> After some more thought I decided to use the drilling process itself
> to put a graphite coating inside the hole. I discovered that by
> putting graphite powder underneath the board so that when the drill
> reaches the other side of the PCB a large amount of graphite powder is
> feed up along with fiber glass swarf. This also must 'press' enough
> powder into the hole wall, because when I did the copper platting, I
> got almost 100% success. 9 out of the 10 test holes at 1 mm diameter
> had 360 degrees of copper plating on the inside wall.
What a cool idea. Any opinions on how it compares to
the conductive ink method described in
http://www.thinktink.com/stack/volumes/volvi/condink.htm ?
I've never used the ink but it looks pretty messy. The
graphite-covered table method could be contained in a
wooden box with another piece of wood laid into the
bottom so it could be replaced when it had too many
holes.
Were you using a wooden table on your drillpress?
If so, could your 10th-hole problem be the result of
hitting no wood? Maybe the wood cut from the table
binding-up the tool provides the necessary carriage
and pressure to impregnate the fiberglass.
-greg
Adam Seychell wrote:
>
> I had an interesting idea on how to make double sided PCBs with
> platted through hole using readily available materials (or things the
> poor hobbyist can get hold of.). Using the electrical conductivity of
> powder graphite, it should be possible to coat the inside the PCB
> holes with graphite power and then electroplate with copper. I have
> managed to get some results with this. As a first test, I applied
> graphite to the board edges and showed that it could make a reliable
> conductive surface for copper platting. I then thought it should be
> possible to do the same with holes. My first attempt was to try and
> rub the powder inside the hole using my fingers. Then tap the board
> hard against the bench to remove the bulk of the graphite filling the
> holes. This method was found to be useless.
> After some more thought I decided to use the drilling process itself
> to put a graphite coating inside the hole. I discovered that by
> putting graphite powder underneath the board so that when the drill
> reaches the other side of the PCB a large amount of graphite powder is
> feed up along with fiber glass swarf. This also must 'press' enough
> powder into the hole wall, because when I did the copper platting, I
> got almost 100% success. 9 out of the 10 test holes at 1 mm diameter
> had 360 degrees of copper plating on the inside wall. One of the holes
> had a small section on inner wall that didn't plate, although still
> had good through hole electrical connection. Next I'll experiment with
> 0.8mm and 0.6mm holes, although for my application only 1.0mm is
> needed.
>
> I am using standard 1.6mm double sided fiber glass PCB material. The
> copper plating electrolyte I'm using is saturated CuSO4 (available
> from a garden nutrient supplier) and 5% sulfuric acid (available from
> my car battery ). I will continue to experiment with the graphite
> powder, and try find out best method on making plated through hole PCB
> in your home. Any ideas ?
>
> Adam
While it sounds like a neat trick as a professional plating engineer in
the industry I have grave reservations as to the reliability of your process.
I have a strong suspicion that repeated thermal cycling of the substrate
during operation is going to lead to multiple opens over time. This may
be useful for prototyping and proof of concept boards but I for one would
never rely on it for circuits you want to run for any continued duration.
I would also suggest to anyone exploring this as a viable process to
consider it only for low power circuits and never for something like
power supply boards. Heat will rear it's ugly head quickly for a process
like this and I can just about gaurantee failures in operation during high
load situations.
To the guy who was using AC for copper plating... Wrong !!!
Use an extremely well regulated dc power supply or a battery.
AC ripple will kill plating process faster than most anything
I know of. The current density for CuSo4 plating should be
kept to around 10 - 15 amps per sq. ft. of exposed copper
surface plating area. The ratio of the solution should be
kept at 10 -15 grams/litre of CuSo4, roughly 10% sulfuric acid
and the rest dionized or distilled water for the given volume
of the tank you are plating in.
--
Ken Tyler
mailto://tyle...@pacbell.net
http://home.pacbell.net/tylereng/links.htm
P.S. Graphite is an intermetallic contaminate for copper and will eventualy
migrate into the copper deposite jeopardizing it's integrity.
> To the guy who was using AC for copper plating... Wrong !!!
> Use an extremely well regulated dc power supply or a battery.
> AC ripple will kill plating process faster than most anything
> I know of. The current density for CuSo4 plating should be
> kept to around 10 - 15 amps per sq. ft. of exposed copper
> surface plating area. The ratio of the solution should be
> kept at 10 -15 grams/litre of CuSo4, roughly 10% sulfuric acid
> and the rest dionized or distilled water for the given volume
> of the tank you are plating in.
Why does it need to be well regulated?
Is that 10-15 amps / sq ft for the anode (coming from) or the cathode
(going to)?
I'm going to save this post, always wondered what the proper electrolyte
was, in my childhood it became obvious that table salt (NaCl) in tap
water wasn't quite right.
But it did conduct.
:)
--
Gene
I immersed the pcb in CuSO4 solution (which I got from a garden shop),
inserted a thin copper wire into the hole to be plated, and applied
voltage. I used enamelled wire so that copper would only plate from
the end of the wire. The end of the wire was positioned to be midway
thru the hole.
The idea was that the copper would begin plating on the edges of the
hole on the top and bottom sides of the pcb. This would build up a
coating of copper that would start growing towards the center of the
hole until the two sides join.
However, the current density needed at the end of the wire needed to
plate a hole in about 5s (to a certain thickness) was that high that
the chemical processes going on would be a bit violent. As a result,
some kind of precipitate would form at the end of the wire forming an
insulation barrier to any current flow. I tried using rectified 240VAC
fed from a variac and through a lightbulb to act like a high impedance
current source, but the insulation breakdown/buildup effect was repeated
many times per sec (produced a loud crackling sound), and polluted the
electrolyte with brown precipitate.
How long did your holes plate? What shape was the electrode and where
was it positioned? What current/voltage was used?
--
Regards,
Russell
Russell Shaw wrote:
> I tried ideas like this myself, but was unsuccessful (I didn't use
> graphite tho).
>
> I immersed the pcb in CuSO4 solution (which I got from a garden shop),
> inserted a thin copper wire into the hole to be plated, and applied
> voltage. I used enamelled wire so that copper would only plate from
> the end of the wire. The end of the wire was positioned to be midway
> thru the hole.
>
> The idea was that the copper would begin plating on the edges of the
> hole on the top and bottom sides of the pcb. This would build up a
> coating of copper that would start growing towards the center of the
> hole until the two sides join.
>
> However, the current density needed at the end of the wire needed to
> plate a hole in about 5s (to a certain thickness) was that high that
> the chemical processes going on would be a bit violent. As a result,
> some kind of precipitate would form at the end of the wire forming an
> insulation barrier to any current flow. I tried using rectified 240VAC
> fed from a variac and through a lightbulb to act like a high impedance
> current source, but the insulation breakdown/buildup effect was repeated
> many times per sec (produced a loud crackling sound), and polluted the
> electrolyte with brown precipitate.
>
> How long did your holes plate? What shape was the electrode and where
> was it positioned? What current/voltage was used?
>
'This sounds like a real experimenter'
The idea you have on getting the copper to grow throughout the hole and meat
in the center will not
work. the copper seems only to deposit to existing conducting surface. Maybe
there are tricks
to force control of copper deposits by pumping electrolyte through hole, but
I'm not sure.
The electrolyte bath I used was a special home made glass tank (270 mm
*270mm * 35 mm) and
held together with silicon sealant.
