Backlash issue, again. Was peaks and valleys in bed map

Having a problem? Post it here and someone will be along shortly to help
Post Reply
clytle374
Printmaster!
Posts: 56
Joined: Wed Apr 20, 2016 1:23 am

Backlash issue, again. Was peaks and valleys in bed map

Post by clytle374 »

First question is on belt tension again as I've had this problem with my V2 build, now its been upgraded to a V3.frankenstien. I had to have the belts extremely tight to solve it. I had actually cracked a skate at the belt clip. Should I be able to get enough tension by twisting the motor with a thumb and forefinger?

The problem I'm having is peaks and valleys in the bed map. It's actually worse than it's was before. https://forum.seemecnc.com/forum/viewtopic.php?f=37&t=12768 sadly all the pictures have timed out. Here is my bed map now. Manual says tight enough they don't slip, I'm passed that to get this good of results.

One other concern is how tight the CF arms are on the metal balls. I might swap the plastic ones back in to see if that is adding to much of a bind.

I kinda hate to call this backlash because it's not that simple, it's related to the last direction moved. Here is a macro I have to test for it a each tower. If someone doesn't mind run this and post the results. This is a duet with a smart effector.

Code: Select all

 ;test at each tower for backlash
G28
;X tower Y tower x-69.2 y-40 to  x-95.2 y-55  then x-121.2 y-70 to x-95.2 y-55
g1 x-69.2 y-40 z10 f6000
g30 p0 x-92.5 y-55. z-99999
g1 x-69.2 y-40 f6000
g30 p1 x-92.5 y-55. z-99999
g1 x-69.2 y-40 f6000
g30 p2 x-92.5 y-55. z-99999
g1 x-69.2 y-40 f6000
g30 p3 x-92.5 y-55. z-99999 S-1

g1 x-121.2 y-70. z10 f6000
g30 p0 x-92.5 y-55. z-99999
g1 x-121.2 y-70. f6000
g30 p1 x-92.5 y-55. z-99999
g1 x-121.2 y-70. f6000
g30 p2 x-92.5 y-55. z-99999
g1 x-121.2 y-70. f6000
g30 p3 x-92.5 y-55. z-99999 S-1

;Y tower x69.2 y-40 to  x95.2 y-55  then x121.2 y-70 to x95.2 y-55
g1 x69.2 y-40 z10 f6000
g30 p0 x92.5 y-55. z-99999
g1 x69.2 y-40 f6000
g30 p1 x92.5 y-55. z-99999
g1 x69.2 y-40 f6000
g30 p2 x92.5 y-55. z-99999
g1 x69.2 y-40 f6000
g30 p3 x92.5 y-55. z-99999 S-1

g1 x121.2 y-70. z10 f6000
g30 p0 x92.5 y-55. z-99999
g1 x121.2 y-70. f6000
g30 p1 x92.5 y-55. z-99999
g1 x121.2 y-70. f6000
g30 p2 x92.5 y-55. z-99999
g1 x121.2 y-70. f6000
g30 p3 x92.5 y-55. z-99999 S-1

;Z tower 80 to 110 and then 140 to 110
g1 x0 y80. z10 f6000
g30 p0 x0 y110. z-99999
g1 x0 y80. f6000
g30 p1 x0 y110. z-99999
g1 x0 y80. f6000
g30 p2 x0 y110. z-99999
g1 x0 y80. f6000
g30 p3 x0 y110. z-99999 S-1

g1 x0 y140. z10 f6000
g30 p0 x0 y110. z-99999
g1 x0 y140. f6000
g30 p1 x0 y110. z-99999
g1 x0 y140. f6000
g30 p2 x0 y110. z-99999
g1 x0 y140. f6000
g30 p3 x0 y110. z-99999 S-1
 
Here is my results now,

Code: Select all

5/8/2020, 7:49:11 PM: M98 P"0:/macros/xyzTowerBacklash": G32 bed probe heights: 0.054 0.074 0.089 0.089, mean 0.077, deviation from mean 0.014
5/8/2020, 7:49:20 PM: : G32 bed probe heights: -0.191 -0.186 -0.191 -0.191, mean -0.189, deviation from mean 0.002
5/8/2020, 7:49:31 PM: : G32 bed probe heights: 0.076 0.091 0.096 0.116, mean 0.094, deviation from mean 0.014
5/8/2020, 7:49:40 PM: : G32 bed probe heights: -0.204 -0.224 -0.229 -0.249, mean -0.227, deviation from mean 0.016
5/8/2020, 7:49:51 PM: : G32 bed probe heights: 0.037 0.052 0.057 0.052, mean 0.049, deviation from mean 0.008
5/8/2020, 7:50:01 PM: : G32 bed probe heights: -0.283 -0.283 -0.308 -0.308, mean -0.296, deviation from mean 0.012
Screenshot_20200508_195524.png
Screenshot_20200508_195524.png (426.48 KiB) Viewed 444 times
Thanks
Cory

