Ink-Jet Printing Custom-Designed Micro Circuits 37
Nerval's Lobster writes "Researchers have demonstrated a technique that produces inexpensive, functional electrical circuits that can be printed using about $300 worth of materials and equipment, including generic inkjet printers. The technique, developed by researchers from Georgia Tech, the University of Tokyo and Microsoft Research, allows circuits to be printed onto irregularly-shaped materials or almost anything able to go through the paper feed on a printer designed for consumers. The chief advantage of the technique is the ability to print circuits using silver nanoparticle ink rather than relying on the thermal-bonding technique called sintering, which is time-consuming and can destroy delicate base materials. Researchers were able to print new circuits in about 60 seconds on almost any material that could go through the printer, though resin-covered paper, PET film and glossy photo paper worked best, while sheets of canvas cloth and anything magnetic were ineffective. Once printed using silver ink on flexible base material, the circuits can be attached to existing hardware by simply laying or taping them in place and making connections using conductive tape or conductive glue. (Soldering would destroy the underlying material.)"
This computer is held together with tape and glue! (Score:2)
Tape and glue. LOL. Does this have any real application outside of rapid prototyping?
Re: (Score:3, Informative)
flexible, odd shapes that are resin encased. Hell think about scaling the units up. You could print a god damn PCB layers instead of the current etching methods. Less Polution and hazardous chemicals needed.
Other possibilities are as they said, flexible plastics (shit used for most cheap keyboards/game controllers) and don't forget OLED displays. Same technique can be used to print them (already being tested).
Re:This computer is held together with tape and gl (Score:5, Informative)
Rapid prototyping IS a real application. And from there, extremely low runs of custom parts. This would have been useful to me before I sent for 10 PCBs that had two wires crossed, because the normal ink printed tests didn't show the problem, as I couldn't run electricity through them. All I could do was test mechanical compatibility.
They went shopping? (Score:3)
They went shopping, bought a silver pen for handwriting electrical circuits [gizmag.com] and attached it to a printer???? (Although a plotter would be a better choice)
Ironing (Score:2)
Sounds a lot better than the home-brew technique I've use a bunch of times in the past: print the pattern on glossy photo paper using a laser printer, use a clothes iron to transfer the toner off the paper to the copper breadboard, then etch in acid bath. The part that usually screws up is transferring the toner from the paper to the circuit board using an iron.
Re: (Score:2)
Sounds a lot better than the home-brew technique I've use a bunch of times in the past:
Dry film negative photo resist is available on ebay (or in New Zealand from me), briefly it's used thus: cut to size, adhere to cleaned board, expose to UV (sunlight fine) with negative artwork (tracks transparent), develop in weak washing soda solution, etch, strip in stronger washing soda solution. No need to work in a dark room just don't do it in front of a window, normal household lighting is fine.
For more details, see my tips for using Dry Film Negative Photoresist [gogo.co.nz]
Yes (Score:1)
But the inkjet cartridges will costs thousands of dollars each ...
Or terrorists (Score:1)
>Hobbyists could also use the setup – which cost about $300 – to produce calculators, thermostat controls or other electronics as well, Abowd said.
Or terrorists could......
I know lets BAN inkjet printers
Re:Or terrorists (Score:4, Funny)
Can't solder to it (Score:5, Informative)
So they managed to make a flexible printed circuit that can't stand soldering. Not too useful.
There are lots of ways to make printed circuits. Etching them photographically is cheap, simple, and produces consistent quality, so that's how it's done commercially. The iron-on transfer thing some hobbyists use isn't that reliable; a substantial number of boards will be defective. There are little desktop milling machines for making circuit boards.
Nobody does that much any more. Commercial board making services [expresspcb.com] take in a file on line and send back a board by FedEx. Prototype board prices today start at $28, so there's not much incentive to do it yourself. You get good quality and plated-through holes to connect traces on opposite sides of the board. The plating-through process is a mess to do on a small-scale basis, but cheap in bulk.
Re: Can't solder to it (Score:2)
Re: (Score:2)
Re: (Score:2)
That's only due to the limitations of the materials they used: had they used a flexible material that could stand the heat of soldering they could solder the parts. Or if they used a printer that had a straight path through it they could use conventional PCB material.
Commercial board making services are great, I use them - but when you want to test a prototype right now this evening, it would be great to have something that doesn't require the usual toner transfer/UV type processing and ferric chloride etch
Other optoins (Score:2)
It might be useful in a few cases, but for most applications the quick turn pcb houses like expresspcb can turn a file into a circuit that is express mailed back in a couple of days for $100 with no human intervention. Since those are standard PCB materials they are reasonable prototypes for real systems.
