Nothing evil about gold for soldered connections - as long as the thickness is kept within IPC-2221 guidelines of less than 0.8um (about 31microinches). Where you get into trouble is with thicker layers of gold. Tin-lead solder normally used in electronics is not fully compatible with gold plating. Gold readily alloys with tin-lead, but when the gold concentration exceeds a few percent the alloy becomes weak during thermal cycling. This process is referred to in the literature as gold embrittlement. As long as the gold plating thickness is small (20-30 µin) this is not considered a problem. In thick gold layers, the only bonds that are formed when soldering are the brittle high percentage gold alloys. When you have a brittle alloy over a soft all-gold underlying layer, naturally something is going to give under stress.
When the gold layer is kept thin enough, the soldering process penetrates to the nickel and forms a more stable and durable transition layer with the nickel. The gold is distributed and diluted (alloyed) throughout the entire cross section of the joint and the percentage of gold in the alloy at the joint is relatively low. The nickel is there for two reasons when you gold plate a board: first as a barrier metal to prevent gold-copper transition alloys that are weak alloys, and second as a hard underlay for the soft gold. Normally you are looking for at least 100 microinches of nickel under gold for mechanical connections, and that works just fine as the barrier/transition for soldered joints as well. Short story - make sure the barrier nickel is thick enough, and the gold is thin enough. All will be well with soldered joints. At 11:34 AM 4/23/02 +1000, you wrote: >Greetings all, > > I have been following the recent thread dealing with gold plating > (WANTED: PCB Expert) where the subject of embrittlement was raised. The > original focus of this thread - PCB warping - is not a problem for us. > The comment made by Brad Verlander got me thinking and a little worried: > (In the extreme cases I have been told of boards having significant > numbers of their components drop off the boards when the board is > manually flexed.) > > This has raised some issues with a couple of the boards we make here > at Zener. The story goes like this... > >We have a number of PCBs designed to accept terminations made via spring >loaded contacts. We have had reliability problems with straight tinned >copper for these terminations, mainly due to surface irregularities of the >solder film. A decision was made to try gold plating to alleviate the >problems. This has proved successful but the PCB manufacturer has gold >plated the entire PCB not selectively in the areas of concern. After >reading the above mentioned thread enquiries were made regarding the >nature of the gold plating, the reply being Au (2 microinches) over Ni >(100 microinches). > > Is this good, bad or indifferent? Do these figures indicate possible > embrittlement problems in the future? > > Can anyone point me towards further reading on this topic - books, > articles, web sites etc.? > > Thanks in advance for any input. > > >Best Regards, > >LINDEN DOYLE >Product Development Engineer >Zener Electric Pty Ltd > >[EMAIL PROTECTED] >Ph: +61 2 9795 3600 >Fax: +61 2 9795 3611 snip * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * To post a message: mailto:[EMAIL PROTECTED] * * To leave this list visit: * http://www.techservinc.com/protelusers/leave.html * * Contact the list manager: * mailto:[EMAIL PROTECTED] * * Forum Guidelines Rules: * http://www.techservinc.com/protelusers/forumrules.html * * Browse or Search previous postings: * http://www.mail-archive.com/proteledaforum@techservinc.com * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
