How little backlash can you get?

So the best I can do with my makergear prusa is .011 in (0.28 mm) this seems high.

Can you get to zero backlash with belts? How little can you get your machine to?

Thank you

Edited because I can convert I swear. What was .028 should be 0.28

Edited 1 time(s). Last edit at 06/13/2012 08:03PM by tjhj2.

That is high :p

Yes, zero backlash is possible with belts.

---

www.Fablicator.com

Can you explain how you measured it so we can be sure to give proper comparisons?

Also what speed and acceleration settings.

Edited 1 time(s). Last edit at 06/12/2012 10:07PM by Sublime.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

You will always have backlash, because no material is perfectly rigid. However, you can minimize it so that it is quite small. I am also interesting in how you measure the backlash.

Quote

You will always have backlash, because no material is perfectly rigid.

Rigidity and backlash are two different things. You can have something stiff with backlash and something weak without backlash. RepRap machines tend to be the latter. Achieving zero backlash with rubber coated belts isn't exactly difficult, low friction bearings, proper pulleys, proper adjustment and you're there.

Getting something stiff without backlash is an art.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Using acceleration will help as well.

For zero backlash you need zero friction or infinite motor torque, hence it is impossible.

A stepper motor will stop a little short where the torque due to the displacement equals the static friction.

---

[www.hydraraptor.blogspot.com]

Hi all,

Sublime,
I used 200mm/min movement speed. For method 1. (relative movement set and federate) G1 X0.01 or G1 X-0.01. For method 2 I used the pointer face dialog box. With 200 set in federate.

So to measure my backlash I measured it two ways. This was how I was taught to check a Cnc mill. So it may be incorrect.
I mounted a dial indicator (it's .001 in precision, analog) to the frame on the guide rods. Touching the bed (y first)
1st method.
Then I moved the bed towards the indicator making sure it moved.
Then tell the bed to move in the opposite direction in very small steps. In my case .01 mm.
Then counting how many movements it takes to get an indication.

2nd method.
Move toward the indicator. Till indication.
Zero the dial
Make a move of a set amount in the same direction you were moving before. I used 1mm
Then move the opposite direction 1mm
The difference from 0 is your backlash or distance lost by changing direction.

Nophead,
Yes true 0 backlash is impossiable. When I say zero I mean within the precision of movement. Ideally less than .001in. (my gauges are standard not metric sorry). Even if I get it to show zero with a .001in gauge it still could have a small amount eg .0001 in. But I would call .001 good.

So considering this,
Since no matter how tight I make the belts any suggestions?


Thank you for your help.

I can't remember anybody measuring it before so I don't know if that is good or bad.

0.011" is actually 0.28mm. I would say 0.028mm is pretty good and 0.28mm not good as the target resolution of reprap is 0.1mm.

A quick test on the Y axis of my Mendel90 using method2 it seems to be about 0.09mm. With method 1 you are relying on the microstepping being accurate and not having any missing steps.

You could try increasing the motor current or lubricating the bearings.

Edited 1 time(s). Last edit at 06/13/2012 08:27AM by nophead.

---

[www.hydraraptor.blogspot.com]

Could you try it with Woodsmoke's Lashmaze?

Maybe I have an electronics issue. But if I turn the motor current up much more the motor start skipping, or chopping. Doing the opposite of what I want.
I don't feel I'm missing steps because if I move I one direction zero the indicator and move 1mm the same direction it moves 1.00 mm or is off by less then I can measure.

I'll try the print.

I think I found part of the problem... It helps if the guide bars can actually tighten down...

Maybe its just me... Fail....

Thanks for all the help.