@Whosa_whatsis Agree, the simple does have slipping problems and it is a design flaw IMO. The Ingentis/Tantillus design allow for the spectra to be tightened to a degree that would break a simple. The key to spectra working is friction and tension so I suppose I should caveat my above statements with that key point.
WRT to the platonic ideal - I hear you and I admire your idealism, I guess I’m more of a pragmatist and (for the time being at least) I’ll stick with spectra. Great discussion though. Sorry for hijacking @Nicolas_Arias
nothing to be sorry about, its a great discussion and we all have learned something. Now, im going to burn my printers and cry on a corner.
@Jarred_Baines you not much of a fitter as your saying its 2x better.
Sorry but if thats the case its more efficient than gears…plainly wrong!
@Whosa_whatsis if its a single line and not looped some of your points become invalid comparing them to a joined belt.
And the design of most belt systems on printers is poorly thought out with no thought to side ways movement or belt tension. And as we are a group of hackers isn’t it worth a try. If it works great if not it’s a lesson learned.
@Nigel_Dickinson what are you replying to exactly? Where did I say it was twice as good?
What is your measure of efficiency? Belts of all types are more efficient than gears - too much friction in gears. Gears do deliver more torque than belts though…
Don’t start deciding how good a fitter I am mate, pull your head in with comments like that… I’m stating facts and backing them up here, its not a competition, just a discussion…
100 % means double as any maths person would know. You mean better…
And as a fitter you should know this
And by the way a geared pulley and toothed belt is a geared system…
Your maths is 50% correct…
I say this because as we all know 100% is double and its impossible to be double right… Right?
Also your understanding of a toothed belt system is wrong… Just because it is a positive drive doesn’t make it a geared system… It has a belt FFS and it can stretch and slip and is tentioned… Gears break - they don’t slip…
@Jarred_Baines sorry thought you were a fitter gears do slip if they aren’t keywayed. And a correctly set up gear system has play to allow for expansion. So a geared system is what we need then instead of belts and line.
Yet others say gears are not good enough for printing.
And the same for lead screw’s…
Seems to many people are applying lack of real use with what they hear.
So either your now saying belts are good because they stretch and slip or we are using the wrong equipment…which is it?
The Grabber spools have a spiral thread that the spectra travels in, so it can never overlap itself and is 100% repeatable, I have never seen any inaccuracy with my large deltas driven by spectra and grabbers threaded spools.
Depends on your ideal, if its speed at the RISK of accuracy, go with a flat belt / friction system.
If its accuracy and speed go for a toothed belt system…
If its high load and accuracy at the expense of speed and efficiency for eg a milling machine, leadscrews are perfect.
We make waterjet cutters at work, they’re (flat) belt driven but they have an encoder with a toothed belt for positioning, why? Because the drive belt CAN give errors.
3d printing is a low load situation but has many MANY movements in a single program. So a fast drive is beneficial, narrowing us down to belts as a obvious first choice… To you the smoothness of the flat belt outweighs the repeatability of the toothed belt… To some people the rigidity of leadscrews outweighs the speed advantage of belts…
If that’s not enough of an answer ask elsewhere apprentice, not wasting any more time arguing the point, its not helping anyone…
Stratasys have around 25 years experience and have no doubt spent millions on research and I am confident they use toothed belts… As do all 2d printers and scanners that I have encountered… Its a logical choice when positioning is critical, speed is favorable and load is low…
@Matt_Kraemer1 , It’s exactly this level of obsession that got us to the point we are today. Power transmission used to be a big problem because of how some printers (ab)used belts. Nothing less than perfection.
@Jarred_Baines well argued so why didnt you say at the beginning instead of bluster and false facts.
Still didnt answer on lead screws considering they are more accurate and as fast. And accurate up to .008 mm over a metre.
Interesting discussion. I think there is a cost/benefit to the various drives we have available to us. Many of us are price sensitive, and there is a point that is ‘good enough’ for the amount of money it will cost to build. Many people are pushing their drive of choice nearer and nearer to ideal.
The high end precision industrial Cartesian motion I am familiar with is done with servos, acoustically tuned tensioned belts, and scales. The kinematics are laser mapped to 21 degrees of freedom and compensated for so they drive nearly true. The costs are many orders of magnitude greater in order to achieve near ‘ideal’ motion. This is overkill for most hobbyists.
For my Rostock, I am compromising with a hybrid spectra/belt system. The actual drive is done by belts. Drive gear to carriage is with a belt. From the carriage around the idler and across the ‘back’ of the carriage is with spectra. Is this better? I don’t know. Maybe because there is no belt teeth running over the idler pulley? I do know that I gave the rostock a larger build volume with minimal extra investment. My ROI was high.
Anyone have links to these Grabber Spools that have been mentioned?
@Nigel_Dickinson
I mentioned leadscrews explicitly in my last post…
They are accurate (degree of accuracy is not “.008mm over a meter”, it’s determined by the tolerance they have been manufactured to as well as other factors like elongation of the material they are made from, thermal expansion, wear etc) But only someone with education on the subject would know that…
And they can be fast, faster if they are high pitch and multiple start threads, slower and more accurate if they are finer pitch. BUT in a HOBBYIST machine (as @Mike_Creuzer pointed out) we would prefer to use $20 motors to drive our machines. The power transfer in a belt system is much more efficient than a leadscrew. This is due to lower friction. Belts are also cheaper making them attractive.
And my 2 belts that I’ve used have been driven by 16 and 20 tooth pulleys, with a 5mm pitch on the belt, that gives 80mm / 100mm PER REV of the motor. I do not think many leadscrews have a pitch that high so they obviously (sorry, I shouldn’t say obvious… “obvious to anyone with engineering skills” is what I really mean) will not be as fast per rev generally speaking although they are more accurate if 1 rev only gives 8mm movement instead of 80 ( in this case, 10 times more accurate, if you do the math… correctly…)
Don’t be such a troll Nigel, you’ve been on my case from my first comment and I’m just trying to help people out here and pass on information and knowledge that I have acquired. You are nit-picking my comments to try and make me seem less credible… who cares if I didn’t mention leadscrews? I didn’t mention hydraulic pistons either, why don’t you hang me for that next?
…
Moving on…
I’m actually quite intrigued by this spectra line now, there’s a lot of people standing up for it, STRONGLY… That has to say something!
@Jarred_Baines I only dug at you as you dismissed it out of hand. If I offended you sorry…but from experience I know various drive systems suit different jobs.
Ive used spectra and belt and find they are good enough in hobby printers.
Ive a delta I’m working on that will have twin pitch lead screws.
If it fails I’ll turn it into a CNC mill.
But like I said we are hackers and sometimes prove industry wrong.
I hope I opened you eyes even if I was slightly to harsh…but dissmisal of an idea is often done with out thought because of preference.