What are the advantages of the back to back belt & pinion systems? Except for slightly better accuracy I don't see the distinct advantage over belt and pinion. I'm sure there is one, its just not obvious to me. Also, both of the examples use two two sets of idlers (one big, one small) anyone know why that is? Both look like this from the side: [_oO*Oo_] Again seems like one set of idlers would get the job done. Thoughts?
2 belts = twice as strong. Belts are cheap so you can double your machine strength for like $2. You can upgrade belts all the way to the strongest possible and then double that with 2 belts. So with the new steel-laced belts, you can probably get close to a rack and pinion strength with 2 belts.
Hi How i understood a vid on the Bell Evermann (sp?) belt system it is mainly to minimise stretch on the belt. With all but the few cm (,that will "go to and back" towards the motor-pinion,) interlocked there is just a percentage of belt actually exposed to stretch. So the main concern is not the strength- or lack of - but the accuracy going with it. greets flo
HI again, if you look closly in this vid: http://www.engineeringtv.com/video/ServoBelt-Rotary-Precision-Posi you will / might notice that there seem to be no reason to even tighten the upper belt, as lon as it is interlocked with the lower one close to where it is pushed down by the bearings... ( aprox 0:50 and 1:30) you can even seee the upper belt lifting into the air when the motor has passed it. greets flo
Interesting observation you have made there Flo. I have watched that video through a number of times now and notice that a lifted portion of the top belt actually moves when one axes, much further away, changes direction. I don’t think the top belt is as well registered onto the lower belt as we may think ? Tweakie.
True, but the belt is only about 40cm long where it goes over the pulley so the short distance means high strength and little stretching. The weakest part is the long distance between the motor and end of rail. This is the part that stretches the most when tension is applied. Anyway there are a few things that need to be done to make it work optimally. First you should have some kind of a gear which runs another pulley on the lower belt so you effectively have 2 pulleys on 2 belts. Second, you need a tensioner with a spring to keep the belts tight around the pulley. These reasons, plus the patent BS, is why I would look to try a better way.
Again, i do think that strengh in it self is not the problem... just use a say 15mm belt instead of a 9mm and so on if you need more strenght. But As the Dude said ( if i got you right ) you just have 2 bearings pushing a defined protion of the belt " into " each other along a defined length (meaning interlocking upper and lower belt) so that your lower portion of belt akts as a rack . Not only this portion is say glued with double sided tape but also the upper is hindered from stretch along this portion. (the protion that matters... 50 cm away nobody cares... ) Try think of it as focussing your point of efficency to a single area ( where it matters). seeems like a clever idea to me. greets flo
I've got to say it could be a small change to one side of the v-slot profile, so it has a shallow depth to allow one belt to be taped down.
So it looks like we all agree that the primary benefit of two facing belts is increased positional accuracy (less belt stretch). This benefit is obviously more pronounced as your belt runs get longer and longer. But why dual sets of idlers? Is that just to insure a specific contact patch size? All of the examples I can find use one big set and then one smaller set further away from the drive pinion, is that just coincidence? Here is a great example of what I'm talking about: https://www.youtube.com/watch?feature=player_detailpage&v=OdJoVh6DRPA#t=62
Hi, that is the same company ( bell everman) showing off their system. The second pair of idlers is neccessary to ensure, as you said , a big enough contact area. If you like to prevent wear on your (upper ) belt from (over-) tightening you simply leave it slack. Just that easy. But now you want the belt to engage at least some more than just say one tooth at your first set of idlers to transfere the force to a big enough protion of fixed belt (acting as a rack). The lower belt wont wear tha easy as it is firmly attached on its whole length. greets flo
Hi, i just wanted to get back on the thought of strength.. It is true that a system is just as strong as it`s weakest piont .... but in this case you`d have to change your perspective in order to fully understand the genious thought behind. Where the belt lifts and goes around the pully actually is just the drive system. A complete seperate subsystem, only inserted into the whole belt setup, while the portion between the two idlers gives the strenght for acting as a rack and pinion portion. One might even "losen" the tension of the belt around the pully as long as the "rack-part" still beeing firmly interlocked by the preassure of the idlers. This can be seen in the vid of the company where the nema motor is taken off. Actuated by hand it works just the same as with motor. I would bet my underwerar that there is still the same "holding power " in the setup as with motor inserted ( no power of course). So actually it could be viewed as twice as strong as a normal setup (two belts). Of course precision is just good as an ordinary setup assuming both beeing executed with the same care and materials. @ "The Dude" it will be hard to come up with something even better. Of Course i will applaud you then. Greets Flo
This could be easily duplicated especially with-out the gear reduction to see how it works out. My guess is that it would work well for projects that require more strength without compromising speed.
You would think that the belt ribs would introduce a bumpy vertical movement in the x gantry as it travels. Bell Everman must not use the idler bearings as a linear bearing surface . Great forum btw, glad to be a part of it.
Hi, on their web page you can download some pdf with dimensional drawings. There you can clearly see that the whole assembly is attached to an ordinary linear guide system. So the whole thing rides on the rail - just as 99% of all cnc machnines. greets flo
Here are my two cents: Instead of a lower belt, I'd use a machined or 3d printed base where belt tooth can mesh. Instead of a top belt, the carriage would use a closed-loop belt (like a tank track) that would run over the fixed track. For the loop either a regular loop with double-sided tooth belt can work, or a single-sided with a bit more complex (four idlers) can work too (this is what Mike Everman does with his LoopTrack model). Instead of a trapezoidal profile, I would use HTD 5M or 8M. (A ballnose of the right size is all you need to machine a track out of a Delrin bar) Bottom line is that 3D printing (or machining) the track saves you some money and most likely drives you away from patent infringement risk, as that patented technology bases its claims on the use of two meshing belts (IANAL). My interest in this technology is to create multi-gantry CNC machines.
