Hello Guest it is March 28, 2024, 01:34:03 PM

Author Topic: Backlash compensation for a rotary axis ?  (Read 21520 times)

0 Members and 1 Guest are viewing this topic.

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #10 on: March 09, 2008, 05:18:47 AM »
I whole heartedly agree with you and so far I have been lucky enough not to have to resort to software compensation.
 There are some settings I think you have to mess around with, just in case you havent seen them look here http://www.machsupport.com/forum/index.php/topic,5523.msg38212.html#msg38212
Hood

Offline docltf

*
  •  354 354
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #11 on: March 09, 2008, 10:22:12 AM »
greolt

      go ahead and build it .i believe you will love it.will it have double chucks,front and rear.make sure when you do the rear chuck that you engineer a space for a drag rotor in case you
      have to install one later.how large is the thru bore, is it over 1.5 ,if it is when you are finished messin with it send it me.

bill

vmax549

*
Re: Backlash compensation for a rotary axis ?
« Reply #12 on: March 09, 2008, 11:21:06 AM »
Jim normal backlash on the linear axis can be compensated for because when you apply the load of the cut it loads up the axis and stacks the backlash up in the cut direction.

But on a rotary when you are cutting on the centerline of a cylinder that cutting load is not constant as the bit works the radius of the cylinder and it can cause the axis spindle backlash to work back and forth as the cutter moves across the centerline and changes the load direction of the bit.

Greg you could always machine in two concentric bearing holders so you could easiely adjust the backlash??? Just like the big boys do.(;-) Or offset bore the case and shim the bearings over.

Just a thought.  (;-) TP

Offline Greolt

*
  •  956 956
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #13 on: March 10, 2008, 02:21:14 AM »

Bill

I went ahead and made a spindle for it.

1" - 10 tpi  which is the size of my wood lathe spindle so my existing chucks etc will fit.

Only one ended though.  I will build a tail stock to suit if it works out.

I have not measured it yet but I think the backlash is going to be a lot better than I thought.


TP

I would not go to so much effort as to rebuild the bearing housings etc. 

For that much work I would go in a different direction. (Harmonic drive)

I'm definitely  not one of the big boys.  Just a carpenter.  ;D


Hood

I have not delved into the backlash compensation settings yet but I will get to that.  :)

Greg

.

Offline jimpinder

*
  •  1,232 1,232
  • Wakefield, West Yorks, UK
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #14 on: March 10, 2008, 08:24:54 AM »
Greg - I'm sorry about the diatribe for backlash - and I am quite sure you understand it. I just get carried away sometimes.

Do I understand that this circular table, using the angular axis, will move to a new position, and then stand still whilst the other axis go to work on it.

If so - I know this sounds crude - but I would put a parking brake on it.

There are several options - a belt round the table that can wind on like a garrot, an electric disc brake that might be better because you could control it through software, or my son ( who weighs 22 stone and loves going down mountains on his bike) puts his faith in bicycle brakes - the latest disc brakes are awsome in their power - I now use them on miniature railway engines. They are compact and easy to fit and have hydraulic or cable operation. Or, in the picture you have posted, a drilled and taped hole for a screw onto the main shaft of the gearbox.
« Last Edit: March 10, 2008, 08:36:20 AM by jimpinder »
Not me driving the engine - I'm better looking.

Offline Greolt

*
  •  956 956
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #15 on: March 10, 2008, 09:23:47 AM »
Jimpinder

"Do I understand that this circular table, using the angular axis, will move to a new position, and then stand still whilst the other axis go to work on it."

That is not how I envisage using it. I'm thinking that it will rotate as one horizontal axis (X or Y) and one vertical axis (Z) move about.

However I will keep that bicycle idea in mind. :)

Greg

Offline docltf

*
  •  354 354
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #16 on: March 10, 2008, 02:32:08 PM »
Greolt
       i see you decided to fill the bore up with an adapter shaft rather than bore that keyway out and adapt from the outside.im sad now!


bill

Offline jimpinder

*
  •  1,232 1,232
  • Wakefield, West Yorks, UK
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #17 on: March 10, 2008, 03:41:26 PM »
Greg -

In that case, get the thing built, with as little backlash in it as you can - but then, I think, you will have to use mach3 backlash compensation. It should be OK because the holding force will always be in the direction of the cut.
Not me driving the engine - I'm better looking.
Re: Backlash compensation for a rotary axis ?
« Reply #18 on: March 13, 2008, 11:13:30 AM »
My .02

Even a rotary table that has adjustable backlash will still have backlash. Software compensation will work fine as long as you are dealing with degree movements. If you convert your rotary axis to act as a linear axis, then you'll have issues since the backlash will vary for each radius that you are trying to compensate for.

All that being said, your best solution esp. since that gearbox doesn't have a way to adjust the backlash is to adapt an electric clutch or brake on the output shaft and then set up Mach to disengage the clutch when you rotate. I think this can be done with an output on your controller and a relay or SSAC and perhaps in your post setup. The second best option is to rig a constant brake that puts drag on the system.  The only problem is if you exceed the brake capacity in a heavy cut, it will move on you and cause your cut to be out of tol. or break tools etc.

If this is something you might consider, look for a lawnmower or automotive electric clutch. Surplus Center is a good source.

Offline Greolt

*
  •  956 956
    • View Profile
Re: Backlash compensation for a rotary axis ?
« Reply #19 on: March 29, 2008, 05:37:32 AM »

I finally got some shed time today so I knocked up a motor mount for the little stepper I temporarily pinched of another machine.

Keen to give it a go I clamped it very dodgy like to the table and "What do you know....It goes!!!!!"

Had to dial in 0.225 degree back lash compensation in Mach3. That amounts to about 0.1mm or 0.004" backlash at 50mm diameter.

Too much backlash to live with but the compensation seems to work OK so far.

I've never done rotary axis stuff before so for a test I used a great little program that bloke called Andrew wrote.

See this link, http://vectric.com/forum/viewtopic.php?t=2920

It takes a gcode file that you have created in 2D and converts it to a rotary axis. Works great.

Anyway as you can see in the pics it all sort of worked as expected. Not bad for a first try.

So I think on the strength of that first try the gearbox should be OK and I will go ahead and set it up properly and make a tail stock for it.

I also need to make another screen set to suit rotary axis work.  Maybe this one will be in flash.  ;D

Greg

.