Delta does not keep Z height on X / Y movement [noob]

denke

Hello All!

I have upgraded an flsun delta XL mk2 printer to the duet wifi board, but I cannot get the configuration right:

Fist, there is a difference between the manual setup guide and the configurator:
The Configuratir sais that the delta radius is the center of the tower mount -> bed center distance
The manual config guide sais that the delta radius is the span of the rod when the carridge is centered: http://reprap.org/wiki/Delta_geometry
Which one should i use?
If the later, than shouldn't I set something about the head? For example the distance between the nozzle and the centerpoint of the rods?

My actual problem is, that the head does not keep the z height when the carridge is moved either X or Y, there is about a centimeter of difference if the head is centered or if it is at the edge of the printable area (it lifts up). The head position is not lost, if I re-center the head it goes back to the same place always.
Which I checked:
-Rod length is correct (268mm)
-steps/mm values are correct (assuming it has to be measured on the pulley) (100)
-Axis are parallel and solid, there is no play (linear rails are used)
-Delta radius - I tried this with according the configurators and the manual setting version (170mm VS 142mm)
-Homed head height (283 mm)
-Wiring and basic hw configuration is fine (motor direction, endstops etc.)
-No missed steps

Thank You for your help,
Denes Dolhay

dc42

The delta radius is the horizontal span of each rod when the effector is centred. The firmware does not need to know the distance between the nozzle and the rods bearings.

Your symptom indicates that either the rod length, the delta radius, or the tower steps/mm is not configured correctly.

See https://duet3d.dozuki.com/Wiki/Calibrating_a_delta_printer for more. Bear in mind that if you are using a config-override.g file, the M665 and M666 commands in that may override the ones in config.g.

HTH David

DjDemonD

Yeah I think the thing often misunderstood with deltas is that any x or y movement is entirely computed and relies on the calibration being precise. A pure z move, relies on nothing but steps/mm. Get the height close, put in a correct printable radius and put in an accurate rod length in M558, then autocalibrate if you don't have a probe use the manual calibration wizard.

denke

Thank You for the help!!
Seems I accidently uploaded the json instead of the zip when modified the delta radius.

The configurator.reprapfirmware.org site states under general -> delta -> delta radius (tooltip) that "Radius between the centre of the bed and the midpoint of the towers" which is false. Since i'm a noob here and I do not know, is it You who I should tell, or should I write someone a mail, or file a report?

BTW, this board and sw is something completely different.
Thank You for the work you put in it!

denke

Oh, another bugreport with the configurator:
When loading the json all motor microstepping changes to "x16" instead of "x16 interpolated to x256" which was exported and the default value

DjDemonD

Sorry I've never used the configurator it seems straightforward enough to just modify the sample kossel config.

T3P3Tony

@denke:

Thank You for the help!!
Seems I accidently uploaded the json instead of the zip when modified the delta radius.

The configurator.reprapfirmware.org site states under general -> delta -> delta radius (tooltip) that "Radius between the centre of the bed and the midpoint of the towers" which is false. Since i'm a noob here and I do not know, is it You who I should tell, or should I write someone a mail, or file a report?

BTW, this board and sw is something completely different.
Thank You for the work you put in it!

Thanks for the reports, I will let Christian know.