Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Soldering problem with Au plating PCB

Soldering problem with Au plating PCB

Subscribe to Thread

Hello All, I have a problem with nonwetting Au finish aft... - Jan 25, 2009 by peterc

This one might help ... - Jan 26, 2009 by emil

We've had this same problem with the solder wicking to the c... - Jan 26, 2009 by robgd3

> Hello All, > > I have a problem with nonwettin... - Feb 03, 2009 by peterc

Hi Peter, The defect you observe during the second side ref... - Feb 04, 2009 by sachu_70

Hi Sachin, thanks for your feedback. According to our su... - Feb 09, 2009 by peterc

Look here ... - Feb 09, 2009 by davef

Hi all We also met the solder nonwetting problem as belo... - Mar 30, 2009 by Jeong Ju-young

Is this a sample board that you ran without components? I... - Mar 30, 2009 by robgd3

Thanks for your answer. For your question, this problem f... - Mar 30, 2009 by Jeong Ju-young

Hello, we had similar problem in the past - when we had to ... - Mar 31, 2009 by d0min0

Hi Peter, Have the plating thickness checked with XRF. If... - Apr 02, 2009 by Vlad

Jeong Ju-young You touched them all up with an iron? Wow! ... - Apr 06, 2009 by C111

peterc

#57875

Soldering problem with Au plating PCB | 25 January, 2009

Hello All,

I have a problem with nonwetting Au finish after second cycle of reflow soldering doublesided boards. Solder nonwetted pcb pads (is wicked to component terminations) Either these pads or some of non component pads became discoloured from yellowish to silver white. This problem is not reworkable with our common hand soldering process. We have 10% error rate.

-According to supplier report surface finish is "Flash gold" where Electrolytic Au thickness is 0,02 to 0,05 um and Nikel 2,5 to 5 um. -Lead process, solder paste Heraus F10

What cause it become discoloured and nonwetted?? Is this sign of Nikel oxidation or Au migration?

Many thanks, Peter.

Attachments:

reply »

emil

#57879

Soldering problem with Au plating PCB | 26 January, 2009

This one might help

http://www.smtnet.com/Forums/index.cfm?fuseaction=view_thread&CFApp=1&Thread_ID=13630&mc=5

Emil

reply »

robgd3

#57882

Soldering problem with Au plating PCB | 26 January, 2009

We've had this same problem with the solder wicking to the component terminations. However, in our case, we did not see the pads turning a silver-white color...rather, they exhibited a tarnishing/darkening of the gold.

We determined that we had some issues with our wash process. When boards were build first side, then washed, then built second side, we found this problem on the second side. The end determination was that either residual flux from the first side build was migrating to the second side, then not being properly washed off the board, and essentially being baked to the board during the drying process. We eliminated the wash step between the two runs, and our situations improved, but still exhibited some issues.

If you can get a sample of the tarnished pads out for analysis, I would highly recommend doing that. That will help you isolate what's going on with the gold finish.

To perform some testing of your own, take a blank board, and screen some paste onto it and reflow. Follow your normal process, and see if it happens. If so, reverse your process (that is, paste the second side and reflow it first) to see if it's something inherrent to that side of the board. If it is inherrent to that side of the board, then it's potentially a board-house/finish issue. If, when you reflow that side of the board, everything works fine, it's likely a contamination issue somewhere in your process. Could be during first side reflow, wash between builds (if you do this), storage, or handling.

cheers ..rob

reply »

peterc

#57976

Soldering problem with Au plating PCB | 3 February, 2009

> Hello All, > > I have a problem with nonwetting Au > finish after second cycle of reflow soldering > doublesided boards. Solder nonwetted pcb pads (is > wicked to component terminations) Either these > pads or some of non component pads became > discoloured from yellowish to silver white. This > problem is not reworkable with our common hand > soldering process. We have 10% error > rate. > > -According to supplier report surface > finish is "Flash gold" where Electrolytic Au > thickness is 0,02 to 0,05 um and Nikel 2,5 to 5 > um. -Lead process, solder paste Heraus F10 > What cause it become discoloured and nonwetted?? > Is this sign of Nikel oxidation or Au > migration? > > Many thanks, Peter.

Thanks for your answers. We have no clean solder process. I have another question: Is Electrolytic Au thickness (0,02 to 0,05 um) O.K. according to international standards e.g.(IPC, MIL, etc.)

Look at the attachments for illustration about my problem.

