American PCB fabs defaulting to lead-free

[Joerg]

Almost got a black eye here on a prototype. Ordered a bunch of large
boards at Advanced Circuits and it turns out that the free upgrade to
lead-free finish is actually a default that cannot easily be changed.
IOW you can normally not decline this "upgrade". However, they were very
understanding that we really didn't want lead-free and will now do our
protos as production runs. No idea why we go RoHS here in the US now
(this is a company that actually produces in the US).

Anyhow, just wanted to let you guys know about that before a nasty
surprise happens to you.

 

[Sergey Kubushin]

Almost got a black eye here on a prototype. Ordered a bunch of large
boards at Advanced Circuits and it turns out that the free upgrade to
lead-free finish is actually a default that cannot easily be changed.
IOW you can normally not decline this "upgrade". However, they were very
understanding that we really didn't want lead-free and will now do our
protos as production runs. No idea why we go RoHS here in the US now
(this is a company that actually produces in the US).

Anyhow, just wanted to let you guys know about that before a nasty
surprise happens to you.

I've got some prototypes from them a couple of weeks ago. Looks like regular
leaded solder to me... I didn't ask for anything specific, they just came
this way by default. But mine are all thru-hole...

 

[John Larkin]

Almost got a black eye here on a prototype. Ordered a bunch of large
boards at Advanced Circuits and it turns out that the free upgrade to
lead-free finish is actually a default that cannot easily be changed.
IOW you can normally not decline this "upgrade". However, they were very
understanding that we really didn't want lead-free and will now do our
protos as production runs. No idea why we go RoHS here in the US now
(this is a company that actually produces in the US).

Anyhow, just wanted to let you guys know about that before a nasty
surprise happens to you.

Some of the coatings are ok, some (like white tin) are awful. The good
news is that lots of people are now offering gold plating at the same
price, and that solders beautifully. The bare boards don't tarnish and
look like jewelry, too.

John

 

[Don McKenzie]

Some of the coatings are ok, some (like white tin) are awful. The good
news is that lots of people are now offering gold plating at the same
price, and that solders beautifully. The bare boards don't tarnish and
look like jewelry, too.

John

Gold plating I feel is the best approach, but be careful, as some
companies are asking 25% more. If you can get them at the same price, great!

I see the future being RoHS, as most companies will eventually gear up
for world wide sales. Sales don't have country borders these days.

Don...


--
Don McKenzie
E-Mail Contact Page: http://www.dontronics.com/e-mail.html

Crystal clear, super bright OLED LCD (128x128) for your microcontroller.
Simple serial RX/TX interface. Many memory sizes.
http://www.dontronics-shop.com/product.php?productid=16460

No More Damn Spam: http://www.wizard-of-oz.com

 

[Joerg]

Some of the coatings are ok, some (like white tin) are awful. The good
news is that lots of people are now offering gold plating at the same
price, and that solders beautifully. The bare boards don't tarnish and
look like jewelry, too.

I probably could have gotten gold coating but I had to make a
split-second decision. The Gerber review took their engineers a while
and it was getting close to 6:30pm in Colorado. Next time I'll try gold.

 

[Joerg]

Elect. gold is our new standard. It doesn't look as rich as the old
gold plated boards, but it's still nice.

I'll check that out. Sure would look more "blingy". One of the problems
was that they just didn't offer tin-lead on prototypes anymore.

 

[Spehro Pefhany]

I probably could have gotten gold coating but I had to make a
split-second decision. The Gerber review took their engineers a while
and it was getting close to 6:30pm in Colorado. Next time I'll try gold.

Elect. gold is our new standard. It doesn't look as rich as the old
gold plated boards, but it's still nice.


Best regards,
Spehro Pefhany

 

[Dirk Bruere at NeoPax]

Almost got a black eye here on a prototype. Ordered a bunch of large
boards at Advanced Circuits and it turns out that the free upgrade to
lead-free finish is actually a default that cannot easily be changed.
IOW you can normally not decline this "upgrade". However, they were very
understanding that we really didn't want lead-free and will now do our
protos as production runs. No idea why we go RoHS here in the US now
(this is a company that actually produces in the US).

The reason is that you cannot sell it in the EU unless it's ROHS

 

[Joerg]

The reason is that you cannot sell it in the EU unless it's ROHS

However, there is a lot of gear that is exempt and many of the parts for
specialized equipment cannot (yet) be purchased RoHS compliant. Even for
some on this board that come in RoHS such as data converters I found
that there was no stock anywhere for the RoHS edition. Had to buy the
tin-lead versions.

 

[Jeff L]

Almost got a black eye here on a prototype. Ordered a bunch of large
boards at Advanced Circuits and it turns out that the free upgrade to
lead-free finish is actually a default that cannot easily be changed.
IOW you can normally not decline this "upgrade". However, they were very
understanding that we really didn't want lead-free and will now do our
protos as production runs. No idea why we go RoHS here in the US now
(this is a company that actually produces in the US).

