Ballscrew tolerances and real world performance

This forum is dedicated to those Hobbyists Interested in CNC machining in their home shops. (Digital Read Outs are also topical, as is CAD/CAM as it relates to CNC)

Moderator: Harold_V

WJH
Posts: 1421
Joined: Sat Jan 16, 2010 9:29 pm
Location: Florida

Ballscrew tolerances and real world performance

Post by WJH » Mon Feb 26, 2018 8:45 pm

So I am educating myself on ballscrews for my 8x14 conversion and am noticing that all the "affordable" ball screws are P7 or C7 tolerance, which means .002 over 12". I don't really care if the Z axis on the lathe is off by .004 at the worst possible case, but I have my concerns about the cross slide Y axis.
In the real world, is this really an issue? Obviously the Y axis moves far less than 12" on my lathe, so I can expect much better than .002? I understand repeatability between parts are not the issue, just hitting tight tolerances when needed? Also, am I correct in thinking that a 5mm pitch screw is the one to get? Also, I am finding most of these cheap ones only come with one ball nut, two is desired for zero backlash, correct?

User avatar
Bill Shields
Posts: 5858
Joined: Fri Dec 21, 2007 4:57 am
Location: Somewhere in the World
Contact:

Re: Ballscrew tolerances and real world performance

Post by Bill Shields » Tue Feb 27, 2018 7:23 am

Yes it is an issue, but only if you are working to tight tolerances - depends on what you definition of tight tolerances is.

This is the difference between machines and rely on 'screw position' to locate the tool as opposed to machines with auxiliary (glass or whatever) scales to feed-back to the control and say 'this is where the tool ACTUALLY is located'.

Bigger issue with 'conversions' is overall rigidity and 'is the tool ACTUALLY' where the screw / encoder say it is.

ball screws and backlash are a function of how things are designed. some ball screws / nuts are designed to be a 'tight' (interference) fit so that there is no backlash with one nut, others do require two (as you have surmised).

Note: most lathes work on XZ coordinates, with Y reserved for off-axis milling.
Too many things going on to bother listing them.

John Hasler
Posts: 1503
Joined: Tue Dec 06, 2016 4:05 pm
Location: Elmwood, Wisconsin

Re: Ballscrew tolerances and real world performance

Post by John Hasler » Tue Feb 27, 2018 10:16 am

Backlash inside a control loop can be a headache, though. It can be dealt with but you can't just ignore it.

WJH
Posts: 1421
Joined: Sat Jan 16, 2010 9:29 pm
Location: Florida

Re: Ballscrew tolerances and real world performance

Post by WJH » Wed Feb 28, 2018 10:49 pm

When I get home, will have to see if my slim line glass encoder could even remotely fit the cross slide on my 8x14 lathe, I doubt it. It would be the closed loop stepper or servo driver that would read the quadrature glass scale input, correct?

User avatar
Marty_Escarcega
Site Admin
Posts: 1708
Joined: Wed Dec 18, 2002 8:17 pm
Location: Mesa, AZ USA

Re: Ballscrew tolerances and real world performance

Post by Marty_Escarcega » Wed Mar 07, 2018 2:22 pm

WJH wrote:
Wed Feb 28, 2018 10:49 pm
When I get home, will have to see if my slim line glass encoder could even remotely fit the cross slide on my 8x14 lathe, I doubt it. It would be the closed loop stepper or servo driver that would read the quadrature glass scale input, correct?
On an 8x14" lathe, the C7 ballscrews with close loop steppers would give you good for performance for that caliber machine.

Adding glass scales add complexity. I assume your 8x14 is just an import lathe?

Like Bill said, rigidity comes into play.....
"Jack of all Trades, Master of None"

User avatar
Bill Shields
Posts: 5858
Joined: Fri Dec 21, 2007 4:57 am
Location: Somewhere in the World
Contact:

Re: Ballscrew tolerances and real world performance

Post by Bill Shields » Wed Mar 07, 2018 8:00 pm

glass scales are a problem on lathes that small. many companies sell other than 'glass scales' that are much thinner...the big problem, even if you CAN get the glass scale on the lathe is loss of tailstock positioning toward the chuck when working with very small parts.

I would be surprised if you can get a ball screw under the slide of a lathe that small (that was not initially designed to support a ball screw).

everyone thinks that you can take a 'hand wheel' lathe and stick a control on it and get good positioning...

please look before you leap....
Too many things going on to bother listing them.