Boring details. This was a v2 with a duet. It's now a V3 frame, ac heated bed, .9deg steppers, metal ball joint, CF arms, smart effector, maker713 effector mount, 24V power supply.

geneb
ULTIMATE 3D JEDI
Posts: 5100
Joined: Mon Oct 15, 2012 12:47 pm
Location: Graham, WA
Contact:

Re: Backlash issue, again. Was peaks and valleys in bed map

Post by geneb »

Flip the glass upside down and run it again. Different results?

g.
Delta Power!
Defeat the Cartesian Agenda!
http://www.f15sim.com - 80-0007, The only one of its kind.
http://geneb.simpits.org - Technical and Simulator Projects

clytle374
Printmaster!
Posts: 56
Joined: Wed Apr 20, 2016 1:23 am

Re: Backlash issue, again. Was peaks and valleys in bed map

Post by clytle374 »

geneb wrote:
Sat May 09, 2020 10:53 am
Flip the glass upside down and run it again. Different results?

g.
I've tried it with and without the glass, same result. Also that backlash macro shows that it's not on the bed.

Thanks
Cory

clytle374
Printmaster!
Posts: 56
Joined: Wed Apr 20, 2016 1:23 am

Re: Backlash issue, again. Was peaks and valleys in bed map

Post by clytle374 »

Okay, I found it. After picking the wrong place to start(belt tension) I decided that the only thing it could be was the tight arms. I swapped the plastic balls in instead of the aluminum ones. Much much looser. The CF arms are from the previous build, the aluminum balls are new. I can't see any difference with calipers between them, so the difference must be in the arms causing some to be so tight. Some have more wear than others obviously. I did use the grease.

Code: Select all

5/10/2020, 4:01:22 PM: : G32 bed probe heights: 0.114 0.094 0.089 0.089, mean 0.097, deviation from mean 0.010
5/10/2020, 4:01:32 PM: : G32 bed probe heights: 0.049 0.054 0.049 0.049, mean 0.050, deviation from mean 0.002
5/10/2020, 4:01:42 PM: : G32 bed probe heights: 0.114 0.119 0.114 0.114, mean 0.116, deviation from mean 0.002
5/10/2020, 4:01:52 PM: : G32 bed probe heights: 0.094 0.094 0.089 0.089, mean 0.092, deviation from mean 0.002
5/10/2020, 4:02:02 PM: : G32 bed probe heights: 0.152 0.157 0.152 0.137, mean 0.149, deviation from mean 0.008
5/10/2020, 4:02:12 PM: : G32 bed probe heights: 0.072 0.077 0.072 0.077, mean 0.074, deviation from mean 0.003

Screenshot_20200510_160713.png
Screenshot_20200510_160713.png (317.82 KiB) Viewed 422 times
That's calibration, pretty sure I can get that bed map worked out. Any know methods to loosen up the ends of the CF arms?

Thanks
Cory

clytle374
Printmaster!
Posts: 56
Joined: Wed Apr 20, 2016 1:23 am

Re: Backlash issue, again. Was peaks and valleys in bed map

Post by clytle374 »

So this morning started looking for the right arm. Not the arm, it's a couple of the aluminum ball ends. Funny thing is that I can't find any bad spots in it, and calipers don't show it being more the .001" bigger than the others. But this one is really tight compared to the others
IMG_20200511_081453_MP.jpg
IMG_20200511_081453_MP.jpg (4.92 MiB) Viewed 409 times

geneb
ULTIMATE 3D JEDI
Posts: 5100
Joined: Mon Oct 15, 2012 12:47 pm
Location: Graham, WA
Contact:

Re: Backlash issue, again. Was peaks and valleys in bed map

Post by geneb »

I'd contact support and see if they'll send you a replacement. Link to that photo above.

g.
Delta Power!
Defeat the Cartesian Agenda!
http://www.f15sim.com - 80-0007, The only one of its kind.
http://geneb.simpits.org - Technical and Simulator Projects

clytle374
Printmaster!
Posts: 56
Joined: Wed Apr 20, 2016 1:23 am

Re: Backlash issue, again. Was peaks and valleys in bed map

Post by clytle374 »

geneb wrote:
Mon May 11, 2020 8:36 am
I'd contact support and see if they'll send you a replacement. Link to that photo above.

g.
They are sending replacements. Asked about that grease they send with them. $199 for 1.75lbs.

Thanks
Cory

Post Reply

Return to “Troubleshooting”