Results 221 to 230 of 396
-
03-09-2015, 11:19 AM #221
- Join Date
- Jul 2014
- Location
- Eastern Colorado
- Posts
- 536
I don't know about that... here's a recent matrix:
Recv: Bed Level Correction Matrix:
Recv: 1.000000 0.000000 0.000637
Recv: -0.000000 1.000000 0.000254
Recv: -0.000637 -0.000254 1.000000
and the topographic report that went with it:
Recv: Bed Height Topography:
Recv: Origin: Front Left
Recv: --0.05153 +0.04247 +0.04872 --0.04903
Recv: --0.04903 +0.03047 +0.07922 +0.03297
Recv: --0.07103 +0.03072 +0.07647 +0.02297
Recv: --0.15053 --0.00853 +0.02397 --0.00828
-
03-09-2015, 12:11 PM #222
- Join Date
- Feb 2015
- Posts
- 371
What exactly does that mean? Would it influence my prints? I ask cause I can print small things fine but larger parts have failed so far. I get some warping and then the layers delaminate. I got the 12" model to print large things. Very frustrating to not be able to use the full build volume.
How do I run a topo report?
-
03-09-2015, 01:34 PM #223
- Join Date
- Jul 2014
- Location
- Eastern Colorado
- Posts
- 536
If you're using Roxy's code, run G28 then "G29 T n4" or whatever n# you prefer. The "T" is what tells it to make a topographic map.
-
03-09-2015, 03:42 PM #224
I also don't understand the bed level correction matrix. I always get the diagonal row of 1.0's as well. doesn't make obvious sense to me. It works ok so no complaints here, just cant' parse the matrix.
-
03-09-2015, 04:46 PM #225
Do you see the Z Axis move back and forth as it prints a layer? If so, the Auto Bed Leveling is doing it's thing. The way the 'new' coordinates are calculated is by doing matrix math and multiplying the coordinate in the GCode times the Bed Level Correction matrix to get the 'new' coordinate. My Bed Level Correction matrix can have a 1.000 on the diagonal, but never 3 of them. That is why I thought the previous matrix was done on a very level bed.
If you are willing to unlevel your bed to learn about this, you can tilt your bed and generate a new Bed Level Correction matrix. As you make the bed more and more unlevel, you should see the numbers along the diagonal decrease in size and you should see the numbers off of the diagonal start getting bigger in absolute size. (And there should be symmetry along the diagonal. One side will have a + version of the same - number the other side has)
-
03-09-2015, 04:59 PM #226
- Join Date
- Jul 2014
- Location
- Eastern Colorado
- Posts
- 536
The matrix is a grid of X Y and Z values. Across the top are
X Y Z
and down the side are
X
Y
Z
............X................ Y............. Z
X 1.000000 0.000000 0.000637
Y -0.000000 1.000000 0.000254
Z -0.000637 -0.000254 1.000000
The coordinates in your gcode gets multiplied by this grid. A move in X gets multiplied by the 1 in X, so the X distance doesn't change. It gets multiplied by the 0 in Y, so it doesn't move in Y at all. It gets multiplied by the number in Z, so the Z axis moves a little bit to make the print level.
Same thing happens for Y and Z. Y moves won't move in X, won't change in Y, and change Z a little. Z moves change a little in X, change a little in Y, and don't change in Z.
Hope this helps.
-
03-09-2015, 06:18 PM #227
- Join Date
- Nov 2014
- Posts
- 522
this made me want to look at my g29 matrix to see how bad my bed is unlevel... :-P
-
03-10-2015, 08:48 AM #228
And back to my point... The 1's along the diagonal don't look right with 'correction' numbers off to the sides. I think we are seeing limitations with the Arduino's single precision floating point math. X*1 + Y*0 + Z*.000637 means X movement in the 'corrected' number system is greater than its magnitude in the original system. It would seem the X number should drop from 1.0 to .999 if there are Y & Z components.
-
09-03-2015, 04:58 PM #229
- Join Date
- Sep 2015
- Posts
- 1
Just to check if i got this right: if you were using the matrix only as a readout to check if your printer is perfectly levelled you would need to achieve a matrix that consists only of 1.000 values right?
also, i want to say thank you to dacb for creating this thread, it totally saved me
-
09-03-2015, 06:41 PM #230
Qidi X Plus 3 Paper thin first...
05-27-2024, 01:15 AM in General 3D Printing Discussion