WJH
Posts: 1421
Joined: Sat Jan 16, 2010 9:29 pm
Location: Florida

Re: Ballscrew tolerances and real world performance

Post by WJH » Thu Mar 08, 2018 1:23 am

I discovered that the “ACORN” and LinuxCNC allow you to map a ball screw to remove the inaccuracies. So I could temporarily mount a glass scale to my axis’s to map each one to form a table to make those lesser grade ballscrews perform as if they were ground to a high tolerance. Only true concern now is just having zero backlash using dual ball nuts
Last edited by WJH on Thu Mar 08, 2018 1:33 am, edited 1 time in total.

WJH
Posts: 1421
Joined: Sat Jan 16, 2010 9:29 pm
Location: Florida

Re: Ballscrew tolerances and real world performance

Post by WJH » Thu Mar 08, 2018 1:26 am

Bill Shields wrote:
Wed Mar 07, 2018 8:00 pm
glass scales are a problem on lathes that small. many companies sell other than 'glass scales' that are much thinner...the big problem, even if you CAN get the glass scale on the lathe is loss of tailstock positioning toward the chuck when working with very small parts.

I would be surprised if you can get a ball screw under the slide of a lathe that small (that was not initially designed to support a ball screw).

everyone thinks that you can take a 'hand wheel' lathe and stick a control on it and get good positioning...

please look before you leap....
This is the constant struggle in my life. While on the road, I dream big, and have to reground myself when I get home and am actually able to measure things... I’ve been thinking of plans B, C, and D just in case... I WILL be able to get a ball screw in there, even if unconventional.

User avatar
DICKEYBIRD
Posts: 176
Joined: Sat Feb 03, 2007 10:45 am
Location: Collierville, TN

Re: Ballscrew tolerances and real world performance

Post by DICKEYBIRD » Sat Mar 10, 2018 9:10 am

WJH wrote:
Thu Mar 08, 2018 1:26 am
This is the constant struggle in my life. While on the road, I dream big, and have to reground myself when I get home and am actually able to measure things.
So very true 'dat! I'm not on the road anymore for work but still spend 9-10 hours a day away from my shop workin' for the man. It's always exciting to make plans during the day for what will be accomplished in the shop that evening. On the long drive home the "mental adrenaline" slowly leaks out like the air from a leaking tire & by the time I eat supper, spend some time with SWMBO & get out to the shop, it's all I can do to just straighten up the shop a little bit or just remember where I was last time. :roll:
Milton in Tennessee

"Accuracy is the sum total of your compensating mistakes."

User avatar
Dave_C
Posts: 960
Joined: Mon Apr 21, 2008 10:34 am
Location: Springfield. MO.

Re: Ballscrew tolerances and real world performance

Post by Dave_C » Sat Mar 10, 2018 9:15 am

just remember where I was last time
I thought I was the only one with that problem! :roll:

Dave C.
I learn something new every day! Problem is I forget two.

User avatar
Marty_Escarcega
Site Admin
Posts: 1708
Joined: Wed Dec 18, 2002 8:17 pm
Location: Mesa, AZ USA

Re: Ballscrew tolerances and real world performance

Post by Marty_Escarcega » Sat Mar 10, 2018 9:17 am

WJH wrote:
Thu Mar 08, 2018 1:26 am
Bill Shields wrote:
Wed Mar 07, 2018 8:00 pm
glass scales are a problem on lathes that small. many companies sell other than 'glass scales' that are much thinner...the big problem, even if you CAN get the glass scale on the lathe is loss of tailstock positioning toward the chuck when working with very small parts.

I would be surprised if you can get a ball screw under the slide of a lathe that small (that was not initially designed to support a ball screw).

everyone thinks that you can take a 'hand wheel' lathe and stick a control on it and get good positioning...

please look before you leap....
This is the constant struggle in my life. While on the road, I dream big, and have to reground myself when I get home and am actually able to measure things... I’ve been thinking of plans B, C, and D just in case... I WILL be able to get a ball screw in there, even if unconventional.
There are kits made for the small benchtop lathes.
Again, don't overstress about accuracy of the screw. Your little lathe will be fine with C7. Just get it together and learn....have fun.
"Jack of all Trades, Master of None"

WJH
Posts: 1421
Joined: Sat Jan 16, 2010 9:29 pm
Location: Florida

Re: Ballscrew tolerances and real world performance

Post by WJH » Sat Mar 10, 2018 9:18 am

Hah, I wake up in the morning, walk out of my hotel room, and ask myself, “Where am I?” So many hotels, and time zone changes, takes me a bit to get my situation awareness going. My brain often thinks I was in the previous hotel initially.

Post Reply