1. 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!
  2. 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
  3. 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

6 DOF from scratch

Discussion in 'DIY Motion Simulator Projects' started by Pierre Lalancette, Dec 18, 2016.

  1. OZHEAT

    OZHEAT Active Member

    Joined:
    Oct 26, 2015
    Messages:
    212
    Location:
    Australia
    Balance:
    2,389Coins
    Ratings:
    +102 / 6 / -0
    My Motion Simulator:
    3DOF, DC motor
    Also as you want to direct drive the ballscrew, use a gearbox efficiency value of 1.0.
    Using .75 will just increase the required power by 25%.

    Those ampflow motors look like relabelled chinese scooter motors, you can buy them for a lot less on ebay.
  2. Pierre Lalancette

    Pierre Lalancette Sir Lalancelot Gold Contributor

    Joined:
    Dec 11, 2016
    Messages:
    817
    Occupation:
    3D teacher
    Location:
    Quebec, Canada
    Balance:
    6,648Coins
    Ratings:
    +791 / 6 / -0
    My Motion Simulator:
    DC motor, Arduino, 6DOF
    Maybe I'll be able to lift up the house. Of course not! I always put extreme scenario, so I know that is I use 6 of them, I won't have to worry.
    I checked your video and I am proud to say that I could follow along. When he said that sin of 0 offer no acceleration and sin of 90 was free falling, it made a lot of sense.

    Agree.

    What really throw me off on that chart is that the power goes up and down. I guess it is because as there is to much torque, the motor simply gives up and the power goes down.

    1Nm torque at 4500rpm, 0.375m/s, sound good to me. Really close to the first target.

    1/5 of 710 = 142 oz-in. Conversion to N-m = 1.00274036033286. I think we are right on the money.

    That is just in case I would eat 2 hot dogs ketchup mustard and a maple sirop pie just before getting up on the rig.

    Make sense. I will keep that load and put a more conventional 75% inclination.

    I was a bit mixed up by the efficiency of the motor that is 76%. What does that mean?

    Do you have an link that I could compare? I could not tell a good DC motor from a refurbish washing machine.

    motor2.jpg

    So now, the motor requires 250 Watts.
    I = 250 / 24
    I = 10.41 Amps, Multiplied by 6 gives 62.5 Amps. I think we are safe.
    But then again, looking at the chart, the power for 1 Nm torque is more 460 Watts.
    Mmmm, that power curve really puzzle me.
    Required Gearbox output torque: 0.531 and 0.797. Still in the safe zone.

    I think we are good. Did I forgot something?
    • Informative Informative x 1
    Last edited: Apr 5, 2017
  3. speedy

    speedy Well-Known Member

    Joined:
    Feb 1, 2012
    Messages:
    1,194
    Location:
    Alexandria , Egypt
    Balance:
    7,723Coins
    Ratings:
    +1,274 / 9 / -0
    My Motion Simulator:
    3DOF, AC motor, Arduino, Motion platform
    Impressive improvement in calculations ... :cheers

    So, NOW .... you are at the exact point where a drawing with dimensions n numbers for the expected simulator is needed .

    If you can work with 3D SketchUp I guess @SilentChill has provided his .skp model for download .

    SketchUp is recommended from the Engineering ToolBox and can be connected to SimTools for model simulation .

    http://www.engineeringtoolbox.com/3d-sketchup-engineeringtoolbox-d_1725.html
    Last edited: Apr 5, 2017
  4. judges

    judges New Member

    Joined:
    Feb 17, 2017
    Messages:
    16
    Location:
    Germany
    Balance:
    309Coins
    Ratings:
    +21 / 0 / -0
    You're confusing mechanical and electrical power. At stall torque mechanical power equals zero, because of zero rpm. Looking at the performance chart (which shows mech power), you'll notice that the motor will then draw like 125 amps (at 24V). So electrical power will be 3000W. Nothing converted into mech power, but everything converted into thermal power. A big heater (and smoke generator after a few secs)... ;)

