Servo hacked linear actuator

[AntonB] has modified a servo into a powerful linear actuator (think: changing rotational motion into linear motion). The process is simple enough, modify a servo for continuous rotation and then add the custom built actuating shaft. You do of course lose the precision of the servo, but a small price to pay to be able to lift ~20 pounds straight up. Inspiration for such a cheap solution came from his Planetary Surface Exploration Rover. Check out a video of both after the break.

[Thanks Eric]

22 thoughts on “Servo hacked linear actuator

  1. If you’re giving up the position sensing of a servo, why not just attach the leadscrew to a cheaper & faster/more powerful gear motor? If all you have lying around are servos it works, just don’t go our of your way to get one if you need a linear actuator.

  2. @EdZ: because RC servos are commonplace and a cheaper and faster gear motor is a relatively rare thing to come by.

    I’m wondering about durability of such contraption though. RC servos have minuscule motors running at high RPM, overheating horribly and their lifetime is short.

  3. Off topic but: @Heliostat hippy: I like your site, it has some good information and has got me thinking. Just one thing, though:

    “As you know tha sun revolves around the earth in 24 hours each day.”

    Can I introduce you to my friends Copernicus and Galileo? ;)


  4. Crap! He beat me to it. I have been working on one of these for a while. One suggestion for anyone planning on building one. Spring for ‘telescoping tube’. Smallparts has it.

  5. @EdZ because you know the thread pitch you can easily calculate how far a certain number of revolutions will move the bar. so you can then purposely move the linear actuator a determined distance by moving the servo a set amount of revolutions.

  6. @svofski
    Having a quick peek in The Laminated Book of Dreams (read: the mail order component catalogue), gear motors of an equivalent power to a servo are about half as expensive. Albeit these are slightly larger motors, not tiny hobby servos.
    The servo he gutted used a potentiometer for position sensing (which was removed). If you had a rotary encoder servo, then yes you can make a precision linear actuator. This is what’s used in CNC mills, though using ballscrews instead of leadscrews, and much beefier servos.

    You could probably retrofit an existing motor with a cheap rotary encoder and a PID controller and get a high power precision linear actuator. Measure it’s uncontrolled motion for a bit, whack the measurements into EurEQa (or enjoy matlab-induced aneurysms), and add the controller equation to whatever you’re using to drive the thing. Or if you want a standalone part, check this out:

  7. My bad EdZ. In that case it is kind of pointless even if you add an encoder. Cuz it costs more for servos. I did not investigate fully enough it appears. Thanks for the link. I did indeed have cnc machines on my mind at the time. I just wondered do you think a hall effect sensor (or a series of them) from brush-less computer fans would work alright for an encoder? I have used the gutted sensors to measure rpms on things before, to a fair degree of accuracy (for my purposes). Just wondering EdZ.

  8. Seems to me like there should be a mechanical way to link the pot to the linear device and retain feedback. Easy but big way is a slide pot…

    @edz: care to share your source of cheap gearmotors? The cheapest gearmotors I’ve seen are about $8, which more than half the cost of an average servo (about $12), and that’s comparing surplus gearmotors of random speed to “new, brandname, get it anytime” servos. Thus the many web pages to convert servos to general purpose gearmotors. (“new” gearmotors, where you get to specify the RPMs and such, tend to be outrageously expensive for some reason.)

  9. EdZ: mail order is ok, but it doesn’t work flawlessly everywhere in the world. RC servos are just commonplace. I’m not such a big fan of them either: they’re noisy, slow, unreliable, have stupid control. But it’s just the way things are.

  10. Hi all,

    Let me make a start by saying that the reason I decided to post this method is not because I found the answer to every operational aspect of the hobby servo – linear actuator combo :)

    As EdZ correctly pointed out, you have to get rid off the pot and hence the feedback element, but coming to think of it, when it comes to continuous rotation the pot is not of much use anyway since there is a dead band of about 10 degrees, which leaves the pot’s output floating – not a very good I/P for the ADC of your micro-controller. If one insisted to use a pot they could very easily affix a potentiometer on the output shaft of the servo. Of course it would be better to use an absolute encoder: :))

    I believe that servos are great for hobby robotics ventures since are way cheaper than standard gearmotors, all the components come enclosed up to IP64 standard, (protection against low pressure spraying water), they come in a small form factor with mounting flanges ready-fitted.

    The idea came to me from my planetary exploration rover (my PhD), where I had to make the linear actuators myself using standard DC gearmotors parallel gearboxes quadrature encoders etc etc
    Have a look:

    The small servo hack would be extremely useful in making a scale model of my rover. I intend to release the rover designs to the community – the servo linear actuator is a first attempt to draw attention :)

    The next step is to hack the servo into a PID speed/position controlled continuous rotation gearmotor with absolute feedback (1024ppr). Schematics are already freely available:

  11. This puts a lot of stress on the motor’s thrust bearing, if it even has one. If the motor isn’t designed to take that kind of stress (pulling the shaft), it could wear out prematurely.

  12. @ramsay
    Wow, just wow, I hope it was some sort of joke or translation error on that sun around the earth thing, else it’s the most embarrassing thing ever, especially since he’s dutch and not american I gather. (dutch are largely non-religious and generally have some decent basic education)

  13. I have now tested the actuator more than 20 times with that payload (10kg). Of course it is NOT intented to be used for such weights. Bare in mind that I’m using a servo that does not even have bearings on the output gear. I believe a 5 kg payload and a better servo would really work miracles. If one wanted faster response from the actuator, they could use a larger pitch thread i.e. an M10 would double the extension rate but would halve the lifting capacity.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s