Timing belts - what's the story?

bend (turning) radius is critical on timing belt for accuracy and lifetime, but also as you increase speed it needs to be increased too.

Bend radius on the spine needs to be at least 1.5x the inner one, or you will break the tension wires.

12 teeths on T5 is indeed a good minimum,16 on T2.5. The value can change depending of manufacturer though

aduy Wrote:
-------------------------------------------------------
> the tooth profile of the t5 is not meant to travel
> in both directions, only one

That is a myth. It is fine for bi-directional motion, and is even recommended for precise linear motion. I don't know where the myth came from, possibly because of the name "timing belt".

I think where people go wrong is having insufficient tension in the belt, which leads to backlash.

the reason i say this is because the square tooth profile does not mesh with the teeth as well as the gt2 belt. [www.gatesmectrol.com] take a look at this and read the description of each.

Edited 1 time(s). Last edit at 05/18/2013 05:19PM by aduy.

691175002 Wrote:
-------------------------------------------------------
> There are no inherent limitations of timing belt
> systems, it is all in the implementation:
>
> [bell-everman.com]
> g/servobelt-linear-sbl
> Has belt systems that can go to 4m/s with 4um/m
> accuracy.
>
> There are 8x4 foot routers that use 2" wide steel
> reinforced timing belts and can rip through metal
> with accuracy in the thou range.
>
> H-belt systems are incredibly fast and precise as
> well: [www.youtube.com]
>
>
> You just need to use belts properly.
>
> Screws of any type (including ballscrews) are not
> a good choice because the inertia of the screw
> makes high accelerations impractical.


man thats what my machine is looking like, im almost done. i havent decided whether its going to be h-bot or core-xy, i think corexy will be better because the belts wont stretch as much.

aduy wrote :

> the reason i say this is because the square tooth profile does not mesh with the teeth as well as the gt2 belt.
> [www.gatesmectrol.com] take a look at this and read the description of each


.That is PR bullshit from a manufacturer pushing his own standard (GT2).

T profile is involute like a gear, and there is a pitch correction to reduce backlash at maximum.

from SDP technical handbook on timing belts :

> The tooth profile of most commonly known synchronous belts is of trapezoidal shape with
> sides being straight lines which generate an involute, similar to that of a spur gear tooth. As a
> result, the profile of the pulley teeth is involute. Unlike the spur gear, however, the outside
> diameter of a timing pulley is smaller than its pitch diameter, thus creating an imaginary pitch
> diameter which is larger than the pulley itself.
> Backlash between pulley and belt teeth is negligible.

GT2 has technical advantages over T profile when you run it at high speeds with high belt tensions, it does wear slower and at high loads perform a tad better. None of those concern us.

But for low loads and high accuracy, I would take T timing belts. The major drawback of GT2 is that it tend to deposit grime on pulleys especially on tights radius. That grime deviate the belt and cause jerk moves. More, the tension wires are not metal like T and dont like much being bended the reverse way.

I used GT2 in industrial systems and dont like them.

Now T5 need normally bigger pulleys than what we use, and T2.5 is a much better choice for our needs. Or we could use the zero backlash AT5. Unfortunately there is no AT2.5 and AT5 needs even bigger pulleys.

The reason many think T belts have backlash is probably that similar belts in imperial dimensions are much swallower and do have backlash but that is not the case for T belts.

Edited 1 time(s). Last edit at 05/18/2013 08:05PM by alj_rprp.

Yes I agree. The diagram shows a gap between the T pulley teeth and the belt which simply does not exist on the belts and pulleys I use. The only exception being some crap T5 8 tooth pulleys I got on eBay from Hong Kong. 8 tooth is not valid for T5 but printed ones work OK.

---

[www.hydraraptor.blogspot.com]

This sounds interesting.
The imperial pitches are supposed to have less backlash, but aren't able to transfer as much power as the T belts because they are shallower. The trade-off of the T is increased space between the flanges of the belt and pulley teeth according to these sources (Gates).
The AT profile on the other hand sounds like it would be ideally suited for our needs and in fact, there even is an AT3 pitch. As with the GT belts, the AT3 pitch is only commonly available as belt from power transmission vendors. Why they don't directly sell the pulleys of the same system is beyond me, but at least there seems to be one vendor that makes AT3 pulleys on demand in the UK.
So, if anyone is interested in trying out AT3, count me in. Maybe we can convince them to make a larger batch for less unit price. Still, I would like to contact them first and ask whether they would recommend MXL, T2.5 or AT3.

