Haas mill "X" axis off by .0002"/ doesn't appear to be a backlash issue

[EMTech]

I need some help, and I am hoping one of you know how to resolve this.

I have a Haas mini mill.

If I set a work coordinate and cut a feature at X0 Y0, my feature is off by X.0002". I know this isn't much for a 2003 Haas, but it is annoying.

I can set XY0, then cut a circle or square on XY0 as center. Then when I position the machine back at XY0 and check with an indicator, the Y is centered, but I have to move the machine .0002" positive for the indicator to show that I am actually at the center of my feature.

It is consistent and repeatable. It is not a backlash issue. It doesn't matter if I come from the X+ or X- direction.

It is always off by the same amount and the same direction.

I could have someone come in and do a ballbar test, but I am trying to figure it out and adjust it on my own.

Maybe it is an X parameter that I am not familiar with?

Any ideas???

 

[D Nelson]

Shorten you indicator length to match your Endmill length see if that helps get you centered up. Maybe your tram is off a touch
Don

 

[EMTech]

Shorten you indicator length to match your Endmill length see if that helps get you centered up. Maybe your tram is off a touch
Don

Thanks for that thought.

I will check the tram, and check the endmill length vs my endmill length.

I think the tram is good, but it has been a while since I have checked it.

 

[Stirling]

I second tram, check with the indicator at the same length is a good test.

Could it also be your backlash comp is out a bit?

 

[friesen]

I suppose you have validated your indicator as well as checking it at 180 degrees?

 

[BT Fabrication]

ball screws wear as it gets older, most say they last approx 1 million inches of travel before replacement. there are parameters to adjust the screws to have the x axis servo to compensate it by moving it back more to adjust for wear. 2 tenths isnt bad, most want from factory just under 1 tenth as you can't have zero free play as it needs to move. think thats bad my x axis is about 0.0109" wear on the ball screw and a new one has been ordered to replace it. Y axis has a couple thou but it can all be electronically compensated.

 

[modelmakerblue]

How are you checking this .0002" error?
Tony

 

[EMTech]

Alright, it appears to have been a tram issue. I checked the feature with an indicator that was approximately the same length from the spindle as my tool was, then it checked good. Previously, my indicator was 3" longer than my cutting tool.

Thank you to D Nelson and Stirling for that suggestion!

As some background, I had recently finished changing out my "Y" axis ballscrew and adjusting backlash and friction comp, so I had parameters on the brain. I completely overlooked the possibility that it could be a tram issue.

It seems obvious now...

Thanks everyone for your suggestions.

 

[EMTech]

ball screws wear as it gets older, most say they last approx 1 million inches of travel before replacement. there are parameters to adjust the screws to have the x axis servo to compensate it by moving it back more to adjust for wear. 2 tenths isnt bad, most want from factory just under 1 tenth as you can't have zero free play as it needs to move. think thats bad my x axis is about 0.0109" wear on the ball screw and a new one has been ordered to replace it. Y axis has a couple thou but it can all be electronically compensated.

WOW! I just changed out the Y axis ballscrew because it had a couple thou of backlash. I can't imagine trying to machine something with over .01" backlash!

 

[D Nelson]

Alright, it appears to have been a tram issue. I checked the feature with an indicator that was approximately the same length from the spindle as my tool was, then it checked good. Previously, my indicator was 3" longer than my cutting tool.

Thank you to D Nelson and Stirling for that suggestion!

As some background, I had recently finished changing out my "Y" axis ballscrew and adjusting backlash and friction comp, so I had parameters on the brain. I completely overlooked the possibility that it could be a tram issue.

It seems obvious now...

Thanks everyone for your suggestions.

I got lucky on that one. But it doesn’t take much tram off to be .0001 off. I’m glad it helped you out
Don

 

[Stirling]

WOW! I just changed out the Y axis ballscrew because it had a couple thou of backlash. I can't imagine trying to machine something with over .01" backlash!

