I have a need for a FAST linear actuator of which I may control position and speed reliably. I have already been looking into pnuematics, however it does not seem that this will fit my needs (I am open to suggestions though).

Currently my wanted motor specs are:

  • >16"/sec
  • about 1lb of force
  • Position control
  • Speed Control

It seems that I may be able to make a DIY linear if you will, however I wanted to ask the forums for a sanity/experience check.

Below is a rough drawing of what I was thinking to create.


  1. I am able to close the loop with position (looking at hacking a caliper for this or maybe even some Linear Pots in parallel, Source Link to Calipers for positioning data)
  2. I have a micro controller fast enough to process this position data (I believe I will be using a STM32F series but sill up in the air on this)

See the picture below for a rough outline of how I am hoping to accomplish this:

Sketch of proposed linear actuator

I will be controlling this with an off-the-shelf ESC.

In addition to the ESC, I will also be buying a ball screw assembly to be used as seen here - Lead screw assembly

I want to get feedback for this idea and anything that experience has shown may not work. This will be my first stab at this sort of project, so while doable it does seem like it can get big quick.

I am worried that I might be attempting to reinvent the wheel here, so what I was hoping to get is feedback on this idea or an alternative.

Thank you all in advance.

  • \$\begingroup\$ For anyone else wondering: 16" per second is about 0.4m/s. Anyway, I think the answer is going to depend on what you want the maximum deflection to be. \$\endgroup\$
    – Hearth
    Feb 11, 2019 at 3:42
  • \$\begingroup\$ On a forum I trawl, there are several people that have built similar actuators from scratch for driving home-built motion simulators. They are usually single phase DC, but you may find some useful information here: xsimulator.net/community/faq/diy-linear-actuators.309 The FAQ link is by no means exhaustive. There are many many threads involving builds very similar to what you describe, but you may have to dig for them. \$\endgroup\$
    – Platytude
    Feb 11, 2019 at 19:13
  • \$\begingroup\$ thanks Platytude!!! This is an excellent resource!! \$\endgroup\$
    – Carlo
    Feb 17, 2019 at 3:55
  • \$\begingroup\$ My partner & I have made such products for clients, if you still need a solution let me know. 3 phase has many advantages for smooth torque and high acceleration. If you have a spec. pls advise. \$\endgroup\$ Jan 14, 2021 at 14:33
  • \$\begingroup\$ to reach 400 mm/s in a 300mm ball screw requires g to be reversed midway to stop with the same -g before 300mm end stop (minus fixture length) thus if 24 turns are available @ 6mm / rev (?) the max RPM = 400 mm/s / 6 mm/rev *3600 s/min = 240,000 RPM. Yet with other methods we have achieved 2m/s speeds with 0.1mm accuracy over a 1m span in an XY array with a 1lb force. \$\endgroup\$ Jan 14, 2021 at 14:48

1 Answer 1


Your approach is feasible, but your drive will be larger than it needs to be and inefficient. You don't need a heavy 600 Kg-f ballscrew and nut to push a one-pound load. Remember that you must accelerate the mass and inertia up to your speed every time you move and brake to stop it. Since your ballscrew is about 11 inches and you want 16 in/sec, you will have to work very hard to achieve this speed and just as hard to stop it. In addition, to provide the required torque, you will need a large motor which will have its own inertia, and almost all of the energy you put into it will become heat from the braking you have to do to stop it.

Go for low mass everywhere - check out coreless or frameless motors and lightweight (plastic?) ballscrews and nuts or a belt drive using a plastic belt. If the object you are pushing has any mass, this may not be feasible and you will be stuck with a high power system. Good luck!

  • \$\begingroup\$ John,I hear ya... the plan currently is to make everything somewhat small scale and then scale up as I need to. This should allow me to keep everything as low mass as possible and allow me to get my control loop down. \$\endgroup\$
    – Carlo
    Feb 17, 2019 at 3:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.