I place two copper sheets on each wall of the tank and both connected to the
positive supply.
The drilled PCB sits in between the copper sheets at a distance about 15 mm
from each. I attach
the positive clamp to both sides of the PCB copper surface.
For a 50 mm * 100 mm board I applied about 1.5A (voltage was about 1.6V)
After abut 60 min the copper layer was thicker than the manufactured copper
on the PCB. I
didn't get to measure the actual thickness but, it was enough to show it
worked.
Adam
cheers
Adam
P.S thanhs for that cool PCB site, it answerd many of my PCB questions.
Greg Wiley wrote:
> Adam Seychell <ASey...@vet.com.au> wrote in message
> news:37952310...@vet.com.au...
>
> > After some more thought I decided to use the drilling process itself
> > to put a graphite coating inside the hole. I discovered that by
> > putting graphite powder underneath the board so that when the drill
> > reaches the other side of the PCB a large amount of graphite powder is
> > feed up along with fiber glass swarf. This also must 'press' enough
> > powder into the hole wall, because when I did the copper platting, I
> > got almost 100% success. 9 out of the 10 test holes at 1 mm diameter
> > had 360 degrees of copper plating on the inside wall.
>
Not comparable with yours. Brilliant! Where do you get the graphite
powder from? You can't just grind up a pencil.
--
Regards, John Woodgate, OOO - Own Opinions Only.
Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
Did you hear about the hungry genetic engineer who made a pig of himself?
PLEASE DO ****NOT**** MAIL COPIES OF NEWSGROUP POSTS TO ME!!!!
No, the deposit doesn't grow like that. It's more likely to grow across
the hole than down the walls.
>
>However, the current density needed at the end of the wire needed to
>plate a hole in about 5s (to a certain thickness) was that high that
>the chemical processes going on would be a bit violent.
And how!
> As a result,
>some kind of precipitate would form at the end of the wire forming an
>insulation barrier to any current flow.
You are electrolysing the water content. The precipitate is copper
oxide.
> I tried using rectified 240VAC
>fed from a variac and through a lightbulb to act like a high impedance
>current source, but the insulation breakdown/buildup effect was repeated
>many times per sec (produced a loud crackling sound), and polluted the
>electrolyte with brown precipitate.
Not surprising. You have **far too many volts**. Current-limiting is not
effective. You have found what happens - pulsating current, boiling
electrolyte, rampant side-reactions etc. etc. The brown precipitate may
contain copper blasted into powder by cavitation in the liquid.
Could you please explain why copper deposited in the way described is
bound to be different from copper deposited by 'legitimate' (more
complex and costly) methods?
It 'might' be different - more porous, less ductile etc., but I doubt
that you can predict that as inevitable.
Pin 2 Hot wrote:
>
> Ken wrote:
>
> > To the guy who was using AC for copper plating... Wrong !!!
> > Use an extremely well regulated dc power supply or a battery.
> > AC ripple will kill plating process faster than most anything
> > I know of. The current density for CuSo4 plating should be
> > kept to around 10 - 15 amps per sq. ft. of exposed copper
> > surface plating area. The ratio of the solution should be
> > kept at 10 -15 grams/litre of CuSo4, roughly 10% sulfuric acid
> > and the rest dionized or distilled water for the given volume
> > of the tank you are plating in.
>
> Why does it need to be well regulated?
I was thinking one thing and saying another :) It should be well filtered.
Any ac ripple carried on the dc supply causes as many problems with plating
processes as it does electronic circuits. The best power supplies are those
that run on 3 phase input. The 3 phases being out of phase from each other
helps create the best line powered ripple free currents for plating that I
personally have used and is common practice in the industry. The only thing
better that I know of is a battery which have their own obvious drawbacks.
If you can't afford a 3 phase poser supply, and what hobbyist can, then
at least use one with the best ripple specs you have on hand.
> Is that 10-15 amps / sq. ft for the anode (coming from) or the cathode
> (going to)?
At the part or the cathode in this case. The surface of the part itself
should be seeing the current. It is recommended that you maintain a 1:1
surface ratio, anode to cathode, for the best distribution of the plated
deposit.
Additionally a .03% phosphorized copper is recommended for the anode
material for best deposit characteristics i.e. ductility and tensile
strength. Brittle deposits can form if the wrong chemistry, anode material,
and current densities are used. Most proprietary plating baths also use a
brightening agent that both improves the appearance of the deposit and it's
physical properties where ductility and tensile strength is concerned.
Brittle or too thin deposits are common causes for barrel cracks caused
by z axis expansion of the substrate materials. These in turn causes opens
in the hole where that occurs. This is more of a problem where there are
internal layers but can affect double sided boards as well.
Considerable care should be exercised if using graphite as a means of
seeding the hole wall for the copper deposition. Copper plating baths are
highly susceptible to organic and metallic contamination and loose carbon
particles can and will create rough undesirable copper deposits. Some
of this can be minimized by plating a dummy panel ( a piece of copper pc
board) at low current densities say about 3-5 amps/sq ft. overnight. This
will help plate some of the graphite out of the plating bath. An even better
approach is make a new solution often and discard the old after just a
few short runs. Just make sure you get as much of the carbon particles
of the board before plating it that you possibly can.
One other issue with the graphite is it's poor initial adhesion to the
hole walls and it's very thin nature. I would recommend when you first
start the electro plating process you start at a very low current density.
If you hit it with a large current initial you run a very real risk of
burning it off much like as if it were a fuse. After 5 minutes or so in
the plating bath the current can be increased to the 10 - 15 amps/sq. ft.
range since the newly deposited copper will now be able to handle the
higher current densities than the thin graphite layer could.
The hobbyist should shoot for a minimum of .0015 in. per side of the hole
wall and 0.002 in. is better. Over plating is almost as bad as under plating
and should be avoided as well. You can use drill bits as a hole size gauge.
For example if your drilled hole size is 0.050 before plating then after
plating a 0.045 drill should fit snugly in the hole but without being forced.
Air agitation in the bath helps replenish the solution in the holes during
the plating process and will yield more uniform hole wall deposits. It is
highly recommended that you provide some form of solution or part agitation
or you will get larger deposits on the surface than the hole wall and you
run the risk of burning the deposit which you should avoid at all cost.
To the guy taking the acid out of your battery... STOP. The battery acid
is highly contaminated with lead metal and lead sulphates which are VERY
bad for copper plating baths. See note above about metallic contaminates.
I would suggest that if you can't get access to commercial grade sulphuric
then at least try to get fresh sulphuric from an auto parts dealer. It is
often sold for motor cycle batteries which are shipped dry and need the
acid added to activate them.
> I'm going to save this post, always wondered what the proper electrolyte
> was, in my childhood it became obvious that table salt (NaCl) in tap
> water wasn't quite right.
>
> But it did conduct.
> :)
>
> --
> Gene
For the record I am against the whole idea of using graphite in this process
and would not ever consider plating a board this way. It is too full of potential
risks to initial and long term quality that it is just not worth the effort.
As I said in another post maybe for proof of design but never take is seriously
for something you want to run reliably and for any lenght of time. Point to
point wiring would be more reliable.
Interesting idea. How well does the copper adhere considering the "loose"
graphite underneath? Have you noticed any peeling?
OK, but why? Obviously the current density spec doesn't need to be that
exact. You said yourself it should be 10 to 15 amps / sqft. So why does it
matter if it repeatedly bounces between say 11 and 14 A/sqft every few mS?
