1. Do not share user accounts! Any account that is shared by another person will be blocked and closed. This means: we will close not only the account that is shared, but also the main account of the user who uses another person's account. We have the ability to detect account sharing, so please do not try to cheat the system. This action will take place on 04/18/2023. Read all forum rules.
    Dismiss Notice
  2. For downloading SimTools plugins you need a Download Package. Get it with virtual coins that you receive for forum activity or Buy Download Package - We have a zero Spam tolerance so read our forum rules first.

    Buy Now a Download Plan!
  3. Do not try to cheat our system and do not post an unnecessary amount of useless posts only to earn credits here. We have a zero spam tolerance policy and this will cause a ban of your user account. Otherwise we wish you a pleasant stay here! Read the forum rules
  4. We have a few rules which you need to read and accept before posting anything here! Following these rules will keep the forum clean and your stay pleasant. Do not follow these rules can lead to permanent exclusion from this website: Read the forum rules.
    Are you a company? Read our company rules

Design of a timing belt actuator

Discussion in 'Motor actuators and drivers' started by jyrki.j.koivisto, May 29, 2011.

  1. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    Hi, my name is Jyrki Koivisto and I'm in a process of designing my first driving simulator and as I don't plan on taking the easy route I'm planning on rolling my own linear actuator as I want long travel (400mm) and a lot of output power. The design would be similar to Force Dynamics 301 http://youtu.be/d1iEtjs6S80 or Motion-sims simulator http://www.motion-sim.com/media/

    I calculated that if I wanted to move my simulator (weighting at presumed 300kg with the rider and all) with 3 actuators at the speed of 500mm/s upwards and with maximum of 2G acceleration I need 1471,5W of power on each axis (no losses taken into account...). That would be the top most upper limit of the simulator and the design would be built that in mind.

    With the maximum of 2G acceleration and maximum of 300kg mass the total thrust needed would be 8829 Newtons or 2943N per axis

    Each axis would be driven with ac-motor and a reduction gearing (something in line of 10:1-20:1, depending on the ac-motors nominal speed and output torque) and HTD timing belt and pulleys and solid pushrod.

    As I'm planning on using 8M-22 HTD pulleys and they have a pitch diameter of 56,02mm giving it a circumference of 176mm and as I want 400mm of total travel the pulley would only rotate less than 2,3 revolutions or with the upward speed of 500mm/s it would need to rotate at the speed of 170 1/min.

    With 3000rpm motor the reduction needed would therefore be 17,7:1 and with the maximum 2G acceleration and the used pulleys the needed moment of force would be 82,43Nm and with the said reduction motor would have to be able to give 4,6Nm I'd say that would be easily achievable.

    [img600px]http://www.netikka.net/jyrki.koivisto/simulator/pulley2.jpg[/img600px]

    Width of the belt is 20mm and the length is 1280mm, rod is 25mm in diameter and solid
    • Like Like x 1
  2. RaceRay

    RaceRay Administrator Staff Member SimAxe Beta Tester

    Joined:
    Nov 8, 2006
    Messages:
    4,656
    Occupation:
    Self-employed | Web and application development
    Location:
    Hamburg, Germany
    Balance:
    23,837Coins
    Ratings:
    +1,958 / 13 / -0
    My Motion Simulator:
    2DOF, DC motor, SimAxe, SimforceGT
    I will follow your very interesting looking project.

    Best regards
    René
  3. kubing

    kubing Member

    Joined:
    Sep 27, 2010
    Messages:
    259
    Occupation:
    teacher, Industrial electronic programmer
    Location:
    kelantan Malaysia
    Balance:
    350Coins
    Ratings:
    +0 / 0 / -0
    nice idea. so what is your plan for i/o interface?
  4. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    VFD's would be controlled by one of these things http://www.sparkfun.com/products/9713. That little big controller has an ethernet capability so it can be set to listen incoming game data directly and do all the calculations too + control of VFD's.

    I'm also planning on using gas springs to aid the movement and to reduce the needed motor power, but for that I need to know first the real mass of the movable part of the simulator, but I calculated that some 2000N per axis would be enough for direct drive from 5Nm motor for 200kg simulator at 2G's. That would reduce the cost of the simulator as reduction gears are expensive...
  5. Briut

    Briut New Member

    Joined:
    Dec 14, 2008
    Messages:
    7
    Location:
    Lithuania
    Balance:
    425Coins
    Ratings:
    +0 / 0 / -0
  6. kubing

    kubing Member

    Joined:
    Sep 27, 2010
    Messages:
    259
    Occupation:
    teacher, Industrial electronic programmer
    Location:
    kelantan Malaysia
    Balance:
    350Coins
    Ratings:
    +0 / 0 / -0
    if i not mistake there is someone use this controller in this forum with no luck..cant remember who.
  7. Frakk

    Frakk Active Member

    Joined:
    Apr 15, 2009
    Messages:
    1,144
    Balance:
    328Coins
    Ratings:
    +4 / 0 / -0
    You can't remember because noone used that controller here before. :)

