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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Immersion Gold over Nickle de-wetting problem

Hi, I'm experiencing a problem on a standard part made many... - Mar 06, 2003 by

Add this forum to your site! Click to learn more. SMTA

O' Connor

#23688

Immersion Gold over Nickle de-wetting problem | 6 March, 2003

Hi, I'm experiencing a problem on a standard part made many times before -Ni/Au finish PCB, Reflow 100% SMD parts. After reflow the gold on the pads does not fully fuse with the solder paste it remains intact on the extremities of the pads, it seems to only form an alloy where the paste has been deposited, the solder paste aperture is 90% size of Pads, despite this the joints formed seem fine. Hope this paints the picture, I carry out no conditioning of the PCBs prior to assy, on this recent batch of boards the gold finish seems duller (dirtier!) that hertofore. Question is should I wash the boards prior to assy? with what? has anyone experienced similar problems before? Thank you for any feedback

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

reply »

John O'Donnell

#23690

Immersion Gold over Nickle de-wetting problem | 6 March, 2003

Greetings Sir, Although I am not qualified to address the issue in detail, I do know first hand that you can obtain very comprehensive and accurate solutions and explanations from George Westby, Director, Universal Instruments Advanced Technology and Surface Mount technology Process Laboratory. George can be reached at 607-779-5258.. he is very knowledgeable, a great guy and easy to speak. I recommend you call him or call me and I will get him on the phone for you.

Considering that I am providing this recomendation, I would appreciate your feedback to the suggestion and to your success.

Regards John O'Donnell 603-433-2125 ODonnell@UIC.com

reply »

#23699

Immersion Gold over Nickle de-wetting problem | 6 March, 2003

O' Connor: Are you saying that: * You solder fine in the areas where you print paste * You solder poorly in the areas where you don't print paste

Aside from that, the gold on these pads seems to be less bright than previous lots.

[if that's not it, please restart.]

Without knowing the answers to these questions, washing these boards will accomplish nothing in terms of improving the appearance of your gold plating.

Tell us about your paste.

reply »

O' Connor

#23702

Immersion Gold over Nickle de-wetting problem | 7 March, 2003

Hi Again, In answer to DaveF - Yes solder is fine where the solder paste is deposited , but the part of the pad visible after pasting is still visible after reflow. Previously the solder paste & all the gold diffused & flowed giving a complete silver finish on the pad. The Solder paste is SN62 MP 100 AGS 90 - from Multicore. Since the first post we have carried out a shear test on the solder joints and find them to be more than acceptable -i.e. 1206 ceramic cap withstands 10lbs force, also found a description in IPC to describe the problem IPC-A-610, section 12.4.4 SMT Soldering Anomolies - Nonwetting, so the problem seems to fall into a cosmetic defect which I can live with but I would prefer to resolve it so it does not re-occur. John O' Donnell, thanks for the contact info. I will let you know what fedback I get if I call.

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

reply »

MA/NY DDave

#23705

Immersion Gold over Nickle de-wetting problem | 7 March, 2003

Hi O'Conner,

I hope you also tell us what Universal-GeorgeW discusses with you. It is a little strange seeing his name on this list and his phone number. Gee I can give you his email address but won't..

Again distance engineering is tough.

I wouldn't worry too much about initial shear. I would worry more about how this particular Gold is going into the solder joint. Also I worry about what this particular finish will look like longer term under high temp and humidity usage conditions. You may need some SEM's (Scanning Electron Microscope) to see what is up and to know whether a longer term reliability failure will now be a greater potential.

Since you have had such good results prior, and seem to NOT have changed your process, paste, stencil, paste control etc the problem seems to be at the PCB shop.

Ask them to check their Gold Bath and do an assay. They may have gotten something into it and are co-depositing GOLD (Au) with something else like copper. Hopefully the shop you are using has regular control charts and assaying techniques in place.

It is not uncommon to dump GOLD baths when junk gets into it and the finish turns lousy. Expensive yes, yet the PCB shop is supposed to keep that process clean.

