Tutorial SMC3 Arduino 3DOF Motor Driver and Windows Utilities

if I press stop in SimHUB and press play again, I make it seem like I'm restarting (without disconnecting the USB), then the engine is working fine. And sometimes the engine freezes out of the central position, the green line is not in the center, if I turn the potentiometer, the lever moves, but the center is not in the zero position... it's not related to electronics, it's something in the firmware or settings, I guess..

Generally the Arduino resets when you reconnect the serial port. You might verify by pressing the Arduino reset button to see if that clears the issue.

Can you reproduce using SMCUtils? There is a check in the codes which trips if it gets a single unexpected encoder reading. This will turn the motor off and you can see that in the SMCUtils screen. I had that problem with a generic Arduino. Noise on the sensor input could also cause this.

Another possibility is that the ibt-2 or PSU has shut off due to momentary overload or glitch. I have this problem occasionally but it clears when I cut 24v power to the drivers. I have my estop turn off the 24v power so it’s a minor inconvenience. It happens based on how hard I push the limits of the drives. In general reproducing with just SMCUtils makes it easier to debug. I had a similar issue that I could reproduce using the “Motion” pattern even when the drive was disconnected from the rig.

Hello this works in Arduino Mega?

If it builds it should run. If it doesn’t build then you will have to fix it.

Attached easy and simple guide for smc3 PID calibration. Only correction to the guide should be the following:

• Step 1: Tune Kp (Proportional) FIRST with Ki = 0, Kd = 0
  
  
• Step 2: Tune Kd (Derivative) SECOND with final Kp and Ki = 0
  
  
• Step 3: Tune Ki (Integral) THIRD with final Kp and Kd values
  

Does your tutorial refer to clean SMC3 or to the one with @Sebastian2 's I-component fix?

Only use SMC3 to set up PID, and it doesn’t matter which SMC3 variant.

Well, with Sebastians fix, the I-component now gets the deviation of the error as input and not the deviation of position. Thus, the principle in your tutorial stays the same and your tutorial is very helpful. But one have to be carful with the values.

@Opast

I respectfully disagree.
The vanilla SMC3 code has a buggy PID implementation. The I and D component are basically broken.

You can test that by turning off your motors and demanding a motor position in SMC3utils thats different from where they are currently at. With a Ki-value > 0 this should result in a PWM-Output thats growing over time. But that doesn't happen.

A similar test can be done for the D component by turning your motors off and turning on the sine signal generator in SMC3utils. For Kd-values > 0, the PWM-Output should be max every time the sine singal goes through 0 and PWM-Output shoud be 0 everytime the sine signal reaches it's max and min value. Again, that doesn't happen.

Without a working I and D component, it doesn't make sense to try a calibration for those components.

My SMC3-version is the only one I know of that hase a working I and D component. I don't recommend flogging to the version I uploaded in this thread a few pages ago, though. This version wasn't perfect by all means. For example: Back then I wasn't aware of a bug of vanilla SMC3 where it was possible for the I component to move the rig way past Clip Input and Max Limits. This bug never bothered anyone, as the I component was broken anyways. But I came across it after I tested my version with fixed I component.
I fixed those bugs and now spend many months without issues with my latest version of SMC3. I attached my latest version to this posting. But be warned: I only tested it on my rig. Like vanilla SMC3, the executed code is different depending if you choose Mode 1 or Mode 2. For my rig I added a Sub-Mode for Mode 2, that is using a Cytron MD30C motor driver. I was only able to test that. I also tried to bring regular Mode 2 as well as Mode 1 to the same standard, but I have no way of testing.

Edit: User @OrtiMai seems to by giving this SMC3 version a try with Cytron MD13S drivers that require Mode 1. He instantly found multiple typos that prevented the code to compile and upload to the arduino. So I guess it's not plug and play for Mode 1 users right now. I'll try to work out the kinks with him and post an update when its done.

This version also has SoftOff and SoftStart and a working "LiftOffset" to help rigs with a heave axis get smooth motion during slow movements.

I only posted a general guide how to setup PID and numbers just for references from ChatGPT. I'd love to try your firmware and should I use your firmware from page 1 modified by you from 1.0 or Cytron? I'm using IBT_2

I'm not sure what you are asking me about.

If you want to try my SMC3 version, you should use the latest, so the one I posted today on page 137 of this thread.

Or are you asking me what Mode you should use in my SMC3 version? The same as in vanilla SMC3. What mode did you previously use on your rig? The vanilla SMC3 documentation says "typical H-Bridge" should use Mode 1 and "Chinese IBT-2 H-Bridge from e-bay or equivalent" should use Mode 2.

If you happen to use a Cytron MD30C motor driver like I do, you should choose Mode 2 and set cytron = true.

However, all this is documented in the code. Have a look at it. I'm glad to help with any remaining questions.

@noorbeast
Could you please update the link in

https://www.xsimulator.net/community/faq/arduino-code-for-use-with-various-hardware-and-simtools.31/

to the version i posted above? Thanks.

Hello. I'm building 2DOF platform for MSFS mostly but also going to play LFS.
It is Arduino Uno with SMC3 mode2, 2x IBT-2 drivers, 2x 24V 250w 75rpm 32Nm motors, 2x 24V 21A PSUs.
Now I'm on the table test stage. When I run SMC3Tools sine I see that the motor reaction is pretty jumping:

The yellow pwm line goes up and down and the crank moves by steps when stable rotation of the crank is expected.
Another problem is that in Motion simulation mode when crank changes its rotation direction the motor consumes huge amount of power and my PSU flashes its led and stops its fan for a moment. I tried to limit pwm change speed in the SCM3 firmware, it works in general but conflicts with PID regulator and doesn't help much.
Tried to limit voltage (21V is minimum for my PSUs), decrease the pwm frequency to 15kHz - the same.
Also tried powerful 24V LiFePO4 battery instead of PSUs but sine movement is still jumping.
Setting Kp to 100 and Kd to 20 helps a bit but PSUs are still experiencing problems on fast rotation direction changes.

All Fpwms 15-25 kHz behave the same, below 15 kHz motor starts to produce noise.
As for pwm - it is starting moving at about pwm 38.