Backlash?

[lasershark1]

Ok, Im seeing backlash everywhere - what is it? Google is no help, it's too common a word to show me in a CNC context...

[calica]

Imagine the X moving in the + direction. Then it reverses and moves in the - direction. Backlash is the amount the machine thinks it moves before there is any actual movement. Its caused by the slop in the drive system. It causes circles to not be round.

Take a look at this object for 3d printers. It's to measure backlash. It'd be really nice to have something similar so we can measure our shapeokos. It could be just gcode. Doesn't matter what the tool width is.

[minorthreat]

Backlash is caused clearance between components in a mechanical drive train.The simplest case is in a pair of spur gears, backlash is the amount of clearance between mated gear teeth. During a direction reversal the driven gear has to first move back through this clearance (dead-zone) which creates no axis movement.

In the ShapeOko world: The Z axis is a screw drive it needs some mechanical clearance to operate. Luckily some smart person minimized this backlash by using a tight tolerance delrin nut. Eventually this nut will begin to wear and will have to be replaced. The X and Y are belt driven, properly tensioned mxl belts have a little backlash, and have significant compliance, switching the belt to a curvilinear or a 60deg tooth would further reduce this if anyone runs into any trouble.

[h0st1le]

This actually brings up something I have been wondering. Is there a specific reason that MXL timing belts were chosen? From what I understand XL belts are designed for use as timing belts, while GT2 belts are designed for linear motion. On my prusa when I switched from T5 belts to GT2 belts I noticed a significant improvement in my print quality. With the Shapeoko being a CNC machine I would have assumed it would require even tighter tolerances. Is there much to gain from a belt switch in this case?

[minorthreat]

(M)XL's are still positive engagement belts, they're trapezoidal and at low speeds tend to have wedge engagement at the root of the tooth. The big benefit you get with curvilinear belts (HTD GT2 etc) is under higher loads and higher tension. On small drives using a low installation tension ,small pulleys, light loads, and low teeth in mesh (like the stock shapeoko) Curvi and trappy belts have about the same performance. At higher tension and on bigger pulleys Cruvi belts are significantly better than trappy belts.

So IMHO if you're extend the load handing capabilities of your mini mill already, go ahead and add the GT2 belts. It's just a sprocket change.

[h0st1le]

minorthreat, thanks for the very informative answer. I was thinking about this a bit more. I believe the reason I noticed such a large difference on the reprap, was most likely due to the tooth spacing. The more densely packed teeth of the GT2 belts were a significant upgrade from the widely spaced T5 belts.

cheers

[calica]

Did you also go from a printed gear to a better one? That would explain the improvement too.

[h0st1le]

calica,

Yes, I swapped out the T5 printed pulleys for commercial GT2 pulleys. To my knowledge there has been little success using printed pulleys with belts of smaller tooth spacing (XL/GT2). I'd venture a guess this is because most current FDM based printers are simply not capable of the required resolution.