Gauging Interest: Printed Circuits on Paper

Hello,

My Name is Jonathan. I work for a large print company in the United States and I’ve been looking into ways to provide services to hobbyist like myself and professionals who want printed circuit boards without having to pay the expenses of purchasing PCB’s or trying to etch their own and still have the flexibility to re-do them fairly quickly and cheaply as part of the prototyping process. I am going to pursue developing this service, at least as a proof of concept, here at my company and was wondering if anyone would be interested in having their circuit designs printed using conductive ink. If this works I should be able to produce sizes up to 3ft x 5ft on standard white paper and possibly other materials based on the quality of the prints and the overall quality of the final product. We will need to work together to coordinate getting the PCB files transferred to me for print. The preferred format is the highest resolution PDF. If you are interested please leave your comments here and provide me with the following information:

Name:
Location:
Final Board Size:
Email Address:
Description of Project:
Comments/Questions:

Mods feel free to redirect me if this post is inappropriate for this forum.

Thanks and have a nice day.

How do you see soldering working with that? Wouldn't the paper burn?
Similarly with SMD and reflow oven - I can't see paper doing too well.
Did you have something besides paper in mind?

How to you solder to paper (or otherwise make reliable electrical & mechanical connections)?

How conductive is the ink?

If this works…

Try a simple circuit a resistor & LED.

CrossRoads:
How do you see soldering working with that? Wouldn't the paper burn?
Similarly with SMD and reflow oven - I can't see paper doing too well.
Did you have something besides paper in mind?

It's conductive ink, you stick the components onto it, using something like conductive paste.
no soldering involved I guess.

If the parts aren't soldered I would suspect the slightest flexing of the paper would lead to connectivity issues.

I've seen this concept around the web once or twice. It's a cute parlor trick but as a fairly serious hobbyist I wouldn't consider it. I assume you must be aware of how inexpensive real PCBs can be had these days.

There exists "metal-jet" to print 2 layer PCB's for years, So printing PCB and glueing components on paper could just be the next step, for prototyping it might be a great speed up.

@jsalassi
Can you elaborate on how the components are attached and where we can buy the "glue"?

Ill try to respond as best i can, positively, to the comments so far. I apologize if your question or comment was not addressed.

The goal of this project is to make a prototyping solution that's cheap, affordable and disposable. Lets say you got a few ideas for a circuit that you've bread boarded out, get a sheet of them printed and test them out before moving onto a more permanent copper clad application. If your at a convention or an event get like 1000 examples of your circuit printed up and pass them out as marketing materials.

CrossRoads:
How do you see soldering working with that? Wouldn't the paper burn?
Similarly with SMD and reflow oven - I can't see paper doing too well.
Did you have something besides paper in mind?

So we have the ability to print on much more rigid material. Its an inkjet printer basically so it supports various sizes of thickness. For the PoC (Proof of Concept) we would stick to the cheap stuff, plain white paper. I don't have any experience with printed circuits but your questions would certainly be things we would try to understand more clearly before offering a product or service. It would also help us discover the limitations of different mediums and thicknesses. Paper would likely burn if you tried to solder something to it. I was thinking copper tape or some clever way to attach both SMD and traditional components to the traces.

robtillaart:
There exists "metal-jet" to print 2 layer PCB's for years, So printing PCB and glueing components on paper could just be the next step, for prototyping it might be a great speed up.

@jsalassi
Can you elaborate on how the components are attached and where we can buy the "glue"?

In what I've read copper tape seems to be the go to adhesive for this. The other thought i had was to print the disposable circuit diagrams and the users have to figure out how to attach the components themselves. A roll of copper tape could be included if needed. What if we just created an epoxy that was filled with conductive ink or create some conductive glue?

Regarding multilayer, would sandwiching 2 sheets together accomplish the same thing? In the print business we can print double sided on the same sheet as well as a booklet type print where we print double sided and fold in the proper order and orientation to read appropriately. If we could design the circuits accordingly we could come up with some pretty interesting form factors.

I agree. I think getting a PCB is not expensive in terms of dollar value but the process is expensive. The design, error checking, and delivery is time consuming, which is inherently expensive. You could etch your own but working with harsh chemicals can be dangerous. Ive heard of people hacking inkjet printers to do the same thing which is totally cool but who still has a deskt-jet that their willing to re-purpose for this. What retail store can i walk into in the United States today and get one, or 100 of my circuit idea printed before i leave the store and would that be something you guys would pay for?

