[Solved] Calibration deviation too high

rkv34
Posts: 5
Joined: Wed Nov 22, 2017 6:47 am

[Solved] Calibration deviation too high

Post by rkv34 » Fri Nov 24, 2017 11:19 am

My calibration deviation is an extremely consistent 0.073 to 0.076. I've double checked the tightness of the frame, pulley tension using the app (from 39 to 43 Hz), updated the config.g with the values below, and have failed to lower my deviation value. I went through the z-trigger height from sublimelayers.com to update the Z value in the G31 command. Any suggestions?

From config.g

M665 R201.384 L374.84 B140 H445 X-0.095 Y-0.061 Z0.0 ; Set delta radius, diagonal rod length, printable radius and homed height
M666 X0.52 Y-0.46 Z-0.06 ; Endstop offset adjustments, these will be adjusted with delta auto-calibration

; Change "H25" to "H3" AFTER commissioning your printer
M558 P4 X0 Y0 Z0 H10 I1 ; FSRs with JohnSL board Z probe behaves as a switch and is not used for homing any axes
G31 X0 Y0 Z-0.185 P500 ; MUST READ: http://www.sublimelayers.com/2017/05/fdffsd.html


Also added the calibration macro to spit out the results of each point. It's going positive to negative values very consistently, hopefully that gives a clue what I'm missing?

Bed probe heights: 0.118 -0.101 0.119 -0.104 0.100 -0.112 0.006 -0.010 0.015 -0.015 0.024 -0.002 -0.029, mean 0.001, deviation from mean 0.075
Bed probe heights: 0.099 -0.126 0.087 -0.138 0.090 -0.122 -0.014 -0.026 -0.010 -0.040 -0.004 -0.016 -0.039, mean -0.020, deviation from mean 0.076
Bed probe heights: 0.113 -0.110 0.105 -0.103 0.101 -0.108 0.000 -0.006 0.013 -0.016 0.015 -0.005 -0.030, mean -0.002, deviation from mean 0.073
Bed probe heights: 0.111 -0.105 0.112 -0.109 0.101 -0.109 0.001 -0.009 0.005 -0.017 0.006 -0.007 -0.030, mean -0.004, deviation from mean 0.074
Bed probe heights: 0.113 -0.110 0.108 -0.104 0.105 -0.104 0.005 -0.015 0.014 -0.016 0.010 -0.006 -0.028, mean -0.002, deviation from mean 0.074

rkv34
Posts: 5
Joined: Wed Nov 22, 2017 6:47 am

Re: Calibration deviation too high

Post by rkv34 » Fri Nov 24, 2017 12:44 pm

Well I found the issue, changed bed.g to S7 so it would calibrate the diagonal arm length and that resulted in:

After bed.g set to S7 instead of S6 to add diagonal rod length
G32 Calibrated 7 factors using 13 points, deviation before 0.074 after 0.005
M665 Diagonal 379.225, delta radius 207.037, homed height 456.386, bed radius 140.0, X -0.122°, Y -0.175°, Z 0.000°
M666 Endstop adjustments X-0.31 Y0.24 Z0.07, tilt X0.00% Y0.00%

I noticed an incorrect arm length in the config_overide.g of 360.1 and finally realized I was using the d300vs config files not the d300vs+ config files. Uploaded the d300vs+ firmware and now everything works great (deviation around 0.080).

User avatar
UltiBots
Posts: 602
Joined: Fri Apr 21, 2017 10:18 am

Re: [Solved] Calibration deviation too high

Post by UltiBots » Fri Nov 24, 2017 2:23 pm

Awesome!
Brad - UltiBots LLC

Crazygamerm2
Posts: 1
Joined: Wed Nov 15, 2017 1:49 pm

Re: [Solved] Calibration deviation too high

Post by Crazygamerm2 » Sat Nov 25, 2017 1:31 pm

HOLY MOLY THANK YOU FOR THIS!!!

I was having the identical issue with the almost identical deviation numbers of .07 -.078 but could not get them lower. Changing the Bed.g file to S6 to the S7 made if peeerfect.

Calibrated 7 factors using 13 points, deviation before 0.333 after 0.007 !!!! Super happy and thank you again for finding this.
Ill make sure to chime in on the facebook page about this as well as it huge for calibration.

