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

DIY 3 DOF pegleg first motion rig

Discussion in 'DIY Motion Simulator Projects' started by Steve B, Jan 20, 2022.

  1. noorbeast

    noorbeast VR Tassie Devil Staff Member Moderator Race Director

    Joined:
    Jul 13, 2014
    Messages:
    21,148
    Occupation:
    Innovative tech specialist for NGOs
    Location:
    St Helens, Tasmania, Australia
    Balance:
    148,584Coins
    Ratings:
    +10,904 / 54 / -2
    My Motion Simulator:
    3DOF, DC motor, JRK
    In SMC3 IBT2s should be run in Mode 2.
  2. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    I started out running them in mode 2, was having heaps of issues which went away when I changed it to mode 1 wired as shown previously.
    Whilst I cannot prove mode 2 was the cause of multiple h-bridge failures I've run the motion rig for over 7 hours on mode 1 without a fault.
    The actuators start to move at a lower pwm duty cycle and without servo whine on mode 1, both an improvement on my rig.
  3. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    Continuing to dial in the rig. This really helped as it's the same actuator layout, car and sim i use(simtools setup starts at 40m):

    The closer i get to how i expect it to feel, the more i notice whats not right with it.

    Having motion certainly hasn't helped my driving, i've lost over 20% of my irating and will likely loose my A license next week at the end of the season.
    I find myself doing OK then noticing something not quite right with the motion and concentrating on that more than racing. (The tracks on the schedule the last few weeks haven't helped either)

    Whilst i've not had an outright H-bridge failure in the last couple of weeks, one did change in behavour significantly. I was noticing after big heave events(going over a sausage curb) the rear left was dropping much more than the rear right then bouncing back to the correct level.
    Testing withing the SMC3 GUI showed the same effect, a massive overshoot of the target position before returning to the correct position, PID tuning made no difference.
    Swapping out this H-bridge solved the issue so it looks like another failure, although not completly this time.

    I've ordered some 100A H-bridges, should arrive in the next couple of weeks to hopefully solve the issue for good.
    Have a couple of 150W motors leftover from a previous project, i may give them a go on the rig when i get chance. The 500w motors in use have so much inertia, they are probably the source of much of my issues.
  4. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    Whilst the actuators haven't actually failed at any point in the last few weeks, there have been opportunities for improvement.
    Continuing to work through the list of identified improvements, the top plates of both left actuators have bent at some point in testing.
    I'm not exactly sure when this happened or why, so have machined some stronger ones.
    IMG_20220215_143043.jpg

    IMG_20220219_204133.jpg

    IMG_20220219_211223.jpg

    I replaced one of the ballscrews that was noisy, this one was noisy from the start. When i first built the actuators i didn't have the correct size tube to flip the ballnuts on the ballscrews, i think a ball bearing or 2 got somewhere they shouldn't on this one.

    I replaced the motor on the rear right actuator with one of the 150W (zy6812) motors i have, these are 12V 0.57Nm rated.
    At 12V it struggles a bit compared to the 500W motors used on the rest of the rig but at 16V it matches or beats the larger motors for speed and motion seems more responsive.
    Unfortunatly it gets a bit warm without a cooling fan directed at it, more testing is required on this one and maybe with some assistor springs to take some of the load.
    • Like Like x 1
  5. zhai1987

    zhai1987 Member

    Joined:
    Sep 7, 2021
    Messages:
    96
    Balance:
    320Coins
    Ratings:
    +11 / 0 / -0
    My Motion Simulator:
    2DOF, 3DOF, AC motor
    Hi. Brother, can you share your wiring diagram.
  6. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,573
    Occupation:
    Retired
    Location:
    Brisbane Australia
    Balance:
    28,370Coins
    Ratings:
    +2,844 / 39 / -0
    My Motion Simulator:
    AC motor, Motion platform
    Been down this rabbit hole...maybe not something you want to hear, but once you start to throw around the full weight of Rig+ Accessories + your body - its time to look at AC servo motors if you want accurate/reliable (competitive) motion.
  7. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    I don't doubt that.
    However as this is my first exploration into motion, spending thousands on AC servo's without testing the waters first to see if/what type of motion suits me seems equally daft.
    The intention from the start was to build cheap and simple motion to see if it's worth it for me before building/buying something better in a year or 2 by which time I should have a better idea what type of simulator I want.