    See above. 76% of the electrical power gets converted into mechanical power. Rest is frictional and thermal losses. Output of the calculator is the required mech power. But you can enter motor efficiency into the gear efficiency input field. Then calculated motor power will be electrical power. Just keep in mind that motor efficiency isn't a constant value. So e.g. at 2Nm torque, efficiency drops to 56%.
    • Informative Informative x 2
    • Agree Agree x 1
  5. OZHEAT

    OZHEAT Active Member

    Joined:
    Oct 26, 2015
    Messages:
    212
    Location:
    Australia
    Balance:
    2,389Coins
    Ratings:
    +102 / 6 / -0
    My Motion Simulator:
    3DOF, DC motor
    @Pierre Lalancette

    @judges pretty much covers power in/out/efficiency.

    Ampflow's prices are reasonable especially if you get them delivered quicker than eaby speed.
    Search for unite motor or scooter motor on ebay.

    I did a calc based on the 150w unite M27-150-P motor.
    Untid.jpg
    I would suggest to forget about spinning the ballscrew anywhere around >3000 for safety sake.
  6. SilentChill

    SilentChill Problem Maker

    Joined:
    Jul 19, 2014
    Messages:
    2,580
    Occupation:
    Railway Maintenance
    Location:
    Morecambe, Lancashire, England
    Balance:
    19,872Coins
    Ratings:
    +3,387 / 33 / -0
    My Motion Simulator:
    DC motor, Arduino, Motion platform, 6DOF
    Ignore that statement, myself and @wannabeaflyer2 run motors at 4000rpm and @SeatTime is over 4500rpm if I remember correctly
    • Informative Informative x 1
  7. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,599
    Location:
    Newcastle Australia
    Balance:
    27,834Coins
    Ratings:
    +2,865 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    My motors run over 5000 Rpm (see image) with no load - but you will not get anywhere near that with a load/PID etc. Never had an issue with resonance/oscillations of the ballscrew shaft.

    [​IMG]
    • Informative Informative x 1
  8. OZHEAT

    OZHEAT Active Member

    Joined:
    Oct 26, 2015
    Messages:
    212
    Location:
    Australia
    Balance:
    2,389Coins
    Ratings:
    +102 / 6 / -0
    My Motion Simulator:
    3DOF, DC motor
    Hmmmm @SeatTime
    Maybe you should of used kevlar instead of carbon fibre for the actuator. :)

    the hobby machinist in me after 25 odd years of lathe and mill work would say it would be very foolish to spin a half inch shaft @ 5000rpm, it ends up wanting to be spaghetti.

    Your in the industry, try ask a machinist or engineer about spinning that ballscrew at that speed.

    What have you got against using a ballscrew with a 10mm pitch?
    It halves the needed ballscrew rpm.

    I would look up what tensile strength a ballscrew has when l get home.
  9. Pierre Lalancette

    Pierre Lalancette Sir Lalancelot Gold Contributor

    Joined:
    Dec 11, 2016
    Messages:
    817
    Occupation:
    3D teacher
    Location:
    Quebec, Canada
    Balance:
    6,648Coins
    Ratings:
    +791 / 6 / -0
    My Motion Simulator:
    DC motor, Arduino, 6DOF
    What did we learn?
    We should throw away all safety advises and put everything to the max! ;)
    That Motors transform electricity into mechanical force, heat, fire, explosion and assurance claims.

    Okay, next question:
    What do you all think about PVC pipe to do actuator main shaft and piston? I don't have access to carbon fiber like the Elites of this forum ;) . Is it a dump idea? What do you all use?