Reminds me that I really need to level.tram mine again. It seems to have drifted.

 

[MiniMillMachinist]

ball screws wear as it gets older, most say they last approx 1 million inches of travel before replacement. there are parameters to adjust the screws to have the x axis servo to compensate it by moving it back more to adjust for wear. 2 tenths isnt bad, most want from factory just under 1 tenth as you can't have zero free play as it needs to move. think thats bad my x axis is about 0.0109" wear on the ball screw and a new one has been ordered to replace it. Y axis has a couple thou but it can all be electronically compensated.

1 million inches of travel is just 170ish hours at 100 inches per minute, are you sure about that? I'd be really upset if my ballscrew was worn out after just a couple of hundred hours!

 

[EmGo]

I think people should have to have a license to buy a tenths indicator ...

 

[rjwalker1973]

I used to work in a ceramic shop and we cut green ceramic on a Haas TL3. The x axis ballscrew is open on it. The ceramic powder would get on the ballscrew as the machine was cutting and just eat the ballscrew up. We had to have them replaced every six months because we would have over .100 backlash. I had to do some creative programming to comp for it lol.

 

[BT Fabrication]

1 million inches of travel is just 170ish hours at 100 inches per minute, are you sure about that? I'd be really upset if my ballscrew was worn out after just a couple of hundred hours!

all depends on how well its lubed. im sure with proper lube it can go 2-3 million before it gets minor out of spec.
seen ball screws with 11 thou of wear and only 6000 hrs, well outside of spec for it.
even bearings wear on things and can cause it to become loose. so a little maintenance is key when keeping wipers in spec etc.

 

[???]

I need some help, and I am hoping one of you know how to resolve this.

I have a Haas mini mill.

If I set a work coordinate and cut a feature at X0 Y0, my feature is off by X.0002". I know this isn't much for a 2003 Haas, but it is annoying.

I can set XY0, then cut a circle or square on XY0 as center. Then when I position the machine back at XY0 and check with an indicator, the Y is centered, but I have to move the machine .0002" positive for the indicator to show that I am actually at the center of my feature.

It is consistent and repeatable. It is not a backlash issue. It doesn't matter if I come from the X+ or X- direction.

It is always off by the same amount and the same direction.

I could have someone come in and do a ballbar test, but I am trying to figure it out and adjust it on my own.

Maybe it is an X parameter that I am not familiar with?

Any ideas???

I think you should replace the machine it's way out of spec. I would expect at least .002" on a Haas it's obviously seizing up

 

[Luke.kerbey]

ball screws wear as it gets older, most say they last approx 1 million inches of travel before replacement. there are parameters to adjust the screws to have the x axis servo to compensate it by moving it back more to adjust for wear. 2 tenths isnt bad, most want from factory just under 1 tenth as you can't have zero free play as it needs to move. think thats bad my x axis is about 0.0109" wear on the ball screw and a new one has been ordered to replace it. Y axis has a couple thou but it can all be electronically compensated.

1 million inches? If my maths is right that’s only 18 hours or so of a machine travelling at rapids of 24 m/min.

 

[Luke.kerbey]

WOW! I just changed out the Y axis ballscrew because it had a couple thou of backlash. I can't imagine trying to machine something with over .01" backlash!

10 thou backlash, that probably sounded like a cement mixer.

 

[MiniMillMachinist]

all depends on how well its lubed. im sure with proper lube it can go 2-3 million before it gets minor out of spec.
seen ball screws with 11 thou of wear and only 6000 hrs, well outside of spec for it.
even bearings wear on things and can cause it to become loose. so a little maintenance is key when keeping wipers in spec etc.

Personally I'd hope for several thousand hours on a ballscrew without significant wear... I've got around 1000 on my mini mill and backlash/accuracy looks to be much the same as when brand new. Roughly a micron of backlash with a 2 micron indicator. Obviously it's hard to test this reliably with a dial indicator!