Thanks, Peter

reply »

sachu_70

#57984

Soldering problem with Au plating PCB | 4 February, 2009

Hi Peter, The defect you observe during the second side reflow could be caused due to improper Au plating thickness in the PCB fabrication process. You can refer IPC-2221 wherein plating thicknes for ENIG mentions 0.08um to 0.23um for Immersion Gold, over the Electroless Ni plating of 2.5um to 5.0um.

- Cheers ! - Sachin.

reply »

peterc

#58042

Soldering problem with Au plating PCB | 9 February, 2009

Hi Sachin,

thanks for your feedback. According to our supplier, surface plating is electrolytic gold over Nickel(not ENIG), but I think that is a still improper thickness for Au. (according to 3rd Working Draft from September 2008 of IPC 2221B "0.45 �m [17.72 �in] is a maximum for gold on areas to be soldered") What a minimum proper thickness for electrolytic Au plating over Nickel? What happened with the Au plating during first and second reflow cycle? It was dissolved to the nikel or cooper plate?

Thanks a lot, Peter.

reply »

davef

#58047

Soldering problem with Au plating PCB | 9 February, 2009

Look here http://www.smtnet.com/forums/Index.cfm?CFApp=1&Message_ID=57106

reply »

Jeong Ju-young

#58453

Soldering problem with Au plating PCB | 30 March, 2009

Hi all We also met the solder nonwetting problem as below. This happend after reflow and this is Rohs cpmpliant. How do i solve this problem?

Thanks

Attachments:

reply »

robgd3

#58455

Soldering problem with Au plating PCB | 30 March, 2009

Is this a sample board that you ran without components?

If so, I'd get it back to your board house, and ask them to explain to you what's happening.

We never did get to a satisfactory root cause on our issue (at least, satisfactory to me). It seems apparent that there is some contamination in either the gold or, perhaps, the nickle at the board house. My board house never found anything, or, at least, never reported what they found when I sent my board back for solderability testing. However, we had bought 10,000 pieces, and found that it was only about 1000 pieces that were exposed to the issue. The board house hadn't changed the date in the screen, so we were unable to determine true lot traceability; but, the balance of the boards ran ok for us, so, we surmised that the issue was isolated to that lot.

What can you do to fix the issue? Other than buying new boards to run, you might try a more active flux in the solder paste, and ensuring that your oven profile does two things: 1. achieve and hold the flux activation temperature for the specified time period, and 2. does not over-activate the paste by going too high in temp, or for too long a period. Can't guarantee that this will work, but, it may be worth a try.

The only other option we came up with was touching up the boards, which is both costly and time consuming.

cheers ..rob

reply »

Jeong Ju-young

#58459

Soldering problem with Au plating PCB | 30 March, 2009

Thanks for your answer.

For your question, this problem first happend on board with components so,we tested a board without components.

As you said(only your case),if i inform our supplier, i will not get any answer from him and this will be very

And i need more information about your idea that some contamination on nickel or Au in the board can affect nonwetting(if you are possible)

We have repaired these boards with iron(The result was good) How can i say to my supplier? I want this problem not to happen again.

Thanks

reply »

d0min0

#58460

Soldering problem with Au plating PCB | 31 March, 2009

Hello, we had similar problem in the past - when we had to change PCB supplier from a batch of hundreds of boards (NiAu plated, SnPb process) we got 2-3 boards with paste wicking to components leads what we did, was to check the thickness of gold on that places ( with our external lab and Henkel lab in UK - thanks guys :-) ) and in both cases it showed close to 0 thicness (with the full gold colour on the pads)

what did the supplier - during process they marked a PCB with defec with a sticker, and that caused a problem during the process - since they change it no more problems with them

best regards

reply »

Vlad

#58492

Soldering problem with Au plating PCB | 2 April, 2009

Hi Peter,

Have the plating thickness checked with XRF. If the layer of Au too thin, then Ni can diffuse all the way through it during the first run and the surface won't be perfectly solderable for the second side. it might also have a thin layer of Ni and Cu can diffuse through Ni and m,ake the Au solderable surface unsolderable.

Regards,

Vlad SENTEC

reply »

C111

#58516

Soldering problem with Au plating PCB | 6 April, 2009

Jeong Ju-young You touched them all up with an iron? Wow! We have also had similar issues. What I did to prove there was a board problem was paste one & reflow one bare board with the same chem (noclean) we saw the non-wetting condition. Then applied our most agresive flux to the entire board ran it through reflow again and still had issues. At that point we pulled the plug and sent all product back the board house. All built product was scrapped as we feard latent failures. we did attempt hand solder as well and everything looked good, But we were not sure what the actual connection would yeild. We took a credit from hte board supplier and never actually got a root cause.

reply »