Need some CoreXY advice

mrehorstdmd

How fast was it going? Acceleration? Jerk?

deckingman

@gtj0 Is the hot end nice and secure. If you put your finger on the nozzle, can you move it in X or Y? Could the drag of your wires and Bowen tube cause the hot end to tilt?

mrehorstdmd

@deckingman I think the tool carriage has a pen mounted on it.

I'd suspect the pen holder is a bit sloppy, or he's running too fast- the distortion in the circles looks like what happens when you try to print small perimeters at very high speed. I suspect the different moving masses in X and Y and flexibility in the X axis lead to the distortion.

I ran my xy stage as a plotter before I finished the printer. I mounted a ball point pen on a short linear guide so there was no lateral slop. The pen could move up and down so even if the paper surface wasn't parallel to the XY plane it worked fine.

plotter operation

gtj0

@mrehorstdmd said in Need some CoreXY advice:

How fast was it going? Acceleration? Jerk?

The issue seems to be independent of speed and acceleration. Currently I have as a test...

M566 X600  Y600  Z125  E6000  ; Set maximum instantaneous speed changes (mm/min)  ***
M201 X600  Y600  Z125  E6000   ; Set accelerations (mm/s^2) ***
M204 P600 T600
M203 X16000 Y16000 Z2400 E2400  ; Set maximum speeds (mm/min) ***

and the boxes are drawn with F6000 but I've tried various other speeds and accelerations.

@deckingman said in Need some CoreXY advice:

@gtj0 Is the hot end nice and secure. If you put your finger on the nozzle, can you move it in X or Y? Could the drag of your wires and Bowen tube cause the hot end to tilt?

The nozzle is rock solid. I'll post some pics of the setup in a bit.

@mrehorstdmd said in Need some CoreXY advice:

@deckingman I think the tool carriage has a pen mounted on it.

I'd suspect the pen holder is a bit sloppy, or he's running too fast- the distortion in the circles looks like what happens when you try to print small perimeters at very high speed. I suspect the different moving masses in X and Y and flexibility in the X axis lead to the distortion.

Nope, no pen holder. I clamped some old fashioned thermal fax paper on the bed and set the hot end to 145C. I'm going to try some super-low speeds and accelerations and see what happens.

I'll also get some pics of the various components up so you can see what's involved.

bot

@gtj0 said in Need some CoreXY advice:

Nope, no pen holder. I clamped some old fashioned thermal fax paper on the bed and set the hot end to 145C.

That's brilliant.

mrehorstdmd

The nozzle isn't dragging the paper, is it?

gtj0

@mrehorstdmd said in Need some CoreXY advice:

The nozzle isn't dragging the paper, is it?

When I first started using this method last year I did a calibration effort that resulted in homing the Z axis directly on the bed (I use an Orion Piezo with no Z offset), raising 5 mm, then moving over the paper and drawing at an absolute height of 0.05. This gives the nozzle enough contact with the paper to activate it but not enough to cause the paper to ripple. I also heat the nozzle high enough to melt plastic first and make sure the nozzle is perfectly clean, then let it cool down.

This is the paper I use...
https://smile.amazon.com/gp/product/B00006B7QA/ref=ppx_yo_dt_b_asin_title_o08_s00?ie=UTF8&psc=1

I tear off a length to match the width of the bed (510mm) then use 4 spring clamps to secure it.

gtj0

@bot said in Need some CoreXY advice:

@gtj0 said in Need some CoreXY advice:

Nope, no pen holder. I clamped some old fashioned thermal fax paper on the bed and set the hot end to 145C.

That's brilliant.

Comes from being old enough to have used fax machines that had handset couplers and a spinning drum you clamped a sheet of thermal paper to.

zapta

@gtj0 said in Need some CoreXY advice:

a spinning drum you clamped a sheet of thermal paper to ...

What is paper?

gtj0

@zapta said in Need some CoreXY advice:

@gtj0 said in Need some CoreXY advice:

a spinning drum you clamped a sheet of thermal paper to ...

What is paper?

Ha!

henrik.basic

@mrehorstdmd
At last I understand the differing opinions on core-xy with twisted crossing.
Some say it works well and runs smoth, and others say it grinds the belts rapidly.
Most agree that the belts need to be back to back as they cross.
But here comes the point:
They have to be "back to back" all the way! Your belts have the teeth facing each other at the pulleys.
You do the 180º twist, and avoid that the teeth face each other in the middle. But the result is that
they scratch edge to edge instead (at two points).

With the usual core-xy with twisted crossing layout, you can do it like the image below.
Green circles marks twisted regions. Black lines next to the red/blue lines marks the side with teeth.

core-xy-twist-and-teeth

gtj0

Just FYI... I've been running my single-level back-twist arrangement for over a year now with the same belt and no wear at all.