Anyhow, just wanted to let you guys know about that before a nasty
surprise happens to you.

No surprises - assuming HASL, they work fine using normal leaded solder.
This includes soldering with irons, wave and reflow methods. It does not
change the process hardly at all. The leaded solder just dissolves it,
creating a slightly non eutectic solder (closer to 60/40). It's the same
with tin plated parts which have been used for quite a few years, other then
lead free BGA's. The downside is you have a possibility of tin whiskers
forming in areas that have not been wetted with the leaded solder. The tin
whisker risk is somewhat dependant on the alloy used for the HASL. Tin
dendrites are from currents forming in moisture and can be avoided. White
tin is really bad, as it compresses the surface during plating and causes
rapid tin whisker growth. We've done many thousands of boards with lead free
HASL without much issue.

Gold plating (ENIG) is bad, as it is a flash of electroless gold deposits a
few atoms thick, over nickel which is plated over the copper traces. The
gold is sometimes hard to wet (BGA's for one - they are so bad we pre tin
the pads). The intention of the process is when applying solder it dissolves
the gold, thus wetting the normally difficult to wet nickel. You now have a
solder joint with a little gold in it, which if in high enough concentration
causes the solder to go brittle. The nickel to solder intermetallic layer is
poor and is prone to cracking (a well known repair problem in the repair
industry). I think the copper to nickel bond is ok, but I never really
looked into it in much detail. Other problems that can arise is weird
thermocouples from all of the dissimilar junctions. This would have been a
potential problem with some research I did awhile back where were measuring
temperature from a thermocouple with a resolution of around 1/1000 of a deg
C. The thermocouple was so sensitive you could peg the graph by breathing
heavy several feet away!

 

[Jeff L]

Elect. gold is our new standard. It doesn't look as rich as the old
gold plated boards, but it's still nice.

This can be bad to solder with small SMT pads, and there are reliability
concerns - see my other post.

Best regards,
Spehro Pefhany

http://www.speff.com

 

[Joerg]

No surprises - assuming HASL, they work fine using normal leaded solder.
This includes soldering with irons, wave and reflow methods. It does not
change the process hardly at all. The leaded solder just dissolves it,
creating a slightly non eutectic solder (closer to 60/40). It's the same
with tin plated parts which have been used for quite a few years, other then
lead free BGA's. The downside is you have a possibility of tin whiskers
forming in areas that have not been wetted with the leaded solder. The tin
whisker risk is somewhat dependant on the alloy used for the HASL. Tin
dendrites are from currents forming in moisture and can be avoided. White
tin is really bad, as it compresses the surface during plating and causes
rapid tin whisker growth. We've done many thousands of boards with lead free
HASL without much issue.

Gold plating (ENIG) is bad, as it is a flash of electroless gold deposits a
few atoms thick, over nickel which is plated over the copper traces. The
gold is sometimes hard to wet (BGA's for one - they are so bad we pre tin
the pads). The intention of the process is when applying solder it dissolves
the gold, thus wetting the normally difficult to wet nickel. You now have a
solder joint with a little gold in it, which if in high enough concentration
causes the solder to go brittle. The nickel to solder intermetallic layer is
poor and is prone to cracking (a well known repair problem in the repair
industry). I think the copper to nickel bond is ok, but I never really
looked into it in much detail. Other problems that can arise is weird
thermocouples from all of the dissimilar junctions. This would have been a
potential problem with some research I did awhile back where were measuring
temperature from a thermocouple with a resolution of around 1/1000 of a deg
C. The thermocouple was so sensitive you could peg the graph by breathing
heavy several feet away!

Thanks for sharing your experience, Jeff. Maybe HASL would then be an
option for us, as long as we wet everything. The latter could be a
problem in thru-hole areas such as DIN connectors.

ENIG looked a bit scary to me from the beginning so I politely declined
when they offered. The nickel-copper is generally good though for
mechanical stuff, I have used that a lot for shield mounts and the like
in order to avoid dissimilar metals touching (nickel plating the shield
studs a well).

Thermocouples would spell disaster for this circuit as it relies on
microvolt level DC paths to be stable.

 

[Kevin G. Rhoads]

The downside is you have a possibility of tin whiskers

forming in areas that have not been wetted with the leaded solder.

Per the Nasa tin whiskers info collection, it isn't just tin that
forms whiskers, it is just about every lead-free alloy and metal,
just some more than others.

 

[Jeff L]

Thanks for sharing your experience, Jeff. Maybe HASL would then be an
option for us, as long as we wet everything. The latter could be a
problem in thru-hole areas such as DIN connectors.

Proper through hole has solder wetting to the top. Problem areas could be
untented vias, mounting pads and holes, etc. Basically if its not a pad,
don't leave it exposed to become plated with solder in the HASL process.

OSP (bare copper with an organic protector) might be another option and it's
cheap, but comes with it's own problems.