THC accel/decel

[cncdrive]

Not the start/stop frequency is infinate, but the acceleration can be about infinate below the start/stop frequency.
And ofcourse infinate is only a theoritical thing, better to say very high, but it is longer than to write "infinate".

[Robertspark]

With servos if there is enough step buffer size then you could use high acceleration and the servo ofcourse will not loose steps and will not fault if the step buffer size is high enough.
E.g. in our DG4S drives it is setable upto 20000 steps.
So, the PID will work out the error and the motor will accelerate and deccelerate and reach target positions based on the position error, so as soon as the error goes high enough the PID loop push the throttle accelerating the motor to it's limits.

So would servos be the BEST solution?? But given they rely on acceleration to correct for step error they will also lag the motion signal
as soon as the error goes high enough the PID loop push the throttle accelerating the motor to it's limits.

[cncdrive]

And yes, it depends on the load inertia.

[cncdrive]

Yes, they will lag, but will lag no more than how fast the command signal minus how fast they can accelerate.
So they can't go faster anyways lag or no lag no matter.
And servos if properly sized for a load can accelerate much much faster than same size steppers.

And so lagging is not really a problem in a plasma application Z axis.

[Robertspark]

"start/stop frequency" >> Maximum starting frequency (fS)

Say 2khz is the pulsetrane number that allows for that high acceleration

would that be 2khz at full step..... 2khz at microsteps ..... or 20khz at say 10uSetp setting

microstepping is only a way of carving up a full step by high speed switching the signal so I presume that "2khz" Maximum starting frequency would be the full step frequency??

[cncdrive]

You should ask the motor manufacturer how they built their graph if it is for full step or what, but I think they usually making it for full step, because there is where the motor is the less powerful on low speeds due to the slow squarewave current waveforms in the coils. Any microctepping makes the motor only stronger.
And it is important to know this because if the graph is for microstepping then one step is considered 1Hz which is then 1 microstepping and so the rotation frequency for the defined start/stop frequency is then lower if the microstepping is higher for the graph.
But I do not think it is for any microstepping otherwise they would show the number of microsteps they built the graph for and microstepping is the property of the motor drive while full step is the property of the motor.
And so it is much more likely that all these graphs are built for full steps if the microstepping numbers are not noted for the graph...

[beefy]

Rob,

when we say "infinate" do not take it literally.
Of course infinite acceleration is not possible in the real mechanical world. It is just a way of saying accel/decel is not applied, e.g. if a certain motor speed required a step pulse frequency of 500 steps/sec, then the cnc controller gives out that step frequency immediately without ramping up (accelerating) to that frequency, and visa versa for stopping, the step pulse train simply stops. And provided the max speed setting is within the motors "start/stop frequency" (motor max speed setting, or THC feedrate setting) the stepper should not lose steps. Of course the applied load will reduce the start/stop frequency so the idea is to have a big oversized stepper with a decent start/stop frequency specification, then the relatively small load minimises the reduction of the start/stop frequency.

Balazs,

yes we do need a separate max speed setting for THC moves but I thought that was already in UCCNC ("THC feedrate" in the I/O tab).
And for those who think it's better to be able to apply accel/decel to THC moves, then maybe an individual THC accel setting, and maybe a tick box to apply or not apply THC accel.

Never knew about the "start/stop frequency". Thanks for the new knowledge

Keith.

[stirling]

At my set acceleration, my Z axis simply moves too far, even from a modest Z speed of 1500 mm/min and a THC update rate of 100 times/sec.

Hi Beefy: Out of interest, what is your Z axis steps per?

[beefy]

400 / mm

5mm pitch leadscrew and microstepping of 10 on a 200 full steps / turn motor.

G203V drive. If I have this right, then it "morphs" from microstepping at low speed to full steps at high speed.

[cncdrive]

Hi Keith,

I was thinking about this a bit and I think that what we have planned for the S-curve planner will resolve this problem.
How we want to make the new planner is that there will be a start velocity parameter which if 0 then it works like it is now, but if it is not 0 then the motor will start with the set start velocity parameter.
So, setting that parameter to around the start/stop frequency of the motor will make the motor to 1(t) velocity jump under the start/stop frequency and will use the acceleration parameter value above that, so it will be ideal for the THC axis and I think it will resolve what was described in this thread and then there is not even a need for separate velocity and acceleration parameters for the THC.
Moreover it is even better than giving a separate velocity and accel parameter for the THC work, because then you would have to limit the velocity to under the start stop frequency of the motor and so it would limit the max.velocity of the axis which might be a problem in long moves, but with the start velocity parameter the axis could still accelerate above the start/stop frequency of the motor, so there is no limitation like that.
What do you guys think?