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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Gold plated PCB's

Steve T.

#19768

Gold plated PCB's | 3 May, 2002

Okay, here goes another thread about gold plating.

I have been using gold plated PCB's for about two years now, with mostly very good results. The main reason I started using gold in the first place was because it is flat, solder wets to it nicely, it doesn't oxidize, and it's the best finish for button contacts. However, I'm starting to have my doubts.

We created a spec for our board house for the thickness, 5 to 15u". On occasion, we notice that the solder isn't wetting to the gold on specific lots. This is an indicator that the gold is too thin. There's nothing we can do with the boards that have been built-up either because when the solder doesn't stick to gold, too bad.

The thing that scares me is if we see some boards that won't wet, what about the boards that were marginal, tested okay, but when their in the field for a couple of months they fail. Holy crap!!

I'm ready to switch back to HASL. Even for as good as gold can be, the process has got to be held very tight in order to give the results repeatably. I haven't experienced a board house that can do it repeatably month after month, lot after lot. HASL is the norm. They can do HASL with their eyes shut.

Any thoughts from the forum? Who out there has used gold and good results for a long time? Who out there won't use gold? What's a good reason to stay with gold?

Thanks for you input.

Steve

reply »

#19770

Gold plated PCB's | 3 May, 2002

Several of the points you make are reinforced in the fine SMTnet Archives. Consider searching them for bedtime reading, until the responses start flowing.

reply »

ianchan

#19776

Gold plated PCB's | 5 May, 2002

Hi mate,

we had a crap experience (once...no more...) with gold plated over nickle (Au/Ni) PCB that had nickle contaminents in the PCB supplier's gold bath.

the Ni caused batch-delivery of the PCB to us for mass production, to exhibit the visible symptoms of solder refusing to fusion wet into the gold pads (we did batches of shear tests to see this).

the solder fusion was great with all the components. on the worse extreme PCB batches, we had components being pushed off the gold pads by "naked human finger" and we were still able to see complete solder wetting fillets on the components. we saw solder fusion to component to be no problem, we had a solder wetting to gold pad as THE biggy problem. we had the supplier do a purge and replacement really quickly.

Mebbe you should go down and get the supplier to do a spectrum analysis of the nickle level content in the gold pad? is your solder wetting problem confined to any specific location or batches? any supplier reports on those affected or suspected (fishy) batches? also is the component that refuses to wet a specific type of component, and is there a pattern of non-wetting component types (possible varient of types but with same terminal finishing)?

All in all, show me the data, show me the data!

Good Luck.

reply »

ianchan

#19777

Gold plated PCB's | 6 May, 2002

as a followup thought, what is your current peak reflow deg-C? and reflow timing (sec)?

Au/Ni calls for reflow temperature of 217 deg-C for solder fusion that produces secondary eutectic alloy.

there was one time we thought 183 deg-C was the magical number and that only applies to the paste reflow temperature, back then we neglected the PCB gold plating metallurgy characteristic...

Hope this helps.

reply »

ianchan

#19790

Gold plated PCB's | 6 May, 2002

Steve, Hi mate,

are you blokes using NC flux paste? or WS paste?

reply »

Steve T.

#19844

Gold plated PCB's | 9 May, 2002

We are using NC. Paste doesn't seem to have an effect on the solderability on a bad batch of ENIG. When ENIG has gone bad, throw the board out.

reply »

#19848

Gold plated PCB's | 9 May, 2002

ianchan

#19859

Gold plated PCB's | 10 May, 2002

Steve, Hi mate,

was doing some text book reading and saw this one paragraph that commented on the PCB fab plating could have residue chemistry inherent in the PCB if the fab process control isn't up to mark.

such chemistry could surface to the Au/Ni pad level and contaminate the solder fusion action during the Reflow processing. seems to be pre-dominent for NC pastes. the text was actually informing the reader the virtues of using WS pastes, as WS paste use will warrent Aqueous Cleaning of the post-reflow PCBA. (my understanding is of limited qualification to pass any judgement on such a statement by the author).

As I dun have lab facility to test and confirm this text based implied suspected root cause, i'd advise an engineering investigation on this theory only if your project penalty costings warrant it.

else, we could all convert over to immersion Ag PCB boards?

reply »

Steve T.

#19865

Gold plated PCB's | 10 May, 2002

Could be. Bottom line is my frustration is high with ENIG. It has it's merits, but if it's not done right, it'll bite you in the butt. Too many other choices to keep this frustration around.

I have never used OSP. I'm requesting 200 boards from my board house with OSP to see how that works.

I have even considered going back to HASL. At present we aren't placing anything under 20 mil pitch, so that's easy enough to work with.

Thanks for your input.

reply »

#19868

Gold plated PCB's | 10 May, 2002

FYI: not all OSP are created equal. I have had expierence, both good and bad w/ OSP boards. To qualify the vendor I would take a bare board and run it thru my oven twice on a reflow profile then flux it with NC flux and run it thru the wave solder for a wetting test and sometimes out for sectioning. Its an issue with OSP for double-sided solder applications.

reply »

#19874

Gold plated PCB's | 10 May, 2002

Kinda makes ya wonder why ya changed, eh? Some of our 20 pitch HASL boards come-out just as flat as a farm pond. Just got ta letum be.

Pete is spot on!!

reply »

Boundary Scan

SMT in-printer dispensing