John Woodgate wrote:
>
> <37952310...@vet.com.au>, Adam Seychell <ASey...@vet.com.au>
> inimitably wrote:
> >Any ideas ?
>
> Not comparable with yours. Brilliant! Where do you get the graphite
> powder from? You can't just grind up a pencil.
You could try defloculated Acheson graphite - a suspension of sub-micron
graphite particles in water, sold as Aquadag. It is a trademark
http://www.joglosemar.co.id/coxinga/mark_aqa-arg.html
shows
> AQUADAG 01 Acheson Industries (Europe) Limited A 02307 302307 May-19-2003
Should be available from chemical supplier. We used it a lot at
Cambridge Instruments, as it gets painted on non-conducting specimens
to stop them charging up when they are being examined in an electron
microscope, and you can obviously get it from suppliers of electron
microscopy consumables.
They also carried silver DAGs (which were colloidal silver in water or
in some non-aqueous solvent) but that is rather expensive.
The professionals use "electrodeless plating compounds" or "electroless
nickel"
to deposit an initial conductong layer on the sides of the holes after
drilling,
before plating and eventual photo-etching.
Not sure that this will help much.
Bill sloman wrote:
> The professionals use "electrodeless plating compounds" or "electroless
> nickel"
> to deposit an initial conductong layer on the sides of the holes after
> drilling,
> before plating and eventual photo-etching.
>
> Not sure that this will help much.
The professionals seed the hole with paladium metallic ions that become
sites for the autocatalytic copper reduction process which deposites from
35 - 60 millionths of an inch of copper thus metalizing the hole walls.
It is then electoplated with further depostites of copper, photo imaged
with a plating resist, pattern plated to the desired thickness, tin/lead
is then plated as an etch resist, then the photo resist is removed prior
to the final etching process. There are different places in the process
where some steps are done differently but this is pretty much the "standard"
process in the industry.
>I am using standard 1.6mm double sided fiber glass PCB material. The
>copper plating electrolyte I'm using is saturated CuSO4 (available
>from a garden nutrient supplier) and 5% sulfuric acid (available from
>my car battery ). I will continue to experiment with the graphite
>powder, and try find out best method on making plated through hole PCB
>in your home. Any ideas ?
I presume that you have contacted several different commercial
board manufacturers to see how the "pros" do it. There's a good
chance that you can use the same techniques as easily as you can with
the graphite powder.
You're already doing the electroplating, which is IMHO the
hardest part. All you really need is a handle on the chemicals used
commercially to get the first conductive layer on the bare board
material in the hole.
Jim
No problem, we're on the same page. That is what I meant, also.
> Any ac ripple carried on the dc supply causes as many problems with plating
> processes as it does electronic circuits.
What sort of problems would a bit of ripple cause?
> If you can't afford a 3 phase poser supply,
Humorous typo!
> ...and what hobbyist can, ...
You'd be surprised.
> > Is that 10-15 amps / sq. ft for the anode (coming from) or the cathode
> > (going to)?
>
> At the part or the cathode in this case. The surface of the part itself
> should be seeing the current. It is recommended that you maintain a 1:1
> surface ratio, anode to cathode, for the best distribution of the plated
> deposit.
I had no idea of that. I've just assumed a square yard of copper anode,
and one tiny 4-40 X 3/4 screw for a cathode hanging down in the bath by
its lonesome, would be ok. I've never tried to plate a board, and don't
really intend to. My experience with plating is more like "Gee, that ol'
rusty Crescent wrench would look cool if it were copper plated, lemme
wirewheel it first. Where's my car battery charger?" And then there'd be
alligator clips hooked to the wrench, and a whole bunch of pennies. :)
But your input on this is quite informative, even facinating. Thanks.
> Considerable care should be exercised if using graphite as a means of
> seeding the hole wall for the copper deposition.
So just sticking a #2 pencil in a drill by the eraser end, and jamming
the point into a hole while spinning, isn't going to do it?
> One other issue with the graphite is it's poor initial adhesion to the
> hole walls and it's very thin nature. I would recommend when you first
> start the electro plating process you start at a very low current density.
> If you hit it with a large current initial you run a very real risk of
> burning it off much like as if it were a fuse. After 5 minutes or so in
> the plating bath the current can be increased to the 10 - 15 amps/sq. ft.
> range since the newly deposited copper will now be able to handle the
> higher current densities than the thin graphite layer could.
This makes sense, graphite is a weak conductor (they make resistors out
of it), so there would be a voltage gradient under current towards the
middle of the hole as you got away from the surface copper. Starting out
with lower currents would give the hole time to increase conductivity
and then you could bring up the current without the voltage drop across
ohms of carbon.
> The hobbyist should shoot for a minimum of .0015 in. per side of the hole
> wall and 0.002 in. is better. Over plating is almost as bad as under plating
> and should be avoided as well.
What happens if it gets too thick?
> To the guy taking the acid out of your battery... STOP. The battery acid
> is highly contaminated with lead metal and lead sulphates which are VERY
> bad for copper plating baths. See note above about metallic contaminates.
> I would suggest that if you can't get access to commercial grade sulphuric
> then at least try to get fresh sulphuric from an auto parts dealer. It is
> often sold for motor cycle batteries which are shipped dry and need the
> acid added to activate them.
Been saving both up for years, old and new (old is for something
different, new is because I can't bear to throw out anything that may
have a use, and I've been through a lot of 14AH bike batteries over the
years, they always give you more 1.265 S.G. than will fit in the
battery). What S.G. of H2SO4 solution would make a good electrolyte for
plating copper? Hope it's not more than 1.265, or I'll have to get out
the hotplate and still. :)
Again, thanks again for the in-depth tutorial on this subject, it is
much appreciated. I crave learning things, even if I don't have an
immediate use for them. You never know.
--
Gene
I did not try much for plating onto the board with external electrodes
because I guessed (sort of from intuition) that there would be little
field inside the holes to attract copper plating, unless some kind of
pressure feed of fluid thru the holes was done.
The only other solution (pardon the pun) I could think of was some
kind of self-plating chemical.
Adam Seychell wrote:
>
> Russell Shaw wrote:
>
> > I tried ideas like this myself, but was unsuccessful (I didn't use
> > graphite tho).
> >
> > I immersed the pcb in CuSO4 solution (which I got from a garden shop),
> > inserted a thin copper wire into the hole to be plated, and applied
> > voltage. I used enamelled wire so that copper would only plate from
> > the end of the wire. The end of the wire was positioned to be midway
> > thru the hole.
> >
> > The idea was that the copper would begin plating on the edges of the
> > hole on the top and bottom sides of the pcb. This would build up a
> > coating of copper that would start growing towards the center of the
> > hole until the two sides join.
> >
> > However, the current density needed at the end of the wire needed to
> > plate a hole in about 5s (to a certain thickness) was that high that
> > the chemical processes going on would be a bit violent. As a result,
> > some kind of precipitate would form at the end of the wire forming an
> > insulation barrier to any current flow. I tried using rectified 240VAC
> > fed from a variac and through a lightbulb to act like a high impedance
> > current source, but the insulation breakdown/buildup effect was repeated
> > many times per sec (produced a loud crackling sound), and polluted the
> > electrolyte with brown precipitate.
> >
> > How long did your holes plate? What shape was the electrode and where
> > was it positioned? What current/voltage was used?
