Software suggestions and requests
I started this thread so that people could have a place to help better the software for the PeachyPrinter.
the first request I have is the ability to import our own STLs into the print library instead of it just having the test prints (I think it would be nice to have our own private "Thingiverse" complete with 3D previews of our prints)
5 Attachment(s)
Peristaltic Pump Exploded
OK, I received my $12 PS pump, and I am impressed with its simplicity. The motor shaft serves as the sun "gear," though it is smooth; the 3 rollers are the planetary "gears," driven by contact with the motor shaft, and press the tube against the U-shaped housing. The rollers have a circular plate with an axis for each to keep them aligned with the motor shaft and tube. I have attached photos and a sketch of the geometry:
Attachment 9427Attachment 9428Attachment 9429Attachment 9430Attachment 9424
I made the shaft and roller (0.435 in.) measurements with a micrometer and a digital caliper, and the outer diameter is derived from them. Calculating the motor shaft turns per 1/3 rotor turn: Given C = Pi D, 1/3 Pi 0.96 = x Pi 0.09, x=0.32/0.09 =3.56 I clamped a vice grip pliers onto the motor shaft end and rotated the pump housing 4 times, resulting in approximately 1/3 rotation of the rotor. Perhaps the difference is due to the compressed tubing at the outer circumference.
I will have to put sensors on the motor shaft and rotor to experimentally determine the precise "gear ratio" between motor turns and rotor turns. Because the smooth motor shaft has the potential to "creep" with respect to the rotor position, unlike gears, counting motor revolutions alone may eventually allow errors due to stopping the pump during a pinch transition (non-constant flow) instead of between pinches. I doubt the "creepage" would be enough to cause significant errors when counting motor turns to provide integer multiples of 1/3 rotor revolutions, but, the software should have a sensor input on the rotor as well, and have a self-correcting algorithm to ensure the rotor position is known within a turn of the motor shaft. I'm assuming the Hall effect sensor on the rotor will be accurate, but not very precise, especially when trying to stop at the same position running forward vs. reverse. Conversely, a sensor on the motor shaft would be more precise, but less accurate due to "creepage."
An optical sensor disc could unambiguously locate the rotor within a fraction of a turn. If the disc supporting the roller axis were installed inversely, opposite the motor rather than between the motor and the rollers, the optical disc pattern could be printed on the outer face of the disc and the optical sensor array could "view" the disc through a slot in the pump housing.