Backlash/hysteresis software correction

3D Modelling, Slicing, Hosts and Management
Post Reply
FranJans
Posts: 41
Joined: Wed Jul 19, 2017 10:35 am

Backlash/hysteresis software correction

Post by FranJans » Mon Nov 06, 2017 9:11 pm

Hi there
I'm trying to get the best possible quality out of my printer. I think I'm pretty much done with optimising mechanics after more than 1,5 years of modifying almost everything on the printer :lol: I recently measured the backlash/hysteresis on the print head and came to the conclusion that there is still 0.05 to 0.1mm of backlash/hysteresis left. This is mechanical speaking the best my printer can do :( So I was left wandering if there is another option I should try. I would love to try software backlash/hysteresis compensation.

I'm currently using Marlin 1.1.0-RC6 firmware with slic3R slicer. As far as I can see both don't support backlash/hysteresis compensation. Does anyone have any idea how to do this?

Before anyone asks. Yes, I have really done anything I could do about the mechanics.

Thanks :D

Trakyan
Posts: 8
Joined: Mon Jul 03, 2017 8:39 am

Re: Backlash/hysteresis software correction

Post by Trakyan » Thu Nov 16, 2017 2:21 am

I recall something similar being implemented in skeinforge. I also recall only hearing bad things about it, like it made things worse or cause other issues. I don't think there is a good software fix for it out there right now but if you're using belts for XY and belts/screw for Z you shouldn't have significant backlash issues as long as belt tension is good.

FranJans
Posts: 41
Joined: Wed Jul 19, 2017 10:35 am

Re: Backlash/hysteresis software correction

Post by FranJans » Thu Nov 16, 2017 5:38 pm

Correct. Skeinforge had backlash correction.
Do you call 0.05 to an absolute maximum of 0.1mm significant backlash? :mrgreen: I think it is rather low. I think it is mostly belt stretch and motor inaccuracy.

I have been busy writing a program to modify Gcode files. It was a lot more difficult than I thought. However, I got it working. The results are as you described. Rather horrible at some points in the print but also rather nice at other points in the print. Strangely enough the printer did skip a step which corrected itself 5mm higher up. This never ever happened to me before.

This is an extremely exaggerate example of my program at work.
Image

The problem is that I can see the jumping points in the finished product. There are two things I haven't tried. One is keeping the extruder running while doing the backlash correction and the other is not purely moving in the X or Y direction while doing the backlash compensation. I can also move more in the direction of actual movement of the print head. The makes the printhead move smoother. This is even more difficult though :shock:

FranJans
Posts: 41
Joined: Wed Jul 19, 2017 10:35 am

Re: Backlash/hysteresis software correction

Post by FranJans » Sat Nov 25, 2017 11:29 am

There are two things I haven't tried. One is keeping the extruder running while doing the backlash correction and the other is not purely moving in the X or Y direction while doing the backlash compensation. I can also move more in the direction of actual movement of the print head. The makes the print head move smoother. This is even more difficult though
I did try this eventually. It still created worse results than no backlash correction. I came to terms with my print quality, which is actually pretty good. No backlash correction for me :)
It was an interesting study though.

Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests