Which is a more accurate linear motion system?
If someone looking for accuracy, which is the better linear motion system:
- Maker slide,
- Smooth rod - bearing,
- Linear guide way,
- V slot,
- Open rail
Please let me know if there's a better one other than above list.
"more accurate" is way too open: in WHAT way more accurate? in the direction? in other directions? under what loads? in which design? there is no "in general" as each of these guidances has its planned applications!
The accuracy of a printer depends way more on the design and engineering of it that on the motion system used. Using expensive super high-precision components won't make a bad design good.
This seems like a relevant topic. What does the question need to be good? Is it 3 questions for each of the 3 comments so far? Obviously the question can link to the current relevant questions, bu that else?
@SeanHoulihane it is relevant, but the question is not framed well for the Q&A format. there might be a question for each design, or there might be a question that might work out a comparison for one specific design or design goal (e.g. "I have a super lightweight toolhead in a bodwden setup. What kind of motion system would serve best?")
What I want is positional accuracy. I agree that belts play a huge role in that. How to improve belts.?? Is there any other way to move the rails?. I read in HIWIN that using ball screws with linear guide ways improves accuracy significantly. But using screws instead of belt, wouldn't it affect the speed???. I think using ball screw for Z axis is a good idea, but it's lead should not be so large that weight of the platform + print shouldn't cause it to slide down(I don't know the technical term for that).
@Athul please edit your question to reflect that. As a side note: the linear guidance has only little to do with the positional accuracy of the belts/screws. The other thing you ask about is a question of its own, a rather good one.
I used a few different linear rail systems in my build:
plain smooth rods for Z motion
smooth rods bonded to a T-haped steel base for a 1200mm Y axis (similar to those at https://cdn.automationdirect.com/static/specs/suremotionlinear.pdf)
igus "low profile linear slide" for X (https://www.igus.com/drylin/profile-rail-guide)
They all seem to work pretty well. The one problem I had was with the igus slides, which I found had a little too much play -- so each change of direction on X had a slight backlash, from the "cars" twisting a tiny bit within the rails. I improved that with careful tuning: shifting the rails a tiny bit farther apart, so they kept the cars under a little tension against the inner sides of the rail -- some call this "preloading".
I think systems that use actual bearings should generally be more accurate than spring-loaded slides. But the igus rails are still pretty good, and they're quite light, compact, and reasonably priced. I still use them, though once in a while I think about swapping them out to do a serious comparison.
The rails aren't the only factor in accuracy, though. I can't detect any play or warp in my Y rails, but that says nothing about accuracy and repeatability of motion along them. That's controlled by the motor and the belt, leadscrew, or other things actually moving them. Leadscrews, in particular, vary quite a bit in accuracy, depending on the shape of their threads, the kind of "nut" riding on them, and other factors.
You can even make your own leadscrew system from just a threaded rod and a nut -- but those threads are not the same, and they allow far more play than a real leadscrew does. Nice article on the "backlash" problem at https://www.liutaiomottola.com/Tools/Backlash.htm.
I recently replaced my linear bearings with Igus polymer bushings. I didn't find the same thing. The bushings were, if anything, a bit tight. I was more worried about them sticking on the rods than there being too much play. Your solution seems like a reasonable one, though, to take up the slop.
I have a 3D Printer (20 x 20) using linear guide way, a clay print(700 x 700) using V-Slot and a CNC using makerslide. I dont't have a positional encoder to check repeatability
License under CC-BY-SA with attribution
Content dated before 7/24/2021 11:53 AM
Davo 4 years ago
If you're using belts, the difference in accuracy will be negligible - provided you have good quality bearings in whichever style(s).