I’m interested in building a liquid handler:
From the image you can see 4 “cores” and for this specific machine I can scale up to 12 liquid handling cores .
Each core:
– 2 steppers
– 1 string potentiometer
– 1 linear actuator
– 2 physical switches
Primary machine ( runs / contains the cores. Grey in color in the schematic ):
– 4 steppers
– 4 string potentiometers
– 1 power supply
– 1 beagleboard / ?? … something that can control up to 12 of the cores mentioned above …
My initial thoughts are to connect all the cores and the primary machine to a single MC. But I need feedback between the cores and the main machine to be fast.
Example:
0) Command is issued
1) The microplate ( the little rectangular cubes with 8*12 array of holes ) locates under the particular liquid handling head
2) The core gets confirmation that the plate is where its supposed to be
3) The core extends the liquid handling head down towards the plate
4) Reagents are dispensed
5) Tips are ejected
So that process iterates numerous times. Can that kind of control be handled through the core processing unit? I guess that all depends on the IO speed of whatever unit I pick?
SO, the questions:
1) How would you set this up?
2) I don’t think I want to be flashing chips every time I run a different protocol… So how does the python web server issue the command? ( My previous iteration was via binary protocol with the MC continually polling for that next protocol )
With the right individual I’d be interested in taking on a co-founder. I think quite a bit can be done in this space, specifically with the advent of synthetic biology and the glut of genomics diagnostics which are to come.
Thanks!
-Carl
carlcrott@gmail.com
carlcrott@gmail.com