> >
>
> 'This sounds like a real experimenter'
> The idea you have on getting the copper to grow throughout the hole and meat
> in the center will not
> work. the copper seems only to deposit to existing conducting surface. Maybe
> there are tricks
> to force control of copper deposits by pumping electrolyte through hole, but
> I'm not sure.
> The electrolyte bath I used was a special home made glass tank (270 mm
> *270mm * 35 mm) and
> held together with silicon sealant.
> I place two copper sheets on each wall of the tank and both connected to the
> positive supply.
> The drilled PCB sits in between the copper sheets at a distance about 15 mm
> from each. I attach
> the positive clamp to both sides of the PCB copper surface.
> For a 50 mm * 100 mm board I applied about 1.5A (voltage was about 1.6V)
> After abut 60 min the copper layer was thicker than the manufactured copper
> on the PCB. I
> didn't get to measure the actual thickness but, it was enough to show it
> worked.
>
> Adam
--
Regards,
Russell
Brown precipitate.
> > As a result,
> >some kind of precipitate would form at the end of the wire forming an
> >insulation barrier to any current flow.
>
> You are electrolysing the water content. The precipitate is copper
> oxide.
> > I tried using rectified 240VAC
> >fed from a variac and through a lightbulb to act like a high impedance
> >current source, but the insulation breakdown/buildup effect was repeated
> >many times per sec (produced a loud crackling sound), and polluted the
> >electrolyte with brown precipitate.
>
> Not surprising. You have **far too many volts**. Current-limiting is not
> effective. You have found what happens - pulsating current, boiling
> electrolyte, rampant side-reactions etc. etc. The brown precipitate may
> contain copper blasted into powder by cavitation in the liquid.
I don't think pulsating current is a problem. I used a variac, so I could
vary the level from no action to lots of action.
> --
> Regards, John Woodgate, OOO - Own Opinions Only.
> Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
> Did you hear about the hungry genetic engineer who made a pig of himself?
> PLEASE DO ****NOT**** MAIL COPIES OF NEWSGROUP POSTS TO ME!!!!
--
Regards,
Russell
Ken wrote:
Thanks for the tips, Ken.
Although I filtered most of the solid lead particles from the car battery acid using
fine filter paper, It still looked slightly discolored (grayish). I'll have to keep
looking around for a supplier of sulfuric acid. Even though its one of the most
common chemicals used the industry its still hard to get hold of in small quantities
for the general hobbyist.
I also am interested in plating tin (or tin/lead) on PCB. I have been told either two
electrolytes tin/lead-tetraflouroborate or methanesulfonate, are used for plating. I
have a lab setup in my home garage for copper plating and had great success. However
the tin/lead plate is required for an PCB etch resistant. Are tetraflouroborate or
methanesulfonate dangerous chemicals to handle ? Maybe there are other tin/lead salt
solutions more commonly available/safer that may work for plating ? For example could
I make ZnSO4 from electrolysis of Zn nuggets in sulfuric acid and then use for
electroplating.
Adam
Pin 2 Hot wrote:
>
> Been saving both up for years, old and new (old is for something
> different, new is because I can't bear to throw out anything that may
> have a use, and I've been through a lot of 14AH bike batteries over the
> years, they always give you more 1.265 S.G. than will fit in the
> battery). What S.G. of H2SO4 solution would make a good electrolyte for
> plating copper? Hope it's not more than 1.265, or I'll have to get out
> the hotplate and still. :)
If your going to concentrate H2SO4 in your home make sure you have adequate
ventilation . I tried this one day in my garage and it produced some extremely
corrosive fumes that effected the surface of just about every metal thing in
sight. At low concentrations (car battery strength) it boils almost like water.
But after a about 10 min when the liquid becomes viscous, it boils at much higher
temperature, about 250 C from memory. It also has a very violently boil,
splattering like water in hot cooking oil. My bench and equipment wasn't pretty
site after the incident and I don't recommend doing such a thing without using a
fully contained flask combined with fume extraction equipment. I did this many
years ago when I was leaning about concentrated sulfuric acid in my high school
chemistry class.
Adam
That's why the hotplate, and with a long extension cord way out back,
outside. Erlenmeyer-type flask with glass thermometer through rubber
stopper in top, and side vent line down into a bucket of water. Nothing
should excape, it's only a small amount of steam coming out, no need to
boil it hard. And yes, concentrated H2SO4 is rather "oily". Do you think
the rubber stopper may break down? It doesn't around lower
concentrations when cool.
Decades ago, I used this same flask the other way around, Budweiser beer
boils at about 97C. You could turn a six-pack into about 3 ounces of the
kind of stuff you could run run your car on. But you have to boil
gently, you dont want to crack C2H5OH into CH3OH, it'll make you blind
or kill you.
Thanks for the advice anyway, might have done me a lot of good, or kept
me from doing a whole bunch of bad.
--
Gene
Pin 2 Hot wrote:
> > Any ac ripple carried on the dc supply causes as many problems with plating
> > processes as it does electronic circuits.
>
> What sort of problems would a bit of ripple cause?
Powdery, loose, and burnt deposites.
> > Considerable care should be exercised if using graphite as a means of
> > seeding the hole wall for the copper deposition.
>
> So just sticking a #2 pencil in a drill by the eraser end, and jamming
> the point into a hole while spinning, isn't going to do it?
Having learned to plate in a professional environment I am naturaly
skeptical of any process that is not in common use. This being one of
them.
> > The hobbyist should shoot for a minimum of .0015 in. per side of the hole
> > wall and 0.002 in. is better. Over plating is almost as bad as under plating
> > and should be avoided as well.
>
> What happens if it gets too thick?
This becomes a problem where the current density at the surface of the
panel is greater than inside the hole. The plating forms a knee at the
interface which can lead to cracking. This can lead to lifted lands causing
stresses on the IC leading to chip failure. Specifialy you want a nice
uniform deposit that is both ductile and has a high tensile strength.
This allows it to flex with the constant z axis expansion seen in laminate
materials and survive due to it's strenght. Over plating acts like a
rigid rivet. After a time the laminate will only tolerate being held in
place for so long and the plating usualy loses the battle.
> What S.G. of H2SO4 solution would make a good electrolyte for
> plating copper? Hope it's not more than 1.265, or I'll have to get out
> the hotplate and still. :)
The specific gravity of common commercial H2SO4 is 1.84 or 66 Be'.
> Again, thanks again for the in-depth tutorial on this subject, it is
> much appreciated. I crave learning things, even if I don't have an
> immediate use for them. You never know.
>
> --
> Gene
--
OK, but why? I don't understand the mechanism behind it.
> > What S.G. of H2SO4 solution would make a good electrolyte for
> > plating copper? Hope it's not more than 1.265, or I'll have to get out
> > the hotplate and still. :)
>
> The specific gravity of common commercial H2SO4 is 1.84 or 66 Be'.
You don't plate at that concentration, do you?
--
Gene
They are both highly poisonous, but so are all soluble tin and lead
salts. I don't think those are especially dangerous, like subject to
skin absorbence, explosion, vapour emission etc.
> Maybe there are other tin/lead
>salt
>solutions more commonly available/safer that may work for plating ? For example
>could
>I make ZnSO4 from electrolysis of Zn nuggets in sulfuric acid and then use for
>electroplating.
