3D Printing for Science

Hi,

By way of introductions, I am a DPhil research student in Condensed Matter Physics at Oxford. Our research group focusses on the fabrication of novel quantum materials by various methods using ulta-high vacuum. As part of this it is often necessary to design and make custom parts for the experiments we wish to run and so it seems 3D printing may be a good way forward in terms of fast prototyping parts for the experiments.

So I'm seeking two things really: 1) does anyone have experience in using 3D printing in a scientific environment? I was wondering for instance about trying to print 3D crystal structures for the purposes of teaching etc. 2) Are there any recommendations regarding a good machine to start with as a test?

Having read around the Wiki pages a bit, it seems that the Prusa may be a good model to start with, for which I would need some printed 3D parts.

Thanks!

P.S. Should anyone be interested in what it is we do the group's web page is here.

If you are not interested in the challenge of building a 3D Printer your best bet is to buy a pre-assembled and tested printer from a member or one of the many people out there that sell their own custom builds.

Prusa has been refined (Mendel90, MendelMax, etc) due to the difficulties in ensuring that parts are lined up properly among other reasons.

If all you want to do is print, I say buy - you'll have enough challenges just as a user and designing your 3D models. If your department has the money, there are commercial 3D printers out there that would increase reliability and ease of use (zcorp/3d systems - more limited personal version at [cubify.com]). There is at least one other competitor I can't think of the name off the top of my head. If you are committed to building, I'm sure others more experienced will jump in and recommend the best option for your needs. Things to consider identifying: required resolution, size of objects, need for quick turn around, etc.

I've printed a bunch of protein crystal structures on my ultimaker. Here's some links:
[www.thingiverse.com]
[www.thingiverse.com]
[www.thingiverse.com]
[www.thingiverse.com]
[www.thingiverse.com]

They are the "blob version" of the crystal structure rather than ribbons structure as you can tell. I've tried printing a ribbons structure but the result was too flimsy and I couldn't get it detached from the support structure. That doesn't mean it isn't doable just the two times I tried I wasn't successful. I've also printed a bunch of small protein structure model kit pieces which could easily be modified to be like crystal unit cell model kit pieces.

[4.bp.blogspot.com]

So yea I think it can be done. I wouldn't recommend Ultimaker though, I've had a lot of trouble getting it up and running. Given a second chance I would have probably gone with prusa. Makerbot is cool and all but their hacky firmware/software is a big turnoff. At least with a reprap-styled design you know you can use marlin, which is a great firmware with awesome features. You could get fully assembled mendels for the price I bought my kit for. Although I'd say if you really don't want to screw around with tweaking you should think about UP! since their support material system seems quite effective.

Thanks for the advice! I think it is most likely that I will go down the route of trying to build one myself due to cost constraints. In the UK we have a rule imposed called full economic costing, we actually have a very high grade 3D printer in the department but research groups are charged about £100 per hour to use it and so no one can afford to do so. Hence, I'm interested in something cheap and light that we can just have a go at and see where it leads. We also have the advantage of plenty of undergrads who do placements with individual groups! I do really like the look of some of those commercial models but I doubt I would every get the expense approved.

Destroyer, I really like your protein crystals! I would be trying to make ball and stick inorganic crystal structures but I wonder how well a 3D printer would cope with printing something with so many gaps in the bulk of the object?

So in terms of building a machine from scratch are there any recommendations? The MendelMax certainly looks a lot more solid and I have access to a machine workshop to make many of the metal parts myself.

Reprap for Science