Usually it is a whole lot simpler and FASTER for you to let them fix their Gold bath then spending time on analysis. As they say FIX it FIRST then analyze. Keep that high Quality Product a Flowing.

LET us KNOW, What GeorgeW tells you, and what you turn up.

YiEng, MA/NY DDave

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

reply »

#23713

Immersion Gold over Nickle de-wetting problem | 7 March, 2003

YiEng, MA/NY DDave: Usually when the gold plating bath goes south, you loose solderability.

reply »

#23714

Immersion Gold over Nickle de-wetting problem | 7 March, 2003

O'Connor: IPC-A-610, section 12.4.4 SMT Soldering Anomolies - Nonwetting was added primarily to accomidate OSP boards. Heated flux is used to remove the 'lacguer'. So, you end-up with a copper colored border [or shadow] around the pads, where there was no paste contact.

In your case, you can "say" that that's what's going on, but it's not correct. You should have wetting across the whole pad [just as YOU expected].

Take cross-sections of the darker [duller] toned pads. Probably, there is no [not enough] nickel on those pads. An XRF reading would probably show 1-2 uinches of nickel and cross-section will probably verify that the problem is there, not the gold.

There are debates as to what causes this: poor cleaning prior to nickel, galvanic reaction, poor initiation of the nickel bath, problems with the activator (catalyst), static charge, and probably others.

We've seen the same problem with a fabricator using two different ENIG chemical suppliers. Each chemical supplier had their own explanations, but neither vendor was able to make it go away for good. The problem occurred on basically the same pads on several panels, making a strong argument for static charge and/or galvanic cell reaction. A third chemical supplier made the problem go away.

In any event, your problem is not the lack of luster of the gold finish. It's the absence of nickel. Immersion gold deposits by ion replacement with nickel ions. There was enough nickel deposited to get the gold to take, but that's it. Immersion gold is sacrificial and is there only to protect the solderability of the nickel. In the case of these boards, you may find that there's no nickel there to solder to.

reply »

neil

#23717

Immersion Gold over Nickle de-wetting problem | 7 March, 2003

Remember the gold is only deposited to provide an air barrier so that the nickel does not oxidise. Normally the gold is absorbed into the solder joint so that the solder forms a joint with the nickle. It is not unusual to see gold remaining on pads if the solder paste has not been in contact with it, if you wish to remove the gold you will need to cover it with solder past and have no stencil reduction.

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

reply »

MA/NY DDave

#23721

Immersion Gold over Nickle de-wetting problem | 8 March, 2003

Hi

Yes DavidF I agree, and that is what I think I am reading in the orginal and follow on post from the poster. I think this problem is just below full failure to solder which was your point, I think.

You, and I particularly, are taking opposing yet complimentary positions in problem resolution. David F right now is searching for flaws in his process or knowledge, and David F #2, right now, is thinking that it has to do with the product he was supplied. I could be off a tad, or you could be off a tad, and I hope we agree at least on that point.

Now a GOLD bath doesn't go bad FAST. It takes awhile for the impurities or improper pre-washes to create difficulty seen by the customer.

I am anxiously waiting for his report positive or negative after his phone call to Universal's GeorgeW. To be honest, even though I respect GeorgeW, I think that you and maybe me will give better guidance.

YiEng, MA/NY MrDave

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

reply »

MA/NY DDave

#23722

Immersion Gold over Nickle de-wetting problem | 8 March, 2003

Hi

The other thing the Ni does is to create a time barrier for the underlying copper (Cu) migration into the joint. Degradation still occurs yet it is slowed significantly.

Diffusion Equations under, electonic current pressures across time, are involved in how a solder joint slowly degrades due to any and all other metals.

SORRY guys and gals, Yet I had to think this out from what I have learned. I believe you can find what I wrote in the books of our industry.

