Variable Voltage Driving an Actuator

Hello,

I'm a bit of an amateur as far as electronics are concerned (my background is civil engineering, not electrical!), but was hoping for some help with the following project:

I am trying to drive a linear actuator at variable speeds. PS - I am yet to make any equipment purchases as I wanted to get my research right beforehand.

The linear actuators I've been looking at state that the speed of the actuator is a function of the voltage up to 24v (i.e. the more voltage, the faster the slider moves). What I want to do is hook up a switch (a sliding switch - like a dimmer but not circular) that will control the voltage input to the actuator, in essence giving me mechanical control over the voltage. I've done some research on the subject but the electrical jargon which I'm not familiar with makes me think that some more knowledgeable input would be a big help.

Can anyone think of a better way to do this than the below?

Mains Input (240V) --> Stepped down to 24V using a transformer --> Voltage controlled via mechanical switch of some kind (any names???) --> Manual control (linear slider)

Any ideas are welcome!! (including switch names or suggestions as I'm not 100% familiar with electrical jargon)

The linear actuators I am familiar with are based on either a DC motor (which your idea sounds like) or a stepper motor.

None of the off-the-shelf stuff is cheap, and your selection will depend on budget constraints.

If you are going to move the actuator to a certain point and leave it there, a stepper motor would be a good way to go. You can buy stepper motor drives for less than $100.

Depending on how much torque/force you need, you'd be better off buying some sort of power supply out of a catalog instead of designing your own.

Can you describe your application in more detail?

No problems, thanks for the reply.

None of the system requirements are totally quantitative, it's more the way that they connect and interact that I want to find out.

The best analogy I can think of is this: Imagine a cable - when the cable is not being pulled, the actuator remains closed (cylinder not extended), when the cable is pulled, the actuator opens, but only proportionally to the amount of cable extension, i.e. if I pull the cable halfway the actuator only opens halfway, and will stay there as long as the cable is in that position. From there, if I was to let the cable go, it would go back to its 'resting' state (closed), and if I was to pull it all the way, it would open fully. I also need it to match the speed of the cable pull - if I pull quickly, it opens quickly, let it go, it closes quickly, slow release, it closes slowly.

To satisfy these requirements I'm guessing that the switching is the tricky part. I was thinking some kind of variable voltage switch to control the speed, but that's as far as I got, so any input is very welcome!!

Reminds me of the words, "servo" or, "Waldo". Also reminds me of the Radio Shack remote antenna rotator. If you can get the moving thing to rotate the shaft of a potentiometer, you can use a similar potentiometer as your control. Put the resulting voltages into an op-amp with gain and it will output a voltage proportional to the difference in the positions.

I'm sure that other people on this site can complicate this all the way up to needing a small computer, but I believe in simplicity.

a relatively easy task. Your slider is a varaiable resistor, your actuator has a slave variable resistor connected to it. An op amp drives the actuator with the objective of matching your sliders reference to the actuator's feedback. Gain, or response is fashioned in electronics, while scale is fasioned in the mechanics of connecting the feedback pot to the actuator.

Initially after reading the responses I had no idea what an op-amp or potentiometer was (yes, that clueless), but after research and understanding, I realise that these solutions are simple and ideal - thanks for the help guys!

Just to clarify, I did a quick paint sketch up with my understanding of how it would work - do you quickly mind running your eyes over the attached jpeg and see if anything seems like it would be a problem?

Note: I thought most op-amps were single output only - how would this work in controlling an actuator with a +ve and -ve wire?

Also, it just occurred to me that the actuator changes direction by reversing polarity - is there a kind of switch that does that that could be inserted into the circuit somewhere? The potentiometer would alter the speed but not the direction of travel...

Thanks again for all the help

You may want to look into one of these...... http://www.electronickits.com/kit/complete/motor/k166.htm

Building kits is one way to get your feet wet in electronics and, at the same time, have success in a project.

