Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: Matthew_Jones on July 12, 2015, 06:50:12 AM

Title: Unequal Backlash
Post by: Matthew_Jones on July 12, 2015, 06:50:12 AM
SO I have been fine tuning and trying like hell to get the flat spots out of my circles. I got a dial indicator went to set the back lash and found something funny. Before I tear the machine apart and tighten everything I want to ask.

The problem is on the Y axis. If you are travel in Y+ the reverse to Y- the backlash is .05mm. If you are travel in the Y- and reverse to the Y+ the backlash is .13mm.

So maybe this is mechanical but I sure would like a few opinions before I go tearing into this machine.

Thanks
Matt
Title: Re: Unequal Backlash
Post by: ger21 on July 12, 2015, 07:11:04 AM
If that's really true, then if you did this 10-11 times, then the machine would be 1mm out of position.
Title: Re: Unequal Backlash
Post by: RICH on July 12, 2015, 07:35:15 AM
Could be due to a number things.
One needs to know the cause of the backlash and that can only be done by trying to isolate each  mechanical component
in a very meticulous way to find the causes. Note that there is no such thing as "zero" backlash and there is no substitute
for quality components.

RICH
Title: Re: Unequal Backlash
Post by: Matthew_Jones on July 12, 2015, 08:25:44 AM
If that's really true, then if you did this 10-11 times, then the machine would be 1mm out of position.


It does stack up like that. You can see it in the circles I am cutting. One side has bigger flat spot than the other and often the start point is not the finish point. Not always though. I have adjusted alot of it out but still get flat spots and thats the only thing I can't get rid of. If I set backlash .05 the lesser reading then only going the other way seems to cause a problem. I am clueless.

Could be due to a number things.
One needs to know the cause of the backlash and that can only be done by trying to isolate each  mechanical component
in a very meticulous way to find the causes. Note that there is no such thing as "zero" backlash and there is no substitute
for quality components.

RICH

The machine seems high quality. I wonder if it could be a combination of things. I am having a hard time trying to figure out what to look at specifically and it what order.

This is the machine (http://www.cnczone.com/forums/attachments/4/6/3/1/3/8/423078.attach)

I'll take any pointers I can get.

Thanks
Matt
Title: Re: Unequal Backlash
Post by: Hood on July 12, 2015, 08:52:23 AM
If you have backlash enabled, turn it off and test again.
Hood
Title: Re: Unequal Backlash
Post by: DICKEYBIRD on July 12, 2015, 11:37:49 AM
Hood
He's b-a-a-c-k!:)  (Yay!)
Title: Re: Unequal Backlash
Post by: Matthew_Jones on July 12, 2015, 12:38:03 PM
If you have backlash enabled, turn it off and test again.
Hood

No it wasn't enabled.
Title: Re: Unequal Backlash
Post by: BR549 on July 12, 2015, 03:37:24 PM
Mechanical backlash will be equal in both directions. This sounds like backlash PLUS a loss of steps in one direction only. That could be DRIVE related or a mechanical problem.

Just a thought, (;-) TP
Title: Re: Unequal Backlash
Post by: ger21 on July 12, 2015, 03:38:04 PM
Quote
It does stack up like that. You can see it in the circles I am cutting. One side has bigger flat spot than the other and often the start point is not the finish point. Not always though. I have adjusted alot of it out but still get flat spots and thats the only thing I can't get rid of. If I set backlash .05 the lesser reading then only going the other way seems to cause a problem. I am clueless.

Then the machine is either losing steps, or something mechanically is loose. Check all the setscrews on all your couplers.
Title: Re: Unequal Backlash
Post by: RICH on July 12, 2015, 03:48:05 PM
I would suggest checking how the Y axis screw is mounted and thrust is controlled.
IE; May have have one bearing , or two bearings, or maybe just a spacer or whatever.
How that is done and adjusted can lead to different backlash.

Again, hard to say, without knowing what's "under the hood" so to speak.

BTW, Car hood as i know Hood can't fit under the cover! :D

RICH
Title: Re: Unequal Backlash
Post by: Matthew_Jones on July 12, 2015, 07:33:01 PM
How do you test for loosing steps?

Matt
Title: Re: Unequal Backlash
Post by: Overloaded on July 13, 2015, 08:41:36 AM
If you have a home switch/sensor, I do it this way.
Home the axis, then uncheck Auto Zero in the Homing dialog. Rapid jog the axis all through it's available area several times (or run a file with many rapids) then rehome the machine. It will go to the same position in relation to the switch and the DRO will show any deviation from the original 0 which will indicate any lost steps.
An axis can lose steps in one direction only due to a tapered gib creating more resistance (wedging) in one direction than the other. I've seen this on some of the lower quality dovetail type machine slides.
Title: Re: Unequal Backlash
Post by: Matthew_Jones on July 13, 2015, 04:16:17 PM
Thanks a bunch. I'll look and see if its loosing steps, I have noticed some funny numbers occasional I didn't understand. I haven't got homing set up but I can set that axis and check that. The machine has its own coordinates displayed (According to what I am not sure) but they tend to line up real well. Mach and the machine, down to the .001 decimal if a millimeter.

It has 1605 precision ball screw for the movement but maybe since its Chinese.... I figure I'll find something loose hopefully nothing worse than that.

I'm also going to spend a little more money on a dial indicator. At least have 2 of them telling me something wrong.

I am grateful for all the advice. I probably will have more questions soon.

Thanks
Matt