Bumpy Stepping Motion (2 Dof Flight Chair)

Hi all, I'm still struggling with my chair giving very bumpy, stair-step-like motion. If I'm flying level and I do a gentle climb or descent, it feels like I'm going up or down steps, instead of a smooth motion. I've tried a bunch of different settings in SMC3. At this point, I don't know if I have a psu problem, ibt2 problem, pot problem, motor problem, or an SMC3 setting problem. Both motors behave this way. Please advise on troubleshooting steps and thank you in advance.

Also, is there any way to see the SMC3 monitor while in-game and connected to SimTools? That seems like it would provide some insight onto what's going on.

Specs:
250w 24v 75rpm motors
IBT-2 drivers
ALITOVE 24v 25A 600W PSU (x2)
360 Hall pots

Not sure why the dropbox images aren't working, but here are the links:

https://previews.dropbox.com/p/thum...TdjhGTRwO4R6ao2819lWRPkNewAVzq15cASCvA/p.jpeg

https://previews.dropbox.com/p/thum...tLt-bRwwuv0cK8SZVY0-/p.jpeg?is_prewarmed=true

https://previews.dropbox.com/p/thum...e5mBlfaSKPwrVUt-uAQN/p.jpeg?is_prewarmed=true

It also occurs to me that I could also have a SimTools config problem so here is my config:

https://previews.dropbox.com/p/thum...S9T5VhBqWNo_hwkVUV_l/p.jpeg?is_prewarmed=true

https://previews.dropbox.com/p/thum...T5bDxR71-lFOJGpTiZ61Np8LvAMqS02XvdIioU/p.jpeg

I just recalibrated the min/max's and it's still very lumpy bumpy. No noticeable difference.

Hi Tim, is it a new setup or the issue just appeared after that the motion rig has been working fine for some time?

On mine, I also notice small bumps when banking, but not pitch/ roll. I use Flypt and i am still trying to tweak the settings a bit. Saying that, i also had some physical rubbing of components against the wooden enclosure which I built and I fixed that - but still some bumps remain, which I am now thinking could be linked to the computed values or even the resolution of the potentiometer.

I'll be eagerly watching this thread for advice.

My first version of the build felt "rough", but not like this really noticeable stair-stepping. I ran across a thread that had me try lowering the Fpwm so I lowered it from 25 to 15 and it really smoothed it out in SMC3 (sine wave pattern). The smc3 image shows my current setup with the exception of the new Fpwm=15khz. I tried 10khz just for kicks but I start getting coil whine.

However, in-game (msfs2020), it is still stair-stepping badly whether it be pitch or roll. As mentioned previously, I went through re-doing the min-max's last night via capture but it made no noticeable difference. I also changed from linear to rotational output (90° limit) just for kicks. Also no change.

Regarding the rebuild... the only thing I did was replace peripherals (swapped out hotas with ffb yoke, hc bravo, and vkb t-rudder with mfg crosswind which is quite a bit heaver) and a car seat which added some weight (and moment to the front end). However, when I'm on it (300lb), it is pretty well balanced. I didn't mess with the electronics or power train.

Perhaps if I attempt to describe the stair-stepping better, it might lead to a faster solution. If I'm flying level, the chair isn't moving at all. Completely still. As I slowly move into a gentle climb (or roll, or whatever), at some point a threshold will be crossed and the chair will "step up" or "jump" to the next level. When I make quick movements, it feels smoother but still quite noticeable. However, in straight and level flight, it is VERY noticeable and completely immersion breaking.

Ooops, I was hoping to show them inline, but let's see if this works...

I did some more tweaking in SMC3 to create the smoothest sine-wave motion within SMC3. Here are current settings. No change in the stair-stepping in-game however. It would be wonderful to be able to see the game data in SMC3. Or even just be able to modify the Sine amplitude in the output mode to see if I could replicate the behavior outside of msfs.

I saw a similar problem. I zeroed all the parameters except pitch and roll. Works great now. Pretty sure that Heave was the culprit.

I was also concerned that the rig was not tuned and the base level. That is that MSFS was sending some small signal that was exciting some resonant frequency. I knew that you could hook up a PC in the middle and I did that. The result is that the rig was following an ugly input signal from MSFS.

