Author Topic: PCB Board Surface Issue, Advice Appreciated!  (Read 1716 times)

0 Members and 1 Guest are viewing this topic.

Offline sbar5781

  • Contributor
  • Posts: 7
  • Country: us
PCB Board Surface Issue, Advice Appreciated!
« on: June 03, 2016, 02:15:12 am »
Hello!

First time poster, so please forgive me if this post is in the incorrect location.

I recently received back a series of prototype PCB boards for IC testing. All the usual assembled components on the board; Resistors, Caps, Level shifters, PGA socket, etc etc. However additionally room was included on the boards to test specific IPDIA capacitors on the board, which the assembly was not able to include as they must be attached to the board with epoxy and wire bonded to 4 pads on the board; basically like a simple Chip-on-board. The PCBs were made with pads places for this Al/Au wire bonding and an ENIPEG surface finish was chosen to ensure the bonding process would be possible.

The problem is when the board was sent from manufacturing to assembly and the IPDIA caps were set as do-not-assemble parts, they still applied a small layer of solder to the exposed ENIPEG bonding pads. Now the wire-bonding team are telling me they cannot bond to this surface.

Can the pads be easily scrubbed?

The Bonding Expert suggested I get some small gold plated squares (60mil x 60mil or so approx) and have them soldered to the bonding pads to provide a suitable surface however I am at a loss as to where to source such metal parts, considering his direct reference was for parts from "SPM", a company 7+ years gone.

Any other ideas? The primary limitation is time (not money, within reason) as we are on a rushed timeline to test the boards.

Thankyou
 

Offline reagle

  • Supporter
  • ****
  • Posts: 528
  • Country: us
    • KuzyaTech
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #1 on: June 03, 2016, 02:45:52 am »
not sure of your board's geometry, but could you make a small adapter board to mount the wirebonded chip on and then solder that on top of the now tinned pads?

Online IconicPCB

  • Super Contributor
  • ***
  • Posts: 1300
  • Country: au
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #2 on: June 03, 2016, 04:52:35 am »
Talk to Your local manufacturing jeweler.. he must have gold leaf to hand.

Gold leaf is very thin, a sharp scalpel will cut it readily.

Cut pieces just larger than the solder covered surface and reflow the goldleaf on top of solder .
 

Offline DimitriP

  • Super Contributor
  • ***
  • Posts: 1055
  • Country: us
  • "Best practices" are best not practiced.© Dimitri
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #3 on: June 03, 2016, 05:30:20 am »
As far as "scrubbing the solder off the pads" goes , a fiberglass pen will do the trick
(depending on the size of the pads and the number of boards ) 


https://www.google.com/search?q=3m+fiberglass+pen&ie=utf-8&oe=utf-8

   If three 100  Ohm resistors are connected in parallel, and in series with a 200 Ohm resistor, how many resistors do you have? 
 

Offline sbar5781

  • Contributor
  • Posts: 7
  • Country: us
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #4 on: June 03, 2016, 09:28:39 am »
Thanks for all the replies.

1. A small adapter board would be theoretically possible but almost certainly not a quick solution. At that point, having spare boards re assembled would be equivalent and just as time consuming and cost impractical.

2. Thanks for the idea for sourcing gold from a jeweler. This question though will expose my inexperience, but how would I go about reflowing the gold leaf? Is it possible with access to only a soldering iron? Our laboratory has limited equipment, however the nearby bonding lab will surely have better tools, but if i understand the bare-bones process then it would help if I need to pass forward instructions to them (as a suspect this will be outside their usual work scope).

3. Scraping sounded intuitively to me to be the simplest option but I need to ask if this is actually possible. That is specifically will any method of scraping/solution cleaning be able to remove all the solder and expose the original surface finish of ENIPEG, NOT the raw exposed copper?

Thanks again
 

Offline ovnr

  • Frequent Contributor
  • **
  • Posts: 658
  • Country: no
  • Lurker
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #5 on: June 03, 2016, 10:26:42 am »
No, any scraping will rip off the gold layer too. Actually, the gold layer in most ENIG finishes is so thin that it dissolves into the solder, so it's probably gone already.

Since this seems to be prototype quantity stuff, any chance you could repurpose some of these gold contacts? Just cut appropriate squares of the metal and reflow them onto the pads. It's not going to be perfectly flat or aligned or anything, but it's probably better than trying to bond to copper...

And you will need to use indirect heat (hot air, for instance) - if you use an iron, you will just coat the surface with solder, after all.
 

Offline sbar5781

  • Contributor
  • Posts: 7
  • Country: us
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #6 on: June 03, 2016, 12:38:29 pm »
Thank you for that link!

This part specifically looks like it should suffice nicely with one dimension of cutting only.

Looks like reflow with a heatgun will have to do. At least I learned this mistake for next time.
 

Offline LukeW

  • Frequent Contributor
  • **
  • Posts: 666
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #7 on: June 04, 2016, 08:30:10 am »
Next time double check the artwork you're sending for the paste stencil - confirm the DNP parts are excluded! :)
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 12479
  • Country: cn
  • Power Electronics Guy
Re: PCB Board Surface Issue, Advice Appreciated!
« Reply #8 on: June 04, 2016, 11:48:42 am »
No, the gold will be dissolved into solder, and nickel/gold beneath will form IMC with solder. The best surface for bonding is nickel, medium phosphorus electroless nickel. However, I found ENIG is also good for wire bonding, but the long term reliability remains unknown as we never did any long term testing. FYI, the bonding machine is Hesse BJ380 with 4~10 mil Al wire. Gold wire should be better compatible with ENIG, but I can not say that for sure.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf