Auto bed leveling for cartesian type 3d printers. Delta print also.

Jeff Nading

Re: Auto bed leveling for cartesian type 3d printers. Delta print also.
« Reply #25, on December 3rd, 2014, 06:59 PM »Last edited on December 3rd, 2014, 07:56 PM
Ok Russ just got this off the Google forum try {G30 A Dnnn} for auto cal. :cool: :D :D


On Tuesday, 2 December 2014 15:23:43 UTC-4, Haydn Huntley wrote:
    BTW, I have a slightly different opinion on using RC's Marlin for making a delta printer print well.
    I've used it on two different printers, with a variety of beds, arms, and hotends on each printer, so I've done it successfully more than 10 times.

    If you just try to use "G30 A", it doesn't do a very good job.  For some reason it seems to have a hard time inferring the horizontal radius and diagonal rod length, even when you start it off with the right values.  I have been able to get calibrations like this, but they tended to print parts which were either too large or too small by a few percent.

    I also bought a 30cm micrometer and used it to precisely measure my diagonal rods (I've been building carbon fiber ones with zero-backlash magnetic joints), so then I could use the "G30 A Dnnn" command, which calibrated within less than 10 iterations, and did an excellent job.


Neal wrote:
I just tried this and you were right, it completed in only 7 iterations, but it doesn't do any compensation for the tower radius errors. They were slightly off when I ran the full G30 A as you can see below. Will this have a dramatic effect on my final prints? Can I use the values that I got from the full G30A or will that throw off the other variables?

G30 A result:
M666 X-2.45 Y-1.65 Z-2.70
M666 A-0.21 B0.55 C-0.34 E-0.88 F0.00 G0.31 R123.00 D206.00 H284.40 P0.00

G30 D198.6 A result:
M666 X-2.75 Y-1.35 Z-2.05
M666 A0.00 B0.00 C0.00 E0.00 F0.00 G0.00 R121.20 D198.60 H284.62 P0.00



Haydn Huntley wrote:
@Neil, When I did it on mine, it did calculate values for ABC and IJK, though they were small.

You should look at the numbers on the left side of your printouts, which showed how close everything was to zero (tram).
On mine, the worst error was 0.110mm, which is 110 microns, which is good enough for me.  I print layers which are 200 microns.

Here is what I got:

>>>G30 A D287.91
SENDING:G30 A D287.91
Starting Auto Calibration..
        Z-Tower             Endstop Offsets
        -12.1250            X:0.00 Y:0.00 Z:0.00
-12.3300        -12.2300    Tower Position Adjust
        -12.2450            A:0.00 B:0.00 C:0.00
-12.1950        -12.1900    I:0.00 J:0.00 K:0.00
        -12.1100            Delta Radius: 160.0000
X-Tower          Y-Tower    Diag Rod: 287.9100
Using diagional rod length: 287.91mm (will not be adjusted)
Iteration: 1
Adjusting Z-Height to: 293.25 mm..
.
.
.
Iteration: 7
Adjusting Endstops..
         Z-Tower         Endstop Offsets
         -0.0050         X:-2.07 Y:-2.12 Z:-1.80
-0.1100         -0.1000  Tower Position Adjust
         -0.0000         A:0.22 B:-0.16 C:-0.06
-0.0050         -0.0150  I:0.00 J:0.00 K:0.00
          0.0300         Delta Radius: 160.5000
X-Tower         Y-Tower  Diag Rod: 287.9100
Auto Calibration Complete
Issue M500 Command to save calibration settings to EPROM (if enabled)


I believe the "G30 E" command will probe and show the results.  The following is what I started with:

>>>G30 E
SENDING:G30 E
        Z-Tower        Endstop Offsets
        1.5000         X:0.00 Y:0.00 Z:0.00
1.2550        1.1900   Tower Position Adjust
        2.0150         A:0.00 B:0.00 C:0.00
1.4050        1.5350   I:0.00 J:0.00 K:0.00
        1.5150         Delta Radius: 160.0000
X-Tower       Y-Tower  Diag Rod: 287.9100




Jeff Nading

Re: Auto bed leveling for cartesian type 3d printers. Delta print also.
« Reply #28, on December 7th, 2014, 07:13 AM »
Ok, I tried it, made it to six iterations, effector came down so fast, it went pass the sensing point of the probe, nozzle hit the bed repeatedly, shut it down. I think what the problem was is that the magnet I am using for the probe is to weak, so there for it's not reacting quickly enough for the speed and force that the effector is coming down. So to fix the problem, or eliminate this possibility I ordered the correct magnets to be used on this probe. 5mm x 5mm x 5mm, I used a 2.5mm x 5mm x 5mm. I'll report back as to my findings, Jeff. :D



~Russ

Re: Auto bed leveling for cartesian type 3d printers. Delta print also.
« Reply #31, on July 7th, 2015, 08:33 AM »
Quote from Alberto on May 4th, 2015, 01:06 PM
Well guys , I'm wondering if anyone was successful with the automatic bed leveling... :huh:
I`m not completely satisfied with my (manual)  results... :blush:
i'm not so sure i'm satisfied with my "auto cal" lol i still manually adjust stuff after the " auto cal"

but if there is any error in your setup at all it may never be as you want it. it can only cal to the precision of your build... something to think about.

~Russ