crispy1 Wrote:
-------------------------------------------------------
> > I have some ongoing experiments I'm doing with
> fishing line in place of belts on a H-Bot, but I
> think I'm gradually coming to the conclusion it
> isn't worth the investment in time. I have a
> workable solution currently, but the lack of
> slippage comes down to holding very high tension
> on the line (relative to reprap belts), and I'm
> not sure how well the system will maintain that
> tension over extended periods.
>
>
> I built an h bot gantry this weekend that uses
> braided fishing line. With a few wraps around the
> drive pulley I was able to get satisfactory
> movement out of it with no more tension than I
> hold on my toothed belts (GT2). Experiments and
> improvements are ongoing, but I see a lot of
> promise.


Hi all,

I have been using braid (fishing line) lately during the last 6 months or so with what I would call 100% success.
I build Prusa i3's and only use braid on both x and y axis.
I also use bowden extruder setup, so the weight is rather small on the x-carriage compared to a "normal" extruder.
Have printed between 800-1000m of abs on first machine I built and have not had to adjust tension once.
You need a proper system for securing the line.
The added advantage is that you can also print your pulleys for the motors as well.
The cost is minute compared to belts and metal pulleys.
Have done tests after tests and can not see any "side effects" compared to belts yet.

Regards
Jan

Edited 1 time(s). Last edit at 07/14/2013 10:13AM by IceMan.

Sounds interesting, Jan. Sublime did the same with his Tantillus printer; going as far as printing pulleys for the Z axis in his cable driven Z design. I never got the prints come out perfectly round, so I pretty much abandoned the thought of having printed fishing line pulleys.
Can you upload some photos and tips of how you printed those things? My interest is rekindled at least.

Edited 1 time(s). Last edit at 07/14/2013 10:57AM by uGen.

uGen Wrote:
-------------------------------------------------------
> Sounds interesting, Jan. Sublime did the same with
> his Tantillus printer; going as far as printing
> pulleys for the Z axis in his cable driven Z
> design. I never got the prints come out perfectly
> round, so I pretty much abandoned the thought of
> having printed fishing line pulleys.
> Can you upload some photos and tips of how you
> printed those things? My interest is rekindled at
> least.


Hi,

To build my first i3 I used my own designed printer. I now make all my parts more or less with the i3 as it runs very quiet.
I use Kisslicer at the moment (full license)
All printing is ABS
My own designed and made hot end/extruder
As you know I use braid for both x and y (working on a way of using it on the z as well)
Have attached some photos of printed pulleys and i3 parts.
I am very happy with the print quality I am getting especially since I use a bowden setup.
Have attached some photos for you to look at. All these photos show parts made with my i3 using braid line.

Regards
Jan

Way back at the dawn of CAD there was the HP 7585 pen plotter. It moved a pen back and forth over more then 36 inches. The total stated accuracy for the toothed belt, pulleys, pen holder, and slide was .004 inches. The repeatability was better then that.
[upload.wikimedia.org]

Frank

Hi Jan,

Nice quality on your parts.
Can you show a picture of your i3. I am curious how you did the wiring.

Regards,

Another Jan.

Whoa, those parts really look impressive. Am I right in assuming that you only use loops and perimeters for the pulleys to minimize loss of concentricity due to the infill pattern?

Hi,
Sorry for the late reply.
I have some photos of my first i3 on my blog, link in my signature below.
And yes I am very happy with the print quality. It is probably no better then any other good printer set up correctly.
I was looking for info some time back regarding i3's print quality and found very little info.
For anyone out there wandering if it prints good...yes it does (mine has got a bowden setup).
Running the printer as much as I can at the moment just to see if the braid setup will show any sign of failure.
So far it is working well above expectations.
I have tried to keep the wiring a bit neat as i can't stand all those wires all over the place.(not always an easy job)
The braid line just wraps around the pulleys. Roll-on-roll-off.
Regards
Jan

Edited 1 time(s). Last edit at 07/22/2013 08:15AM by IceMan.

---

======================================
[3dprinting4u.wordpress.com]

this is cool and i like how smooth the finish is, my dad used to have this yellow french horn string that was very strong with no strech, i wonder if that would be any good for this sort of thing.

Edited 1 time(s). Last edit at 07/22/2013 05:53AM by aduy.

I just come to this thread, and I agree many thing that has been already written. Especialy I agree both 691175002 said. But some sayings about ballscrew supremacy made me jump over my seat.

