Molten-Core: e3d tool changing type corexy

I've been a delta speed racer for a while and only decided to switch over to corexy when I saw e3d's tool changing. I'm using their method of locating the tool with the ball & groove but I'm switching out the servo for an electromagnet. I prefer to have less moving parts. The current build volume is 300x300x500. I have currently modeled regular 12mm linear rails but I will probably be switching over to a larger linear guide from igus for the gantry at the very least.

Key Features:

-Up to 5 or 6 tools depending on how I arrange the linear guide for the z axis.

-.5" MDF frame... Now many people might not like that but its cheap and I can laser cut it at my local maker space. It just needs to be sealed with the right paint and it will be darn stable

-DC motor drivers for X-Y & extruders. I've managed to find enough drivers between 13 & 25 dollars to make it worth doing. I will probably test using them for the z axis later but I have a feeling the steppers might actually be a better choice for slow moving axis.

-Encoder for the dc motor extruders is placed on the tool to allow for jam detection & proper filament movement with a bowden.

-I'm using a nice belted linear guide I found on robodigg for the Z axis. It has 500 travel, very well built, about an inch wide belt, 150 after shipping cost, & only took a week to get to my door. I'm excited to try it out.

What boards support the servos and actual closed loop control? I could make a printer print a lot faster and more accurately if there was something out there that supported codes loop control.

Quote
Leafy1
What boards support the servos and actual closed loop control? I could make a printer print a lot faster and more accurately if there was something out there that supported codes loop control.

The usual choice is step-servo drives that emulate a stepper motor + driver, by accepting step and direction inputs.

The idea of using BLDC motors + encoders is appealing and has been done, but is not particularly straightforward. You need a gearbox to provide sufficient torque and precision, and two encoders: a simple one before the gearbox to do commutation for the BLDC motors, and a precision encoder after the gearbox to measure the distance travelled. You could perhaps manage with just on encoder before the gearbox, but then you wouldn't be able to compensate accurately for backlash in the gearbox.

The sorts of low-cost BLDC motors used in quadcopters are not suitable, because they rely on back emf for commutation, and that doesn't work at low speeds.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I'd want scales on the axis directly for the precise control. You could easily use cheap digital calipers with half thou resolution. They already have outputs on them for a display and people already use them as ghetto dro.

Quote
Leafy1
What boards support the servos and actual closed loop control? I could make a printer print a lot faster and more accurately if there was something out there that supported codes loop control.

So there are several directions you can go with this.

Here are the projects I have used or found:


used it. PID tuning is a pain & doesnt get better resolution than +/- 5 encoder pulses

[www.crowdsupply.com]


I like these below better because they have designed ways of PID tuning that are simpler than tarocco's method. The other nice thing is you can order a control board + hdrive for these boards for really cheap on aliexpress. My STM32 + 2 amp hdrive for extruder build im hoping to test was 4 dollars.

[github.com]

[github.com]


BLDC :

[odriverobotics.com]