    Having used motion for around 25h now. Probably the last 10h of that with the motion somewhere near dialled in, I'm still undecided as to if the benefits outweigh the drawbacks of motion.
    The biggest drawback being monitor placement, without motion I had my centre screen between the wheelbase and wheel as a low as possible. With motion I've had to raise my monitors quite a bit, this leaves me with a choice of not being able to see the instrument cluster or barely being able to see the road over the dashboard without straining my neck. The other option is to move the monitors way forward behind the wheelbase which would reduce my FOV quite a bit.
    I have tried VR and didn't get on with it, was like trying to drive whilst looking through a murky periscope.
  8. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    Wiring diagram is slightly modified version of SMC3, can be seen a few posts back. Only thing not shown there is the power supply which is 2 HP server power supplies in series, a diode and some capacitors with a simple precharge circuit, I'll post a drawing of that in a few days.
    Last edited: Mar 8, 2022
  9. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,573
    Occupation:
    Retired
    Location:
    Brisbane Australia
    Balance:
    28,370Coins
    Ratings:
    +2,844 / 39 / -0
    My Motion Simulator:
    AC motor, Motion platform
    Often less motion is more for car sims - It's more about extra feedback on what the car is doing to help you drive better, not trying to actually simulate the real forces in a car with big motion. The best cheap motion sim to build is a balanced 2DOF rig with Traction Loss, as your not loading up the small DC motors too much keeping the response fast and accurate. Once you start to throw allot of weight around at speed as happens in a fully supported/driven rig, the inertia can easily slow the response of the low power motors/drivers etc and give poor results. Maybe see if you can try a well setup AC powered rig to see what is actually possible.
    If you want to feel forces, IMO G systems are the way to go (IE. active harnesses and pressure paddles).
    • Agree Agree x 1
    Last edited: Mar 10, 2022
  10. zhai1987

    zhai1987 Member

    Joined:
    Sep 7, 2021
    Messages:
    96
    Balance:
    320Coins
    Ratings:
    +11 / 0 / -0
    My Motion Simulator:
    2DOF, 3DOF, AC motor
    Thank you. I also want to make a 3DOF with a DC motor
  11. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    I think when i a said the small 150W motor was struggling you got the impression it could barely lift the rig? By struggling i mean it was 60-70C after 30 mins racing an SR8 around Nordschleife.

    Few laps around Knockhill, this is with 3 500W motors and a 150W on the rear right.
  12. zhai1987

    zhai1987 Member

    Joined:
    Sep 7, 2021
    Messages:
    96
    Balance:
    320Coins
    Ratings:
    +11 / 0 / -0
    My Motion Simulator:
    2DOF, 3DOF, AC motor
    Steve B: what's your discord,
  13. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    The other 150W motors arrived yesterday, brackets made to fit them in place of the 500w ones.
    Completely reversible if it doesn't work out, just the PID tuning to change.
    IMG_20220310_114127.jpg
  14. zhai1987

    zhai1987 Member

    Joined:
    Sep 7, 2021
    Messages:
    96
    Balance:
    320Coins
    Ratings:
    +11 / 0 / -0
    My Motion Simulator:
    2DOF, 3DOF, AC motor
    150W ` 4, how many can you be responsible for... Is it enough... I look forward to your effect!
  15. fzxj520

    fzxj520 Active Member

    Joined:
    May 14, 2015
    Messages:
    176
    Location:
    SimTools is a generic Motion Simulator
    Balance:
    1,341Coins
    Ratings:
    +90 / 1 / -0
    My Motion Simulator:
    2DOF
    Reasonable use of springs can reduce the load of the motor and improve the response speed
  16. Billel

    Billel Member

    Joined:
    Mar 29, 2022
    Messages:
    37
    Balance:
    195Coins
    Ratings:
    +2 / 0 / -0
    My Motion Simulator:
    Arduino, Motion platform, 6DOF
    how much torque do those motors has ? both the 150W and the 500W, i'm gathering infos to start my own 6DOF stewart platform build with linear actuators, based on IBT2's and Arduinos, i can't deside how much power and torque do i need for the choice of motors.
  17. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    From memory the 150W motors are 0.56Nm and the 500W ones are 1.9Nm.
    The thing to remember is that these torque values are just the rated values at speed, 2500RPM in the case of these motors.
    The actual torque will be proportional to the amount of current (Amps) going through the motor.
    So from stationary assuming you can supply and switch enough power motors usually draw between 5 and 7 times their rated current.
    Thats around 45-63A for the 150W motor and 132-186A for the 500W motor, however as soon as the motor starts to turn the current will start to fall and reach it's rated running current when it reaches the rated RPM.
    If the motor tries to reverse direction without decelerating to a stop then reversing first then the current will be considerably higher still.

