Fabbing a Circuit Board with Electrodeposition

While researching electroplating online, a really dumb idea smacked me on the forehead. I'm sure it's not possible, but it was crazy enough to convince me to create this account so that I could share it with the smart people.


Electroplating is set up somewhat like basic electrolysis; a cathode and anode submerged in a water-electrolyte solution. Metal from the anode is deposited on to the cathode.
check>

In electrolysis, I've heard that graphite rods have been used as the electrodes, because they do not corrode, yet still conduct well enough. Because modern pencil 'lead' is mostly graphite, could it be possible for the cathode to be a surface on which a the traces for a circuit were drawn, and the anode to be a hunk of copper? Could the copper be plated to a non-metallic cathode? Could this signal the end of the world?

If this is possible, then reprap could just plot out traces onto any piece it had built, and 3D circuits could be electroplated right on!

Try it!

---

-------------------------------------------------------

Hell, there are no rules here - we're trying to accomplish something.

Opportunity is missed by most people because it is dressed in overalls and looks like work.

Thomas A. Edison

Well I have heard you can electroplate copper onto zinc galvanizing spray. Might be useful for spray on n' plate circuit board fabrication.

There are techniques in industry which print slighly conductive traces and then plate them with copper.

I don't how you can electroplate though without connecting all the tracks together first.

You could mill away the bridges afterwards I suppose.

---

[www.hydraraptor.blogspot.com]

Turn the problem around and do micro EDM instead?

That way I think you could you work your way out from the center of the board disconnecting traces from the common ground.

Thank you Joe S. for starting this thread!
I have been haunted by similar ideas for some time now, so here are a few of them:

Graphite deposited by a pencil might not be the best choice of material, as it might not stick to the board very well. Zinc galvanizing spray sounds good. Conductive silver or copper paint would probably work (but more or less expensive). There is also graphite paint around. You could also draw the circuits with silver nitrate solution and expose to sunlight (this should deposit metallic silver).

Here, a problem arises: None of these candidate "paints" seems suited for beeing filled into any kind of pen, as the conductive particles would probably clog it.

You don't have to draw "bridges" between the traces if you connect them externally, for example by putting the whole board face-down on some kind of metallic sponge (f.e. steel wool?) in the electrolyte solution.

If you're still thinking about graphite... I got this thing for Xmas a few years back, and it's got big beefy graphite sticks that can be used.

[www.duluthtrading.com]#

Hope it's helpful, at the least.

I actually just posted a question similar to this. It's not electrodeposition, but, how about just mounting a resist-pen as your tool and have it draw the PCB pattern directly on the copper board? It seems like the fidelity of the trace pattern would be superior to using heat-transfer or photo sesnetive techniques.

Or, you could fabricate your own UV LED light Pen, and work with photosensetive boards that way.

We've got far too many people saying "how about" and far too few people saying "I just tried this and it...."

---

-------------------------------------------------------

Hell, there are no rules here - we're trying to accomplish something.

Opportunity is missed by most people because it is dressed in overalls and looks like work.

Thomas A. Edison

... some 20 years ago i did ( ) some "photo-plotting" of PCB's with light-sensitive coating and a green LED (no UV-LED's then ) - the main problem was finding spray-coating with the right sensitivity-range for green - the single vendor didn't survive very long ...

For drawing thin photo-lines with the green LED's in a normal XY-plotter i lathed the LED-housing in form of a thin sharp cone, then painted the cone black and grinded the tip until the light could pass with the right aperture - last inserted some LED-tips with different apertures in old HP-pens (lights when pressed down) and used them like normal plotter-pens in the width-range from 0,1mm to 2mm

This Will be the same with actual UV-LED's too. On monday i ordered 6 UV-LED's with 400nm wavelength for testing with UV-curing of fluid PMMA, but they'll be capable of activating sensitive PCB-coatings too.

But then too i wasn't fond of DIY-PCB-making with all the chemical mixtures (especially the residues), so i bought my first CNC-mill and (because of lack of any appropriate software on Atari ST) programmed my own CNC-drivers, a simple CAD/CAM and some screen-copy- and isolation-routing strategies.


