g540 microsteps

[Numb Nut]

I'm using a G540 with Keling 381 oz/in motors (KL23H2100-35-4B) powered by a KL-4813 PSU rated at 48V/13A.

As the X2 CNC fusion kit hasn't arrived yet, I was playing about with a DOS program that I wrote (in borland turbo c++) to step one of the motors with a laser pointer connected to the shaft and noticed that the microsteps were variable sized - some larger than others. This was very repeatable and the pattern repeated every full step (10 microsteps).

Is this expected? Does the trim adjuster control this? I realise that a mass produced controller can't be expected to be perfectly matched to every stepper motor out there but am intrigued to know a little bit more about this.

I realise that a microstep will be giving me 0.0001" movement at the milling table so I'm not too concerned about non-linearity.

Finally, is it normal for a g540 powered stepper motor to "sing" a little bit? I assume this is due to the pulsed nature of the power circuits. The exact note changes as microsteps are applied. Are the bipolar outputs both run in sync from the same switched mode pulses?

Thanks for any info you can provide.

[Mariss Freimanis]

Microstep placement depends entirely on motor linearity (electrical angle vs. mechanical angle curve) when drive is accurate (accurate sine and cosine phase currents). Some step motors designs are optimized for high holding torque and this optimization is at the expense of motor linearity.

Take a look at Vexta PK268-03A motors; they have excellent linearity and your microstep placement will be nearly perfect. These motors have 250 in-oz of holding torque which is reasonable for an accurate motor.

Mariss

[Numb Nut]

Thanks for the reply, Mariss.

Since the linear movement per microstep is around 0.0001" (i.e. a tenth) I'm not too bothered about achieving perfect linearity - was just pondering the underlying cause. Your explanation makes perfect sense. I guess the controller has a built in idea of how it will achieve microsteps and that won't always match the motor - some motors might work well with a sine/cosine current model and others might not.

Any thoughts on whether there should be any sound emitted from a motor when it's stationary? It's noticeable but not loud. It certainly won't be audible over the noise of the X2 spindle motor is turning, once it's all assembled.

[Mariss Freimanis]

Even microstep placement and motor smoothness are two sides of the same coin. A motor that doesn't have evenly spaced microsteps will vibrate at certain speeds. Here is the reason:

If you send step pulses at a constant rate, the motor turns slower where the microsteps bunch up and turns faster where they spread apart. This means the motor decelerates where it turns slower and accelerates where it has to turn faster. This acel and decel gets repeated every 1.8 degrees of travel and is perceived as vibration.

If the microsteps are evenly spaced, the motor turns at a constant speed without any vibration.

Mariss

[chicof]

Are there bigger motors (800in-oz) with good linearity? Could you recommend a brand or a type?

regards,
Francis