[Pandinus]

This may sound crazy, but I was thinking about top speeds of a machine while tuning the motors for acceleration.

As most of you know stepper motors lose torque at higher speeds. If your acceleration "CONSTANT" is lower the final speed can be higher. In a plasma system you want as high an acceleration to cut speed after pierce as possible, but it would be nice to have the maximum available jog speed between cuts.

So here is the question, would the idea of a dynamic acceleration as a function of velocity be crazy. Even better, two acceleration sets for cutting and jogging.
It could be as simple as two values. From 0 to mid speed acceleration is one value, then from mid speed to max speed the acceleration lowers to a second value in a linear fashion or a complete formula entered as a function of velocity counteracting the loss in torque.

This is what I get for thinking too much.

[jmelson]

This may sound crazy, but I was thinking about top speeds of a machine while tuning the motors for acceleration.

As most of you know stepper motors lose torque at higher speeds. If your acceleration "CONSTANT" is lower the final speed can be higher. In a plasma system you want as high an acceleration to cut speed after pierce as possible, but it would be nice to have the maximum available jog speed between cuts.

So here is the question, would the idea of a dynamic acceleration as a function of velocity be crazy. Even better, two acceleration sets for cutting and jogging.
It could be as simple as two values. From 0 to mid speed acceleration is one value, then from mid speed to max speed the acceleration lowers to a second value in a linear fashion or a complete formula entered as a function of velocity counteracting the loss in torque.

This is what I get for thinking too much.

This is S-curve acceleration. There was a module for this developed by a visitor at NIST, but it never got into the public version, as far as I know. I think some other people have worked on it a bit since, I don't know the current state. I think the math might get deep when you combine this with cutter offsets, smoothing and everything else.

Jon