Firefly
Posts: 42
Joined: Tue May 30, 2017 10:29 am

Re: [Solved] Calibration deviation too high

Post by Firefly » Fri Oct 19, 2018 7:46 am

I been printing with a .6 nozzle and mostly in center of bed and never noticed how uneven my first layer is until I went to print something bigger.

so my G32 initial results:

Calibrated 6 factors using 13 points, deviation before 0.147 after 0.068
Calibrated 6 factors using 13 points, deviation before 0.113 after 0.053
Calibrated 6 factors using 13 points, deviation before 0.176 after 0.058
Calibrated 6 factors using 13 points, deviation before 0.761 after 0.082

with S7 results:

Calibrated 7 factors using 13 points, deviation before 0.044 after 0.043
Calibrated 7 factors using 13 points, deviation before 0.038 after 0.037
Calibrated 7 factors using 13 points, deviation before 0.034 after 0.029
Calibrated 7 factors using 13 points, deviation before 0.128 after 0.050
Calibrated 7 factors using 13 points, deviation before 0.174 after 0.071


what code did you add to which file to spit out the each points?

looks like I may have a different issue, any suggestions?

Firefly
Posts: 42
Joined: Tue May 30, 2017 10:29 am

Re: [Solved] Calibration deviation too high

Post by Firefly » Fri Oct 19, 2018 8:53 am

I figured out to use S-1 to spit out each points. So I got the following...

Bed probe heights: 0.830 1.101 0.887 0.159 -0.220 0.086 0.554 0.807 0.640 0.310 0.162 0.268 0.432, mean 0.463, deviation from mean 0.365
Bed probe heights: 0.814 1.130 0.872 0.159 -0.217 0.095 0.538 0.814 0.640 0.314 0.161 0.264 0.435, mean 0.463, deviation from mean 0.366
Bed probe heights: 0.798 1.094 0.862 0.151 -0.221 0.087 0.537 0.816 0.629 0.295 0.154 0.268 0.416, mean 0.453, deviation from mean 0.361

does this means my belt might be too loose?

garyd9
Posts: 28
Joined: Wed Sep 05, 2018 1:44 pm

Re: [Solved] Calibration deviation too high

Post by garyd9 » Fri Oct 19, 2018 10:20 am

Each of the bed probe heights is telling you how far the nozzle is "pressing" (from a Z0) in order to get the FSR/JohnSL board to show a trigger. Assuming that your bed.g has the center (x0 y0) last, the first oddity is that your underbed FSR's are triggering at a Z height greater than 0. (With under-bed FSR's, most of the values are typically negative.)

Of course, the first question is "what is z0"

Normally, your FSR's won't trigger at Z0 (because no additional force is being applied to them.) Once the nozzle pushes down on the bed (Z less than 0), the FSR's are being subjected to additional force, and so they trigger.

Can you post the G31 line from your config.g (and verify that G31 only occurs once in config.g and not at all in config-override.g)? As well, the contents of your bed.g would be useful (to check if you're using the H params with the G30 commands and to know the order of the probing.)

Also, check your FSR's. On my "plus", the red pads would constantly slide off the FSR's and I ended up using a staple gun to secure them. (Not really. I used super glue.)

Edit: This page talks about setting the Z parameter of G31: http://www.sublimelayers.com/2017/05/fdffsd.html (The Z trigger height.)

Firefly
Posts: 42
Joined: Tue May 30, 2017 10:29 am

Re: [Solved] Calibration deviation too high

Post by Firefly » Fri Oct 19, 2018 8:54 pm

In config.g:
G31 X0 Y0 Z-0.3 P500

config-override.g:

Code: Select all

; This is a system-generated file - do not edit
; Delta parameters
M665 L379.000 R207.501 H456.175 B140.0 X0.174 Y0.220 Z0.000
M666 X-0.950 Y0.905 Z0.046 A0.00 B0.00
; Heater model parameters
M307 H0 A90.0 C700.0 D10.0 S1.00 B1
M307 H1 A512.9 C267.0 D9.0 S1.00 B0
M307 H2 A340.0 C140.0 D5.5 S1.00 B0
M307 H3 A340.0 C140.0 D5.5 S1.00 B0
M307 H4 A340.0 C140.0 D5.5 S1.00 B0
M307 H5 A340.0 C140.0 D5.5 S1.00 B0
M307 H6 A340.0 C140.0 D5.5 S1.00 B0
M307 H7 A340.0 C140.0 D5.5 S1.00 B0
; Z probe parameters
G31 T1 P500 X0.0 Y0.0 Z0.70
G31 T3 P500 X0.0 Y0.0 Z0.70
G31 T4 P500 X0.0 Y0.0 Z-0.30

Red pad were ok, but repositioned one during the troubleshooting
Last edited by Firefly on Fri Oct 19, 2018 9:14 pm, edited 2 times in total.

