| |
 |
|
|
Science Forum Index » Electronics - Design Forum » Gold-PLated PCBs?
Page 1 of 2 Goto page 1, 2 Next
|
| Author |
Message |
| Robert Latest |
 Posted: Thu Feb 22, 2007 6:27 am |
|
|
|
Guest
|
When I'm doing small runs or singles of PCBs I use a manufacturer who pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
robert |
|
|
| Back to top |
|
| Phil Allison |
| Posted: Thu Feb 22, 2007 8:23 am |
|
|
|
Guest
|
"Robert Latest"
Quote:
When I'm doing small runs or singles of PCBs I use a manufacturer who
pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot
be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
** Gold plated connectors that require normal, 60/40 SnPb, solder to use
have been around for many DECADES.
No problems in sight as of 2007.
Gold plate is just soooo easy to solder.
........ Phil |
|
|
| Back to top |
|
| krw |
| Posted: Thu Feb 22, 2007 10:55 am |
|
|
|
Guest
|
In article <slrnetqs3g.3nh.boblatest@kir.physnet.uni-hamburg.de>,
boblatest@yahoo.com says...
Quote: When I'm doing small runs or singles of PCBs I use a manufacturer who pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
It's likely you have only a gold flash on the boards. IIRC, gold
will pollute the solder bath but should be no problem to your
boards. If it is only a gold flash there shouldn't be any problem.
Gold is generally only used for contacts (and more than a flash).
--
Keith |
|
|
| Back to top |
|
| James Beck |
| Posted: Thu Feb 22, 2007 12:23 pm |
|
|
|
Guest
|
In article <545ga1F1vecaoU1@mid.individual.net>, philallison@tpg.com.au
says...
Quote:
"Robert Latest"
When I'm doing small runs or singles of PCBs I use a manufacturer who
pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot
be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
** Gold plated connectors that require normal, 60/40 SnPb, solder to use
have been around for many DECADES.
No problems in sight as of 2007.
Gold plate is just soooo easy to solder.
....... Phil
Must have hitched a ride with a RoHS product.
My PCB fab uses ENIG for their RoHS PCBs.
http://www.trianglecircuits.com/ENIG.html
Jim |
|
|
| Back to top |
|
| John Larkin |
| Posted: Thu Feb 22, 2007 12:29 pm |
|
|
|
Guest
|
On 22 Feb 2007 10:27:53 GMT, Robert Latest <boblatest@yahoo.com>
wrote:
Quote: When I'm doing small runs or singles of PCBs I use a manufacturer who pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
robert
Gold and copper diffuse into each other, messing up the gold layer.
The only way to prevent this is to use very thick gold (too expensive,
usually) or to first plate a diffusion barrier layer, usually nickel.
So the "gold" plated boards you buy are probably electroless nickel on
the copper followed by a tiny layer of immersion gold. What the gold
is doing is making the nickel solderable!
There's only microinches of gold on a pcb, 10 maybe, so when it
dissolves into the solder, it pretty much vanishes.
We did some gold plated boards when we were experimenting with
lead-free assembly. They look beautiful and solder great. But we
dumped the lead-free thing.
Gold-copper and gold-aluminum intermetallic compounds (like Purple
Plague) are often brittle, but there's too little gold on most gold
plated boards and connectors to make trouble.
John |
|
|
| Back to top |
|
| Eeyore |
| Posted: Thu Feb 22, 2007 2:08 pm |
|
|
|
Guest
|
John Larkin wrote:
Quote: But we dumped the lead-free thing.
Why was that John ?
Graham |
|
|
| Back to top |
|
| John Larkin |
| Posted: Thu Feb 22, 2007 2:20 pm |
|
|
|
Guest
|
On Thu, 22 Feb 2007 18:08:14 +0000, Eeyore
<rabbitsfriendsandrelations@hotmail.com> wrote:
Quote:
John Larkin wrote:
But we dumped the lead-free thing.
Why was that John ?