    And last, until I can afford any more equipment, I started to do 3D design of my simulator. Since I am learning Houdini, this was a good project for it.
    motor3.jpg
    • Like Like x 1
  10. adgun

    adgun Active Member

    Joined:
    Jan 28, 2008
    Messages:
    424
    Occupation:
    mechanic
    Location:
    Netherlands
    Balance:
    5,017Coins
    Ratings:
    +102 / 3 / -0
    Speed - Ballscrews
    Ball velocity in a Ballscrew should not exceed 3,000 rpm x in. (rotational speed (rpm) times the nominal diameter (in.)). For example, a 3/4 x .200 size Ballscrew should be limited to 4,000 rpm (3,000/.750 = 4,000 rpm). For applications requiring speeds beyond 3,000 rpm x in., use a larger lead, a larger diameter, or contact Roton Application Engineering. Freewheeling Ballscrews work best at 350 rpm and less. This limit is imposed by the dynamic action of the stop pins contacting the ball retainer. For operation of Freewheeling Ballscrews beyond 350 rpm, contact Roton Application Engineering. In addition to the above guidelines, each screw drive system should be evaluated for safe rotational speed so that natural frequency vibrations are avoided (see Critical Speed section).
    Taken from www.roton.com application engineering
  11. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,599
    Location:
    Newcastle Australia
    Balance:
    27,834Coins
    Ratings:
    +2,865 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    Nope - PVC won't do. Steel pipe or something similar - A bit of history, like when I built my first actuator a few years ago. Pipe was close in size, but had to be hand finished to properly run in the linear bearing.


    Actuator stage3.jpg
    • Informative Informative x 1
  12. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,599
    Location:
    Newcastle Australia
    Balance:
    27,834Coins
    Ratings:
    +2,865 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    You did not read all of my post...How many actuators have you built....Me out.
    • Agree Agree x 1
    Last edited: Apr 6, 2017
  13. OZHEAT

    OZHEAT Active Member

    Joined:
    Oct 26, 2015
    Messages:
    212
    Location:
    Australia
    Balance:
    2,389Coins
    Ratings:
    +102 / 6 / -0
    My Motion Simulator:
    3DOF, DC motor
    Untitled.jpg
    source http://www.thomsonlinear.com/downloads/screws/Leadscrews_Ballscrews_Splines_cten.pdf

    A 16x5 ballscrew has a root diameter of approx 1/2"


    Roton should correct their app notes.
    "Ball velocity in a Ballscrew should not exceed 3,000 rpm x in. (rotational speed (rpm) times the nominal diameter (in.)). For example, a 3/4 x .200 size Ballscrew should be limited to 4,000 rpm (3,000/.750 = 4,000 rpm)."

    3,000 rpm x in. (rotational speed (rpm) times the nominal diameter (in.) does not equate to their example 3,000/.750 = 4,000 rpm
    Nor do they take in to consideration length of ballscrew.
    • Funny Funny x 1
  14. adgun

    adgun Active Member

    Joined:
    Jan 28, 2008
    Messages:
    424
    Occupation:
    mechanic
    Location:
    Netherlands
    Balance:
    5,017Coins
    Ratings:
    +102 / 3 / -0
    Length from ballscrew is whritten in critical speed section.
  15. Pierre Lalancette

    Pierre Lalancette Sir Lalancelot Gold Contributor

    Joined:
    Dec 11, 2016
    Messages:
    817
    Occupation:
    3D teacher
    Location:
    Quebec, Canada
    Balance:
    6,648Coins
    Ratings:
    +791 / 6 / -0
    My Motion Simulator:
    DC motor, Arduino, 6DOF
    Damn, PVC won't do...
    With all the ballscrew arguments, maybe Kevlar is not a bad idea after all. :)
    Will have to totally rethink my actuator.
  16. SilentChill