    Good luck Jyrki!
  8. kubing

    kubing Member

    Joined:
    Sep 27, 2010
    Messages:
    259
    Occupation:
    teacher, Industrial electronic programmer
    Location:
    kelantan Malaysia
    Balance:
    350Coins
    Ratings:
    +0 / 0 / -0
    you're right frakk... wrong forum. my bad. good luck Jyrki :D
  9. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    Hi guys, I'm not so much worried about the controller as I've played with it a lot (previous version of UBW32 actually). What I'm worried is the mechanical part of the actuator. Taking a closer look at those 2 simulators I referred earlier one can see that the push rods are hollow to save weight (FD301 is a ballsrew design). Those rods that I'm planning on using weight 3kg/800mm, not a massive amount but still they add to the movable mass, I chose them as they are genuine hydraulic piston rods and have hard and smooth finish for quiet operation and longer bearing push live. They will loose some weight after I cut and mill them though.

    I'm still hoping to get away with the gas springs. One gas spring (gas strut or whatever they are called) manufacturer that I found over the internet claims a maximum speed of 35metres/minute (=580mm/s) and that would just about be enough for the 500mm/s designed top speed.
  10. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    Hello everyone!

    Now I've made some sort of a wooden mockup of the actuator on a pine plank (800x200x20)mm and started prototyping how to mount the push rod to the belt and I've come up with one sort of suggestion. I've made this clamp (50x48x20)mm with 5,2mm half holes and a smaller counterpart for it (48x20x20)mm to push the belt[img600px]http://www.netikka.net/jyrki.koivisto/simulator/DSC03879.JPG[/img600px]
    On top of them will be bolted an aluminium plate (74x48x20)mm that will hold the lower two pieces and the belt sandwiched in between them plus hold the push rod in place. It may not show up very well but there is 5mm room between the free end of the clamp and the adjacent belt portion. Also I'm thinkin of adding a second stationery linear guide between the pulleys to give extra support for the lower end of the rod.

    Top end will house the idler or turn pulley and the bearing pushing on top of it. Underneath the pulley there will be a pair of 6304 bearings.
    [img600px]http://www.netikka.net/jyrki.koivisto/simulator/DSC03881.JPG[/img600px]
  11. value1

    value1 Nerd SimAxe Beta Tester SimTools Developer Gold Contributor

    Joined:
    Jan 9, 2011
    Messages:
    2,184
    Location:
    Zug, Switzerland
    Balance:
    14,463Coins
    Ratings:
    +3,318 / 11 / -1
    My Motion Simulator:
    2DOF, DC motor, JRK, Joyrider
    Hi Jyrki
    I'm very interested in your progress and your experiences during your project. From what I've been looking at so far, building such a type of sim based on a timing belt construction is possibly an affordable way to achieve the movements. You might consider reducing the moving weight to the absolute minimum. Every kg counts!
    What type of motors do you intend to use?
  12. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    I'm thinking of using comman ac-motor, either directly coupled to the belt or via a reduction gear (10:1-20:1 or so...). I'd prefer direct coupling and gas spring aids if that would work... but that remains to be seen!

    It will not be cheap, I've spend some amount of money to this already... :no:

    [img600px]http://www.netikka.net/jyrki.koivisto/simulator/DSC03882.JPG[/img600px]
    [img600px]http://www.netikka.net/jyrki.koivisto/simulator/DSC03883.JPG[/img600px]

    Back plate of the finished actuator will not be made of pine wood but aluminium or steel, this pine here is just for prototyping of overall geometry... also the clamp will be simplified and modified further to allow for 2nd smaller auxiliary guide and easier manufacturing.

    Thank you every one for your support and chears!
  13. Frakk

    Frakk Active Member

    Joined:
    Apr 15, 2009
    Messages:
    1,144
    Balance:
    328Coins
    Ratings:
    +4 / 0 / -0
    Looking good so far Jyrki.

    Did you think it was going to be cheap? :) This is not a wipermotor junkyard project, this is serious power.

    Whatever you do, always make sure to BE SAFE! An actuator like this will kill you if you are not careful.
  14. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    I modified that pine proto as to actually accept an ac-motor on it and did some quick testings and one of the things I noted was that those hydraulic cylinder rods are just too heavy and I will use them as the second stationery guide rail and use hollow tubing for the real working push rod. This gives me a few nice added benefits one of them being that the backplate can be made out of light sheet metal as I intend to use the heavier solid rod as a structural support, second good outcome is that the movable mass reduces and thirdly I can hide the gas spring INSIDE the hollow tube rod!

    I also noted that it is actually quite leathal looking thing when moving and although I only revved the motor to a maximum of 5Hz (motor I installed was 1410rpm@50Hz) the belt really moved when unloaded. Otherwise it was a weak thing, it could only just about lift itself (weighing at 15kg) it really needs gearing... Part of the blame can be set for the used VFD which was only rated for 370w, but still...

    I'm not sure if the simulator needs to move as fast as I originally planned (500mm/s), what speeds should I aim for? I have no experience from simulators so I don't know where to set the goal... some advice please!