I recalled that if you hook up a second computer to the Arduino you can configure SimTools to send the motion commands to a UDP port on that second computer. It's super easy to configure as a second interface. SimTools figures out what is active and does the right thing.

Here is a screen shot from that session on MSFS 2024. Level flight clear day. It is jarring. This is before I turned off heave,

What impact does setting the extra values have? Just gives the system a place to flush the crap data down?

Hi Tim, if simtools is the suspect here (quite unlikely though), then try the rig with flypt and see the response; or even try another game, dcs or a racing game. Just to see the behaviour.

By the way, I have the same setup as yours by the looks of it, motor and pot.

The Extra 4, 5, and 6 correspond to Roll, Pitch, and Yaw velocities or acceleration. These more accurately reflect the actual forces in a cockpit than just roll pitch and yaw.

The smc3 graph looks like turbulence to me perhaps? Therefore turning off Heave making it go away makes sense.

Going on Joe's hunch, I turned off all motion except for pitch and did a test flight in both 2020 and 2024. The stair-stepping motion persisted with no noticeable change. I tried with myself both off and on the chair. However, if I make fast movements, it feels smooth, no stepping.

I tried SMC3 Sine again and it still feels quite smooth. No sign of the stepping. Note that it could be possible that it's smooth because the SMC3 sine is fairly fast and there is no way to adjust that that I can see. If there was a way to change the parameters of the sine wave to make it much shallower, that would be helpful to troubleshoot my case.

I do use the min/max capture and then make adjustments while in flight to each motion independently. With the exception of the stair-stepping, the sensitivities and overall motion feels good and reasonable.

I think I may have found something. I took off the pot belt and the linkage from the motor to the chair so it can spin freely. I turned on smc3 and set the pot so that it will turn at it's slowest speed that I could manage. I noticed that I could stop the lever with my hand. If I turn the pot up a bit, at some speed I can no longer stop it. It seems either my Kp or PWMmax isn't high enough. I'm reassembling now to see if that solves the issue.

No dice. I tried raising either/both Kp and PWMmax and it just made the stepping more abrupt. I tried lowering it and it made it "mushy" but still obvious. It's like the output is low resolution, making it behave like an etch-a-sketch.

I'll try flypt today.

I wonder if there could be a problem with my data from pc to ard? I did switch out my USB hub as I needed more ports for my peripherals. It is a USB 2.0 hub (all peripherals are 2.0). I may try plugging the ard directly into pc as well when I get home.

Wow, I found zero info on how to set up and use fly pt. Their website is all but useless.

Uploading my ST3 interface settings just in case something here isn't right.

Here's a video of the motion...

https://www.dropbox.com/scl/fi/o4yk...ey=4gjn7golgk7rpo75thttwur59&st=jqnr1bqv&dl=0

EDIT: I just tried adding 100% smoothing to pitch and roll with all other motion disabled... it just made the step-up slightly "softer", but still the same amount of step up.

EDIT 2: Just tried running ard usb straight to pc, no change.

After much reading, I becoming convinced that I'm experiencing cogging and I suspect the gear ratio is wrong for my flight sim. This set up would probably be great for racing, but I either need to use a shorter ctc, or get a different motor. Or come up with a gearing solution. First step will be to drill a shorter hole in my lever to see how much of an improvement that makes. Fortunately I designed it with quite a bit of range of motion to play with.

It is turbulence but at times an order of magnitude more than you want. Even with heave at zero you still get some of that.

I went to set up the Extra values and found they were labeled Pitch Velocity, Roll Velocity and Yaw Velocity in SimTools 3. I turned off pitch and roll and set up the Velocity instead to the same value. With this setting I didn't need any filtering although I will likely add some in. I tested over the Grand Canyon so it was windy and the turbulance was mostly reasonable.
EDIT: I guess I didnt hit save on those settings. Using the Velocity settings is even worse in terms of smoothness.

In the game there is a turbulence setting low/medium/high. Like so much else in MSFS 2024 it doesn't do anything. Or at least doesn't have impact on this issue. Sounds like OP has something different going on.