Project: Speed and force control of a linear actuator

[ramzerimar]

My background is in mechanical engineering, but I'm working on a project for which I need to design a control system for a linear actuator. I want to design the actuator using a roller screw mechanism, which in turn will be conected to a 3 horsepower electric motor through gear reduction. I need to control both the force that the linear actuator will be applying or the speed with which it it moves.

I know I can use strain gauges to measure applied force and then use a PID algorithm to control it to a desired level. Also, I can use hall effect sensors attached to the motor to control speed by similar means. The problem is: the AC motor which will be used (probably something like this) needs a variable frequency drive (VFD), and I'm not familiar with those. I've used Arduino to control speed of small DC motors, but never AC motors this size. Would it be better if I use something like a PLC to accomplish this task, or do you have any other suggestions?

Any hints would be helpful.

 

[Nidum]

Tell us what you actually want this system to do .

 

[ramzerimar]

Tell us what you actually want this system to do .

The system is supposed to compress some material (soil) inside a box for an experiment. Think about a piston, with the shaft on the moving part connected to the linear actuator. It's something like that. One of the specifications for this project is that the motor should deliver between 2-3 horsepower, and I have to be able to control the force that is being applied on the material according to the experiment.

 

[Nidum]

Ask whether you can use hydraulics .

High power electro-mechanical servo systems are notoriously difficult things to design and get to work reliably .

A solution using hydraulics with the motor only running at constant speed to provide primary power would be ever so much easier to do .

In any case start by estimating the force and speed of movement needed for the application .

 

[ramzerimar]

Ask whether you can use hydraulics .

High power electro-mechanical servo systems are notoriously difficult things to design and get to work reliably .

A solution using hydraulics with the motor only running at constant speed to provide primary power would be ever so much easier to do .

In any case start by estimating the force and speed of movement needed for the application .

To be fair, I don't have any experience with hydraulics, but if it's a better idea than using a electro-mechanical system, I could give it a try. How would that work? I would still need a electric motor. How would the hydraulics system couple with it and with the linear actuator?

Could you explain why would it be so difficult to implement a electro-mechanical system? The VFD system would easily control the motor speed, right? Controlling the force applied through an electro-mechanical system would be so much difficult? I've thought about using a PLC or a microcontroller to send signals to the VFD through a PID loop. I would use strain gauges to measure the force being applied by the actuator.

 

[ramzerimar]

I need to check what will be the forces required for the application, but let's say I want to apply a force of 5000 N (just an example).

I've imagined something like this actuator (http://file:///C:/Users/Admin/Downloads/3600-4166_0202_rsa-gsa.pdf ).

Maybe I could replace the AC motor with a smaller DC motor and apply 5000 N of force? Have in mind that, for the project, I'm supposed to design the whole actuator (using roller screws, gear reduction, etc...).

With a DC motor, I wouldn't need a VFD and it would be easier to design a control system.

 

[Nidum]

My computer doesn't like that link and keeps blocking it .

Where the requirement is for a high force , low speed , medium positional accuracy servo system hydraulics almost always give the simpler and cheaper solution .

With high power electromechanical servo systems there is always a problem with response times . Big motors take a long time to change speeds . This problem is made much worse by the inertia of heavy gear box and other mechanical components attached .

There is usually a further problem caused by backlash between components .

These types of system can be made to work but the control logic and electronics needed can often be very complex .

Hydraulic servo systems can have very rapid response times and do so with lower cost components and more basic control logic and electronics .

 

[dlgoff]

Where the requirement is for a high force , low speed , medium positional accuracy servo system hydraulics almost always give the simpler and cheaper solution .

Indeed. And there's a lot of "off the shelf" products that would probably fit his needs.