You don't need electrolysis! Sulfuric acid will do a job on zinc nuggets
with *no* encouragement. BUT copious amounts of hydrogen are evolved,
and that can easily go BANG! I would advise against, because ZnS0/4 is
not noted as a plating salt.
Electroplating is like cooking: you do it by taste (NO!!!). You need
just a pinch of acid - no, a dash. You certainly don't want an SG of
1.265 due to the acid alone.
Pin 2 Hot wrote:
> > Powdery, loose, and burnt deposites.
>
> OK, but why? I don't understand the mechanism behind it.
Let me dig around through the literature tomorrow for the official
explaination.
> > > What S.G. of H2SO4 solution would make a good electrolyte for
> > > plating copper? Hope it's not more than 1.265, or I'll have to get out
> > > the hotplate and still. :)
> >
> > The specific gravity of common commercial H2SO4 is 1.84 or 66 Be'.
>
> You don't plate at that concentration, do you?
10% - 12 by volume for acid cu plating.
> Maybe there are other tin/lead salt solutions more commonly
> available/safer that may work for plating?
Soluble tin/lead (or other heavy metal) compounds are quite toxic in
general.
> For example could I make
> ZnSO4 from electrolysis of Zn nuggets in sulfuric acid and then use for
> electroplating.
As John Woodgate already pointed out, the sulphuric acid will take care of
the Zinc without further encouragement. But the process will give you zinc
(Zn) sulphate, no tin (Sn) or lead (Pb) compound.
--Daniel
--
"The obvious mathematical breakthrough would be development of an easy
way to factor large prime numbers." -- Bill Gates, "The Road Ahead"
One of the reasons why I love the Usenet so much are exchanges of almost
literal quality like the one above. Where else can one have those genuine
hobbyist-professional clashes? Personally, I'm partial to the pencil
approach.
Still laughing,
Daniel Haude wrote:
>
> On Thu, 22 Jul 1999 02:15:31 -0700,
> Ken <tyle...@pacbell.net> wrote:
> > Pin 2 Hot wrote:
> > >
> > > So just sticking a #2 pencil in a drill by the eraser end, and jamming
> > > the point into a hole while spinning, isn't going to do it?
> >
> > Having learned to plate in a professional environment I am naturaly
> > skeptical of any process that is not in common use. This being one of
> > them.
>
> One of the reasons why I love the Usenet so much are exchanges of almost
> literal quality like the one above. Where else can one have those genuine
> hobbyist-professional clashes? Personally, I'm partial to the pencil
> approach.
>
> Still laughing,
> --Daniel
Would you purchase a computer mother board constructed with the "pencil"
technique ?
He who laughs last,
> Would you purchase a computer mother board constructed with the "pencil"
> technique ?
There are two things I can say to that:
1) I don't give a damn how it was made as long as it works. Of course,
more likely than not the manufacturer will be out of business before the
warranty expires, so there is nothing I can do when it breaks no matter
how it was made.
2) I respect and value your professional opinion and knowledge about the
matter of PCB manufacture. The process that you describe, however, does
not sound more or less reliable than the pencil method by itself -- it
just happens to be "the way it is done" which is enough proof to me that
it works reliably. There is also the possibility that the pencil method
(or using graphite as a 'primer' in general) works just as reliably but is
not feasible on a large scale production due to time/cost consideration.
Frankly, I don't have much of an opinion about this at all (lacking both
knowledge and experience). I just loved that pencil-in-a-drill suggestion
and your answer to it.
Yes. What is a #2 pencil? Here in UK, we have grades from 6B (very soft)
to 7H (scratches diamonds), with HB at ground-zero, and the excellent
but little-used F grade off to the side. They vary in clay content, and
6B is a pretty good conductor.
No, but I'm seriously considering it for stuff I make myself, some of
which is special-purpose test equipment, with serious applications. It
seems likely to be no less reliable than stripboard with 'klingon'
interconnect wiring.
There is also the possibility that it's a genuine new idea that would
have earned the originator $$$ if he had not published it here already.
Compared with the standard industry process that Ken described, it
sounds as though the potential time and material saving would justify
quite a lot of research to *make* it work reliably.
#2 is the "normal" pencil. Higher numbers are harder. You sometimes see 2
1/2, but I don't recall ever seeing anything above 3, which make fairly
light lines on paper unless you really press hard. For the PC board
purpose, I think you would want the softest you could get. I used to use a
#1 pencil for making resistors many moons ago. These were specialty pencils
made for marking the electronically scored tests where the machine sensed
the conductivity of the mark. Today's tests work optically, so these
pencils may be harder to come by. Perhaps artists use this stuff regularly?
John Woodgate wrote in message <+DSc9CBO...@jmwa.demon.co.uk>...
><slrn7pe481...@insitu.physnet.uni-hamburg.de>, Daniel Haude
><ha...@physnet.uni-hamburg.de> inimitably wrote:
>>One of the reasons why I love the Usenet so much are exchanges of almost
>>literal quality like the one above. Where else can one have those genuine
>>hobbyist-professional clashes? Personally, I'm partial to the pencil
>>approach.
>
>Yes. What is a #2 pencil? Here in UK, we have grades from 6B (very soft)
>to 7H (scratches diamonds), with HB at ground-zero, and the excellent
>but little-used F grade off to the side. They vary in clay content, and
>6B is a pretty good conductor.
the pencils are split in half, graphite is laid in, and the 2
halves are glued back together.
nathan
> <slrn7pe481...@insitu.physnet.uni-hamburg.de>, Daniel Haude
> <ha...@physnet.uni-hamburg.de> inimitably wrote:
> >One of the reasons why I love the Usenet so much are exchanges of almost
> >literal quality like the one above. Where else can one have those genuine
> >hobbyist-professional clashes? Personally, I'm partial to the pencil
> >approach.
>
> Yes. What is a #2 pencil? Here in UK, we have grades from 6B (very soft)
> to 7H (scratches diamonds), with HB at ground-zero, and the excellent
> but little-used F grade off to the side. They vary in clay content, and
> 6B is a pretty good conductor.
A #2 pencil in Canada and the USA is yout average every day pencil.
Equivalent to HB in the Netherlands, as well as -I presume- the UK.
Marcel
Better still, mount one of those thin (0.3mm I thinks) pencil leads
between metal contacts (that aren't likely to melt) connected to a
gruntish power supply. The lead can be made to glow brilliantly
white hot and it becomes pliable and droops like a long piece of
solder. After about 30s it extinguishes. On examination of the taper
in its diameter, you can see that the carbon has actually evapourated
away.
"Al, N2NKB" wrote:
>
> Have you ever wondered how a pencil is made?
> Do you think that a long thin drill cores out the wood?
> Is the wood real or composite?
> Closely examine your pencils now, and you will discover how they
> are made. I was surprised last week when I took up the task.
> Al
>
> John Woodgate wrote in message <+DSc9CBO...@jmwa.demon.co.uk>...
> ><slrn7pe481...@insitu.physnet.uni-hamburg.de>, Daniel Haude
> ><ha...@physnet.uni-hamburg.de> inimitably wrote:
> >>One of the reasons why I love the Usenet so much are exchanges of almost
> >>literal quality like the one above. Where else can one have those genuine
> >>hobbyist-professional clashes? Personally, I'm partial to the pencil
> >>approach.
