Nozzle hitting AND lifting off of bed after G33 - Delta

powerhobo

I'm using a He3D K200 Delta with Repetier 0.92.9 FW, and the most recent Repetier Host.

I'm following along with this video exactly:

I'm making sure that my probe height is correct (1.5mm), and that after G32 S2, the distance between nozzle and bed at X0 Y0 Z0 is about 0.1mm (thickness of a piece of paper).

I've done all the steps of the video about 20 times now, and the nozzle is still consistently hitting the bed in some places, and lifting off of it in others. 

The bed height map shows deviation of only 0.1mm after Z correction, but when manually measuring my heights I have areas that are 0.5mm off the bed, and areas where Z2.7 is hitting the bed. That's a big variance, and makes it absolutely impossible to print.

What am I doing wrong? I've been working on this every night for three weeks, and I can't find an answer or any tutorial that works.

Here are my 2 most recent height maps:

Bed Height Map

X:		-50.00	-33.33	-16.67	0.00	16.67	33.33	50.00
y:	50.00	4.03	4.48	5.06	5.79	6.65	7.60	8.64
y:	33.33	4.50	4.80	5.21	5.80	6.54	7.39	8.35
y:	16.67	4.88	4.99	5.24	5.68	6.28	7.00	7.86
y:	0.00	5.10	5.01	5.10	5.38	5.80	6.43	7.18
y:	-16.67	5.15	4.86	4.76	4.88	5.18	5.68	6.33
y:	-33.33	5.01	4.55	4.28	4.21	4.36	4.71	5.25
y:	-50.00	4.76	4.09	3.63	3.38	3.39	3.61	4.03

after g33

X:		-50.00	-33.33	-16.67	0.00	16.67	33.33	50.00
y:	50.00	6.99	6.99	7.01	7.05	7.04	7.01	7.04
y:	33.33	7.00	7.00	7.00	7.05	7.04	7.03	7.06
y:	16.67	6.99	7.00	7.01	7.05	7.03	7.03	7.04
y:	0.00	6.95	6.98	6.98	7.01	6.98	6.98	7.00
y:	-16.67	6.99	6.98	7.01	7.04	7.00	7.03	7.04
y:	-33.33	7.01	6.99	7.00	7.05	7.01	7.03	7.00
y:	-50.00	7.00	6.98	6.99	7.00	7.00	7.00	6.99

Please note that no matter how I have the bed physically set, Z-probing ALWAYS shows the back-right side of the bed (around X66 Y38) as way too low, but when watching the hotend move it is clearly the hotend assembly moving up away from the bed, not that the bed is actually physically lower. I have tried to compensate for this by physically raising that side of the bed, and it doesn't really change anything when it comes to Z-probing.

powerhobo

**EDIT** probe height is 0.15mm, not 1.5mm.

Repetier

Wow. Very impressive how good G33 is working even with error it was never meant for.

First you should now that your basic calibration is very wrong. Without G33 you have errors up to nearly 5mm, which simply means that your configuration does not match your real geometry. G33 was more meant to fix the last 0.1-0.2mm So first you should fix your geometry definition, meaning angles, diagonal length etc. Then your uncorrected height map should already be much closer. I guess your current config will also have distortions in xy plane from this geometry error. There are many tutorials available on how to calibrate delta printers.

G33 is configured to remove correction at a certain height. So as soon as you are higher 0.3mm or whatever you have define in configuration the corrections gets reduced and after a higher height completely reset. Update this to handle the remaining error. The more you need to correct the bigger the transition zone must be.

I guess that your problem lies in that fact. 

powerhobo

I never said G33 wasn't working. I asked what I was doing wrong. So now at least I know that the preloaded firmware from He3D is not set up correctly with its geometry like they said it would be. I guess I'll spend another couple weeks searching for more tutorials that I've already been searching for. Thanks.

Repetier

My guess is that the 90° angle must be something like 90.2°. Disable G33 then run G32 and make bed map. Then increase angle a bit, G32 and bed map and see how values change. After a few iterations you get a feel. Tutorials may show a faster way like this one

http://escher3d.com/pages/wizards/wizarddelta.php

with a calculation tool on what to update.

powerhobo

I will give these a shot once I'm home from work. I'm sure this will be very helpful (all information is). Thank you.