re:Re: magnetic spindle bearings

Jon Elson wrote:

>Elliot Burke wrote:

>>Has anyone looked at magnetic spindle bearings?
>>IBAG ( http://www.ibag.ch/e/index.shtml makes these things, they claim
>>stiffness similar to ball bearings, 200,000 rpm, up to 125 HP (maybe not

all

>at the same time).
>>Evidently there is servo action to adjust the magnetic field and keep the
>>spindle centered. One could imagine capacitive sensors, servo circuitry,
>>and coils to push the spindle back to center made as part of the motor.
>>Capacitive sensors have been made good to nm, and 200,000 rpm is only 3000
>>Hz, not a very high speed for analog circuitry.
>
>
>This is really complicated stuff. Especially when adapted to high speed
>rotating machinery.
>Air bearings would be a LOT easier to make. Using barrel laps and ring
>laps to bring the
>bearing and the shaft, respectively, to correct size is sufficient to
>get VERY accurate fits.
>If you can get clearance down to about .0002" per side, you get an
>amazingly stiff bearing.
>Air bearings in the 1 -2" diameter range routinely get stiffness of 1
>million Lbs/inch
>deflection. That means a 1 Lb load deflects the bearing only one
>micro-inch. This is
>MUCH stiffer than ball or typical oil-journal bearings. Needless to
>say, an air
>bearing has VERY low friction, and it doesn't need micro-inch sensors,
>power amplifiers
>servo loops, etc.

>>air bearing-> high accuracy & speed, limited stiffness
>
>If you check the literature, when designed for good radial stiffness,
>air bearings do VERY well!

>magnetic-> high accuracy, speed, and stiffness. Cost is way up there!
>
>
>Yes, I'm sure the last sentence is right!

>Jon

Jon, I'd sure like to know what air bearings have that sort (1Mlb/in) of
stiffness!
My search found numbers of 40 klb/in and below.
And yes, these were carefully designed and made bearings.
AFAIK, the larger stiffnes numbers are associated with liquid bearings,
which are limited to lower speeds.

Any references?

I have read an older (1970) book on the subject of hydrostatic bearings, and
it seemed in agreement with the results of my web search. At that time the
complexity of air bearing stiffness calculations made the author omit them
from the book, so I don't know if he included the hydrodynamic effects with
the hydrostatic effects in his conclusions.

The air bearing applications I've seen have been been rather low force so
the deflections were small.

The single point diamond turning lathes stopped using them a long time ago
because of the limitation in stiffness.

Maybe there has been some great design improvement in air bearings over the
last few years?

We don't have to be afraid of servos controlling spindle axis location,
consider a CD player, which has two axis servo control of the read head to
sub micron tolerance with very simple and inexpensive apparatus. A
capacitance bridge and some force coils should do it. It is possible that
the forcing function could be built into the motor, further reducing
complexity.

Elliot B.