I3 X Belt Idler

Hey all, I was reading a post on wavy surfaces and one of you genius's mentioned a grooved belt rolling around a smooth idler bearing and the possible vibration caused. I looked at mine and it was obvious that it was not a smooth ride. I cut down a spare pulley and here is the result. Something to think about.

Sure metal idlers are the most precise way to guide a belt.

I printed mine based on this thingi.
They have belt retainers, which can be useful in some cases.
-Olaf

Ive ordered a bunch of toothed idlers for my printers very cheap url=http://pages.ebay.com/link/?nav=item.view&id=311508862244&alt=web they should be slightly more accurate but I'm not sure they'll always be quieter.

Edited 2 time(s). Last edit at 01/30/2016 06:19AM by DjDemonD.

Quote
DjDemonD
Ive ordered a bunch of toothed idlers for my printers very cheap [link] they should be slightly more accurate but I'm not sure they'll always be quieter.

I've made very similar ones myself out of some regular drive pulleys that I've cut down to leave just the teeth+flange and some flanged bearings, the end cost was ~1.75usd per idler. I couldn't find anything like the ones you've linked to at the time of purchase, otherwise I would have bought them

Edited 1 time(s). Last edit at 01/30/2016 01:25PM by Dalius98.

Hey Olaf, nice touch with the racing stripes. Beauty print by the way.

Toothed idler pulleys won't remove your waves. It's caused by resonance in the stepper motor, that correlates to driver amperage, and motor rotation degrees per rotation.

I chased this issue for 2 years, toothed pulleys, flipped belts, yadda... Changed from 1.8 to 0.9 degree motors, and the wave pattern in my prints became half the size of the 1.8 motors. Regardless of microstepping, steppers have a higher 'torque' as it internally moves past it's 1.8 or 0.9 degree angle.

If you have ever hand turned one of your motors while under power, the single 'click' the motor moves is the motor going that 1.8 degree of rotation.

Also no difference in print pattern between 5mm and 2mm drive belts.

Edited 1 time(s). Last edit at 01/31/2016 11:55AM by Dirty Steve.

Useful time saving info thanks, so in theory if someone made a 4000 steps per revolution motor and we had the electronics to drive it we would not be able to see these waves? What about geared motors?

Edited 1 time(s). Last edit at 01/31/2016 12:28PM by DjDemonD.

Yes, a 4000 step per revolution motor would do that. Gear reduced motors would decrease the wave pattern in relation to the gear ratio.

Here's a 1000 steps per revolution motor link could run it without microstepping to keep the processor demand down?

Those would reduce resonance. You could probably get away with 1/4 or 1/8 microstepping without too much processor demand, depending on your setup. I'm running 0.9 degree (400 step/rev) motors at 1/16 microstep without issues.

Quote
Dirty Steve
Toothed idler pulleys won't remove your waves. It's caused by resonance in the stepper motor, that correlates to driver amperage, and motor rotation degrees per rotation.

I chased this issue for 2 years, toothed pulleys, flipped belts, yadda... Changed from 1.8 to 0.9 degree motors, and the wave pattern in my prints became half the size of the 1.8 motors. Regardless of microstepping, steppers have a higher 'torque' as it internally moves past it's 1.8 or 0.9 degree angle.

If you have ever hand turned one of your motors while under power, the single 'click' the motor moves is the motor going that 1.8 degree of rotation.

Also no difference in print pattern between 5mm and 2mm drive belts.

This matches my experience also.

If you want to really keep the motors under control, you need higher voltage power supplies and DSP drivers. I built 32V power supplies from junk parts (no need for regulation- just a transformer, rectifier, and filter cap) and bought a couple DSP drivers for about $30 each (http://www.sainsmart.com/cnc-2m542-stepper-motor-driver-controller-4-5a-support-nema17-23-34.html). I get absolutely no print surface artifacts.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Different controllers and stepper drivers can induce more or less 'wave' in the print depending on how clean the drive signal they output is.

Edited 1 time(s). Last edit at 01/31/2016 01:14PM by Dirty Steve.

Those 1000 steps per revolution motors look nice but they should be at $160 each!

Very informative gentlemen. Thanks.