Belt drive Z-axis

Discussed here are the various tilt/ slide and Z-Axis mechanisms
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable
User avatar
@A2
Posts: 51
Joined: Wed Apr 23, 2014 7:58 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Belt drive Z-axis

Postby @A2 » Thu May 01, 2014 10:57 am

I'm looking for an example of a belt, and pulley Z-axis drive being applied to a DLP-SLA printer.

My math indicates it's not possible, but maybe there is a workaround that I have not considered. I have read of one person saying they saw an example of a belt driven DLP-SLA printer, but a reference was provided, or notes of it's performance.

With a PDMS film (Silicone) on the vat bottom, the separation force is ~100 oz (27.8 N) for an image area of 625 mm2 (25 mmx25 mm). The separation force can go as high as ~140 oz, and it goes higher as the exposure time increases.

Digital Material Fabrication Using Mask-Image-Projection-based Stereolithography
http://www-bcf.usc.edu/~yongchen/Resear ... l_2012.pdf

Using a force of 100 oz, and my proposed build platform area of 7.5 in x 4.25 in x 100 oz = 3187.5 oz force, or 199 pounds of separation force. My NEMA17 stepper motor is rated at 76 oz-in, and a GT2 belt is rated for 5.8 lbs. So it's easy to see that it will not work on it's own, and why a tilting, and/or sliding vat to over come the Van der Waals force is employed.

I'm also aiming for a Z-axis resolution of 0.010 mm, (0.0004 in).

If I use a threaded rod I'm considering using a Roton 10 mm x 2 mm pitch Trapezoidal thread.
http://www.roton.com/Mating_Components. ... ly=7060746
2 mm pitch/200 steps = 0.010 mm

GT2 belt: 2 mm pitch, 6 mm wide.
Allowable Working Tension Per 1 Inch of Belt Width 25 lbs (111 N). 25.54 mm/6 mm = 4.26, 25 lbs /4.26 = 5.8 lbs max working load.
http://www.sdp-si.com/D265/PDF/D265T015.pdf
Last edited by @A2 on Tue May 20, 2014 9:26 pm, edited 2 times in total.

johnrpm
Posts: 128
Joined: Sun Dec 01, 2013 5:25 pm
Location: UK
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby johnrpm » Mon May 05, 2014 6:14 am

Just an observation..... if the weight is always downwards, then with a leadscrew backlash
is not so much of an issue, provided that friction is not too high, but maybe I have missunderstood.
Random Precision

Szzer
Posts: 20
Joined: Tue Jan 28, 2014 5:17 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby Szzer » Tue May 06, 2014 11:59 am

I agree with John.

Other thought: The seperation force can become quite high (it can bend my 2mm thick glass quite a bit), possibly stretching the belt. Meaning loss of accuracy.

sitzme
Posts: 58
Joined: Wed Dec 04, 2013 12:19 am
Location: Colorado
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby sitzme » Tue May 20, 2014 10:03 am

There should not be a significant permanent stretch of the belt. Usually belt drives equal faster motion with less resolution. That said, it would be interesting if someone tried it. Motor torque might be the limiting factor. Personally, I will stick to lead screws.

hp_
Posts: 301
Joined: Tue Dec 24, 2013 10:25 am
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby hp_ » Tue May 20, 2014 3:13 pm


User avatar
@A2
Posts: 51
Joined: Wed Apr 23, 2014 7:58 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby @A2 » Tue May 20, 2014 7:18 pm

@johnrpm:
A screw does not address Z-banding or other types of surface artifact generation, and that is a concern of mine that I want to address by using a belt. My fall back to the belt-pulley-gear approach is a Tr10x2 Roton lead screw, with an anti-backlash nut or a loose fitting nut.

Banding and failed/successful hacks to resolve it
http://forums.reprap.org/read.php?279,285943

I'm not presently concerned with backlash because I believe there is enough weight loading the Z-axis belt, and I'm going to use a geared drive. backlash is also some what negated with a high gear reduction ratio, as more turns of the gear are required to initiate significant change.

Software can also be used to negate backlash, e.g. Slic3r has the ability to account for Bowden tube extruder backlash. My design also allows me to add a tethered weight to help maintain the load on one side of the gear tooth.

Does Creation-Works have a backlash option?

@Szzer:
The GT2 belt is pre-stretched before use, and is considered to have zero stretch after this has been performed. It's assumed that you are not exceeding the recommended loading, other wise assume the belt has stretched. I'm concerned with the separation force of an object the size of the build plate.

If the belt stretches during the object separation phase, it's occurring when I don't need accuracy. I'm going to employ a peeling action, (e.g. tilt, slide, jog, Teflon membrane, etc) to counter the separation force.

My first concern is printing a large object or raft the size of the build platform 7.5 in x 4.25 in, and the separation force required. For example, a 7.5 in x 4.25 in build platform will need a Z-axis drive that could potentially need to lift 200 lbs.

I have not seen any DLP-SLA printers taking advantage of the full area of the build plate. No one has figured out how to solve the separation problem for very large objects with a straight pull, i.e. no tilt or sliding, etc. I also think it's a little bit misleading to state the build plate size, and not included the maximum surface area that the Z-axis can separate from the build plate.

A quote from Kudo acknowledging that the separation force is the limiting factor to maximum build size:

“the separation force between the cured resin layer and the bottom of the resin container still limits the minimum feature size, printable mechanical structures and build volume.”
http://www.kudo3d.com/products

My second concern is the distortion of small objects caused by tilting, or sliding, of the object to overcome the separation force. To avoid distorting small objects I want to employ a dead lift, i.e. no tilt, slide, or jogging side to side of the build platform, but that does not seem possible at this time.

A dead lift should allow for more accurate object creation. I'm looking for solutions that address both concerns, fine detail generation, and high separation force. Maybe both can't be done on the same machine. Because I don't have a clear solution I'm going to incorporate a passive tilt mechanism, and I'm going to experiment with a Teflon membrane.

@Sitzme:
I'm using a gear drive to achieve a fine resolution, and I have calculated the lifting force, Z-axis resolution, and the retraction speed, and I believe it will be as fast as most SLA systems.

It's possible to achieve the following Z-axis resolution with a gear drive with a pulley and belt:

60:1
Full step: 0.00333 mm
32 microsteps: 0.0001041 mm

20:1
Full step: 0.010 mm
32 microsteps: 0.00031 mm

I doubt very much my system can take advantage of such a fine resolution, but it's there if you can.

@hp_:
What belt are you using, stepper motor size, and drive train? Any issues using a belt?
Tks for posting a pic of a belt driven DLP-SLA!

darrin3d
Posts: 18
Joined: Mon Mar 10, 2014 12:42 am
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby darrin3d » Tue May 20, 2014 8:42 pm

I built a big 4' x 8' cnc router that is belt drive on the x and y axis'. I have been running it for 3 years and still no significant belt stretch. They are steel belted. I used heavy belts that are 3/4 inch wide and they work great. I have a gear reduction setup as well that is belt driven and no issues with those belts either.

If you buy decent industrial grade belts that are made for industrial robotics, you should be able to build a nice setup. They are quieter than screws too.

Finhead
Posts: 277
Joined: Fri Jan 31, 2014 12:56 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: Belt drive Z-axis

Postby Finhead » Tue May 20, 2014 9:27 pm

From all I've read belt drive will not give the resolution you would like on the Z with enough torque to pull from the build plate.
There is also the fact that I have not seen anyone finish a build with a belt on the Z that has started or experimented with it.
I use a 10mm metric screw with 2 delrin nuts and spring with zero issues, cost me all of $4 so it would also cost more to run a belt.

Clinton


[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable