High Speed Rotary Stage for Polarizers

[crgorman]

Stuck on a problem. I need to spin a 1" polarizer optic 0 to 90 degrees in 0.007 seconds. If my math is correct (please correct me if I'm wrong) and I'm evaluating this in 360 degrees or 0.028 seconds/revolution, that is 2142.857 RPM or 35.714 RPS/HZ.

ω = 2142.857 ( Revolutions)/minute × 6 = 12857.142 degrees/second = 224.399 rads/second

I'm looking at frameless motors, rotary stages, hollow-shaft motors. Can anyone think of any other systems that may aid me in my experiment?

I must be able to mount this polarizer whilst transmitting light through it and must be able to obtain the 0 - 90 degree requirement. Open to ideas. Thanks.

Chris

 

[3DDave]

Is this to be in one of two positions or is it necessary to be in continuous motion?

 

[crgorman]

3DDave, it will have varying degrees to reach. Anything between 0 and 90 degrees. Example, I could move to 23 degrees, then to 70 degrees, and then back to 10 degrees. This will occur for about 30 seconds. So precision bi-directional repeatability is a must. Thanks.

 

[drawoh]

crgorman,

The fun part of your problem is the acceleration. Work out [α]. Imagine you have a not very small kid throwing rocks very hard at this thing. The two analysis models are the same, and the kid with the rocks may be your lesser problem.

This is a challenging mechanical engineering problem. You need a drive mechanism. You need to manage mass and moment of inertia. You need this thing to remain in one piece as it operates. You need this thing to not shake apart everything in the vicinity as it operates.

Do you have mechanical engineers in house? It is time to talk to them.

--
JHG

 

[crgorman]

drawoh,

I've been looking at calculating the acceleration but it has been a minute since I have done it. acceleration = w/t correct? Change in angular velocity / change in time right?

This seems to be the only viable solution, given they have absolute encoders and low propagation controller characteristics.

https://www.kollmorgen.com/en-us/products/motors/direct-drive/cartridge-c-and-ch/cartridge-ddr/

We don't have any ME guys or gals in the building. Man do I need one. Thanks.

 

[drawoh]

crgorman,

You are rotating a 1" polarizer 90[°] in 0.007s. Let us assume constant acceleration to some speed, followed by an equal constant deceleration back to zero.

[θ] = 90[°] [×] [π] [÷] 180deg/rad = 1.57rad

[ω]_ave = [θ]/t = 1.57rad/0.007s = 224rad/s

Since your accelerations are linear, your maximum angular velocity must be double the average velocity...

[ω]_max = 2[ω]_ave = 2 [×] 224rad/s = 449rad/s (I am doing the arithmetic in Octave here).

You have half the time period to accelerate in either direction...

[α] = 2[ω]_max/t = 2 [×] 449rad/s / 0.007s = 128[×]10³rad/s²

At the outside diameter of your polarizer...

a = [α]r = 128[×]10³rad/s² [×] 1in[×]0.0254m/in / 2 = 1630m/s²

That is a tad over 160Gs.

Your drive cannot be axial because you have a clear aperture. You need to attach a gear, a belt or a link, and keep it attached as the thing slams back and forth. The resulting forces must be isolated from any delicate components in your system. Hopefully, those delicate components are not part of the polarizer's optical schematic. You have to make your highly stressed components as light as possible. You have to minimise the radii to any masses.

Electronically, this is simple, unless you cannot read the encoder in real time. As I noted above, the mechanical engineering is a challenge. You need a mechie to work on this.

--
JHG

 

[itsmoked]

This should be done with a voice coil actuator NOT a motor. You'd want a motor if it simply rotated like a scanner but since you don't want it to turn more than 90 degrees and want partial moves voice coil is the correct ticket.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[moon161]

You can probably put it at the optical 'waist' where the imaging rays mostly converge. That will allow you to have a polarizer diameter of 'conceivably pretty small'. This makes the problem a whole lot more tractable than the first image I had, of spinning a 2" polarizer filter in front of a lens assembly.

If you do this, then you just have a hard drive voice coil with a hole through the bearing, and a polarizer in the hole. Don't forget to paint it black.