Everything here is melt in a bucket about ballscrews drive and belt drive. If you compare a chinese GT2 system with a C0 grade ballscrew system, yep, the ballscrew wins indeed. But you compares performance of a walker versus a Ferrari. They don't cost the same, and chinese GT2 are the worst quality ever. It's as true as they are the cheaper ever. It's also as true as you will need to spend $1500 in quality ballscrews to work a 8x8x8" printing area, if you want to disqualify belts.

First, while putting the quality factor away, you cannot compare a leadscrew motion with a direct drive belt system. The screw system multiply the resolution and torque and simulteaneously divide speed and acceleration rate. If you wanna be fair, you should apply the belt system a reduction to match the same travel by motor rotation. This will raise the belt system to a very close performance compared to a screw system.

The most important comparison point about the two systems, is load. Unlike the screw, the belt can stretch under load and this can introduce precision issues. You definitely need an accurate belt selection to prevent any stretching. Some belts are armoured with various materials, including carbon fibre, metal aloys or kevlar to limit stretching. Somes rubbers are loaded with charges to reduce friction and vibrations. All this have a high impact on the quality of the belt and its price. You want to choose different belt pitch, width, and specification to match your needs. While a belt is limited by load, a screw is quickly limited by speed. Each screw have a critical speed you don't want to reach. To push the limits, you need to reduce the screw lenght and increase the diameter. There's a way to use fixed screw with rotary nut, but it last shorter and it's mecanicaly more complicated to do.

Ballscrews requires high mounting accuracy. Any alignement problem will quickly introduce wear in a balscrew system, and will make the whole driving system don't last, from motor to the end. Well mounted, ballscrews still needs to be greased frequently to last. A belt just need to be tensioned. It can be twist and light missalignement is no big deal. A belt system does not require anything until the periodic check-up. The belt is definitely simpler to manage, from design to use.

Now, let's reintroduce the quality factor. At comparable price/travel, a chinese C7 ballscrew system is no better than a good belt drive system, even if the screw still cost a bit more. It's louder, have more wear, and will last shorter than the belt. If you apply the same milling precision to pulleys as needed to make a ballscrew, the belt system will be ways better. In the extreme configuration of a C0 ballscrew system (that cost a fortune), it will be difficult for the belt to stand the same load with the same accuracy. But you must appreciate the overall quality of the machine must match the ballscrew precision. On my big CNC Mill, I reach a 200µ precision along 2500mm(6.5ft) with a belt system. My cuts are pretty clean, so I believe I have no vibration issues, that's enough for me. I don't need a 20µ accuracy.

What about 3D printing ? Well, even if there's some nice attempt to make a reprap with ball screws (Lautr3k), that driving system is not what you want for a 3d printer. You need velocity and there's no significant load to manage. You could run leadscrews at high speed with 3000 rpm servos, but the screws needs to be at last 16m, and you will have a slow acceleration anyways. On a single 8" axis, the C0 screw, the 3000 rpm servo and his digital driver will cost you about $550. AND, you'll need linear rails with equal precision (of course heavy), which will cost about $300/axis. And the overall machine design/stiffness must match, that's additional expenses. Hmmm. Worth it the money ? Definitely no. The FDM technology makes that amount of precision totaly useless, until you'll be able (and have enough time) to print with a 0.05 nozzle. For a little bit of that price, you'll combine a 0.9° motor with unit manufactured precision pulleys, a 3:1 belt reduction, and low friction HTD gt3 quality belts. The precision will be ways enough for your printer and it won't be difficult to install.

A belt system is what a 3D printer needs. With the CoreXY or Delta technology you can run much faster than your filament can extrude anyways. So you could install a moderate belt reduction to improve printing quality without increasing significantly the printing time. If your stepper motors will run faster, they will introduce less vibrations too. Win both ways, and spend less.

Edited 1 time(s). Last edit at 03/13/2015 06:24AM by Zavashier.

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)

Hello
I had made a scaled up prusa i3 recently, and experience layer offsets in my x axis (the 400mm) axis when the print reaches the height of about 1-2cm. I believe its not for under or over currenting and there is no problem with my stepper drivers and motors, becuase the layer shift doesnt happen at every layer, and only happens after a while, and the shift is about 1-2cm. The belt on x axis vibrates like a guitar string, and I believe thats the cuase of the problem. As I used GT2 belts, is there any other alternative which can handle the 400mm axis?