> >
> >Yes. What is a #2 pencil? Here in UK, we have grades from 6B (very soft)
> >to 7H (scratches diamonds), with HB at ground-zero, and the excellent
> >but little-used F grade off to the side. They vary in clay content, and
> >6B is a pretty good conductor.
> >--
> >Regards, John Woodgate, OOO - Own Opinions Only.
> >Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
> >Did you hear about the hungry genetic engineer who made a pig of himself?
> >PLEASE DO ****NOT**** MAIL COPIES OF NEWSGROUP POSTS TO ME!!!!
--
Regards,
Russell
--
Regards,
Russell
Oh, paper! Our draftsmen use 7H pencils with the stone tablets. (;-)
They have to, the tablets won't go through the printer.
Burned, actually. Into CO2.
That was Thomas Edison's original incandescent lightbulb filament, some
cotton thread inpregnated with carbon black (soot from a kerosene lamp).
Once he figured out to put it in a vacuum to keep the carbon from
oxidizing, it worked rather well. (Don't know why he didn't try
tungsten, he tried just about everything else).
As the originator of the "#2 pencil" idea, it was made "tongue in
cheek". I'm glad many of you got a "guffaw" out of it, but it was just a
reference to the graphite powder-in-the-hole plan that proceeded it.
But if someone really wants to try it, a #1 pencil has a higher graphite
to clay ratio, you can even drill the holes on the board with a taper
drill with the same angle that matches the angle of a sharpened pencil
(it'd be easier to insert component leads into a funnel). The burnishing
action of a rotating graphite/clay tip would smooth out the rough edges
of the fiberglas, and leave a smooth shiny semiconductive surface for
the copper to stick to. Reversing the "graphite burnishing" direction
(reverse the drill rotation a couple of times) would further serve to
break off any glass hairs sticking into the hole, for a smoother and
more flaw-free plated finish.
The #2 pencil idea was a bit of off-handed levity, but under scrutiny it
just may work.
To respond to some other things in this thread:
--------------------
Pencil leads go from #1 to #4. 4 is for very light drafting, but most of
you probably don't remember having a full arsenel of leads for your
mechanical pencils.
--------------------
If I were in this for the money, I might have withheld the pencil idea.
;)
But my idea of usenet is that it can be very effective for
"brainstorming". And there is no greater satisfaction then when a group
of people bounce ideas around, and we collectively meander our way to
brilliance. Everyone goes for the ride, and we all learn. Hence my
badgering Ken with a lot of questions to draw out some knowlege, for the
common good. Being rich doesn't mean having money, it means having
knowlege. And Ken (among others) has filled in some holes in my
understandings of everything that is the universe.
--------------------
Ken, if you use about 10% of the concentrated 1.84 SG H2SO4 in your
plating solution, I calculate an SG of 1.084, about like a dead car
battery, is this about right? Or is SG of H2SO4 in H2O non-linear
compared to percentage?
--------------------
--
Gene
I'm reading this thread on and off, but suddenly I was
inspired by a blown fuse lying on my table here. The inside
is
completely covered with shiny metal. Would that be a $$$
idea, sticking a piece of thin wire through a hole and shoot
30 amps through it ?
Met vriendelijke groeten,
Frank Bemelman
(reageren per email ? verwijder dan de 'x' uit mijn
emailadres)
> I'm reading this thread on and off, but suddenly I was
> inspired by a blown fuse lying on my table here. The inside
> is
> completely covered with shiny metal. Would that be a $$$
> idea, sticking a piece of thin wire through a hole and shoot
> 30 amps through it ?
Why not? Metal sputtering is used for lots of stuff, but doing it in a
vacuum may help.
--
Gene
The wood is real. They let it grow around the lead.
This is a bit long winded
I have used a small plating line which costs in the about £5000 while this
is beyond the reach of hobbyists the actual process is quite simple someone
with a bit of mechanical ability could easily make something similar maybe
by using a frame to hold plastic gallon cans with the tops cut off,
Then making a simple frame (like a ladder) with some threaded bar and some
2"x1" timber battens
This frame would sit over the cans and the PCB would be clamped to each
treaded bar
and the frame moved over and back to agitate the PCB
The agitation could be performed by an Auto wiper motor or similar
The chemicals work better if heated but provided the room is not very cold
they will still work okay
I drill the PCB then plate the holes
I then use an LPKF machine to mill out the PCB if you are etching the PCB
you will need to apply photo resist to the holes to stop the etchant from
attacking the PTH copper
(If you etch the PCB then try to plate it, Copper will deposit over all the
PCB shorting out the tracks)
The system uses 5 tanks
Tank1, Weak soloution of sulphuric acid
used to clean the PCB
Tank2, Water
Tank3, Palladium catalyst
This is the critical tank, copper will not plate easily onto FR4 material
this catalyst
speeds up the process, virtually anything dipped in this tank can be plated.
plastic, paper, wood etc
Tank4 salt remover
Tank5 Plating tank
This consists of a soloution of copper sulphate at each side of the tank are
2 sets of 2 electrodes (slabs of copper 6" x 3" x 1/4") hung on hooks to two
bars which pass over the tank at each side. These are the Anodes Over this
tank is an agitating frame that looks like a ladder, with a bar which is
connected by a cable this is the cathode the PCB is clamped to this frame
and the frame oscillates over and back above the tank
This tank uses a variable current of up to 25A per side for a PCB 12"x12" at
about 6V
This is the only hard bit the current needs to be variable unless you only
use size of PCB
the cost of the chenmicals is as follows
Tank1 £30 for 5 litres
Tank2 distilled water
Tank3 £180 for 5 itres
Tank4 £30 for 5 litres
Tank5 £30 for 5 litres
If anyones intrested let me know
Regards,
Tom Allen
>A pencil connected to a 10Ampish power supply by its ends can make for
>some rather spectacular smoke effects!
>
>Better still, mount one of those thin (0.3mm I thinks) pencil leads
>between metal contacts (that aren't likely to melt) connected to a
>gruntish power supply. The lead can be made to glow brilliantly
>white hot and it becomes pliable and droops like a long piece of
>solder. After about 30s it extinguishes. On examination of the taper
>in its diameter, you can see that the carbon has actually evapourated
>away.
>--
>
>Regards,
> Russell
Ah... That takes me back to university. I used 0.5 mm leads across the
binding posts of a fairly skookum power supply. Crank the voltage all the
way up and the current limit all the way down. Then turn the current limit
up until the middle of the lead glows a dull red. As the graphite
evaporates from the middle, the resistance increases and the lead will
start glowing white hot. After a while, the lead will break and an arc
will form. At this point the light intensity shoots up by about an order
of magnitude. It burns like this for about 5 seconds before the gap gets
too big and the arc extinguishes. Now that was an educational experience
:-)
More likely oxidized away. The melting point of carbon is quite high. Now
if you could only use this technique to crystalize the carbon ...
Actually, I've just found it in the Farnell catalogue (order code 451-071) it's
made by Multicore and called "Copper Set".
Check out : http://www.multicore.com/product-index.html
Tom Allen wrote:
> Apologies cos' I can't access all of this thread, but I take it someone was
> looking into the possibility of plating their own PCB's
>
> This is a bit long winded
>
Snip...
--
Dr John Mitchell
IRC in Biomedical Materials
Queen Mary and Westfield College
University of London
Mile End Road
London E1 4NS
Tel 0171 975 5547/5574
Fax 0181 983 1799
Thanks Tom
Jon
Usualy HCL at 5 - 10% to help persipitate out the stannous tin present
in the paladium bath. Tin and paladium are natualy occuring together
in nature.
