The need for speed - NEWBIE microstepping quesiton

Hi all

I am running Mach 3 using KTA 261 controllers from Ocean Controls in Melbourne. I currently have these running in 1/16th mainly because that's how they were shipped. <g>

They are driving 8TPI 1/2" ACME.

My question:
Is there an inverse relationship (or any other for that matter) between the level of microstepping and the maximum RPM that any given motor / voltage / current combination can produce ?

In other words if I change to 1/4 step do I get four time the maximum RPM and what does this do to my torque ? I expect my accuracy will reduce since I am not moving in such fine steps but a 1/4 step on a 1.8 degree motor doing 8TPI would be negligible for the basic wood routing I am doing.

Thanks

David

The reason I ask is that when I was running 1.5 MM all-thread, I could not set Mach 3 to a feed rate of over 250mm/minute. My steps per unit in Mach 3 was 2133. My motors would "vibrate", resonate","flutter" (pick one) at anything above that. When I changed to 8TPI (3.175mm) and hence changed the steps per unit to 1008 I could increase the feed rate by a factor of three and still no motor problems.

So, if I drop my micro step count, that means I drop my steps per unit further, can i ramp up the feed rate ?

Anyway, your thoughts please.

There is a maximum set by the pulse rate that can be generated by the PC and handled by the stepper controller. If you need 16 pulses to micro-step through one full step, you have to generate 16 times as many pulses in the same time to turn at the same speed as you would if you were full stepping.

Having said that, what you describe your motors doing sounds more like mid band resonance, and that usually improves with /higher/ microstepping.

So try it, and if it gets worse, add mid band resonance compensation instead.

Quote:

---

Originally Posted by James Newton

There is a maximum set by the pulse rate that can be generated by the PC and handled by the stepper controller. If you need 16 pulses to micro-step through one full step, you have to generate 16 times as many pulses in the same time to turn at the same speed as you would if you were full stepping.

Having said that, what you describe your motors doing sounds more like mid band resonance, and that usually improves with /higher/ microstepping.

So try it, and if it gets worse, add mid band resonance compensation instead.

---

Higher microstepping means reduced torque. You can stay with 1/4th microstepping unless there is a need for higher...

Quote:

---

Originally Posted by leaveme

Higher microstepping means reduced torque. You can stay with 1/4th microstepping unless there is a need for higher...

---

I'm very curious why you would say that?

From my experience, resonance issue goes away at 1/4th mode. But now look into the pulse time. Stepper is just getting 1/4th power to be energized for each step (at 1/4th mode.

Quote:

---

Originally Posted by leaveme

From my experience, resonance issue goes away at 1/4th mode. But now look into the pulse time. Stepper is just getting 1/4th power to be energized for each step (at 1/4th mode.

---

Again, very curious as to why you would say that... In the Linsitepper driver, (and all others that I'm aware of) the stepper still gets FULL power no matter how much microstepping is applied. It just takes 2, 6, or 18 time more steps for each /cycle/ of power. So the motor needs a faster stepping /rate/ to move at the same speed, and will turn slower if the step pulse is not adjusted to come faster, but should have no less /power/ in any case.

In my experience, the more you microstep, the less resonance you experience as a result of starting and stopping the motor after each step... smoother motion means less vibration.

Full stepping (with both phases fully on) can have a higher torque than all microstepping modes because the microstepping modes have reduced current in one or both phases.

But all microstepping modes including 1/4th stepping and above should have similar holding torques.

And if the microstepping driver has been properly set up it will use higher current to give similar wattage and torques;

For instance a 5v 1A rated motor in FULL STEP uses 5W * 2 phases on = 10W.
Set up for microstepping you would use 7v 1.4A which gives the same heat dissipation and very similar torques.

Geckodrive.com has a good explanation of microstepping and power at their website.

John

Quote:

---

Originally Posted by johnmac

Geckodrive.com has a good explanation of microstepping and power at their website.

John

---

Yes, it's at:
Geckodrive Frequently Asked Questions FAQ
and then scroll down a bit. As it explains, the full step driver only has an advantage when the motor is standing still, which is really not what anyone cares about when the goal is to have axis flying from point to point. It goes on to show that the microstepping drivers have a /slight/ advantage at low speeds and are about a wash at high speeds.

And it doesn't account for the fact that, as Roman points out, one would normally adjust the drive voltage and current to account for the difference so the resulting torque is the same.

One thing I'm curious about is the comments on that page about mid band resonance. Both full step drivers and cheap chopper type microstepping drivers can induce the types of resonance oscillations that result in huge power loss at certain speeds, and the gecko's have some tricky methods for avoiding that. But a linear drive with fully analog drivers shouldn't have that problem at all, as far as I understand. I know I've never encountered mid band resonance from my Linistepper drivers as long as I stay in the higher microstepping modes.

Microstepping is one way to help eliminate mid band resonance. At higher speeds, microstepping doesn't serve any purpose. High speed microstepping does require a computer fast enough to produce the pulse stream at a high rate. Also, in mach, that means selecting a higher kernel speed.

John