Boss8 Series II Y Axis grows after warm-up?

My Boss8 series II has developed Y axis zero problems that started to became detectable
about 3 years ago.

First of all the Y ballscrew was rebuilt about 10 years ago, however usage has been
prototype only, and sometimes isn't used for months. I estimate a max of maybe 3,000 light
duty hrs on the ballscrew. I have checked oiling, and that seems to be okay, at both the ball
nut, and the ways.

It developed into a definite problem about two years ago. After (approx 20 or 30 min)
warm-up, the Y axis would grow approx .0013 in the direction of the column (as if the y
screw was increasing in length from the thrust bearing, as it warmed up) even if the axis
cutting load was virtually zero.

Two months ago I speculated that loosening the y axis gib (I had readjusted it when the ball
screw was rebuilt) 1/8 turn may help (loosen 3/4t then tighten 5/8t) ,,, it did not. In fact, it
made it worse, "the amount of growth doubled!" So now I'm seeing .0025 shift after
warm-up. I instantly moved the gib adjustment to the original position, but this seemed to
changed nothing? I still have about .0025 change in position after warm-up.

And, once things cool down completely, or I restart the machine again at a later date, it
seems to be at the same location it was at the time of the first setup, when the machine was
cold, (not previously running) "y minus .0025" from the warmed up position???

This is a PAIN, and a waste to have to run my machine for 30 to 60 min doing nothing but
cutting air, so the axis is normalized before critical cutting! Then as the machine cools, or
sits, the y axis shrinks, so I have to keep the machine running (and verify y zero) to keep
things accurate!??

Once its warmed up (and I keep it warm) the shift normalizes.


***Any experienced input, or experiences relating to "what could be going on" would be

greatly appreciated***

This may be way off, BUT how is the motor/encoder itself. I replaced the brushes on my Y axis BOSS 9 a year ago. It's more accurate now, and I did originally assume it was something mechanical.


Sent from my iPhone using Tapatalk

The ball screw is NOT in tension (between 2 sets of bearings). Thus due to heat, it will grow. It only has a precision bearing on one end. Work produces heat. The way oil is used to reduce friction and in a small degree to take some heat away. Gib adjustment if too tight makes for more work thus heat. Of course a gib too loose creates its own accuracy issues.

I understand the front thrust bearing (2 back to back) arrangement, and that the rear (roller cage bearing) driven end (allowing expansion) floats, but it should not grow. I just don't see where the heat if any is coming from? The thrust bearings are huge, and beefy as hell, how could these produce enough heat to consistently shift in the direction of the columb at such low RPM?

And the ballnut would have to produce more heat than is expected in normal operation to cause expansion like that? It happens at load levels of almost zero. (3/16" end mills, and smaller) And feeds of less than 10 IPM.

I will use a thermal IR gun to check the Y ballnut temp, before and after warmup next time I run it, however, I can't imagine this would be the source of any significant heat. (?)

Has anyone actually experienced this problem, and its cause?

"Load levels of almost zero"

Some machines use a significant portion of servo torque just to move the table (~25%). If it moves too easy, your gibs are too loose.

Assuming your screw is 15 inches long and it heats up 20 degrees, you will get 2 thou movement at the end of the screw. 6.7*10^-6*15*20

Commercial machines conquer this with glass scales, warmup routines, and some even circulate coolant through the ballscrews to stabilize temperature.

but the machine only has 12 inches of travel, so .8 of that and I doubt he is at the end of travel. These machines are not known for this, certainly not mine, same iron, different control

Do you have backlash comp on the machine and what is it set to?

Is the belt over tight or over loose?

does the screw turn easily?

What ops are you seeing the error in? milling hole drilling.............

It matters not what the travel is but the length of the screw from the bearings to the ballnut. I guessed 15" as a worst case.

I agree it seems excessive, a good first step is to make sure the screw turns free...

If the ball screw is bent, either do to a crash or because og to Little pre tension it Will grow and your ball screw carrige Will move. Check the run out of the ball screw on the top and on the side.

First thing i would Try is to redo the ball screw pre tension

You are seeing 0.05 mm growth.
Over 0.5 m, 15C thermal gives you 90 microns, or 0.09 mm.
So .. you need about half that, or 7-8 degrees.

Any chance the screw warms 7 degrees ?