I ran across this a little while ago:

[www.bbman.com]

Apparently MXL belts are about 25% stiffer than GT2 (2mm pitch) belts. I use MXL belts because they are a little more available, but I know most people are using GT2 belts because they theoretically have less backlash. Does anybody have experience comparing the two in the same machine?

The resistance to stretching is mostly down to what sort of stiffener a belt has embedded in it. Some have steel or carbon fiber embedded.

There is wide variation between the different variations, even within the GT2 range, let alone comparing it to other types of belt.

Quote
DragonFire
The resistance to stretching is mostly down to what sort of stiffener a belt has embedded in it. Some have steel or carbon fiber embedded.

There is wide variation between the different variations, even within the GT2 range, let alone comparing it to other types of belt.

I'm just looking for an apples-to-apples comparison of MXL vs GT2 (2mm) - standard neoprene belts with fiberglass core for both.

Given the same materials used for production, there should be little to no difference in equal width/length belts. Any difference at that point would come down to quality control or amount of embedded fiberglass which I imagine would vary based on factory it was produced in. The real difference would be in looking at other core materials as mentioned above.

Quote
obelisk79
Given the same materials used for production, there should be little to no difference in equal width/length belts. Any difference at that point would come down to quality control or amount of embedded fiberglass which I imagine would vary based on factory it was produced in. The real difference would be in looking at other core materials as mentioned above.

Yes, you would think. But in my original post, I included a link to this Gates document (presumably for both types of belts manufactured by Gates) that shows per inch of belt width, an MXL belt has a stiffness of 23,000lb/in/in but that a GT2 belt has a stiffness of only 18,000 lb/in/in. I'm inclined to believe Gates' numbers (being the designer of the GT2 belt), but there is also the tradeoff with the potentially lower backlash of the GT2 belts. So my question was if anyone has tried both on the same machine for comparison.

Fair enough, being the manufacturer I'd lean towards believing their numbers as well. I am going to be switching from mxl belts to gt2 in a couple of weeks hopefully. But it will be coming along with a redesign of my mechanical motion as I frankenstein my CubeX into a CoreXY machine I've dubbed the CubeXY Project. I'm not sure if it would really be useful as a direct comparison between the belts in terms of real world performance. Additionally it'd probably just be best to do a true stretch test. take 1M of each type of belt and attach weights to them to test the amount of stretch for a given weight. It would be relatively simple to do. If I have any GT2 left over after my conversion I'll gladly set a simple test up to make a comparison and provide my results here.

I'm not an engineer, but the other question you need to ask is what kind of dynamic force is being applied to your belts while printing? This is something which is going to be uniquely different to each printer due to moving masses, print speeds, acceleration rates etc.

I'd be more interested in proving if both belts qualify as 'good enough' I'd postulate that the answer would be yes. I get very high quality prints using my MXL belted printer as it is. I doubt backlash concerns for this type of application have any appreciable affect. I chose GT2 because it was available cheaply on amazon and I'd rather not take the risk with the backlash as I also imagine that backlash in your mechanics could have an additive effect which may become noticeable.

Quote
obelisk79
Fair enough, being the manufacturer I'd lean towards believing their numbers as well. I am going to be switching from mxl belts to gt2 in a couple of weeks hopefully. But it will be coming along with a redesign of my mechanical motion as I frankenstein my CubeX into a CoreXY machine I've dubbed the CubeXY Project. I'm not sure if it would really be useful as a direct comparison between the belts in terms of real world performance. Additionally it'd probably just be best to do a true stretch test. take 1M of each type of belt and attach weights to them to test the amount of stretch for a given weight. It would be relatively simple to do. If I have any GT2 left over after my conversion I'll gladly set a simple test up to make a comparison and provide my results here.

A CoreXY machine would be a really good test of belt stiffness because of the relatively long belts. The main thing I'd be looking for in a stiffness comparison is the amount of "ringing" that appears on a flat surface just after you take a corner. In other words, print a plain cube but set the "jerk" value up high enough so that some "ringing" appears. And then compare the results from the 2 different types of belts. For a backlash test, printing out a cylinder works pretty well (back to normal speeds and jerk values). If there is any backlash, you can measure flat spots across the cylinder at the points where each axis changes direction.