YiENG, MA/NY DDave

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

reply »

iman

#23729

Immersion Gold over Nickle de-wetting problem | 9 March, 2003

Which specific title books from under your pen, would you recommend for the avid process reader/learner?

specific - talking about info/process/data wealth books, and not which books give the most royalties *grinz*

reply »

MA/NY DDave

#23739

Immersion Gold over Nickle de-wetting problem | 10 March, 2003

Hi

Well I don't have all the titles right handy, yet let me give a few places for you to check. All my books aren't exactly at my finger tips.

Author: Tumalla (??spelling) He is now at Georgia Tech Author: Lasky, Seraphim Author: Ning-Cheng Lee or Lee Ning-Cheng with an upbeat Author: Manko (former top honcho in soldering Author: Coombs Author: Charles Swall (?? spelling

Good Luck

YiEng, MA/NY DDave

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

reply »

Jim

#23760

Immersion Gold over Nickle de-wetting problem | 11 March, 2003

MA/NY DDave

#23762

Immersion Gold over Nickle de-wetting problem | 11 March, 2003

Hey Neat,

We share some of the same books, although I would strongly recommend he expand his collection. That Flex book is only one, the most pricey one, of many on that technology.

JUST Kidding THANKS!!

If we started listing all our magazine, technical articles, an other stuff we would get fined for web abuse. I can imagine that David F also has one heck of a library.

YiEng, MA/NY DDave

P.S. Next post is going to be more serious and back to the topic at hand. P.S. Gee I hoped on this visit I would see what Universal-GeorgeW gave as a telephone reply. UIC/GeorgeW used to do analysis of these types of things first for customers and then otherwise.

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

reply »

MA/NY DDave

#23763

Immersion Gold over Nickle de-wetting problem | 11 March, 2003

Hi

As I got thinking about your problem, as you stated it, I am wondering if maybe we have some oxides or too thin of a Gold Immersion layer so that some of the underlying NI is being oxidized. The colorization that you described is what intriques me. Maybe it is a bath problem as I initially thought or maybe it is storage creating oxidation before you get to use the PCB.

I will let David F attack your side of this problem if he wants to go furthur, since these kinds of problems sometimes can be tackled best if approached in both directions.

As long as I have them before I delete them and move on here are some articles on Gold Plating in general. A little on Immersion Gold yet not a lot.

http://www.gold.org/discover/sci_indu/GBull/2002_1/P%20Goodman.pdf

http://www.gold.org/pr_archive/pdf/IEE_Review_article.pdf

YiEng, MA/NY DDave

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

reply »

O' Connor

#23780

Immersion Gold over Nickle de-wetting problem | 12 March, 2003

Everybody thanks for all your help, the problem has been solved. Before I describe the solution I've got to own up, I'm the PCB Design Eng., The Process Eng. lobbed this problem onto my desk last week, said the problem lay with the PCB finish, whats the solution?. At the time I questioned the reflow profile but was told no problem there. I log on regularly here and find this a useful forum - so my first post. To cut a long story short the Process Eng. checked the reflow profile after a request from the Operations Manager to do so and found the problem lay there. Apologies for the waste of your time, I've also got to apologise to the PCB vendor for sending him on a wild goose chase as well.

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

reply »

#23782

Immersion Gold over Nickle de-wetting problem | 12 March, 2003

O' Connor

Haaa. It's usually the design guys that foist the problems onto my desk. Now that is one shrewd process type that you have working at your company!!!! Can I have his / her name and number to get some tips on the technique that he / she uses?

We ignored questions on the process side based on the creditability of your solder shear testing results. So, what specfically did your process engineer find after checking the "reflow profile ... and found the problem lay there."

Still if there was a reflow profile problem, it seems a bit unusual that you'd: * Get good solder flow. * Obtain reasonable shear testing results.

reply »

MA/NY DDave

#23783

Immersion Gold over Nickle de-wetting problem | 12 March, 2003

Hi

Thanks more than you know for your honesty.

Honesty adds 200-300% value in my book of all times.

I am glad your problem was so easily solved, and I agree that you did the right thing by appologizing to your PCB supplier.

Great, Lets move on

YiEng MA/Ny DDave

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

reply »

PCB Cleaning