[SRT Mike]

I am working on a project - it's not a CNC project, but rather a little electro-mechanical device. I hate to be secretive but I am working on a design for someone else and I can't say exactly what it is, but I can say that it has a platter that will rotate. The platter will be about 4" diameter and will be mounted "upright" so that the axis of the platter will be horizontal and parallel to the ground. The platter weighs 400 grams and will rotate at low speed.

It would be ideal of the rate of rotation could be varied from about 4rpm to 8rpm and it would be great if I could control not only speed but also position. But the most important factors are noise (it must be close to silent) and power draw - it must run on batteries so it needs to be quite efficient.

I can use a stepper or a small precision DC motor with a gearbox. I've seen some microstepping demonstrations and they are very quiet, especially at low RPM's... but they were also with unloaded motors. I wonder if I can keep it smooth at low RPM with a load attached?

I also have read steppers are 50-70% "efficient" whereas coreless DC gearmotors can get up to 90% efficiency. However, small coreless motors run at 10,000RPM or thereabouts, and require a gearset to reduce the speed, and most of those gearsets are about 50% efficient.

So it is my understanding that a stepper may actually be more efficient? Am I correct there? When I am not moving the platter, I don't need to hold it in place - so the energy used would only be during rotation. But I am worried about noise or jerkiness with such a low speed. I can also gear it down a little and run the stepper faster if anyone has any ideas on optimum RPM's for energy efficiency and noise?

[Al_The_Man]

A couple of things to consider, DC servo is generally quieter and more efficient than a stepper, the stepper consumes the same power whether stopped or moving, so you would need to disable the drive when not in use in a portable application, the DC servo is quieter and only consumes power according to the load and move profile, with zero current when stationary.
Both DC motor and stepper have maximum torque at zero rpm.
But the DC motor would require a feedback device for positioning, and if building something from scratch, may involve a bit more in the way of electronic developement or If you are able to use a small microprocessor there are app notes on the PicMicro site.
http://ww1.microchip.com/downloads/e...tes/00532c.pdf
Also there are a few kits out there that you could adapt, this one has allowance for a pulse input feedback, but IMO it would be best to use it with a small low res encoder for accuracy.
http://store.qkits.com/moreinfo.cfm/KT-5194A.pdf
I just noticed that this particular kit is speed control only not position.
If you are only moving at a slow rate, with low accel/decel a gearbox should not be needed.
Al.

[hub]

Quiet = servo.
4-8rpm = Gearbox. Maybe a dual stage cyclo. Or a planetary..
Not need holding torque = servo.

For that, I would not go with steppers.

Hub

[SRT Mike]

I understand the drawbacks to servos and steppers.... but the problem with a servo is I need to use an encoder, which adds a lot to the cost, and the motor and gearbox itself is expensive. I know a stepper is open loop and steps can be lost - but this isn't a big deal, I don't need much accuracy on rotation (remember it's not a CNC machine).

Also, a small DC precision coreless motor with gearbox is well over $100. A stepper is a lot less. I realize you can buy crap gearboxes for a few bucks and motors for even less, but the stuff I am looking at is MicoMo, Portescap and Maxon. Most of the motors are $60+, the gearbox is $30+ and the encoder is $20+. These are precision parts and are quiet... but they are expensive.

With a servo, the motor will be running 5,000RPM and the gears will reduce it down to maybe 150RPM.

So... can a small (Nema 8) stepper being microstepped at 150RPM and rubber mounted be almost totally silent?

...and...

Is there any numbers on effiency comparisons? Let's assume the stepper is microstepped and does not need to be energized when the rotor is not actually stepping... what sort of efficiency can be achieved? 50%? 70%? Less? more?

...and...

Finally, given that a servo has a geartrain which adds a lot of drag, it lets the motor hold position easily... but the steppers I have bought previously usually have quite a bit of resistance to turning when off. You can feel each click as you turn the shaft, even when powered off. Is there a name for this trait and is it measurable or disclosed on datasheets? I haven't seen it. I think a stepper, if it is like steppers I have lying around my shop, may have enough resistance to movement for my application without being energized, but I am not sure. The steppers I have are quite a bit larger than what I would be using for this project, if I go with a stepper.

And finally, anyone have any sources for very small (like less than 20mm diameter) servos or precision DC gearmotors other than those I mentioned?

Thanks

[Al_The_Man]

Finally, given that a servo has a geartrain which adds a lot of drag, it lets the motor hold position easily... but the steppers I have bought previously usually have quite a bit of resistance to turning when off. You can feel each click as you turn the shaft, even when powered off. Is there a name for this trait and is it measurable or disclosed on datasheets?
Thanks

The term is detent torque.
What makes you think you need a gearbox with a servo? especially at that rpm.
Servo's are quite capable of very low rpm, the torque is fairly flat over the rpm range, it is just that steppers generally have a higher torque for a given Nema size.
You would not even hear a DC servo at the rpm. Agreed the feedback puts the price up a bit, but you have positioning down to the resolution of the encoder, and x4 that for systems that detect the quadrature pulse edges.
Al.
Al.

[SRT Mike]

The term is detent torque.
What makes you think you need a gearbox with a servo? especially at that rpm.
Servo's are quite capable of very low rpm, the torque is fairly flat over the rpm range, it is just that steppers generally have a higher torque for a given Nema size.
You would not even hear a DC servo at the rpm. Agreed the feedback puts the price up a bit, but you have positioning down to the resolution of the encoder, and x4 that for systems that detect the quadrature pulse edges.
Al.
Al.

Hi Al,

Sorry, I wasn't clear in my previous post. When I was talking about a gearbox if I don't use a stepper, I meant a DC motor feeding a gearbox and "something" for feedback (either an encoder or feedback from the gear teeth that the motor would drive). As I understand a servo is basically the pre-packaged combination of these things, so if I used a servo, it wouldn't have a gearbox external to it.

But a DC gearmotor can be had for $70... all of the very small servos I have seen appear to be a lot more. Of course there are 'continuous rotation' hobby servos for $10 if you buy individually, but my experiece with hobby servos is that they are pretty darn loud. Is there something in between?

I would love to be at $20-30 in 100 unit quantities for a solution that is quiet, reliable and allows me to position it (doesn't have to be very accurate, if I could get within 5 degrees, it would be great). Is there something between hobby servos and those used on precision CNC machinery?