Delta Calibration Help
-
Knowing how the grid probes, this seems to be related to direction of travel, which then brings to mind something very subtly loose, or "preferring" a direction, or similar.
Given the mean error is .002, this could be VERY subtle.
On the other hand, given this is a delta, and the variance seems to be associated with the square grid, not a triangle, that bias is probably in the effector, not in the towers.
In short: Push on everything, medium hard, with your fingers, and see if you can find anything that shifts or "takes a set".
-
Ah, our posts crossed, tilted view looks very different, and now getting devaations totaling 1.5mm (approx).
Something very wrong here. Either that BLtouch is really wacky, or something is really loose.
-
@Danal Sorry but my English is not good...
The first image refers to bed probed after delta calibration... The second 2 images refer before delta calibration...
Do you understand? -
Yes, that makes sense. Your english is plenty good enough.
So, on that image after calibration, do check for something just the littlest bit loose somewhere.
-
@Danal said in Delta Calibration Help:
Yes, that makes sense. Your english is plenty good enough.
So, on that image after calibration, do check for something just the littlest bit loose somewhere.I check all the screws on effector and carriages and all is tight enough...
-
@samu_87 Almost certainly backlash/something loose causing your effector to tilt a little, combined with the offset probe. Most likely in you X or Y tower, as it shows when moving in the X axis.
Ian
TronXY X5S with Duet 3 MB6HC and Roto toolboard : Cartesian bed-slinger with Duet 3 Mini 5+ WiFi and 1LC : RRP Fisher Delta v1 with Duet 2 Maestro : Polargraph with Duet 2 WiFi
-
can the belts be tensioned?
-
@droftarts Maybe the error is in the Z axis, because is the only axis that is firm when the effector move in X direction, so if is backslash for me is in the Z axis, right?
-
I update the firmware to the latest version, and I check all screws and belt...
This is the result without Auto Calibration... I can't see the pattern like in the first pic..
So it can be the Auto Calibration that give me this issues?
Anyone have a method to calibrate the printer looking the bed??
Thanks!
-
Have you seen this? https://duet3d.dozuki.com/Wiki/Calibrating_a_delta_printer
-
@Phaedrux Yes thanks! but the strange effect in photo one?
Thanks!
-
The ripples in photo #1 indicate backlash.
-
I measure the back slash and is from about 0.02mm on each axis.
I made this 2 bed analysis one with double point of each other and the other parameter is equal.
So I can't say why one line is up from 0 and the other under... The only difference is that in line 0 the effector move from left to right and in line 1 from right to left etc...Any idea??
-
This is pretty common on deltas that use probes offset to the nozzle. The effector is tilting slightly when moving one direction versus the other which causes the BLTouch to be slightly higher or lower depending on which direction it's going.
This is why probes like the smart effector or piezo are preferable for delta printers.
-
I'm thinking effector tilt. Because your probe is offset from your nozzle, if your build isn't exactly perfect your effector may tilt slightly at different xy locations, and because the offset probe, this will impact the probe results. Highly recommend probing with the nozzle on deltas.
-
Ok Many thanks!! Now is more clear..
What can I check to minimize this error? Thanks! -
This post is deleted! -
@samu_87 said in Delta Calibration Help:
Ok Many thanks!! Now is more clear..
What can I check to minimize this error? Thanks!I guess that's where it comes down to build quality to limit effector tilt based on position. Normally I think auto calibration can correct for it, but the auto calibration is also affected by the tilt because it uses the probe. Catch 22. The best solution is to change the probe type to one that uses the nozzle. Other than that, go through the build and ensure it's as close to perfection as you can manage.
-
Maybe is my error, I have this for construction:
So I put the command G31 P500 X27.45 Y-15.85 Z 2.144 for BLTouch..
BUT when I go to calculate the point to probe here:
http://www.escher3d.com/pages/wizards/wizardbed.php
I must put the same probe offset?
If I put the same probe offset I think that the probe point are shifted on the left..I thing that this 2 areas are problem to compute the data...
-
You can use H offsets in your bed.g for each probe point to account for the tilt when probing.
