Precision Control of Linear Actuators Using Force Limitation and Feedback Loop

[Borka]

This might be a simple and pretty basic question, but i have not succeeded on finding any relevant info online, so hopefully someone can help me out.
Is it possible to pull and actuator and it resists being pulled with a preset amount of force?
What I'm thinking is e.x you have set a preset a limit of 50N and unless the force used to pull exceeds 50N the actuator will not move, and as soon as the force reaches 50N the actuator will move and basically work as a 50N weight.

 

[Nidum]

A solenoid and a linear motor will both do that with some load sensing and control electronics . Roughly speaking solenoid for short travel and linear motor for long travel .

 

[jim hardy]

Look up 'torque motor'

 

[Nidum]

Jim beat me to it - a suitable motor with a piece of string around a pulley can do the same thing as the linear motor and solenoid .

 

[Hesch]

I don't quite understand what you mean. ( You must express yourself more accurate ).

If you use a linear motor connected directly to some rod ( not a rotational motor with some worm gear ) you can do it very precisely. I will suggest a synchronous linear 3-phase motor.

Make a 3-phase drive that converts force into phase currents.
In front of this drive, you connect a speed- or position controller. The controller+drive is enclosed in a loop feeding back speed/position.
An input to the speed/position controller signals the allowed maximum output force to the driver.
In front of the speed/position loop, you connect a function generator, that generates ramps, steps, whatever.

Now, if some condition involves a cross of the force limitation, the speed/position controller will truncate the force output to the driver, and will feed back a correction of the function generator.

Say you are using a speed controller, and the function generator signals speed = 0. The actuator is halted.
Now you increase the load of the actuator, so that the force limit is exceeded, so the actuator will be moved and the speed controller will feed back a correction to the function generator that it must follow or "let go". Hence you will not have a "wind-up" in the overall system. The system will just accept the conditions.

To do all this you must use digital control, using a micro processor, which have to:

1. Sample speed/position
2. Calculate forward the force to be used.
3. Limit the force.
4. Back calculate the excess of the force to the function generator
5. Correct the function generator.
6. Calculate the whole system with the new function generator values.

That's it.

 

[jim hardy]

Such constant torque motors were commonplace thirty years ago for magnetic tape drives , to keep constant tension on the tape.

As Hesch said, the more clearly a question is asked, the better the answers it will inspire.

 

[Borka]

I don't quite understand what you mean. ( You must express yourself more accurate ).

If you use a linear motor connected directly to some rod ( not a rotational motor with some worm gear ) you can do it very precisely. I will suggest a synchronous linear 3-phase motor.

Make a 3-phase drive that converts force into phase currents.
In front of this drive, you connect a speed- or position controller. The controller+drive is enclosed in a loop feeding back speed/position.
An input to the speed/position controller signals the allowed maximum output force to the driver.
In front of the speed/position loop, you connect a function generator, that generates ramps, steps, whatever.

Now, if some condition involves a cross of the force limitation, the speed/position controller will truncate the force output to the driver, and will feed back a correction of the function generator.

Say you are using a speed controller, and the function generator signals speed = 0. The actuator is halted.
Now you increase the load of the actuator, so that the force limit is exceeded, so the actuator will be moved and the speed controller will feed back a correction to the function generator that it must follow or "let go". Hence you will not have a "wind-up" in the overall system. The system will just accept the conditions.

To do all this you must use digital control, using a micro processor, which have to:

1. Sample speed/position
2. Calculate forward the force to be used.
3. Limit the force.
4. Back calculate the excess of the force to the function generator
5. Correct the function generator.
6. Calculate the whole system with the new function generator values.

That's it.

Sorry for the lack of accuracy in the question, but what you described is exactly the answer i was looking for. Thank you very much!.