Mumtaz" is an Indian/Arab name. As for me, my wife converted me to become non-muslim...probably the non-practicing muslim here in Riyadh...

reply »

#42389

Oop's Pb BGA's in a RoHS process? | 22 June, 2006

"We certainly live in a better world now."

Regarding lead-free solder I assume?

If so, then...............

Oh god....... Another one sucked into the EU black hole of ignorance.

Sorry for the thread drift, couldn't help myself.

Hope your parts don't fall off.

Paul M.

reply »

PCB Soldering Tools

PCB Rework / Repair Services