    I've no idea how the load will be spread on a stewart platform but in the case of my rig i'm guessing the rear actuators are lifting around 110KG between them and the fronts around 50KG.
    My front actuators are using the 150W motors running at 16V (Rated @12V). The front of the rig moves really fast(Probably 200mm/s), i think it's jumped off the ground on more than 1 occasion.

    The rear is still running 500W motors, these are rated at 24V however i can only use 19.5V without blowing up H-Bridges (Big H-Bridges claimed to be 100A/260A peak). With IBT-2's i could only use 16.5V before the H-Bridges would fail. Due to running at lower than rated voltage the rear actuators are slower that i would like, probably only 90-100mm/s.
    I do have another of these H-Bridges on order and plan to use 1 for each rear motor (Both channels for 1 motor).

    This is the big H-Bridge i'm using, it's 2 channel so does both the rear actuators. This is the second one, the first blew up within 2 seconds of power on, i assume due to too much current.
    IMG_20220314_223023.jpg

    150W motors fitted to the front.

    IMG_20220409_204605.jpg

    I did try 150W motors on the rear, they just didn't have enough torque and ran hot. They ended up being slower than the 500W motors so i swapped back.
    I'm still planning to add spring assist to the rear, but as everything is working and we're half way through the iRacing season so i'm not messing for now.
    So far the rig has worked for around 70 hours without any issues(except caused by me changing things).
    • Like Like x 1
  18. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    With iRacing 2022 season 2 finished and season 3 about to kick off i thought it was time to update my build thread.
    The motion rig has managed 140 hours total power on time (time on track will be a little less), no failures to report.
    I did have 1 day when simtools failed to connect to iRacing and needed a reboot but i think this was caused by a windows update as one of my monitors changed resolution and my sound settings were also messed up that day.
    I did remove and replace the ballscrew in the front right actuator as i had a noise that just wouldn't go away, it sounded like aluminium creaking but extra bracing to the rig, checking everything was tight and so on all failed to eliminate it. My last resort was to tear down the actuator and replace the ballscrew, in the process of doint this i found the noise(not the ballscrew but changed that anyway whilst everything was apart).
    The rubber spyder between the 2 halves of the motor coupling was making the noise as it flexed against the coupling, this one was of a red material whilst the others i have are orange and silent. The fix for now was to cover it with grease which seems to have solved this issue but eventually i'll try and get another coupling with orange spyder.
    The photo below shows the floor pads i made during the season to spread the load and allow the rig to slide slightly within the cups to prevent binding, to the right is an actautor leg with a bearing bracket ready for adding spring assist. Hopefully testing that later today.
    IMG_20220610_221317.jpg

    So i guess the big questions are:
    Is it reliable? Already answered above.
    How has it effected my driving?
    Was the motion worth building?

    End22S2.png
    Above is my iRating from the begining of time. The yellow arrow indicates when i added motion, the red shows the start of 22S2.
    Yes it looks bad for motion but that's not quite the whole story, i started racing Radicals this season as well as my normal 2 X GT3 series and ajust don't have enough seat time in the Radical to be competitive. The low participation in Radicals also hurt my iRating with generally ony 1 or 2 splits making a top half finish challenging.
    GT3 also had the biggest shakeup since i joined iRacing, tyre model was updated and completly updated aero.
    The changes made the GT3 cars much easier to drive, now unless you put a wheel on the grass under braking spins are a thing of the past (although Porsche, Lambo and Audi drivers still had issues).
    The downside of all this grip was a massive loss in terms of force feedback which made extracting decent laptimes really hard to do consistantly. I struggled with this all season.

    By week 10 i was still stuck for exactly why i was struggling to match my old times with a much grippier car, not helped by having so many weeks racing on tracks i'd never driven before.
    Interlagos was a bit of a turning point, a track i was previously fast on and now 1.5 - 2s off my pace, i tried running a medium downforce setup and the force feedback came to life telling me what the front end was doing for the first time all season. (Had other issues at interlagos including 3 races with a 40s penalty for speeding in the pits and another ruined by a driver blocking and taking both of us out).

    Week 11 at Virginia international raceway, another track i'd never raced. Whist medium downforce wasn't ideal here it was still worth it for the feel and ended up with some decent results.