Firefly
Posts: 42
Joined: Tue May 30, 2017 10:29 am

Re: [Solved] Calibration deviation too high

Post by Firefly » Fri Oct 19, 2018 8:56 pm

bed.g:

Code: Select all

; Auto calibration routine for delta printers
; Before running this, you should have set up your zprobe Z offset to suit your build, in the G31 command in config.g.
;
; You can generate the probe points using the bed.g file generator: http://www.escher3d.com/pages/wizards/wizardbed.php
;
; About the G30 S parameter
; Used to specify what calibration computation to perform.
;
; S=-1 Don't adjust anything, just print the height error at each probe point
; S=0 Equivalent to S=<number_of_points_probed>
; S=3 Adjust homing switch corrections only
; S=4 Adjust homing switch corrections and delta radius
; S=6 Adjust homing switch corrections, delta radius, and X and Y tower position offsets
; S=7 Adjust homing switch corrections, delta radius, X and Y tower position offsets, and diagonal rod length 

M561						; clear any bed transform, otherwise homing may be at the wrong height
G31 X0 Y0					; don't want any probe offset for this
G28							; home the printer

; Probe the bed and do 6- or 7-factor auto calibration
G30 P0 X-86.60 Y-50.00 Z-99999	; X tower
G30 P1 X0.00 Y-100.00 Z-99999	; between X-Y towers
G30 P2 X86.60 Y-50.00 Z-99999	; Y tower
G30 P3 X86.60 Y50.00 Z-99999	; between Y-Z towers
G30 P4 X0.00 Y100.00 Z-99999	; Z tower
G30 P5 X-86.60 Y50.00 Z-99999	; between Z-X towers
G30 P6 X-43.30 Y-25.00 Z-99999	; X tower
G30 P7 X0.00 Y-50.00 Z-99999	; between X-Y towers
G30 P8 X43.30 Y-25.00 Z-99999	; Y tower
G30 P9 X43.30 Y25.00 Z-99999	; between Y-Z towers
G30 P10 X0.00 Y50.00 Z-99999	; Z tower
G30 P11 X-43.30 Y25.00 Z-99999	; between Z-X towers
G30 P12 X0 Y0 Z-99999 S7		; center and auto-calibrate 6 factors

G1 X0 Y0 Z150 F12000			; get the head out of the way of the bed

garyd9
Posts: 28
Joined: Wed Sep 05, 2018 1:44 pm

Re: [Solved] Calibration deviation too high

Post by garyd9 » Fri Oct 19, 2018 9:25 pm

Take backups of config.g, bed.g and config-override.g

Then delete config-override.g (or at least get rid of the G31 lines in it.)

On the last G30 in bed.g, it should be S6. S7 is explicitly discouraged unless you have probe points that cause the delta arms to go nearly horizontal (and have a massively large build plate.) Otherwise, it's inaccurate.

Heat up your heated bed to whatever temp you usually have it for printing. Heat up your nozzle to your normal hot end temp as well. You might as well remove any filament you have in there as this will take a bit of time and there's no need to burn up filament (and make a mess with oozing.)

Finally, re-do your z-height. Use the link I provided earlier for figuring it out, or just make a macro containing:

Code: Select all

M208 S1 Z-3
M564 S0
G90
G1 X0 Y0 Z0.5
G91
Run that macro and it'll place your nozzle at x0,y0 and at a height of around 1/2mm. Now it gets to be fun, depending on what you have available to measure nozzle height with. Ideally, you have a set of calipers, and can measure the thickness of a piece of paper. If not, just assume 0.1mm. Slide a piece of paper under the nozzle. Go to the g-code console, and type/send "g1 z0.2". See if the paper drags on the nozzle. (Hopefully, it doesn't.) Type "g1 z0.19" and test again. Keep doing this, lowering the "z" value each time until the paper is dragging. It's perfectly okay to use a negative number for the "z" value (for example, "g1 z-0.1"

Once the paper starts dragging (but not so much that you can't move the paper), stop and type the following command into the gcode console: "g92 z0.1" (If you were able to measure the thickness of the paper, substitute "0.1" with the actual thickness.) That command tells the printer what the "real" height is. Now, move the nozzle up a bit with "g1 z5" and remove your piece of paper. Next, in the gcode console, send "g30 s-1" and it'll slowly move the nozzle down until the FSR's trigger. It'll then report a value in the gcode console similar to "Stopped at height -0.137 mm" Again, type "g1 z5" followed by another "g30 s-1". Do this 3 times. If the numbers are all fairly close to each other (within 0.015 or 0.02), take the average. If they are far apart, do the "g1 z5" and "g30 s-1" sequence a couple more times to try to get more consistent numbers and take the average. That average number should be written down (it's your z trigger height and will be a negative number.)

Go to edit config.g and find the G31 line. (You've taken it out of the config-override.g, so the one in config.g should be the only one.) Set the "z" parameter to the z trigger height you wrote down. So, if it was "-0.123", then your G31 line would look like "G31 X0 Y0 Z-0.123 P500"

Save config.g, let the duet reboot, and re-do delta auto calibration 3 times. Finally, check to see if things are working better.

Post Reply