Scaling Down

Someone made a microfluidic device by printing a pattern on a Shrinky Dink, shrinking that, then using it to mould the channels. [blog.wired.com]

I can't help thinking that there's an idea there that needs to be expanded upon. (no pun intended)

Edited 1 time(s). Last edit at 12/04/2007 04:45PM by Steve DeGroof.

Hi Steve,

... the idea is good --- cover a sheet of shrinky-dink with a thin slice of not shrinking material and heat the compound to bulge out and form a 3D-form.

The next logically steps are complex multilayer-sheets with separation-areas between, so when heating the printed part it would morp to a complex 3D-form with embedded cavities and bended structures.

What's not so good is the poor accuracy - the shrinking-rate is mostly determined in some percents until 2/3 of the original size, but the end-accuracy is in some percents too, so it's not a very exact method ...

Viktor

How about printing the part on stretched latex then?

... it should behave in same manor - but the accuracy isn't much better, as ticker latex-sheets arent very homogene in bulk.

Anyway, the idea is good enough to test with bulging/morphing structures

Viktor

As Viktor noted in this thread, they both deal with aspects of the same topic.
[forums.reprap.org]

The thread above seems to be talking more about fabricating Shrinky Dink templates in the RepRap, whereas was immediately more interested in how you could use structures fabricated by this method to enhance the capabilities of the device.

My idea was using this fabrication method to make a microfluidic deposition head. My polymer chemistry is rather rusty though.

Lots of rapid prototypers appear to based around the inkjet principle. I would perceive microfluidic channels to be one of the core requirements of constructing an inkjet head. Although I'm not sure how you'd insert the electrical elements, or how well a rubber inkjet head is going to work (the technique is a casting technique though, so it shouldn't be too much of a stretch to use harder materials).

Perhaps it could be used for any application where a very fine line of fluid would be useful as well... etching?

Heat-Shrinking Polymers are really cool stuff. They are kind of in the middle between thermosets (like Furfural) and thermoplastics(like the filiment we are currently using).

Actually I suggested some time ago that we look at crosslinkable thermoplastics (the things that you need to form heat shrinks) since they would allow us to extrude as a thermoplastic, recycle any mistakes/low tempature use parts, or add a crosslinker and thermoset them for high tempature applications.

If you were to patially crosslink them prior to and then stretch them during the actual build. The result would be a large part that when reheated would return to the unstretched size.

Of course they all cost about $4/lbs the last time I really looked at them.

Mike

$4/lb is not bad. we're looking at $3.50/lb for HDPE filament and $5/lb for ABS filament. last i heard CAPA was going to be $30/lb for filament. barf.

How about something that shrinks, and hardens, as it dries? Not cools.

Of course, drying a largish object evenly would become as much a challenge as figuring out the future shape of an object that's shrinking as you lay another layer on top of it.

Hi Sean,

... remember nopheads curling-thread - with highly shrinking pastes you would receive a bended pyramidal structure, when you extrude a cubicle.

So you have to fab filigrane bendable supports first, which would remove the stress between the bed and the object (this would be a solution for the curling of normal extruding too)

Viktor

CAPA filament should workout closer to $15/lb,

---

Ian
[www.bitsfrombytes.com]

I didn't say something that shrunk as it cooled. That's exactly the problem I referenced. I was thinking of something that could be extruded in a viscous solution, then boil off the solvent. If the material shrunk by a large factor as the solvent was boiled off, and if you could boil off the solvent in such a way that the object was dried evenly, you might get something that could be "printed" large, then dried to a smaller size.

Sean,

... should be possible - print a paste with a very high amount of solvent, print with a second dispenser a paste on top (and/or vice-versa for multilayers) which didn't shrink, then separate from the bed and let it dry or heat it up to evaporate the solvent.

I had shortly a similar problem with soldering glass-powder ad waterglass+water as solvent - when dispensed it was good, when firing in the oven the mess shrinked so heavy down, that the material rip partially apart and several widely separated droplets remained instead of a continuous glass-tray ...

By the way: - if you followed the thread with paperwork and unfolding STL/folding sheets to 3D-objects, then you can test making complex 3D-structures or cavities and covered/embedded capilaries by folding plain stripes of material by shrinking material in the hinges or with embedding partally heavy stress in (separated) areas, so the material bulg/morph and forms cavities too ...

Viktor

Why not print on a hot plate set on low so the plastic stays hard, but warm until finished and then cools off as one piece.

cheap: thats one of the things i want to try once i get my repstrap up and printing. i think it has great potential.