If we use your analogy of pulling on a rope to move the actuator, if you were to connect an arm to the potentiometer and connect a spring from the actuator end to the arm on one side and your rope to the arm on the other side, it could be set up so that when you pull the rope, the potentiometer would turn one way and the actuator extends and when you release the rope, the potentiometer turns the other way and the actuator would retract. The spring tension while extending would provide tension feedback to the rope you are pulling on. You would also want to install limit switches to protect the motor and drive system when the actuator reaches full stroke in either direction. Some actuators have that protection already built in.

Gordonov,

Something I'm not clear on... say you have a linear actuator with a travel of 12" and a variable resistor with a travel of 1". As I understand your request you'd like to pull a rope 1" and have the actuator extend 1". If that's the case then you would need a proportional mechanism, one in which a 1" travel of the rope results in a 1/12" travel of the variable resistor. This can be accomplished in several ways... open gearing, string and pulleys or small chain and sprockets.

williamj

Referring to post #7

I was thinking more like this:

It's a terrible piece of artwork and the circuit lacks many details, but it's an idea about how to reverse polarity on a motor without any switches.

It looks like that would work as long as the amplifier can supply the current needed to run the motor.

I was expecting maybe a push-pull output stage with nice fat power transistors added on.

Gotta go now. Real life beckons.

williamj: Indeed, a good point - it becomes more of a problem when the travel of the actuator is much larger than the range of the variable resistor, but I think in this case the two won't be vastly different, and what small difference there is could be made up with pulleys or gearing as you suggest.

BillB3857: That sounds like exactly the idea I had in mind, however won't there still be the problem of reversing polarity?

As I understand it, the feedback and control pots will alter the resistance - this feeds to the the op amp, which varies the speed of extension of the actuator via voltage changes, but if the polarity isn't reversed, the actuator won't change direction, just speed. Is this correct?

#12 - Thanks for the diagram...it looks interesting, unfortunately my circuit diagram knowledge being what it is, I don't quite understand it...

The concept is that you make both positive and negative voltage with your rectifier arrangement. Give that to an op-amp and it can deliver positive or negative voltage, as needed. Thus, the op-amp reverses polarity in a single chip and eliminates using switches. Another poster put up this link to an op-amp that is a good suspect for this job.

http://www.ti.com/product/opa549

To the OP--- Have you looked at the link I posted in Post #8. It provides the reversing through what is called an H-Bridge and operates on one power supply. That unit, with its control pot connected to a lever/spring/rope set up should do what you want. Again, caution is needed to properly limit the extreme ends of travel.

#12 - That looks like a clean and efficient way to do it, but I suspect my lack of electrical knowledge would let me down in making it work. The potentiometer-op-amp setup makes sense to me as it's not overly complex.

BillB3857 - I did originally but thought it didn't involve a potentiometer - I look closer at it now and realise it does! However, I would still need to run the output from this to an op-amp to regulate the speed of the actuator via the voltage, is this correct? As I understand it, this unit would take the place of the standard control and feedback pots in my original diagram, but otherwise nothing would change - correct? Otherwise this seems spot-on for what I need it for...thanks for the link, I'll definitely be in the market for a couple of these.

One other problem that just came into my head now - so say the cable was pulled say at e.g. half-speed (so the pot has been rotated half-way), the actuator would move at half-speed, which is fine...and if I released pressure on the cable and allowed it to slide back until spring tension, it would reverse the polarity through the H-bridge and the actuator would reverse direction, which is all good.

However, if I was to halt the cable in a half-way position, the actuator would still continue its travel at half-speed until it reached the fixed limit without stopping. Can you think of any way that if the cable was halted in position, the actuator would also halt?

You design the electronics so they drive the actuator until it finds the equal setting on its potentiometer.

You are thinking of the pull string as an accelerator pedal. It's not. It's a position locator. Whatever position you pull the string to, the actuator seeks that position. The farther the 2 pots don't match, the faster the actuator tries to find the equal point.

Hence the need for two pots instead of one...of course. I should have thought about that...thanks for the heads-up!

My thought on your rope controlling the position of the actuator would be similar to the speed governor on a lawn mower. The lever (your rope) pulls on a spring. The wind from the fly wheel of the mower balances the position of the spring with pressure (a second spring). Two potentiometers would be by far the better approach and wouldn't be too hard to implement.