You're not being realistic.
Since this is paper, you're not talking about punching lots of holes in for thru-hole parts, only sutface mount parts.
Look at page 11 here
http://www.fairchildsemi.com/ds/MM/MM74HC595.pdf
This is a typical shift register - the pads the part goes on are 1.75mm x 0.65mm, and this part is on the larger size of pads in use.
There's no way copper tape is practical for attaching leads.
To put down conductive epoxy, a stencil would be needed, same as for solder.
Inexpensive stencils are made from 3mil or 4mil thick mylar; the solder paste is fairly "thick" (not runny) and is squeegied into place. The gel would need to be also, and then you'd have to wait for a yet to be determined time for the epoxy to cure. The solder paste, you stick it in a reflow oven at temps to 225C (for lead solder paste, and I think higher for unleaded) for a 4-5 minute cycle, when it cools, it's done. I haven't looked into conductive epoxy or other glue with that kind of property.

a coworker of mine just showed me this

http://agic.cc/about/

jsalassi:

[quote author=Jack Christensen link=topic=259267.msg1831421#msg1831421 date=1407262706]
I’ve seen this concept around the web once or twice. It’s a cute parlor trick but as a fairly serious hobbyist I wouldn’t consider it. I assume you must be aware of how inexpensive real PCBs can be had these days.

I agree. I think getting a PCB is not expensive in terms of dollar value but the process is expensive. The design, error checking, and delivery is time consuming, which is inherently expensive. You could etch your own but working with harsh chemicals can be dangerous. Ive heard of people hacking inkjet printers to do the same thing which is totally cool but who still has a deskt-jet that their willing to re-purpose for this. What retail store can i walk into in the United States today and get one, or 100 of my circuit idea printed before i leave the store and would that be something you guys would pay for?
[/quote]

I don’t see how the design and error checking can be any different; the process you suggest supposedly fills the same role as a PCB so the design process has to be pretty much similar. I’ve etched my own and I don’t consider the chemicals particularly dangerous. But with the inexpensive options available, I can get much better boards made for me. The turnaround time is the tradeoff. But if time really matters, turnaround time can of course be bought at a price. And the more boards, the less expensive it is. I would not consider paper PCBs as good for prototyping/PoC as the electrical characteristics have to be different as are the mechanical characteristics. The vast majority of today’s designs are SMT; can the paper process handle 100-pin SMT devices with 0.5mm lead pitch?

It’d be a real good idea to get some experience with existing PCB technology before investing a lot into this idea. I only see it as suitable for very low-tech, low parts-count, non-critical circuits. I’d have to ask if that’s a niche I wanted to play in.

Cost hasn’t been discussed, what is the estimated price range for this technology? As a hobbyist, I can get good quality double-sided boards for $0.25/in2. I do have to wait 2-3 weeks for them, but I’m a hobbyist so basically nothing is very urgent.

This is probably coming off badly, close-minded and negative. I don’t mean it that way, but I am being blunt and candid. Perhaps you see this a some sort of breakthrough, but I think it’s more a solution in search of a problem. As ever, I am willing to be convinced, but I’m not there yet.

maybe you should first study how a real pcb (copper layer over FR4) is made, and then research how you can replace the FR4 with a cardboard & copper with conductive ink.

i believe printing onto various surface wouldn't be a problem, even with thin traces like 0.08 or even 0.06 mil.
but I presume you will encounter other problems:

  1. how do you stick the components onto the cardboard? soldering requires a temperature over 200~220 deg C, and even higher with lead-free process (240~250C). conductive glue might be a good idea.
  2. traces got resistance too, a segment of 20mm 0.08 mil trace on 1oz copper is like 5 ohm, how well does your ink conduct? silver particles are EXPANSIVE!!!
  3. drilling, how do you create a through holes on the cardboard? you want flat surface around the hole, not some volcano-looking holes.
  4. how do you connect different layers? like through holes? well this can be done with conductive glue tho, you just squeeze the glue into the hole and make it conduct.
  5. multilayer pcb, for multilayer pcb, you got to line-up each side. discover that you drilled over the trace on the other side of the pcb after you drill is actually a pretty disappointing experience.

Well your idea is very interesting, compared to traditional pcb making process, no ironing, no UV exposure, no all those chemicals (like FeCl3)... etc.
I just hit the "print" button and it'd be ready in minutes.

btw some china companies are making 50x50mm PCBs at RMB$30 (per 5pcs), lead-free are like RMB$60~$80, gold-plated are like $130...

I have seen something similar before.
My first question was to hand him a sm atemega and ask if he could attatch that.
The glue was basicaly just some more ink.

Current capacity of the ink was miniscule.

It was i nice idea but to do anything except light alaed i thought it to me somewhat impractical.

Also paper is flexible, not much need for flexi circuits imho.

Maybe I missed where the original poster gave the conductivity of the ink?

polymorph:
Maybe I missed where the original poster gave the conductivity of the ink?

The folks with the idea on Kickstarter say 0.2 Ohms/square.

Lots of good feedback here. I appreciate the thoughts and ideas. I'll consider these things as I move forward with the project and keep this thread updated.

Ok. Good luck.

I'll be interested to hear how it goes.

I've wondered about this myself....

I'm wondering if the solution to mounting chips and keeping them there would be some kind of conductive glue, print the circuit then glue on all the components, think solder only for paper.

I actually think printing chemical-resist ink onto copper-plated FR4 is a good idea.

I'm just getting tired of the transfer (either ironing or UV exposure) things...

etching is not really that time consuming, i just throw the board into FeCl3 and go watch some simpsons, and it's ready when I'm back.