--
Ken Tyler
mailto://tyle...@pacbell.net
http://home.pacbell.net/tylereng/links.htm
So why not just get a through-hole kit (Farnell/Multicore)? For
a man of your means the cost is trivial...
Walter
Disclaimer: My employer is not responsible for the above.
If you want to email me, please use a valid From: address.
Pin 2 Hot wrote:
> Why not? Metal sputtering is used for lots of stuff, but doing it in a
> vacuum may help.
>
> --
> Gene
I was discussing this thread with our engineering mgr. today and I can
only describe the look on his face as one of sheer horror and disbelief.
I rather enjoyed it because it must have mirrored my own on a number
of occasions while reading through some of these messages.
I'm laughing to hard to go on sorry....
John Mitchell wrote:
>
> Don't know if anybody's mentioned this yet but I recall seeing somewhere (don't
> remember where) a kit consisting of tiny little hollow rivet type things. Rivet
> thing goes through hole in PCB, component lead goes through hole in rivet (or
> not), solder both sides and BINGO! Anyone know where this came from?
>
> Actually, I've just found it in the Farnell catalogue (order code 451-071) it's
> made by Multicore and called "Copper Set".
>
> Check out : http://www.multicore.com/product-index.html
Yeah, that process was used commercialy about 30 years ago. It is also the
primary reason that plated thru holes were inveneted because of their terrible
reliability.
Ralph & Diane Barone wrote in message ...
>In article <3797D6FD...@martin.com.au>,
>Russell Shaw <rus...@martin.com.au> wrote:
>
>>A pencil connected to a 10Ampish power supply by its ends can make for
>>some rather spectacular smoke effects!
>>
>>Better still, mount one of those thin (0.3mm I thinks) pencil leads
>>between metal contacts (that aren't likely to melt) connected to a
>>gruntish power supply. The lead can be made to glow brilliantly
>>white hot and it becomes pliable and droops like a long piece of
>>solder. After about 30s it extinguishes. On examination of the taper
>>in its diameter, you can see that the carbon has actually evapourated
>>away.
Isn't that stuff very costly? And does it have a short storage life?
You might be able to make it work reliably in a nitrogen atmosphere.
Oxygen is bad news, I suspect. You can get the nitrogen by burning
alcohol in a closed volume of air until it goes out, then passing the
gas over soda-lime to dry it and take out the CO/2.
Yes, so I've learned. That means that a #1 pencil would be softer and
more conductive. I wonder how a #1 compares with a 6B, electrically?
I suspect he balked at the idea of trying to draw it into a fine wire.
It's horrible stuff to work with. Probably also excessively costly, in
those days. Soot was much cheaper!
> <3797930A...@bright.net>, Marcel <mvu...@bright.net> inimitably
> wrote:
> >A #2 pencil in Canada and the USA is yout average every day pencil.
> >Equivalent to HB in the Netherlands, as well as -I presume- the UK.
>
> Yes, so I've learned. That means that a #1 pencil would be softer and
> more conductive. I wonder how a #1 compares with a 6B, electrically?
I don't know if this is true, but I have been told pencils using the HB
markings are available in the US in stores selling artists supplies. Never
actually looked for them, though.
How does HCl manage to precipitate stannous tin, when SnCl/2 is very
soluble?
I've only seen HB/B/4H type markings on Canadian pencils - dunno what
a #2 pencil is...
--
Peter Bennett VE7CEI
GPS and NMEA info and programs: http://vancouver-webpages.com/peter/index.html
Newsgroup new user info: http://vancouver-webpages.com/nnq
John Woodgate wrote:
>
> <3798976D...@pacbell.net>, Ken <tyle...@pacbell.net> inimitably
> wrote:
> >Usualy HCL at 5 - 10% to help persipitate out the stannous tin present
> >in the paladium bath.
>
> How does HCl manage to precipitate stannous tin, when SnCl/2 is very
> soluble?
I have no idea. It is what the literature on the process relates to me.
John Woodgate wrote:
> Isn't that stuff very costly? And does it have a short storage life?
Yes, no.
Peter Bennett wrote in message <379be90f....@newshost.ucs.ubc.ca>...
><slrn7pe481...@insitu.physnet.uni-hamburg.de>, Daniel Haude
><ha...@physnet.uni-hamburg.de> inimitably wrote:
>>One of the reasons why I love the Usenet so much are exchanges of almost
>>literal quality like the one above. Where else can one have those genuine
>>hobbyist-professional clashes? Personally, I'm partial to the pencil
>>approach.
>
>Yes. What is a #2 pencil? Here in UK, we have grades from 6B (very soft)
>to 7H (scratches diamonds), with HB at ground-zero, and the excellent
>but little-used F grade off to the side. They vary in clay content, and
>6B is a pretty good conductor.
When I have a question, I consult the "be-all" and "end-all",
the world wide web.
An Alta-Vista search turned up a site that should keep the
discussion going.
http://www.pencilpages.com/articles/grades.htm
Jim
Aw, guys. Don't tell me I'm the only one who's ever seen "HB" on a pack of
mechanical pencil leads. (Always wondered what it meant.)
--
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Cliff Sharp | "Speech isn't free when it comes postage-due." |
| WA9PDM | -- Jim Nitchals, founder, FREE |
+-+-+-+-+-+-+-+-+-+- http://www.spamfree.org/ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
The discussion isn't about mechanical pencil leads - you can find those in a
dizzying array of varieties, usually in a display that has a reference chart
telling what each grade is intended for. The discussion was about wooden
pencils (at least that is my understanding) that seem to come is a very
restrictive and geographically dependent selection.
No. There is a surfactant in there, maybe not much. DAG is
'Deflocculated Acheson Graphite' and it is a colloidal solution. The
graphite is not truly dissolved, but dispersed in very small particles
indeed, each surrounded by surfactant molecules that prevent the
particles from settling as a precipitate.
It comes in many grades, most of which have quite a high electrical
resistance. For plating holes, you would need a pretty low-resistance
grade.
Richard Kilsby wrote:
>
> In article <SnaX0cAi...@jmwa.demon.co.uk>, John Woodgate
> <j...@jmwa.demon.co.uk> writes
> >It comes in many grades, most of which have quite a high electrical
> >resistance. For plating holes, you would need a pretty low-resistance
> >grade.
> >
> IIRC the latest "greener" technology (no Palladium) uses an electrically
> conductive polymer as the seed layer down the holes.
Do you know of a location where more information is available on this
technology ?
> The drilled boards are conditioned in a hot bath of Potassium
> Permanganate solution to make the epoxy glass ready to accept the
> polymer and presumably also leave a good key for the Cu subsequently
> plated onto the seed layer. I think the conductivity does not need to be
> that high since once a layer of Cu is on top it does not matter.
> Some plating lines use a pulse technique when plating really tiny holes.
> Bit like charging\rejuvenating batteries?
> Rick Kilsby
The permanganate process is an old one and has two purposes. First and
foremost it is to remove epoxy drill smear from the exposed edges of
the copper inner and outer layers. It also as you mentioned "conditions"
(charges the polymers) the epoxy substrate which later attract the seed
material long enough to catalyse the electroless copper plating process.
Pulse DC plating has it's attractions for both high speed plating and
as you say for holes that have a high aspect ratio (small holes thick board).
