Belt Drive Expected Resolution vs. Accuracy With MicroStepping.

How much accuracy are you giving up for resolution, are you really getting the resolution that you calculated using microstepping?

Using a belt drive, and microstepping, what is your expected resolution for half, eighth, sixteenth, and thirty-second?

After reading the following paper, I'm unsure how a belt drive is able to achieve fine resolution moves, with a decrease of accuracy. I see a lot of routers using a belt drive, and they appear to be machining aluminum to a high resolution. Should I just not worry about it, and expect that I'll get the resolution that want?

Half-step driving is not for improving accuracy, but for solving other problems like vibration and irregular rotation.
[web.archive.org]

I'm going to use large NEMA 23 stepper motors, I have not decided on the torque, but between 280 oz-in to 425 oz-in, I plan to switch between resolutions from 0.100mm, to 0.006mm.

Circumference of a GT2 Pulley 20 Teeth = 40mm

Min-move at full step: 40mm/200 steps = 0.200mm
Min-move at half step: 40mm/(200*2) steps = 0.100mm
Min-move at eighth step: 40mm/(200*8) steps = 0.025mm
Min-move at sixteenth step: 40mm/(200*16) steps = 0.013mm
Min-move at thirty-second step: 40mm/(200*32) steps = 0.006mm


This is a picture of the type of belt drive I would like to employ.

Tks!

Edited 1 time(s). Last edit at 07/17/2014 06:05AM by A2.

Quote
A2
How much accuracy are you giving up for resolution, are you really getting the resolution that you calculated using microstepping?

With the stepper driver properly adjusted, yes, that's possible at the pulley. Reasons for lower resolution when actually printing:

- Weaknesses in the belt drive, like guiding the belt over flat bearings with the tooth side.

- Frame and extruder assembly weaknesses, like bending when acceleration forces appear.

- Friction: the more force you need to move something, the more the actual pulley position will diverge from the ideal one.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

@Traumflug:

Quote
Traumflug
With the stepper driver properly adjusted, yes, that's possible at the pulley.

Thank you for the explanation, that's what I needed to hear.

Quote
Traumflug
Frame and extruder assembly weaknesses, like bending when acceleration forces appear.

I'm going to experiment welding the external structure of my extruded aluminum frame.

Tks.

Edited 1 time(s). Last edit at 07/17/2014 08:00PM by A2.

Another one is: there are 14 tooth pulleys. Raises accuracy by 43% (compared to 20 tooth) and gives more acceleration force on the belt.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Great suggestion to use a smaller pulley to improve accuracy, I will look into it. Tks.

Edited 1 time(s). Last edit at 07/19/2014 09:31PM by A2.

I opened a topic about it - but maybe it is suitable here also:

What are the hard-limits for reoslution and speeds?

Is it rather the motor 8i suspect not, since I believe to having read datasheets of Nema 17 and nema 23 where there were 3500-4500RPMs mentioned...)

Or is it he electronics/drivers which have to give enoug pulses/instructions fast enough towards the printer`s7Cnc`s motors?

I am planing on using ball-screws with a slopepitch of 5mm.

that combined with 1/64 or 1/128th microsteping would for sure need lots of pulses/s to achiee fast movements.
So where are the hard-limits in such an application?
I will be using an arduino DUE,
RADDS Board actually

Alex

Its the electronics and the firmware, a mega type micro just is to slow see [reprap.org]