Where can I find "modified" inkjet printers?

March 26, 2005 7:32 AM

Scientists (and others) are using modified inkjet printers to print a wide range of really nifty things, such as working electronic circuits, 3D architectural models, and living tissue like organs and skin. My question is, do you have to modify one yourself, or are they for sale? Where would I get an inkjet printer that was capable of printing out a layer of resin or wax or whatever? The only one I can find is the kind that prints out edible birthday cake photos.
posted by RylandDotNet to technology (9 answers total)

What you're looking for isn't exactly a modified inkjet, rather it's a purpose-built 3D printer. I can't speak for printing circuit boards and human tissue, but in product prototyping, 3D printers that build up a 3D model in thin layers of resin / elastomers / composite etc. are used to generate mid-quality appearance & testing models.

They have dropped significantly in the last few years, but you'd still be looking at a couple thousand minimum. A quick search turned up ZCorp.
posted by id girl at 8:19 AM on March 26, 2005

Well, if it's printed circuits your after, your gonna be looking at a pretty expensive setup.

It's a little more complicated then modifying an EPSON inkjet printer. Due to the materials being jetted, you would require a piezoelectric inkjet head and EPSON is the only major commercial printer using piezoelectric heads. HP, Brother, etc. use a thermal inkjet heads.

However, I have heard of people using modified EPSON printers but almost always these devices are using either single nozzle inkjet dispensers, like the ones made by MicroFab Inc., or commercial inkjet heads like those made by Spectra Inc. or Xaar Inc.

Even if you can find someone who will sell you a complete system, it will be very expensive.

The real technological leap isn't the printer but the material being jetted. If your making a printed circuit board you essentially need a liquid that is viscous enough that it can be inkjetted but then will cure into a solid conductive trace.
This is usually organo-metallic inks or colloidal nanoparticle solutions.

Some companies have created specialised inkjet heads that can jet solder though (See the MicroFab link above).

For some more background, here are some papers on actual setups used to print conductive material (They are a little old...):

Solar Cell Contacts [PDF]

Metallic MEMS devices [PDF]
posted by toftflin at 9:10 AM on March 26, 2005

As an alternate tack, you may be able to find a university that has one and will print things for you. University of Michigan has a 3D lab with a Zcorp printer, and they will print your 3D model at pretty reasonable rates.

This page also lists a number of national firms that will provide that service, some providing online quotes.

As for buying your own, unless you're made of money, it ain't gonna happen anytime soon. This is an industry niche where a 'low cost' printer retails for $60,000.
posted by ulotrichous at 9:29 AM on March 26, 2005

There is an MIT class
posted by Good Brain at 9:32 AM on March 26, 2005

Do a google search for "Rapid prototyping"
posted by wezelboy at 11:00 AM on March 26, 2005

do you want to buy one to start your own business? I am guessing that you just want to print some stuff out? If so, there are plenty of places that will do it for you.

The general name is Stereo Lithography, and there are a variety of methods, from resin baths activated with lasers to deposit systems which use starches or plaster.

Most places require you to provide an .stl file, which is a stereo lithography standard, most 3D CAD systems will output them.

Costs add up as you want better finish, stronger structures, etc.
posted by jonah at 11:08 AM on March 26, 2005

Actually, what I had in mind is a lot simpler, closer to the edible birthday cake end of the scale than the rapid prototyping end. But this is all good info.
posted by RylandDotNet at 12:15 PM on March 26, 2005

COTS inkjet printers can be modified to print a lot of stuff; the trick is to generate an 'ink' formulation that can tolerate the printing process and won't gunk up the printing mechanism. For cells in particular, there have been a few amazingly hackish solutions using thermal inkjet printers: despite being ejected from the head at high speed in a droplet with a boiling surface, about 90% of cells survive. Though you'd probably need institutional access to get the papers, you can check out the abstract on pubmed or go to Dr. Boland's bio at Clemson University. There's supposed to be a link to his lab from the latter, but in the grand tradition of biomedical engineering web sites it is broken in a horrible way.

You should try buying a sacrificial print cartridge, cleaning it out really good, and injecting what you want to print into the reservoir. The viscosity needs to be close to that of the ink being printed, and make sure that nothing will sediment out. I think, but am not sure, that a piezoelectric inkjet will probably be your best bet for liquids that have combustible components in them. This will let you do 2d onto rice paper or the like; you'll have to kitbash the printer hardware and its drivers quite a bit, though a bet a CNC hobbyist might be able to give you some good advice. It may be possible to find people who have done the 2d step online; I remember seeing a New York Times article about a chef who uses inkjet printing to selectively pattern flavors onto his foods, and I'd imagine a few other people have tried something like this.

On preview:
Jonah, stereolithography only refers to the method you describe with the lasers. Many other systems will take .stl files but the general term is rapid prototyping, as others have mentioned. STL is a really good method for making complex shapes but it's also hideously expensive. I think a liter of the photopolymer used costs somewhere in the thousand dollar range, unless it's come down recently. The deposit systems would be more sensible for casual use.
posted by monocyte at 12:34 PM on March 26, 2005 [1 favorite]

We just recently got a bunch of rapid prototyping equipment at the architecture faculty at university of toronto, including a 3-d printer. Ours shoots out a supposedly non-toxic plastic, as opposed to some other models that basically combine cornstarch with superglue. But either way it's pretty damn expensive - the cartridges are a few hundred dollars US each, and it generally takes most of a cartridge to do a small print. I think there're plans for the lab to accept outside jobs eventually, but right now they won't even let us kids touch the new toys.
posted by slipperywhenwet at 11:29 PM on March 26, 2005