Electric motor spec 1

[bitblender]

Hi chaps,

I'm building a belt driven linear actuator because I can't find one that meets my required spec within my budget. My design has a 300mm stroke and needs to run at 400mm/s. The belt runs round two 40mm diameter pulleys which need to run at 192rpm to move the belt at my required speed. The actuator will need to change direction quickly and accelerate fast due to its use in a motion simulator. Each actuator will be moving a load of around 10kg - 20kg.

So, anyway, I need a motor which is capable of powering it, which is why I'm here. I know nothing about electric motors, so maybe someone could give me some advice as to what I need?

THanks all...

 

[itsmoked]

This 20kg load includes the linear actuator? I ask because some can be pretty stout = heavy.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[bitblender]

No, the 20kg load is not including the actuator, but the load more realistically is likely to be closer to 10kg.

 

[jraef]

I'm all for not reinventing the wheel. Have you looked at IAI (Intelligent Actuator Inc.)? They have a belt driven linear actuator that can handle that kind of load, stroke and speed. I have used them in the past, they are great and super easy to program and interface.

link to IAI

 

[bitblender]

Thanks, I took a look but they don't seem to have any rod actuators. I should have said before that my design is a belt driven rod.

 

[itsmoked]

Hopefully cwilson will show.

You say 400mm/s over a range of 300mm.

Does this mean you need to be able to move your 20kg(worst case) its 300mm in 0.75s?

This is implying that you must reach a speed far above 400mm/s as you must accelerate then decelerate the load over that distance.(more power) Correct?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[bitblender]

Thats correct. The ideal actuator would be the SCN5 (

http://www.automation4less.com/store/proddetail.asp?prod=SCN5-010-300AS03-NA)

but its way over my budget, so I want to build something myself that would get close to those specs.

 

[roydm]

Sounds like a pneumatic cylinder c/w speed control valves would be better suited, it will be hard to find an inexpensive electrical solution that fast.
Roy

 

[bitblender]

I'm not sure if pneumatics will change direction fast enough. I need to be able to simulate rumble effects.

 

[mikek10]

What is your budget?

Realistically anything operating that fast and to any degree of accuracy is going to be expensive. Off the shelf is likely to be cheaper than building your own with seperate servo drive, motor etc.

 

[itsmoked]

I've just been on a pneumatics positioning job. Same thing happened. They user thought it would be cheap but ultimately it wasn't after nickeling and dime-ing, it ended up being $500 and frankly not all that great. That job didn't have any velocity issue(slow) either. So I am just agreeing that this is going to be, "not cheap".

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[roydm]

Bitblender,
I read post again I, I am still a little unclear about your application, could you describe it a little more.
I had a thought about the old servo recorders, some of those were extremely fast e.g. could draw out a 60 Hz sine wave. The input signal was compared to the slidewire feedback then amplified to drive a motor up or down scale. The motors were either AC or DC. Later designs used servomotors or linear motors.
I'm not sure why pneumatics is considered too slow, too springy perhaps?

Regards
Roy

 

[mikek10]

Its the old triangle.

Pick 2 out of good, fast and cheap

If you want good and fast it sure as hell isn't going to be cheap!

 

[rsteves]

Here are a few calculations to help you spec an appropriate motor to build your actuator.

Assumptions:
1. Worst case is accelerating from stop at position 0 to max speed at midpoint, and decelerating to stop at position 0.3m.
2. Average speed must be at least 0.4m/s.
3. Assume constant acceleration with a linear velocity ramp (peak velocity at the midpoint).

Due to linear velocity ramp, peak velocity will be twice the average velocity. So peak velocity will be 0.4m/s * 2 = 0.8m/s. Time to reach the midpoint will be 0.15m / 0.4m/s = 0.375s.

Acceleration = Peak Velocity / Time = 0.8m/s / 0.375s = 2.13m/s^2
Force = Mass * Acceleration = 20kg * 2.13m/s^2 = 42.6N (or about 10lb)

If the pulley is 0.04m diameter then peak pulley speed = 0.8m/s / 0.04m / pi = 6.366 rev/s = 382 RPM
Torque = Force * Radius = 42.6N * 0.02m = 0.852N-m (or about 0.63 ft-lb)

Most motor manufacturers will show torque vs speed curves for their motors. Your minimum requirement is a motor that will produce over 0.852N-m (0.63 ft-lb) of torque at speeds up to 382 RPM. Multiply those values by a safety factor of 1.5 or so to account for friction and additional inertia of pulley, belt, and rod. If you know the belt and rod masses, add them to the 20kg in the above equations for more accurate results. The pulley inertia is probably negligible.

Cheers.
PS. No guarantees I didn't make any mistakes in the above calculations but the results feel about right. Use at your own risk. Oh, and whether it's worth building a custom actuator vs buying one off the shelf all depends on how much you value of your own time which only you can decide, so don't be discouraged if others tell you it's not worth the trouble.

Rob Steves - Zaber Technologies Inc -

http://www.zaber.com