Nowadays it's much easier as there are many tools and PP's for my actual CNC-mill (same vendor as with my first one, but better/faster controller and more power), so i won't start with photochemistry again - better make it with Eagle and isolation-routing when in need ...

Viktor

Edited 1 time(s). Last edit at 02/04/2009 04:40PM by Viktor.

I have tried Joe's graphite idea now, and it works.

The pencil lines didn't stick to normal PCB substrate, so I used wood instead. To make a conductive line, I had to trace it several times with the pencil. The resistance of such a trace is about 1-5KOhm per mm.

I contacted the graphite trace on the wooden board with a small clamp, then I submerged it in a solution of copper sulfate. I connected it as cathode to a single NiCd cell (~1,4V). For the anode I used a length of copper wire.

A copper coating was visible on the graphite trace after a few minutes, but to be sure I left the whole thing running for hours.

The resistance of the successfully copper-plated trace was to low to be measured. As expected, the trace was not rub-proof.


I also tried to contact several unconnected graphite traces at once using a bed of steel wool, but that didn't work.

My first attempt to the whole thing used a mix of vinegar and salt as an electrolyte, but it failed, producing only blackish brittle stuff (copper oxide?) on the cathode.

---

--Matthias Jacobs

Interesting. When you say "not rub proof" I assume it would not be bonded well to the wood, but was it solid copper or did it disintegrate when rubbed?

---

[www.hydraraptor.blogspot.com]

I think it would make more sense to selectively remove copper traces, because I can't see an easy way to add copper to a non-conductive surface. I found that by connecting one terminal of a hefty capacitor to the copper of a copper plated board, I could make ~1mm holes by tapping the other terminal against it. I think a toolhead consisting of a sharp electrode and a capacitor charger would work. Its not something I have time to try yet, but it seems like quite a cheap solution. I reckon smaller blast holes would be possible with a smaller capacitor, the voltage could always be varied to have this same effect.

@nophead:
The copper seemed to be in small crystals; stable to a light touch, but disintegrating when scratched with a fingernail. I guess mechanical stability could be improved by using conductive graphite paint instead of pencil, but I don't have any at hand.
Plating tin onto the copper and then sintering might also work.

@james glanville:
Interesting. How hefty a capacitor?

---

--Matthias Jacobs

Had another go just now: [picasaweb.google.com]#

This was using a 20V power pack connected in parallel with a 2200uf cap (I suspect the psu has some more inside.) To make the tracks I grounded the sheet of copper, and used an old nail to draw the tracks. It was a bit like arc welding, if the nail was too close or quick it stuck and had to be tugged off. As I see it:

Advantages: Cheap, easy to make toolhead, uses common blank pcbs
Disadvantages: Erosion of electrode, electrode may stick, care must be taken to work from center of board so the copper can always be grounded

Feedback would be useful, not sure how practical people think this method would be.

You are using EDM [en.wikipedia.org], which is certainly a viable way of accurately machining. The wear of the electrode is solved by using wire and feeding it from a reel.

It is normally quite slow but with just a thin layer of copper it should be reasonable.

You do have to start with PCB material, an additive process would be nicer, IMHO.

---

[www.hydraraptor.blogspot.com]

Could you extrude solder onto a pcb as a protective layer? I'm not sure how well solder resists etchant, but it might work. This is still not an additive process though. The other thought I had was extruding solder directly onto plastic sheets. I forget which plastic I tried ages ago, but i found it melted the plastic and sunk in quite nicely, with minimal smoking plastic. If this seems a workable idea i can try and dig up the board and identify the plastic.

During my appreticship in the early 70's this process was know as spark erosion, it was used mainly for creating high precision moulds for injection moulding.

I seem to remember that the spark erosion machines opertated in a bath of oil.

Im supprized how easy it was to remove the copper from the PCB was this standard 1oz copper clad fiber glass or resin bonded paper board?

