Cheap Laboratory Liquid Handling

I am a graduate student at the University of Missouri. Today me and two others spend several hours transferring small amounts of liquid from a hundred small containers into another hundred small containers. I thought "This sucks, I have better things to do with my time!" Then I thought of RepRap.

RepRap is basically a small liquid-handling system. Several companies sell flexible liquid-handling robots designed to transfer liquids from one small container to another. Basically they are robots which use micropipettes with disposable tips to withdraw liquid from one place and deposit it into another. These robots cost thousands of dollars, most or all are over $10,000 and are quite large and for fairly specific applications and use non-standard consumables (tips, reagent containers, etc.).

Has anyone thought of using RepRap as a platform to do exactly this, only being much cheaper? From how I see it (and I am not an engineer) you would simply replace the syringe with a micropipette and add a servo to eject the tip into a waste bin between samples. You can buy electronic pipettes which could theoretically be hacked into the system to allow for programmable adjustment of the volume to be withdrawn/dispensed (~$500). This would be an incredible project for some undergraduate engineering students and would allow for exciting collaboration between chemistry/life science researchers. I work in a lab, and I can think of dozens of labs (and I am sure there are thousands) which would buy a system like this if it were under $1000, reliable, and easy to use. See [www.youtube.com] (about 1:00 in) for an example of what I am talking about, though this machine is far more capable (and expensive!) than what we need (the chores I have in mind would be comparatively small, but would still save valuable man-hours and frayed nerves).

I am not saying this would be easy, but I think the RepRap project has already developed nearly all of the necessary technology.

Any thoughts on this? I am glad to answer any questions.

this is a good idea but the major issue with devices of this type is liability insurance. the cost of most is not in the electronics or programing or anything else. so most buisnesses hospitals or universities that might need something of this nature would not be able to buy something that was not UL approved or bonded/insured. They could make their own but then would be responsible for the liability facor themselves. which i guess is the point here . as long as the end user is willing to accept their own liability then using the developed processes in whatever way they need it is a good thing.

I think we need to look at third world countries to adopt this sort of technology. Just watched a TED Talk by Ernest Madu. He was talking about providing world class health care in poor countries by implementing different administrative practices and modalities in health care centers.

[www.ted.com]

One of the things they did in the Heart Institute that he heads was to buy the equipment to make a given chemical--not totally sure what it was but sounded like some sort of radioactive thing for imaging--instead of buying it. A dose here in the States costs over $200. When made in house it costs $2.

That's the kind of thinking that allows our RepRap to be used in the kind of environments Clayton suggested. I'm down for it but have no idea how to implement it or get it accepted in an actual lab or hospital environment.

Demented

Good point, I guess I see this as something labs would put together themselves. I could also see a lab purchasing the materials (perhaps more than needed for only one? )and having a volunteer build the device.

Or it could be made more simple for a non-super engineer if a working device was built and had the kinks worked out.

Also, let me be clear how I see this used. Our lab studies plants, other labs I know that would be into this study plants, or bacteria, or other non-medical things. There is no big liability involved in handling these samples. No one will die if something messes up. The worst that would happen would be an angry grad student. Labs which have super-valuable samples, or where the samples are involved in important medical applications, they will just go ahead and buy the expensive machine.

I can see a lab using this as a more flexible solution, especially if it is cheap.


Can anyone think of any problems with this idea?

The device would have to have a little bit of power pushing-down in order to attach a pipette tip, and would need to be able to sense the bottom of a container, say by going down until it can't anymore, and then backing off a tiny bit to withdraw liquid. Is the current RepRap hardware capable of this?

I think I am going to see if there are any undergrad engineering students at my school who may need a cool capstone project to do.

Yeah, a cheap cartesian robot would have a lot of uses in a lab environment, especially if it's easy to program. I could see the extruder head being replaced with a pipette manipulator. You'd probably be able to keep most of the electronics as-is and just reprogram the PIC / Arduino and write some new host software.

One idea I have kicking around in the back of my head is to develop a paint head. Same sort of thing: build the head, reprogram the microcontroller, write new host software.

This is one of the major advantages of Reprap that most have overlooked. The core system is great for artists and for hobbists and will maintain a major focus on versitility.

But for more serious applications there exists the possibility of making specialized reprap like robots that are not in fact repraps.

Take your liquid transfer system as one example. Add a second arm that transports a load cell, or a light gauge, or any type of sensor and you start having a means of running a massive number of tests for some property. This is the brute force research method of drug development that has always been far too expensive for non-drug applications, at least until reprap.

OR use the reprap to build parted molds and then build a reprapped injection system for investment wax. Pack in plaster, bake the wax out and then pour aluminum. Use that to create layers of wiring that can then be layed onto a build to form basic circuits. Heck, build a standarized ground bar into each layer and have a spring loaded magazine that drops the pre-built layers of circuity onto a running build.

OR use the reprap to build vented molds, then build a reprapped magazine for composite blend injection, use a gas catayst to cure the composite and then eject it from the mold. 100X increase in build speed and a lot larger list of possible materials. Use heavy filler loads for dirt cheap builds and exotic composites for speciality applications.

Or combine all of these ideas, use a casting wax in the reprap, plaster cast tooling for the injection system, then use the tooling to make massive numbers of small simple test robots out of heavyly filled materials that cost next to nothing.

You get the idea.

A lot of pattern and job shops are going to jump on this as a way to speed up and improve existing sytems rather than as an end unto itself.

Mike

The views expressed in this post are my own and in no way reflect those of my employer.