The principle is that pulsing the current with a given duty cycle of on/off
cycles gives the hydrogen that forms inside the hole walls during the plating
process a chance to be released back into solution. Otherwise the formation
of hydrogen bubbles creates voids or vacates the solution from within the
hole preventing plating from occurring.
High dispersion nozzles are generally used along with this process
impinging directly onto the boards surface. This keeps fresh solution in
the holes and helps dispel the build up of hydrogen gas bubbles within the
holes. The increase of solution flow through the holes and the subsequent
removal of hydrogen both help to increase the current throw into the hole
and aids in evening the distribution of the plating.
In Britain, the same grading scheme applies to both leads and wood or
plastic pencils.
IIRC there is no electroless Cu in this process. The copper clad boards
with conductive polymer down the holes are photo masked with dry film
resist and then electroplated with Cu. Then SnPb. Then stripped of
resist. Then the exposed Cu areas are etched away. Voila plated through
board. (I have missed out all the rinsing, micro etching, rinsing,
rinsing, rinsing, rinsing etc stages for clarity, oh and rinsing!
The boards prior to plating have black holes! I think the process is
called "Black Copper" or "Black Oxide"
I will check Monday.
Rick Kilsby
Any residual KMnO/4 will be very keen to mop up any hydrogen it can.
There's no fundamental problem with DAG or any colloidal graphite *if*
you get the right grade. But I doubt if it's necessary. The idea of
mechanically impregnating the epoxy with graphite powder is said to work
and it is certainly worth pursuing.
If water is added to anhydrous KMnO4 the reaction can be quite violet!
Rick Kilsby
John Woodgate wrote:
> Any residual KMnO/4 will be very keen to mop up any hydrogen it can.
With all of the processes that follow that stage there is little or no
chance that any will be present. The KMn0/4 process is followed by a
nuetralizing stage, several rinses, micro etches, a desensitizing stage,
catalyst, and rinsing in between all of these. Then before it would
see the plating stage it has also gone to the photo printing room
and been processed in the developing solution. From there it goes down
another pre-clean line prior to the acid copper plating. Like I said
not much chance for any residual permanganate by this time.
Yeah, but the ratings (#1 / #2 / HB / etc.) are for the "lead" in the
pencil, not the pencil itself, which usually has the same shape and is
made of the same wood.
Typical of the situation where a contaminant would do no harm. If it
did, there would be 1% residue that nothing would shift!
>John Woodgate wrote:
>
>> <slrn7pe481...@insitu.physnet.uni-hamburg.de>, Daniel Haude
>> <ha...@physnet.uni-hamburg.de> inimitably wrote:
>> >One of the reasons why I love the Usenet so much are exchanges of almost
>> >literal quality like the one above. Where else can one have those genuine
>> >hobbyist-professional clashes? Personally, I'm partial to the pencil
>> >approach.
>>
>> Yes. What is a #2 pencil? Here in UK, we have grades from 6B (very soft)
>> to 7H (scratches diamonds), with HB at ground-zero, and the excellent
>> but little-used F grade off to the side. They vary in clay content, and
>> 6B is a pretty good conductor.
>
>A #2 pencil in Canada and the USA is yout average every day pencil.
>Equivalent to HB in the Netherlands, as well as -I presume- the UK.
>
>Marcel
I'm from Canada, and as far as I know, an HB is the plain old,
middle-line pencil. I'm assuming a #2 would be a 2H, as it's simply
less likely to smudge than an HB, much less so than a 2B, and thus
would be better for writing/doing diagrams/whatever. So no, a #2 is
NOT the same as an HB...
John Woodgate wrote:
>
> No. There is a surfactant in there, maybe not much. DAG is
> 'Deflocculated Acheson Graphite' and it is a colloidal solution. The
> graphite is not truly dissolved, but dispersed in very small particles
> indeed, each surrounded by surfactant molecules that prevent the
> particles from settling as a precipitate.
>
> It comes in many grades, most of which have quite a high electrical
> resistance. For plating holes, you would need a pretty low-resistance
> grade.
>
The graphite powder I used for electroplating holes was from grinding up some
old carbon rods used for arc gouging. I also tried graphite flakes used for add
to oil for improved lubrication. Both forms of graphite powder worked equally
well, the trick is all in getting the drilling process to coat the PCB material
with graphite. At the moment I'm still setting up my copper and tin plating
apparatus. Then I'll be able to experiment more with graphite power method. The
low resistance is not necessary. The initial plating current is kept low
(100mA/sqr ft) Even if its several megaohms per millimeter on surface then ions
will still get attracted to the surface (will just take longer time). Once
coated with some copper the surface will dramatically reduce in resistance and
the current can be turned up to full.
Adam
Let's just wait for your results. A long time could be a very long time.
On second reading, it sounds like the boards you're speak of have a
limited number of through-plated holes and are of rather low density.
If this is the case, wouldn't the old technique of using soldered-in
eyelets rather than through-plating make practical sense? (This comes
from a guy who absolutely detests eyeleted boards!) Other solutions
are simply to solder pads on both sides of the board to component leads
for vias, or simply use jumper wires either across or through the board.
For medium to high-density boards, the through-hole plating solution
was perfected at least twenty years ago, but unfortunately doesn't lend
itself to home use in most situations. That's why small PC shops exist,
and using them when you need thru-plated holes on your boards makes sense.
(It does to me when you describe plating thru-holes one-at-a-time using
a wire centered in the hole! Good grief, try that with a #68 or #72
drill hole!)
Harry C.
--------------------------------------------------------------------------
Adam Seychell (ASey...@vet.com.au) wrote:
:
: I had an interesting idea on how to make double sided PCBs with
: platted through hole using readily available materials (or things the
: poor hobbyist can get hold of.). Using the electrical conductivity of
: powder graphite, it should be possible to coat the inside the PCB
: holes with graphite power and then electroplate with copper. I have
: managed to get some results with this. As a first test, I applied
: graphite to the board edges and showed that it could make a reliable
: conductive surface for copper platting. I then thought it should be
: possible to do the same with holes. My first attempt was to try and
: rub the powder inside the hole using my fingers. Then tap the board
: hard against the bench to remove the bulk of the graphite filling the
: holes. This method was found to be useless.
: After some more thought I decided to use the drilling process itself
: to put a graphite coating inside the hole. I discovered that by
: putting graphite powder underneath the board so that when the drill
: reaches the other side of the PCB a large amount of graphite powder is
: feed up along with fiber glass swarf. This also must 'press' enough
: powder into the hole wall, because when I did the copper platting, I
: got almost 100% success. 9 out of the 10 test holes at 1 mm diameter
: had 360 degrees of copper plating on the inside wall. One of the holes
: had a small section on inner wall that didn't plate, although still
: had good through hole electrical connection. Next I'll experiment with
: 0.8mm and 0.6mm holes, although for my application only 1.0mm is
: needed.
:
: I am using standard 1.6mm double sided fiber glass PCB material. The
: copper plating electrolyte I'm using is saturated CuSO4 (available
: from a garden nutrient supplier) and 5% sulfuric acid (available from
: my car battery ). I will continue to experiment with the graphite
: powder, and try find out best method on making plated through hole PCB
: in your home. Any ideas ?
:
: Adam
:
:
Clifton T. Sharp Jr. <cli...@corecomm.net> wrote in article
<379A3101...@corecomm.net>...