Question: Does anyone know how if there’s been work done with V slot bearings running on T-slot? Obviously the T-slot would need to be rotated by 45 degrees around it’s axis so the bearings were aligned on the corners of the profile. I see @Shauki is making progress with his roller bearing based x-y axis on the quadrap and I’m thinking of having a play with the cable-bot design @David_Moorhouse was working on as an alternative to core-xy.
I’m aware a steel bearing and aluminium surface might not be ideal. Long run, the T-slot could have a thin UHMW strip applied to the running surface or nylon bearings could be used.
http://www.aliexpress.com/item/20pcs-624VV-V-Groove-Sealed-Ball-Bearings-Vgroove-4-X-13-X-6mm-2-2-5mm/1679655671.html
@Tim_Rastall this has been tried many times over and over and although with some success but in the long run the only real way to go is with proper linear guides&bearings. They are not that expensive any more as they used to be, much more wallet friendly and for sure more durable and accurate. Remember, any machine is as good as its weakest part.
What worries my about the V-groove bearing is accelerated wear on the guide. A v-groove running along a corner is not pure rolling motion. They have to slide some because of the difference in diameter running from the notch out to the flange. Since these are steel, the guide would need to be slick and wear resistant. Adding a UHMW strip might fix that.
You still end up with a compromise between rolling and sliding friction though.
I use IT for rail system… Not perfect but works…check video on my g+ and this my system https://www.youmagine.com/designs/624vv-vgroove-rails today print item for make test with corexy table
http://buildyourcnc.com has done lots with v bearings and aluminum angle. Lots of nay-sayers but they have build a business on it so it cant be that bad.
@Matej_Rozman Without wanting to appear that I doubt your advice - any examples please?
Specifically, What are the failure modes that make a roller bearing less viable than linear slides?
I’m trying to understand what durability/accuracy payoffs are versus cost/printability etc. I’ve used linear slides, linear bearings (LM*UU etc) Self aligning bushings etc and they all have their down sides, on face value a rolling bearing is a good choice for something that go’s back and forth a lot :).
AFAICT accuracy is constrained by the straightness of the static member on which the carriage runs and the tolerances of the interface between the carriage and the static member.
@Shauki has a number of viable looking use cases for roller bearings and he’s got a doctorate in Mechanical Engineering so I’m interested in exploring the idea if nothing else.
@Mauro_Manco did you consider an adjustable pre-tensioner in your carriage to dial in the pressure/slop?
@Tim_Rastall I agree with your conclusions about v roller bearings. But you need to remember you need to make sure your run for the bearing allows for even wear.
Or you’ll start to get artifacts if the bearing wares unevenly.
@Tim_Rastall there plenty of them in the delta Google group, first iteration of kossel had run on bearings carriages and aluminium rail. Look the Kossel today, linear rail&bb carriage. I’ve done plenty of short cuts in my life and though some were successfull none even came close to proper linear rail. And please note, im not talking about 1$ LM8UU, but even that will work better if used wisely.
@Matej_Rozman , I think what @Tim_Rastall will want to know is why it’s not used on Kossel anymore. Was it not precise enough?
@Dale_Dunn Agree. My immediate thought is that rolling component allows you to reduce the chances of binding problems and an easilly replaced sacrificial UHMW strip or whatever might not be the end of the world…
Yep.but not apply…because no find simple way for it. Surrely with pre-tensioner make this better. Try To publish something in weekend for XY table currently waiting fish line. But surely is a real ultra low cost idea and system for make rails. Bearing not have great quality but i love it. Good night guys
With these loads, the UHMW may not wear out. I am a little worried about slop in the axial direction though. Unless the bearing is internally preloaded, there will be some movement. Not that I can’t think of a way to fix that…
@Dale_Dunn yes and no. All kind of solution are ok at the beginning, but then over time stuff happenes. This is why i am promoting proper solutions, because they just works, even years later and even many abuse.
@Matej_Rozman
@Dale_Dunn is correct. I’d say a delta is the worst use-case for this application due to the variable force vector applied to the effector arms by the carriage. I’d expect that a Cartesian system where the force vectors are static would be less troublesome.
@Dale_Dunn I had a similar concern. without testing it’s going to be hard to tell. I’d though of third bearing oriented at 90 degrees to the plane on which the opposing bearings sat could solve this. ‘Plenty of possible solutions to this’ was my gut feeling.
@David_Moorhouse 's cablebot will have some of that variable vector stuff going on. That is, axial slop in the bearings will appear in the print, if not mitigated.
@Matej_Rozman Interestingly, the ‘stuff happens’ point is one of the reasons why I’m interested in this. I’ve recently had some challenges arising from bronze bushings wearing out and introducing slop/binding issues. I’d like to have a solution that allows for some adjustment with wear, which isn’t practical with LMs or bushings.
I’m also interested in something that can scale hugely. Linear rails and shafts have to increase in diameter as travel distances increase with significant cost implications.
“Scale hugely”. Uh oh.
Thanks for posting this, @Tim_Rastall . I do see some trouble with a dry metal on metal contact with these bearing since there will always be a speed difference between where the innermost and outermost part of the bearing touch the rail. Makerslide uses polymer v-groove bearings, but having a low-friction (and consequentially less wear-resistant) lining on the larger partner (the rail) makes more sense imo.
What i see in the -vv bearing types, though, is an ideal idler for spectra line. Alignment-wise they are practically identical to 62* bearings with a machined groove and only have a slightly larger outer diameter.