    SilentChill Problem Maker

    Joined:
    Jul 19, 2014
    Messages:
    2,580
    Occupation:
    Railway Maintenance
    Location:
    Morecambe, Lancashire, England
    Balance:
    19,872Coins
    Ratings:
    +3,387 / 33 / -0
    My Motion Simulator:
    DC motor, Arduino, Motion platform, 6DOF
    Just buy 1 tube of carbon fiber it will only be around $20-$30
  17. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,599
    Location:
    Newcastle Australia
    Balance:
    27,834Coins
    Ratings:
    +2,865 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    I had walked away from this conversation as nowadays I cannot be bothered arguing when I am sitting in a simulator that works quite fine :rolleyes:. But to think that there is something really dangerous going on, or I and others just blindly connected a motor without any real thought or testing could not be further from the truth - just read my thread. The load of theory put forward above from Google is not incorrect, but unfortunate is very much a generalization which needs to be tempered by reality and actual testing. Reality like follows:
    1. My motors may have read at 5376 at no-load but once connected to the actuator and rig they do not get close to that figure due to friction/inertia/load etc, so there is not issue with ballscrew resonance, or things throwing themselves apart:).
    2. Why did I pick a high speed motor and 5mm pitch ballscrew? I needed the torque which I got from the 5mm pitch (remember the ballscrew is just taking the place of a gearbox/reducer), but then I also needed good speed to ensure accurate tracking - so a high speed motor was chosen.
    3. I did test 4 different motors and the one I chose was the best performing that I could easily source at a reasonable price. So yes cost and availability was a factor - well it is DIY and my money :).
    4. Noise and weight is a factor to me and had an impact on actuator speed due to me changing my linear bearings to poly bushes - see my thread. Now this obviously slowed down the actuators due to the extra friction of bushes over a linear bearing, but I was prepared to take the hit for the reduction in noise and weight (another reason for high speed motors to offset this).

    I could go on and on but the main thing from all this is - @Pierre Lalancette don't be scared of what we are doing here or think you need to wrap your actuators in Kevlar for safety - please everyone this reminds me of my Doctor talking about 'Doctor Google', everyone reads what is on Google and immediately thinks they are an expert - the information on Google and the internet as a whole is a great tool, but needs to be tempered by reality and often real experience and actual training.

    Thanks for reading my rant and in the future when I do not answer a question or participate in a 'discussion' you will know why :) .
    • Agree Agree x 3
    • Like Like x 1
    • Winner Winner x 1
    Last edited: Apr 6, 2017
  18. speedy

    speedy Well-Known Member

    Joined:
    Feb 1, 2012
    Messages:
    1,194
    Location:
    Alexandria , Egypt
    Balance:
    7,723Coins
    Ratings:
    +1,274 / 9 / -0
    My Motion Simulator:
    3DOF, AC motor, Arduino, Motion platform

    Agree and my motors are 1/4 HP 3-phase direct coupling with 1605 ballscrew running 3000 RPM with no issues at all ... willing even to go faster with VFds :D .
    • Like Like x 2
    Last edited: Apr 6, 2017
  19. Pierre Lalancette

    Pierre Lalancette Sir Lalancelot Gold Contributor

    Joined:
    Dec 11, 2016
    Messages:
    817
    Occupation:
    3D teacher
    Location:
    Quebec, Canada
    Balance:
    6,648Coins
    Ratings:
    +791 / 6 / -0
    My Motion Simulator:
    DC motor, Arduino, 6DOF
    Wow, Did we touch a sensitive cord here!
    Don't worry @SeatTime , I have read all your post with great interest and I know the time you put in your build. I take your advise quite seriously. Even your rants are informative.
    Still, I understand the worries people may have. It is up to me to take the risk or not. It may be out of norms, but due to the records and testing here, I doubt it is that dangerous.

    Hey @speedy, I really like the simplicity of your system. It is not something I would have though. Going strong at 3000rmp.

    Of course, buying carbon tube... Simple solution. Thanks @SilentChill .
    • Friendly Friendly x 1
  20. SilentChill

    SilentChill Problem Maker

    Joined:
    Jul 19, 2014
    Messages:
    2,580
    Occupation:
    Railway Maintenance
    Location:
    Morecambe, Lancashire, England
    Balance:
    19,872Coins
    Ratings:
    +3,387 / 33 / -0
    My Motion Simulator:
    DC motor, Arduino, Motion platform, 6DOF
    If I listened to people like OZHEAT and nay sayers I would of not built my first 12v 6DOF, go with what you think, if it doesnt work F'it and try again. Its all good fun