Graham
Some of our aerospace customers are explicit in *not* wanting
lead-free. And the process temps are a lot higher, with possible
reliability issues. And the solder joints look like hell, making them
harder to inspect.
We now know that we can do it if we have to, but so far nothing is
forcing us to do so. The whole thing is silly; the solder on pcb's is
not soluble, is easily identified and kept out of landfills anyhow,
and constitutes only a couple of per cent of lead usage.
John |
|
|
| Back to top |
|
| Eeyore |
| Posted: Thu Feb 22, 2007 3:15 pm |
|
|
|
Guest
|
John Larkin wrote:
Quote: Eeyore <rabbitsfriendsandrelations@hotmail.com> wrote:
John Larkin wrote:
But we dumped the lead-free thing.
Why was that John ?
Graham
Some of our aerospace customers are explicit in *not* wanting
lead-free. And the process temps are a lot higher, with possible
reliability issues. And the solder joints look like hell, making them
harder to inspect.
We now know that we can do it if we have to, but so far nothing is
forcing us to do so. The whole thing is silly; the solder on pcb's is
not soluble, is easily identified and kept out of landfills anyhow,
and constitutes only a couple of per cent of lead usage.
I agree with you 100%.
I was just intertested to see what your decision was based on.
Graham |
|
|
| Back to top |
|
| Robert Baer |
| Posted: Fri Feb 23, 2007 1:30 am |
|
|
|
Guest
|
Robert Latest wrote:
Quote: When I'm doing small runs or singles of PCBs I use a manufacturer who pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
robert
From what little i have seen, the problem appears at/near the solder
connection of the wire from the case (negative electrode) of an aluminum
capacitor; classic "purple plague".
The solder "pool" around the lead lifts off with no gold apparent
anywhere on the PCB or the solder, and ther is a lot of whitish "dust" /
crystals in the region, and the cap end is discolored.
In certain circumstances, it can be worse that tin whiskers.
Almost any metal other than tin or lead will dissolve in eutectic
tin/lead solder; hence the advent of Savbit (2% copper added if i
remember right so the copper bits would not be leached away) as well as
the special roll of silver-bearing solder found in all old Tektronix
scopes (to protect the silver plating on the ceramic standoffs).
Gold looks sexy, but can cause rather expensive problems. |
|
|
| Back to top |
|
| Robert Baer |
| Posted: Fri Feb 23, 2007 1:32 am |
|
|
|
Guest
|
Phil Allison wrote:
Quote: "Robert Latest"
When I'm doing small runs or singles of PCBs I use a manufacturer who
pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot
be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
** Gold plated connectors that require normal, 60/40 SnPb, solder to use
have been around for many DECADES.
No problems in sight as of 2007.
Gold plate is just soooo easy to solder.
....... Phil
You must not have been in the industry much; i have seen too many
corroded connections over the years. |
|
|
| Back to top |
|
| Robert Baer |
| Posted: Fri Feb 23, 2007 1:33 am |
|
|
|
Guest
|
krw wrote:
Quote: In article <slrnetqs3g.3nh.boblatest@kir.physnet.uni-hamburg.de>,
boblatest@yahoo.com says...
When I'm doing small runs or singles of PCBs I use a manufacturer who pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
It's likely you have only a gold flash on the boards. IIRC, gold
will pollute the solder bath but should be no problem to your
boards. If it is only a gold flash there shouldn't be any problem.
Gold is generally only used for contacts (and more than a flash).
Gold for contacts is OK, as long as it is kept away from solder
connections...and most especially solder connections to aluminum
capacitor (negative) leads. |
|
|
| Back to top |
|
| Robert Baer |
| Posted: Fri Feb 23, 2007 1:38 am |
|
|
|
Guest
|
John Larkin wrote:
Quote: On Thu, 22 Feb 2007 18:08:14 +0000, Eeyore
rabbitsfriendsandrelations@hotmail.com> wrote:
John Larkin wrote:
But we dumped the lead-free thing.
Why was that John ?
