backlash

K

kregg slack

Guest
i have 20 thounsands backlash on my old leblond lathe, the cross feed. i heard if yo splite the tee nut it will take care of the backlash. if so do you cut it clear in half, so there is two tee nuts or just part of the way so you can try to tighting it up. i use gages on important work but working on lathe and mill cleaning them up and trying to get the backlash out.
 
You can split the nut then separate the two pieces with a shim to tighten it up. Some lathes are made with a two piece nut with a wedge between them to provide adjustment. The thing is, you will never remove backlash entirely for more than a very short period of time so you always have to be aware of it anyway. Regards, Bill.
 
ok thank you, i will just leave it alone, some day i would like to own a $12000.00 new jet and have zero backlash on every thing.
 
You cannot have -0- backlash. And .020" ain't bad at all for an old lathe, got another 20 to 50 years of life in it.;)

Bill
 
i guess then it will last me my life time. sence i have all the tooling for it i wont try to up grade to something else and still have backlash in something new.
 
With amount of hours a typical hobby/home machine gets in a year, should last a very very long time.

My Monarch has .020" backlash in the cross slide and this machine is scarry accurate.

Bill
 
I split the nut on mine and hold it together with two screws tightened just "so" and with blue Loctite for future adjustment... as Bill said you can not have 0 adjustment... just realize it is there and work accordingly...
 
Good Grief

That is just about brand new.
Backlash on a cross feed is simply a non issue. Heck, even if you had "zero", it sure wouldn't stay there very long.
Just aways keep the slack one way, and worry about something that counts........jackie
 
ok thank you, i will just leave it alone, some day i would like to own a $12000.00 new jet and have zero backlash on every thing.

You can easily make a machine tool with zero (less than 0.0001") backlash. They are made every day. But, the feed screws and nuts are not Acme thread, they are ball screws and nuts. Just like ball bearings.

One problem, good quality ball screws are very expensive for linear positioning where 1.000" position is 1.000" +/- 0.0001" and each progressive inch is the same true position. This would pertain mainly to milling and jig drilling machines. On a lathe where true position is not as critical these ball screws are much cheaper but still would be 3X-5X more than a regular Acme screw and nut.

In earlier days (late 0'60's) a precision ball screw and nut cost about $4,000 per foot (in 0'60's dollars). Then came linear scales that did the true position measuring. Nowadays, the actual ball screw is measured every unit of resolution, the true position is recorded in the CNC control, so that positional error is corrected by the processor and fed back to the screw motor-controller.

Just 0.020" backlash, hey, give thanks!!
 
Wait Till This Happens

One of your Machinist walks up to you and says, "I turn the handle on the cross slide on the 32 inch Lehmann Lathe, and the crosslide doesn't move". If you use one long enough, it will happen........jackie
 
Nowadays, the actual ball screw is measured every unit of resolution, the true position is recorded in the CNC control, so that positional error is corrected by the processor and fed back to the screw motor-controller.

The better CNC machines still use linear scales as a form of feedback to the computer. The encoder on the motor only counts revolutions and segments of revolutions not true position of the table. The linear scale compares true position to revolutions of the screw and adjust accordingly. If the error is too large you'll get an alarm code, then you have a choice of bumping the parameters to a larger window of error or changing the ballscrew. A machine that has only motor encoders will have a positional error as the screw wears. You have to adjust the parameters to compensate for the error. The CNC mill I run has the capability to dial in a correction factor for every 1/2" of travel through the table range.

Bottom line, a CNC with encoders only on the motor will have positional error as the ballscrew wears.
 
Back
Top