Showroom 6DOF Motion Platform á la Knaufinator

As @noorbeast says, start your own build thread and send a link to me. I will be happy to help. I'm also interested to see how you manufacture it

In the meantime, it sounds like you're making good progress I can post some information about my first single motor test, which might be helpful:


As you can see the movement of a single motor is irregular and difficult to correlate with the image. If you are doing it with jus the motor then you will see much bigger and more irregular-looking movement.

Now I have the platform built I have found that the output from some games can be very noisy. That is when the filters in Simtools become very useful. It's possible that this is what you're seeing. Even with the platform fully built, I have found that some of the outputs are difficult to correlate to start with.

When I could see that the Axis Testing 'Heave' dof was working correctly in all of the motor positions on the circuit board, that was good enough evidence for me to progress to the next stage.

what board are you using to interface between simtools and the servo drive ?

Its an esp32 with level shifters and I/O expansion.

Hi flyingonion, I have built this platform. Can I ask, do the servos respond correctly from SimTools->Output Testing?

Hello, separate axis test is normal, it's just connected to the game, no response. This is my setting. I uploaded a lot of pictures. Can I display them completely?attempt 6DOF Rotary platform (xsimulator.net)

Thank you very much for your help. Now the single servo motor works normally. It is because my computer is not set to English. I will go on to the next step. I am going to buy welding machine, welding arm and support rod. Is your support rod ordinary steel pipe?

Yes, I used mild steel tube 25mm diameter and wall thickness 2mm.

Be careful to check the buckling capacity. My first design (Ø20) buckled a little bit so I have just done larger diameter and shorter length. I will post photos of the new support rods later,

Hello folks - it's been a while since I've posted but I have finally got this in the house, upgraded my pc and graphics card, and upgraded the controls.
Getting the motion, the VR and the motion cancellation dialled in with AC has been a long process but today I felt like I finally got it sorted:

What is the name of a case like yours, and where can I buy it

Hello, what is the name of the enclosure that houses the servo AASD15A driver, can you give me the link to buy this enclosure? Thanks!

Unfortunately it is not a commercial product. I made it all myself by welding the frame together and getting a piece of glass cut to fit onto the front. I made the top, sides and back from aluminium.

Very nice machine you got there. Apart from the squeaking it seems REALLY silent. I think I need to ditch my ballscrew actuators and go down the route of a rotary Stewart platform..

How much range and speed do you get out of this design? I like the dialed down profile but there are some tracks where I'd like to get some pitch and roll out of the system and also especially some surge and sway impulses.

If you ever get around to it do you mind making a vid with your profile around a track like Laguna Seca or Watkins Glen? (I'd prefer Laguna Seca). Not sure what your sim of choice is but I'd gladly donate the track to you in iR of rF2 if needed.

Thanks bberger - yes, that squeaking comes from the Heusinkveld pedals so it really is very quiet apart from that! I will work on solving that so you can get a better idea of the sound. In the meantime, I've done a video around Brands Hatch (working on my editing skills to make my videos a little bit more informative)

Hopefully this gives a bit more of a sense of the way it moves over gradients. I'm doing some work on the filtering at the moment but I'll happily get a video together around Laguna Seca in the next few days for you.

In terms of range, I limit it to around ±100mm in each direction and about 20° rotation although I find I'm using a much smaller amount of both than I initially thought I needed. Not sure about the speed but on paper it's capable of 5g acceleration and it can certainly exceed a threshold where driving a lap becomes a painful endurance test!

Further to my exploits in filtering, I have a question for @yobuddy on summing Axis(a) and Axis(b) in the Interface - Output definition.

The problem I'm trying to solve is that high gain gives good feel for low frequency undulations and cornering when it's smoothed but it introduces latency in the high frequency surge events such as hitting the brakes or running over kerbs. Removing the filtering sorts out the the latency but makes the high frequency events painfully harsh so the gain needs to be reduced to get this under control. I think a combined high gain smooth signal plus low gain unfiltered might be an interesting avenue to explore. Any thoughts on this?

So I'm trying to sum unfiltered+low gain with smoothed+high gain for the same signal. I think I can do this simultaneously by using Axis1a and Axis1b, but can I sum them in the output? Something like <Axis 1a+Axis1b>,<Axis2a+Axis2b>,....etc.

Or am I better off outputting them separately and summing in the microcontroller code?

Turn the filtering off and get raw data - then adjust the pid for your controller (if you can)

the controller just tells the AC drivers to goto a position as fast as it gets an update from SimTools, if you want to control the PID do that in each of the AC drivers.

Thanks guys - yes, adjusting the PIDs on the controllers has made a big improvement to latency so everything is starting to feel a bit more balanced now.