Week 9 Update
Last updated
Last updated
What was Accomplished:
Experimented the capabilities of the newly arrived syringe(by directly inserting liquids into it)
The syringe was able to pump out acetone, photoresist, IPA, and DI consistently
The syringe was not leaking regardless of liquid
However, these liquids can only be consistently pumped WHEN the syringe is directly attached to tubing, and not to a needle(since the liquid flowing through the exit of the needle actually flows onto the surface of the syringe)
Took measurements of the syringe(since we, and re-designed CAD models accordingly)
The measurement of the syringe hub was different than advertised(around 4mm), in comparison to our 3mm inner diameter syringes
To mitigate this issue, while testing the syringe, a luer lock needle was directly attached onto the syringe, instead of PTFE tubing
3D printed spin coater container for the lab automation assembly
3D printed revised CAD components for pump assembly
Roadblocks:
Measurements provided by the syringe(as advertised) were inaccurate
Not ALL of the glass syringes are leak proof, so we have to be very careful selecting the right syringe
Plans for next week:
Wait for the actual linear actuator motor to arrive to finalize assembly, and compile the code
Order parts for 3-4 additional syringe pumps
3d print the rest of the spin coater