Motor conventions for Polar printers

I have been asked to implement Polar kinematics in RepRapFirmware. I would appreciate some input from those who have experience of polar printers. In particular:

- What is the convention for connecting the motors to the motor outputs of the electronics: is the radius motor connected to the X motor output and the turntable motor connected to the Y motor output, or vice versa?

- Should G28 X home the radius or the turntable? Likewise, what should G28 Y home? Or are letters other than X and Y used in G28 commands to home the radius and the turntable?

Thanks in advance for replies.

---

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].

Hope you figure it out! X & Rotation like a record deck sounds right to me, swap-able if you prefer Y& Rotation but why would that be different? In my initial concept of a printer like this I thought of a bed like a photo booth chair that went down as it went round, but then with a linear axis above that rotated.

I released RepRapFirmware 1.20beta2 with polar printer support about an hour ago. Configuration instructions are at [duet3d.com].

In the absence of any replies before I froze the code prior to release, I went with using the X motor output for the radius.

---

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].

Cool, be interesting to know how it performs with 0.9 stepper at least on the rotation (and see what happens at the center),
and of course 32bit control, spiral vase mode too.
Also a machine that does Rotation on X on Z with a fixed head would be a good project, i'm also thinking about covering larger areas with minimal motion, wondering if its possible, ie flipping bed into different positions(not describing it well) perhaps be combining some kind of scara motion(where the scara is not the arms but the bed motion /linear/rotary... maybe i'm just over complicating whats the simplest principle.

Quote
MechaBits
Cool, be interesting to know how it performs with 0.9 stepper at least on the rotation (and see what happens at the center),
and of course 32bit control, spiral vase mode too.

Build one and let me know!

Quote
MechaBits
Also a machine that does Rotation on X on Z with a fixed head would be a good project...

You would be limited to layer heights that are multiples of the thread pitch. Getting a large enough screw and nut with a fine enough pitch might be difficult, but I guess you could machine them yourself using a lathe. Also i think you would still need to be able to move the head in Z by at least 1 thread pitch if you wanted to do anything other than spiral vase prints.

---

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'm virtually ready for the polar bed, though the the stepper & board elude me for now,
I thought we where already limited to layers that have to match with thread pitch,
if needed maybe head could have additional movement (lever/servo) just to clear bed for some kind of Z-hop or head switch rocker,
but I dont quite understand what you said in the last part.

Edited 2 time(s). Last edit at 10/26/2017 08:45AM by MechaBits.

In Cartesian and CoreXY printers, layer heights are invariably smaller (often very much smaller) than the thread pitch. What you do often see recommended is to make the layer height a multiple of full steps of the motor, which normally means a multiple of 1/200 of the thread pitch.

What I meant in the last part is that when you rotate the bed to get the current bit of the print under the nozzle, the height of the bed would change, so you will need to adjust the nozzle height to compensate.

Edited 1 time(s). Last edit at 10/26/2017 04:38PM by dc42.

---

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].

You mean if Z was tied to rotation it would move up as it rotated, but if rotation and Z are decoupled and Z is a rotating platform on seperate Z (could be a platform with 3 bearings & rods, which rises or falls im thinking something short like 100mm) or on an arm on z or simply like a record deck, where arm moves up.

I dont have Z on full steps(1/16th), maybe I should? a little more accuracy/repeatability, but I thought with the duet you could go lower(&with 0.9's)...though i've never run the numbers, to figure if its just totally not worth it.

Edited 3 time(s). Last edit at 10/26/2017 08:55PM by MechaBits.

Quote
MechaBits
You mean if Z was tied to rotation it would move up as it rotated, but if rotation and Z are decoupled and Z is a rotating platform on seperate Z (could be a platform with 3 bearings & rods, which rises or falls im thinking something short like 100mm) or on an arm on z or simply like a record deck, where arm moves up.

Yes.

Quote

I dont have Z on full steps(1/16th), maybe I should? a little more accuracy/repeatability, but I thought with the duet you could go lower(&with 0.9's)...though i've never run the numbers, to figure if its just totally not worth it.

Using microstepping on Z is good; but the layer height should preferably be a multiple of the full step distance. For example, if you use x16 microstepping and your Z steps/mm is 800, then you have 50 full steps/mm, so you should preferably choose layer heights that are multiples of 1/50 mm.

Edited 1 time(s). Last edit at 10/27/2017 03:03AM by dc42.

---

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've been searching for firmware that will handle cylindrical coordinates for my printer. I have a rotating gantry instead of the usual rotating bed and wasn't sure how much work would be needed to accommodate for the difference - I would have a translation of (x,y,z) to (r,θ,z). I will give RepRapFirmware a try.