Welcome! Log In Create A New Profile

Advanced

X Y axis accuracy doubled.

Posted by Madkite 
Re: X Y axis accuracy doubled.
February 20, 2012 07:03AM
My X-axis is happy with it. But that's mostly due to the modified x-motor-bracket.




reprapworld.com
PLA, ABS, PETG multiple colors (3 and 1.75 mm) €16,49 / kg
Megatronics 3.0 €79.99 / Minitronics €37,19
Re: X Y axis accuracy doubled.
April 15, 2012 07:07AM
A few people have mentioned about only half the torque being required with this set-up and some have even mentioned about adding gearing to make this a quarter.

I'd just like to point out that with stepper motors, their torque rapidly degreases as the speed increases.

This belt set-up is effectively a 2:1 gear ratio, so it requires the stepper motors to run at double their normal speed for the same carriage speed. This reduces the torque at the top speed greatly.

In addition to this, a lot of people seem to run their stepper motors on 12V, which also limits their high-speed torque.

In terms of accuracy, you are adding a number of extra idlers - The run-out of each of these will decrease the accuracy. (For example, the loose fit between the bolt and the bearing.) Additionally, you have a much longer length of belt which can stretch more.

I can't argue with the results you have seen however, so there must be another factor that I haven't thought about. I thought it was worth mentioning the above though so it can be taken into account for future iterations.
Re: X Y axis accuracy doubled.
April 16, 2012 06:27PM
I'm considering going back to a normal setup on my VertX, sometimes on a fast move it was too fast for the motors and it skipped many steps (at 200mm/s, but not at 150mm/s travel)
Re: X Y axis accuracy doubled.
April 18, 2012 08:04AM
Been meaning to do some accuracy test videos but I'm bogged down with work and life stuff right now. Also started work on my metal/plastic printing system. Not sure if it will work but when I have it completed and some results I will let you know.

The belt is not quite twice as long but only carries half the torque as its effectively using two strands instead of one. I have not had to to adjust it much only for wear. I get the feeling the belt is way stronger then the kinds of loads I'm putting on it and it stretching or springing has not been an issue. As for slack in the bearings and fixtures. All the bearings are clamped tightly so they are not loose and as the belt is tensioned slightly when you set it it should not suffer from slack anyway. Unlike a gear train where every piece of clearance adds up the preload in the belt removes that problem.

I don't think my motors are running anywhere near there limit but my printer doesn't go that fast. I have a very heavy bead so I can't and my extruder is not up to it.

If you really want a printer to go fast it just means you need bigger more powerful motors and drive circuitry. As well as stronger components. The whole point of my idea was to try and get the best compromise out of the stuff that's easily and cheaply available without it getting too complicated and technical or out of the experience of the average Reprap user. If you have built a reprap then you can do this mod. That was the limits I put on this particular mods scope.


Make your Mendel twice as accurate.
[www.thingiverse.com]
Re: X Y axis accuracy doubled.
April 18, 2012 11:32AM
I've been thinking about and designing a CNC router with a 1.5m x 750mm bed, that uses the belt doubling layout for the X and Y axes. I think it will need doubling when using NEMA23 motors to provide enough torque to move the gantry, and allow the use of larger pulleys on the motors for better grip, while keeping positional accuracy. Speed is less of a concern when cutting, as feed rates are lower. I'm a bit worried about introducing lash, but it's so much cheaper than proper leadscrew for a router of these dimensions. There have been contributors on CNCZone who have used belts, and some commercial machines have changed over too. Haven't seen anyone use belt doubling yet, though. I'll hopefully be building over the summer.
Re: X Y axis accuracy doubled.
April 18, 2012 02:35PM
I would suggest using a bigger belt if only so you get a better grip on the drive gear teeth as it will be under real load. In theory there is no reason why it cant be a triple pulley system if you want more torque and speed is not a concern. As long as the belt is tensioned properly there should be no backlash as all the slack is took up already. There should be virtually no spring too if a properly sized belt for the job is used. If you are using big stuff you may need some kind of tensioning rig as pulling it may not be enough.

The other factor in not having things move where they shouldn't is a strong frame and general construction. If the brackets and frame distort under load it will show up in the machining.


Make your Mendel twice as accurate.
[www.thingiverse.com]
Re: X Y axis accuracy doubled.
April 18, 2012 08:03PM
Thanks Madkite for the advice. I'm going to use wider belt, like 15mm, probably GT2 5mm profile, or possibly HTD 8mm. How would a triple system work?! I think doubling will be enough, and I'm designing idler pulleys, mounts and adjusters. The frame will be made from aluminium extrusion, 40mm x 80mm.
Sorry, only registered users may post in this forum.

Click here to login