Graham
Some of our aerospace customers are explicit in *not* wanting
lead-free. And the process temps are a lot higher, with possible
reliability issues. And the solder joints look like hell, making them
harder to inspect.
We now know that we can do it if we have to, but so far nothing is
forcing us to do so. The whole thing is silly; the solder on pcb's is
not soluble, is easily identified and kept out of landfills anyhow,
and constitutes only a couple of per cent of lead usage.
John
Gee, go down to Radio Shack and get some Silver Bearing Solder and
use that; it looks spiffier than old tin/lead! |
|
|
| Back to top |
|
| Guest |
| Posted: Fri Feb 23, 2007 6:07 am |
|
|
|
|
Robert Baer wrote:
Quote: Gold looks sexy, but can cause rather expensive problems.
Assuming no cost constraints (customer paying thousands
of dollars for the design, deliverables are ten small PWBs),
no Rohs requirement, and a normal office environment, which
would you choose? Gold over nickel over copper? Some sort
of exotic alloy for the solder? Or good old copper traces
and 63/47 solder? |
|
|
| Back to top |
|
| Rene Tschaggelar |
| Posted: Fri Feb 23, 2007 6:27 am |
|
|
|
Guest
|
Robert Latest wrote:
Quote: When I'm doing small runs or singles of PCBs I use a manufacturer who pools
many small batches into one large one to save costs. Consequently, some
minor aspects (like the color of the solder mask and silk screen) cannot be
specified because you'll get whatever the main batch gets you're hitching
your ride on.
Another unspecified aspect is the finish: They'll either do tin-plated or
gold-plated. Usually it's tin-plated, but today I got a gold-plated board.
Wowwwy. Talk about spiffy.
But aren't there issues with the gold dissolving in solder or some such
thing? I seem to recall that tin and gold form some unfavourable alloy.
Personally I don't care because I dont have millions of units in the field
to watch out for, but if these issues are extant, why gold-plate a PCB in
the first place?
robert
Robert,
Silver has a limited lifetime. Is becomes black
from fingerprints (sulfur) and it is said to
become harder to solder when oxidized.
HAL is said to becoem hader to solder after a
certain time too.
Gold aparently is keeping fresh and solderable
for an unlimited time.
I never had any issues, also not with any I
came across. It might be the warnings come from
manufacturers.
I'm fine with silver. Gold costs 10% more.
Leadfree is no problem at all. I had to replace
my old Weller with a new one though. The old one
went through tips at an unbelievable rate.
Rene
--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net |
|
|
| Back to top |
|
| DaveM |
| Posted: Fri Feb 23, 2007 6:01 pm |
|
|
|
Guest
|
<me@privacy.net> wrote in message news:Y5ednY68Pr72I0PY4p2dnA@giganews.com...
Quote:
Robert Baer wrote:
Gold looks sexy, but can cause rather expensive problems.
Assuming no cost constraints (customer paying thousands
of dollars for the design, deliverables are ten small PWBs),
no Rohs requirement, and a normal office environment, which
would you choose? Gold over nickel over copper? Some sort
of exotic alloy for the solder? Or good old copper traces
and 63/47 solder?
In my experiences of long ago while working as a cal-lab tech and as a tech in
the Navy, I have noticed a significant number of equipment failures related to
gold-over-nickel plated terminals, soldering posts, component leads, etc. The
reason was that the gold would delaminate from the nickel and create an open
circuit, or even worse, an intermittent connection that was a nightmare to find.
After recognizing this problem, when I had to replace a component or terminal
that was gold-over-nickel, I would always heavily tin the gold area, then wick
away all the solder, taking the gold with it, before making the final solder
joint. And, if I noticed any joints that looked suspicious, I would always try
to leach all the gold away from the joint and resolder.
Never had a repeat failure by doing that.
--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in the
address)
Some days you're the dog, some days the hydrant. |
|
|
| Back to top |
|
| |
Page 1 of 2 Goto page 1, 2 Next
All times are GMT - 5 Hours
The time now is Fri Aug 29, 2008 4:24 pm
|
|