If these circuits could be soldered it would be a lot more interesting. Too many modern components are in tiny packages that couldn't really be connected with tape or conductiv glue.
Re: (Score:2)
Apparently you haven't seen Z-Axis tape [3m.com] yet.
The tape in action [sparkfun.com]
Re: (Score:1)
with the price of Copper continuing to rise, who will be able to afford copper based PCB's in the near future? Silver has been consistently in the 5-12 dollar per ounce range and copper demand has reached the point that it's starting to hit the same levels with the expectation that copper will eventually level out at 100 per ounce or 10x the historical price of silver.
Silver also has the interesting property of being a much better conductor then copper along with being even more maleable, thus easier to wor
Re: (Score:2)
Um...silver's been above $20 per troy ounce for a while now. Copper, OTOH, has been closer to $3-$4 per pound. Copper has had a run up in
Except that conductive glue is insanely expensive (Score:3)
You're probably looking at $25+ or more per project just for that stuff.
I was pricing it out once because I have an essential hand tremor that prevents me from being able to solder.
Cool! ...er, wait... (Score:1)
screw circuits; it's gates that count (Score:2)
This would be far more interesting if they could produce even low-performance transistors. But I suspect you'd want to start out with a flatbed, and you'd wind up focusing on non-flexible devices that you could build up through many layers. Interestingly, big, low-performance transistors would change some of the typical features of VLSI: you could do incremental testing (before layering on more circuits - perhaps even printing replacement devices if certain already-printed components didn't work. You'd p
Current state of the art (Score:3)
My group has been looking into this for a couple of months.
Lots of laboratories have achieved inkjet-built circuits in the past, to the level of "proof of concept". To date, all of them require exotic materials or expensive materials or have relatively high trace resistance, or some combination of the three. None are suitable for low-end hobbyist application yet. The project from the article is a good starting point for interesting research.
The issue with silver ink is cost: silver ink is massively expensive, even by inkjet standards. To date AFAIK, no one has been able to lay down copper traces with good (meaning: low) trace resistance suitable for prototype boards. A lot of people are working on this.
Inkjet printers can be easily modified to accept thicker material by mechanically raising the head transport mechanism. This usually involves cutting something apart and inserting shims and spacers (machine-screw washers work well), but this is not terribly difficult. Then cut away the angled paper feed mechanism (that bends the paper from vertical to horizontal) and add horizontal rails to guide the media through the unit. Also not terribly difficult. You can then print on just about anything: phenolic, glass, plastic, &c.
Inkjet print heads use one of two mechanisms: thermal and ultrasonic. Thermal vaporizes some of the ink to accelerate the droplet, while ultrasonic uses a piezoelectric mechanism to "squeeze" the droplet out. Almost all printers use thermal heads, Epson being the notable exception. Check the specs to see if the unit you're using has the type of head you want & if your ink can withstand being vaporized without clogging the pores.
Clean unfilled cartridges are available from InkSupply.com [inksupply.com] for experimenting, and you may need a cartridge chip resetter [slashdot.org] to reset the counter to "full". You can directly lay down etch resist by using a wax-based ink such as Mispro Yellow [inksupply.com].
I've got a modified printer next to me. You can use it to print just about any liquid onto any flat, thin material... and not just conductive traces. You can print fluorescent dyes, or solvents that make microchannel arrays in CD-rom disks (place in spinner to force a liquid through the microarray channels). A colleague at MIT claims that they are printing biologicals as well; ie - laying down micro-organisms on patterned nutrients. (NB: I don't know that the microorganisms are inkjet printed, his may be a hybrid system.)
Lots of potential for interesting research here.
Re: (Score:2)
Those are some very interesting observations.
What's the smallest etch resist you've been able to work with?
Most of my recent projects have involved small surface mount components. It seems like most non-prepackaged radio modules all use QFN. So, trace size and accuracy are a big hindrance to at home board creation.
I don't make boards (Score:2)
What's the smallest etch resist you've been able to work with?
I don't make boards with my printer, my post was an overview of things other people are doing with the inkjet process. Google "inkjet PCB" will return lots of hobbyist sites that talk about it, such as this one [hackaday.com].
I understand that laying down etch resist is a bit harder than my post would imply. The inkjet is accurate enough that the drops form a mosaic of circles with voids between, so the board has to be heated while printing (or after) to get the wax to flow-cover continuously. Also, prepping the board tak
Antennas (Score:2)