    EDIT:

    I calculated the output power to have been around 60W so it's not a suprise that it felt so weak. I have noticed that ac induction motors don't work so well at such a low rpm. I reckon that a reduction of 1 to 3 would give better results as the motor would then start to work better.
  15. value1

    value1 Nerd SimAxe Beta Tester SimTools Developer Gold Contributor

    Joined:
    Jan 9, 2011
    Messages:
    2,184
    Location:
    Zug, Switzerland
    Balance:
    14,463Coins
    Ratings:
    +3,318 / 11 / -1
    My Motion Simulator:
    2DOF, DC motor, JRK, Joyrider
    I have not built my sim yet – so the following is more of a theoretical view.

    I have measured accelerations of ±0.6g driving my car, not a Rally or F1 car though. The later cars of course have much higher g-values.

    By tilting the seat 35° you can achieve 0.6g gravitational acceleration in tilting direction (backwards, forwards or sideways, i.e. acceleration, deceleration, centrifugal acceleration) and still keep 0.8g downward acceleration (weight). With 500mm/sec you can simulate 0.1g. A portion of this you can add to the g's above – for a fraction of a second when you have 500 mm actuators.
    I'm not sure anymore if this is worth the – substantially higher – effort of a 3DOF.
  16. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    The 2G acceleration I mentioned is only used to calculate how fast I could make the thing move to a new position and how much force would it take to make the movement, there is no way that a stationery simulator like this could ever sustain such accelerations while playing. I've let myself think that the sensation of simulated force comes from the fact that the brain is fooled to think that a tilting of the seat is a real acceleration and not just gravity pulling which it really is.

    I'm thinking of dropping the belt drive idea and change the thing into a ball screw design... That would eliminate the need of a reduction gear. I searched over the internet for a planetary reduction gear but I think they would be too expensive for me. One gear that I really liked was this thing http://www.planetroll.com/en/antriebstechnik/planetdrive_pd_planetengetriebe/index.php

    I originally planned to make this thing with ball screws and it just so happens that I have few meters of 32mm diameter 5mm pitch ball screws...
    • Useful Useful x 1
  17. adgun

    adgun Active Member

    Joined:
    Jan 28, 2008
    Messages:
    493
    Occupation:
    mechanic
    Location:
    Netherlands
    Balance:
    5,540Coins
    Ratings:
    +131 / 3 / -0
    Hi
    In my opinion ballscrews are the way to go
    With an 5 mm pitch you get a 250mm/sec topspeed (rpm is limited @ 3000)
    Fore cheap 10 mm lead ballscrews see www.slidesandballscrews.com
    Fore technical info about ballscrews see application engineering @ www.roton.com
    I,am sorry if y tell nothing new to you
    best regards AD
  18. sn4il

    sn4il New Member

    Joined:
    Oct 26, 2009
    Messages:
    55
    Balance:
    2Coins
    Ratings:
    +1 / 0 / -0
    Hi,
    There was a video somewhere on this forum (if I remember correctly) with simulator operated by actuators based on two (2) linear slides (ball screws). Two ball screws were joined in the middle and operated by one motor.
    Something like this:
    [​IMG]
    Sorry for the quality, I'm at work and it was super fast drawing :D
    Maybe someone will remember what I'm talking about.
    sn4il
  19. adgun

    adgun Active Member

    Joined:
    Jan 28, 2008
    Messages:
    493
    Occupation:
    mechanic
    Location:
    Netherlands
    Balance:
    5,540Coins
    Ratings:
    +131 / 3 / -0
    It,s a Quanser hexapod

    Attached Files:

  20. jyrki.j.koivisto

    jyrki.j.koivisto New Member

    Joined:
    Aug 30, 2010
    Messages:
    85
    Location:
    Finland
    Balance:
    314Coins
    Ratings:
    +3 / 0 / -0
    That Quanser hexapod would be extremely costly for a hobbyist to do with ball screws (left and right hand screw + nuts for them x 3), BUT with belts it could be done quite cheaply if one could source cheap enough planetary gear reductions! Think about it! On a belt there are always two opposite directions that the belt travels! How much torque does a general wiper motor produce? Maybe someone could source strong dc motors? If so there would not be need for a reducer gearing.

    EDIT:

    I think I like the quanser thing a lot, it would be compact in size but still offer the same kind of movement as I originally wanted (travel would be less but still quite a bit of movement with the same belts)

    ((((1280mm-176mm)/2)-100mm)/2)=226mm of total travel with the same 1280mm long belt, or +-113mm. Next standard belt size woul be 1440mm giving 266mm or +-133mm of travel. After that there's 1600mm long belt giving 306mm or +-153mm :? Actual travel distance will of course depend on the angle the legs make, I'll have to do some calculations on that and what relationship it has with the power requirements.

    Thanks you guys! I'll go with the Quanser thing now :yippiee:

    EDIT2:

    I made a mistake calculating the belt travel, correct way to calculate it would be: ((belt length-pulley circumference-pulley diameter)/4)-belt clamp width. 1600mm belt would then give 291mm of free travel on the belt and 1280mm belt would give 211mm of travel