X axis belt upgrade?

1

Hi all!

I’ve talked about the deflection issues I’ve had with my machine before. I often do some very complicated cuts with this and i’m pushing it pretty hard. I’ve already gotten at least 100 hours of cutting no problem but at while I was making some complicated speakers I was cutting pockets for wood to slot into and they consistently came out 0.05” under the proper width.

I fixed this issue by doing spring passes but my cuts already take 20 to 30 minutes and I’m trying to cut down on the time and want to cut faster. I may also be doing more advanced cuts and 3-D carves for large speaker horns soon and I’m going to be doing lots of calibration to my machine as well as making a new wasteboard.

The y-axis doesn’t seem to deflect as much since there are two belts being used. i’m wondering if upgrading the x-axis to a steel core belt would be a good option. The Y axis doesn’t deflect that much and is pretty stout. I wouldn’t mind upgrading just the X axis belt for a steel core one even though they don’t last as long. since I would only be doing The X axis belt, changing belts shouldn’t be too big of an issue since I only need about 5 feet of it so I’ll just get 20 to 30 feet so I have extra. not too bad of a cost.

My main questions are, how long do steel belts last exactly? What would be a conservative timeframe between changing my belts.

Do they currently have 15 mm 2GT belts? Does anybody have a link to where I can find them?

And finally, Do they make 15 mm 2GT Kevlar belts? I saw another form on here or somebody posted a bunch of graphs about the deflection in the various types of belts and Kevlar seem to really close to steel. At the end of the day I need the most rigidity

edit
I just had another idea. What if I change the pulleys for similar diameter ones that take wider belts? I’m thinking I could maybe do a 20 mm. Basically find whatever Kevlar belts I can in a different tooth/width and then buy a pully to match

Thank you All!

2

The steel belts were sent out by c3d on machines and in maintenance kits for a while. C3d went back to the regular belts because of high failure rates.

The steel core belts work fine until they dont. If you feel the steel core belts are better for your application then try some. Plan to periodically change them protively to stay in production.

The folks at c3d chose the belt sizes for a reason. When buying in bulk for manufacturing the cost would be about the same for 15/20 mm belts. The strain on the stepper and pulley would be higher with 33 percent more width.

You may just have to try it to find out of it os better or worse.

1 Like
3

Yep, steel belts are very much a consumable.

OTOH, at AliExpress prices for rolls of Powge 15mm steel core 2MGT, they are a consumable so, as Guy says, if you’re willing to replace and retension frequently to get to them before they snap a core and mess up a job…

I spent ages looking for Kevlar GT2 in 15mm and found nothing, which is odd because this belting is made in giant strips and cut down to width so it’s not a supply side problem, the market seems to be mostly 3D printers though.

I am quite happy with the 9mm kevlar belts on my XXL, still have them on both Y axes.

I switched my X axis to linear rails and dual 15mm 3MGT on larger pulleys.
The linear rails made a big difference, but nothing like the change that users of the small machines see.

I have not found the bigger belts to be really any better than the kevlar in terms of final cutting performance, the deflections in the extrusions, V wheels, frame etc. all limit cutting performance anyway, especially on the XXL. Stepper deflection also comes to bite you, not a lot of point in adding more and more belt to a stepper which deflects anyway.

If you want precise final dimensions you’ll realistically need to run finish passes with low cutting loads. In this mode the belts don’t deflect much and the frame rigidity isn’t a problem.

Where I need really tight fits I cut, measure and then adjust my offsets in CAD and re-run those final joint faces to get them to final dimension.

2 Likes
4

Yes but my thought process is, do they make 15mm Kevlar belts with different tooth styles where I could just get a new pulley?

1 Like
5

Aha,

Down the rabbit hole you are following, you know there is a mad hatter and we’re having a tea party down here? :wink:

The GT2 / GT3 belt arc tooth profile by Gates is the only one I’ve seen with small enough backlash and enough rated tension load to be effective in this design, it was well chosen by the Shapeoko designers. There are both 2mm and 3mm tooth pitch versions, the 2MGT and 3MGT, the 3mm pitch comes in much wider versions but not at the 40mm circumference pulley size we want for the Shapeoko.

Going up to any of the HTD type belts increases the backlash by at least an order of magnitude as well as requiring larger pulleys (like the 3MGT belts), see the charts on pp11 of the Gates doc (beware the right hand chart is mis-labelled, the Y axis should be the same as the left hand chart).

To use a larger pulley without increasing minimum step size or reducing the apparent motor torque through changing the gearing you need to gear down the stepper (or replace it with a 0.9 degree step version or a bigger torquier one, by which time you might as well replace the controller and your upgrade just went a bit HDM).

IMG_8511
IMG_85111536×2048 200 KB

I geared down my stepper with 2MGT pulleys to drive a 20 tooth 3MGT pair of 15mm pulleys on paired 15mm 3MGT belts, which are thicker and less stretchy as well as wider than the 2MGT I took off. This was a mod to Dan Story’s excellent set of Fusion CAD for the linear rail conversion and all cut on the SO3 XXL.

My finding was - don’t bother, the rails are a good upgrade, the belts, meh, not so much.
Also, if you are going rails and not buying the cheap Chinese knock off rail blocks with no preload, you need to specify no preload with Hiwin or any of the proper manufacturers otherwise you’ll introduce backlash with the axis stiction.

If you can find somebody who will cut you 15mm Kevlar, buy all of it and resell it here, I’ll buy it…

Otherwise the 15mm fiberglass is OK and pretty equivalent to the 9mm kevlar.

Other than that, for the XXL size machine I would say that upgrading to the Pro or the HDM is probably a better route. The SO3 has well matched components and, other than the horrid V Wheels, no real obvious weak spot to upgrade without just hitting the next weakest link.

HTH

4 Likes
7

This topic was automatically closed after 30 days. New replies are no longer allowed.