    Week 12 the last week of official racing was at Donnington park, the only track on Iracing i've driven IRL.
    Running Medium downforce here gave me decent laptime but i was a sitting duck on the first half of every straight due to poor exit speed so it worked for qualifieng but not in the race.
    Back to square 1, out of desperation i disabled iRFFB(a program that sits between iRacing and FFB wheels to add fidelity and effects to FFB) and turned up the FFB to 12Nm (Usually run at 10Nm). This gave me a resonably reel as to what the front was doing and achieved some decent finishes.
    Not liking the harsh feel of Iracings native FFB so this is still a work in progress.

    So back to the motion, well for the first 50 or so hours it's distracting. After that i only conciously notice the bumps in the track and hitting curbs.
    Surge gives a nice kick in the back on gear changes which does help but could be done with a good bass shaker almost as well.
    Pitch, at least the amount i'm runnung is subtle and only really noticable at Knockhill and Bathurst.
    Roll kinda helps and works well in conjunction with heave when hitting big curbs.
    The only time i noticed roll from track camber was the banking at Daytona, that took me by surprise the first lap or 2.

    The big thing with motion for me was immersion and getting up to speed quickly at the start of an evenings racing.
    Previously i've had times during races when when my brain has gone out of sink with the simulation and caused issues, a bit like taking over someone elses paused game mid corner and you don't know how much steering lock was applied or if the car even has traction.
    With motion i seem to be connected with the sim on a more subconscious level and have not had these issues at all.
    The other major benifit has been the time taken to get up to speed in an evening, without motion it could take me 30 to 40 mins to get within 0.5s of my prevoius evenings times.
    With motion i'm usually within 0.5s by my third lap.

    I have to say it was well worth the effort to build the motion rig and whilst i dould definatly do it again i don't think i could ever justify the costs of the high end motion rigs out there.

    Aynway i've gone on for far too long, will hopefully have an update on the spring assist (not actually springs) later in the week along with hopefull a decent video of the rig in action. (Video unlikely this week due to GT3 being Hockenheim at night. Likely many multi car crashes and very few clean laps)

    Steve
    • Winner Winner x 1
  19. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    Finally fitted my spring assist to the actuators, it's taken a couple of months to figure out exactly how and what to use and a bit more time to make the brackets to install it all.

    I initially was going to use springs, however springs are not easy to find in the required rate and suffer with the issue of the ammount of assistance they provide changes with how far they are compressed.
    Second idea was bungee cords but there was no obvious way to mount them that wouldn't cause other issues.

    So i settled on pneumatic cylinders, 2 at the rear and 1 at the front. Due to not wanting them to lift the rig off the ground when i dissmount the rig, they had to be fitted in parallel with the actuators pushing down against the same feet as the motors.

    Below is the beam between the front actuator legs with a single cylinder in the center.
    IMG_20220615_101613.jpg
    IMG_20220615_101620.jpg
    IMG_20220615_101631.jpg

    Front cylinder assembly installed.
    IMG_20220615_230229.jpg
    IMG_20220615_230242.jpg

    The rear uses 2 cylinders which push against the underside of the seat frame.
    IMG_20220614_150617.jpg

    IMG_20220614_150631.jpg

    In the background of the above image is an empty helium tank i am using as an air receiver.
    Due to the volume of the receiver compared to the cylinders the pressure should only change by 3 PSI from fully compressed to fully extended so the assistance provided will only vary by 1.5Kg (from memory) over the full range.
    I tested the rig for a couple of hours last night with just the rear cylinders pressurised.
    I now have the front cylinder working as well and have refitted the small 150W motors to the rear actuators as they now have much less work to do.
    A quick 10 min test and all is working fine with the small motors all round, much more motion on the rear actuators than i had before due to the small motors being considerably faster.

    Just have a massive mess to clear up and a lot of cable managment to do (the cable management can wait).
    Steve
    • Creative Creative x 1
  20. Steve B

    Steve B Member

    Joined:
    May 8, 2020
    Messages:
    40
    Balance:
    248Coins
    Ratings:
    +32 / 0 / -0
    My Motion Simulator:
    Arduino
    Quick test of the rig with the small 150W motors all round and pneumatic assistance.
    At the end of the video i was never going to make the chicane so demonstrate what happens when you hit a wall at speed, yes that's the front of the rig jumping off the ground.

    • Like Like x 1