Hi,
I have been trying to calibrate my home made delta printer for a few weeks, and I am experiencing some difficulties.
The system is composed by Radds, Due, TMC2100 and Repetier 1.04dev.
I currently don't have a z probe and would prefer to complete the calibration without using a z probe.
I find the paper sheet method quite precise. (I use 0.05 mm thin paper, the one used for the payment receipts of the shops).
In any case, to fix small mechanical alignment defects of the towers and a not perfectly flat bed
(indeed, let's say slightly convex), I decided to use the Escher calculator, also for the good results obtained by many.
Unfortunately, to understand how to use it correctly with Repetier required dozens of attempts and hours of reading in the forums.
In the beginning I got worse results than the basic parameters.
Then I realized that there are two key steps:
1 - Enter the nozzle height error values in the calculator with inverted sign.
2 - Enter the values "New tower position angle corrections", subtracting them directly from the corners of the columns a / b / c
In this way it worked.(10 probe points for 6 factors).
One step was enough to obtain a deviation value of 0.03mm. I then repeated the test 21 more times.
The parameters changed slightly every time, but the deviation error remained at 0.03mm.
I am however happy. Obviously that value is the limit given by mechanical defects, such as the slightly convex table.
To complete the calibration I thought of adding a distortion map via Host, but I did not understand if there is a way to activate it manually.
As suggested in other posts, I still activated the pin (Z min) and if necessary I can connect an on / off switch, to manually activate the signal, therefore not phisical connected to the effector.
My procedure was:
- M322 S3 Reset auto level matrix permanently
- G33 R0 Delete distortion map
- G28 Home
- G131 Remove any extruder offset
- M99 X0,Y0, Z0 Adjust sliders using a stick, one by one
- G132 S1 Measure tower endstop offsets & stores
- M320 S2 Activate auto level permanently
Then I used the Escher calculator and after each iteration I updated the EEPROM values with M500.
I guess I can't use G29 without z probe.
I tried to correct the distortion with G33, first with G33 L0 to read the syntax, then I probed manually with G0,
point by point, detecting a 5x5 matrix as suggested in the Repetier guide, in order to probe 21 points,
except the four outside the table, which should be interpolated (option active in configuration.h).
Once detected, I added all 21 points with G33 X<xpos> Y<ypos> Z<newdistortioncorrection>.
I store the values in EEPROM, which are correctly listed with G33 L0, but the correction matrix does not make any correction.
" Transformation matrix: 1.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 1.000000"
I'm sure, I'm making some mistakes...:-(
So, can anyone suggest me the correct procedure to perform the manual bed/bump correction without Z probe,
after the Escher correction?
Thanks in advance,
Marco