Making a tool to do this would be quite interesting Im not sure how you would limit the RFI generated in the erosion process though. How the rest of or electronics would cope with that kind of interferenec.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

It was standard copper clad board. If I ever make a toolhead based on this approach, I think I would use two power supplies so the electrode could be optically isolated from the control circuitry. However, I definitely need to get my almost-darwin to print better parts before I start experimenting more.

hmm.. another idea might be to "print" low melting point alloy onto the copper, then immerse the board + alloy into copper chloride or other suitable etchant.

once done just remove board, ultrasound wash and melt off the alloy.

bonus is that the prepared board will be less likely to corrode as the bare copper is "tin" plated

i also looked into using the same alloy to make a rolled filled core polymorph "wire" for making motors and other interconnects. the alloy is brittle but if correctly heated using an induction heater will repair itself a la "Terminator" and conduct again.


email me for more info at testing _ h at the yahoo

There is a semi-additive process of making PCBs. [en.wikipedia.org]. See last paragraph under "paterning" at that site.

Briefly, the idea is to obtain blank boards with a super-thin layer of copper cladding. Add a thickness of copper (topped with tin) traces to that thin cladding. Then run the board through a brief etching process to remove the remaining exposed thin cladding (where traces are not).

The advantage is that very little etching is required to remove that super-thin layer of cladding. Not so bad for the environment.

Sorry, didn't get whole url above!
[en.wikipedia.org]

That is the process PCB-pool use, I had some boards made there recently.

[www.pcb-pool.com]

Although the starting layer of copper is half the 1oz final thickness, so perhaps not super thin.

---

[www.hydraraptor.blogspot.com]

i know of a "green" etchant (in more ways than one ) Here

Here is another interesting article regarding the additive process. It suggests that a thickness of electo-deposited copper is less conductive than rolled copper sheet. [www.mail-archive.com]

Anyway, what attracted me to your great forum here is was some discussion of possibility of using EDM (Electrical discharge machining) technology to engrave a circuit board. Rather than starting in the middle and working outwards, I was thinking of applying the ground to one edge and engraving in a scanning pattern from the opposite edge, working towards the grounded edge.

Electroplating is a good idea, and I understand we are trying to build as much of the robot on our own as possible, but won't we still need to purchase the components that sit on the board? Why not just start with copper clad boards.

I believe ABS is resistant to HCL (if not we could find some other thermo-poly), we could start with copper clad boards as the base then:

cnc the holes with the reprap
deposit ABS in where we want circuits
then simply etch the boards using hydrochloric acid & peroxide
clean ABS off
tin board & solder components

Hi, Old thread I know but it seems even google takes a while to get to my place.

I use artists fine graphite paint to coat the interior of holes in PCB for pth and vias then electroplate from a 6v source with series resistor. The resistance should be determined experimentally, I am too lazy to research proper rates so I use trial and error over the years and have settled on 120R for boards up to 4"x2" and 60R (2x120R ||) for up to 8x4.
The solution I use is a CuSO4 with a dash of HCl and H2SO4 with added Thiourea for smoothing and a drop of radiator coolant for brightening.
Various formulas are available online just dont do it too fast or the plate wll be coarse and lumpy. My process takes about an hour.

The same process can be used to copper plate anything that can be painted with the graphite, just rember to make a contiuous connection to the graphite.
Hope this helps.

... how strong is the bonding to the surface?

Could it be good enough to coat a ceramic plate with copper or gould, so that it will withstand some shear force?

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hi Viktor, never actually tested shear force but wire leads soldered into the holes stay with heat and pulling force, eg swinging board by wire while trying to test something. Also I smetimes forget the board is plating and end up with a very thick plate and have to sand some surface off if I am real lazy and do not only the holes but the whole surface of the board.

If plating plastic for sheilds, the stuff sticks enough to be drilled and bent and still adhere to surface. I dont know about gold but if it can be plated to copper I would do the copper first then gold.

I forgot to mention that the paint has to dry first, I even put it in a toster oven fo 15 mins on 120-150C to make sure it is set.

I normally wouldn't suggest chinese plastic, as it doesn't seem to print particularly well, but

[www.repraper.com] is a conductive ABS, which might conduct well enough to copper plate..

Soldering to it would be an entirely different problem.

Edited 1 time(s). Last edit at 07/16/2012 07:36PM by raldrich.