Hi

Briefly (without all the details)


nMotorPIDSpeedCtrl - no
nMotorEncoder[] - yes
nMotorRunState[] - yes
nSyncedMotors - no
nSyncedMotors - no
nSyncedTurnRatio - no

You've probably read in the other thread that the PID speed control only seems to work if the controller gets to initialize the controller.

I believe (but have not tested) that the nMotorEncoderTarget[] array works with the tetrix motors.

I would reset the encoder to zero each time, if it was me..

the section of manual you quoted from relates to the standard LEGO NXT motors...

If you've got high ampere ratings that are burning up the motors, then you need to look at a mechanical fix.

If you don't want to write your own PID, but you want the most accurate distance stops, slow down as you are approaching the target.

Just some ideas... hope that helps!

Thanks for the information....

alanlevezu said,

..."You've probably read in the other thread that the PID speed control only seems to work if the controller gets to initialize the controller."...Can you tell me what steps, insures that controller gets initialized?

Is this the known download issue that sometimes occurs?

Does RobotC create a software PID or does it "communicate with the on controller PID" and use the cmotor controller's internal PID - because it seems that the bus delay does come into play when RobotC is in control. Correct?

I have a situation where drive motors running at 100 % for 10 seconds after Logitech Control is stopped still move under power for a few seconds - but at 50% power when drive motors are stopped by Logitech Control they stop in just a fraction of a second. A fast stop. Can this be a delay type issue with commands to start and stop some how delayed or queued?

I was not clear if RobotC uses a "write and read" on the I2C bus to the controller and these commands timings are then subject to a 25 ms delay which is the standard communication pause when communicating? All of this is because I am trying to understand where the inaccuracy comes from in the motor control?

Briefly

moving forward right motor 2 turns
then left motor 1 turn

is not the same location on X and Y axis as

move forward left motor 1 turn
then right motor 2 turns

so bus or programming delays produce different results. Timing does matter.

IF the controller could start both motors at same time a straight line could "produced with best possible accuracy".

What could help in the environment?

Deep_Blue

From experimentation, the PID appears to be located in the controller.

Also from experimentation - the only way to guarantee the Tetrix controller PID is working is:
1. NXT off, 12v Off
2. 12v On
3. NXT On
4. Start program

It appears that the PID in the controller stops being active whenever the 12V is cycled
It appears that when this happens, the controller does not let the NXT know, and thus the NXT does not re-initialize the PID.

debugging your "situation" is impossible without seeing your actual code...

I have no clue on the I2C bus comm stuff. (sorry)