Hello Guest it is October 18, 2019, 10:45:03 PM

Author Topic: backlash for both direction on same axis  (Read 3582 times)

0 Members and 1 Guest are viewing this topic.

backlash for both direction on same axis
« on: September 14, 2008, 09:43:57 PM »
is there a way to get two figures in backlash <---.004 and -->.0015 for one axis . This would save me a lot of head ache and money.

David
Re: backlash for both direction on same axis
« Reply #1 on: September 14, 2008, 09:47:48 PM »
For instance in the backlash screen, could "B" and "Y" be changed into a +Y, -Y inorder to separate the different backlashes

Offline Chaoticone

*
  • *
  •  5,607 5,607
  • Precision Chaos
    • View Profile
Re: backlash for both direction on same axis
« Reply #2 on: September 14, 2008, 10:33:00 PM »
David, do you know where your backlash is coming from? I have never seen more backlash in one direction than the other on the same axis before and trying to imagine how you would get that.

Brett
;D If you could see the things I have in my head, you would be laughing too. ;D

My guard dog is not what you need to worry about!
Re: backlash for both direction on same axis
« Reply #3 on: September 14, 2008, 10:39:30 PM »
Lead screw has more play one direction than the other the person I got the machine from built in flaws that I need to rebuild some parts remove the glitch.
Re: backlash for both direction on same axis
« Reply #4 on: September 14, 2008, 11:47:42 PM »
davy

No way can you have 2 different backlash  # on the same screw.
 Whatever backlash you have will always remain the same wheather + or -.
You better check nut mounts and end bearings .

Ed
Ed VanEss
Re: backlash for both direction on same axis
« Reply #5 on: September 14, 2008, 11:58:18 PM »
I am looking at a dial indicator showing it. I have .003 one way and i am estimating on the other amount because it doesn't read to the fourth decimal but i estimate .0001 the other way.  when it goes to .007 and comes back to a fraction before zero. it takes like ten times before it clearly doesn't return to zero.
Re: backlash for both direction on same axis
« Reply #6 on: September 15, 2008, 12:10:34 AM »
I am looking at a dial indicator showing it. I have .003 one way and i am estimating on the other amount because it doesn't read to the fourth decimal but i estimate .0001 the other way. when it goes to .007 and comes back to a fraction before zero. it takes like ten times before it clearly doesn't return to zero.

That doesn't sound like backlash to me.  I just don't see how a mechanical fault could possibly behave that way.  More like an electrical problem when you change directions.  Are you sure you're getting step/dir timing and polarity that's compatible with your stepper drivers?

Regards,
Ray L.
Regards,
Ray L.

Offline jimpinder

*
  •  1,233 1,233
  • Wakefield, West Yorks, UK
    • View Profile
Re: backlash for both direction on same axis
« Reply #7 on: September 15, 2008, 03:32:32 AM »
Backlash is a mechanical "shortcoming" made up of all the little spaces between gears and other things in your drive train. If you are down to 0.0001 (1/10 thou) in one direction, it doesn't sound like you have any and 3 thou in the other is perhaps a little high, I like to be + - 2 thou, but whatever - it must by definition be equal in both directions.

I would first check you steps per unit and make sure that is accurate. This should be calculated althoug if you are saying someone else made the machine you might not have all the detail necessary. Check the steps per unit by moving the axis one way and checking the dial indicator and zeroing the DROs, then moving the axis the same way by typing in the MDI line G0 X1 or X-1 as the case maybe. It is essential you move in the same direction, to take up any backlash. Check the distance moved. It should be exactly one inch.

I am assuming you checked the backlash by moving one direction, zeroing, then moving (with the MDI ) in the same direction one inch, and coming back (with the MDI) and checking if you are at zero. (You should be but will not be, there must be some backlash, although whether you can measure it or not is another matter.) This should be done with backlash compensation OFF.
You can, of course check backlash in bother directions. Enter your backlash in the table, and switch on.

If movement is consistantly short in one direction, this coud be missing steps, as your BOB board (if you have one), changes direction, since many BOBs have voltage and signal changing electronics which upset the timings of the signal from your computer. I do not know how many steps per unit your are using, but missing a few could make this difference. I am on 6000 steps per inch (ignoring microsteps, so I woud need to lose 18 steps for a 3 thou error - on your system it may be less)

I do not know what drivers you are using, but most driver boards will drive directly from the 25pin output from the computer. (Mine are Geckos and drive through a non-powered BOB board) Try making a direct connection and see if this alters things at all. You can leave all the rest of the wiring intact, just use a direct link from your computer to your driver for the step and direction pins. Probably the simplest way is to unscrew the output wires to your driver board and fasten these onto your input wires from the computer.

To go back to the opening, however, if you are down to 3 thou and zero the other way, you are doing well, you should see mine (although backlash compensation takes care of it)





Not me driving the engine - I'm better looking.