Which motor to choose?

[Harmvd91]

Hi everyone,

For a project in Linear Fresnal Mirrors - Solar power at the universiry UFSC (Brazil) I need a motor to position the mirrors. I have estimated the desired torque, which is about 30 Nm. The desired rpm is very low. An rpm of order 10^-1 would satisfy. The desired accuracy is about 0.1 degree. However, our budget is very limited.
I think servo motors (controlled via an arduino board) would be a very good choice to control the position of the mirrors, but I have no bloody idea where to buy cheap decent servos (max $100 each); what is a good site or servo brand/type? Or is it maybe better to choose for another type of motor?

Regards,
Harm

 

[Skogsgurra]

Stepper motors with a gearbox seems to be the natural choice. At least for me. Microstepping and PWM is easy at those low speeds. And if you use unipolar drive and resistors for current limit, you can simplify the Power Electronics a lot.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Harmvd91]

@Skogsgurra; thanks for your reply.
The most common stepper motors have an accuracy of .9 or 1.8 degrees. The more precise but powerfull steppers are actually equivalent to servos right?
And is there any website where you can find a motor on base of specs? Like min. torque, max. price, min. accuracy et cetera?

Thanks,
Harm

 

[itsmoked]

Definitely. Servo motors are way overkill for solar tracking.

Though you can get lots of cheap gear box versions:
DC Gear motors

Another possibility is standard HEAVY DUTY R/C servos.
Heavy duty classic R/C servo
You can get Arduno shields that drive armies of them.
Arduino Servo Sheilds

Cautionary tail of R/C servo usage

But back to steppers. Look at "gearbox steppers".
Ebay geared steppers

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

You need some sort of speed reducer (gear, timing belt, screw or whatever) to get the torque you need from a small stepper.

The speed reducer will, at the same time, better the resoluion så you can easily have the 0.1 degree that you need. The accuracy is often a lot better than 0.9 or 0.18 degrees - it is the resolution that has those typical values.

Microstepping improves resolution a lot. It is not difficult to have 0.01 degre resolution when motor shaft is unloaded. A speed reducer, without backlash, will improve that further.

But, do not try without gear/speed reducer with any motor. It can be done, but at a high cost - you are using less than one percent of the motor's capacity if you run without speed reducer at these low speeds.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[iop995]

Gearbox steper in open loop. Cheap and sure. I think you don't need continously moving, steps of 0.1 degree being enough, step motor is one more best solution. Sure, step pulses are send accordingly solar tracker sensing.

 

[Harmvd91]

Okay, thanks a lot to all!
Two more questions;
1. Is the accuracy related linear to the gear ratio? (I do think so, but just as a check)
2. Would I need metal or titanium gears? As I think plastic gears will break under these forces.

Regards,
Harm

 

[Skogsgurra]

You need to understand the difference between resolution and accuracy.

Resolution gets better with higher gear ratio.

Accuracy may get worse, if there is backlash and/or stiction in the gear. But, in general - accuracy improves also. But that is highly dependent on the quality of the gear.

Re plastic or titanium (why titanium?, there is also steel and brass to chose from): You must, of course, select a device that can handle the forces involved. Titanium sounds quite expensive. I usually see steel and brass in different combinations.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Harmvd91]

Thanks a lot again.
So the resolution increases a lot, but what I need is an accuracy of 0.3 degrees.
I found one very cheap servo with integrated gearbox here (I also found two geared steppers on robot market place, thanks itsmoked, but unfortunately they were out of stock). However, the backlash is >=1.5 degrees. That means my accuracy is 1.5 degrees (so not even close to the .3)?

If I search for other motors / gearboxes, what is the direct influence of the backlash?

 

[BrianE22]

That's a lot of backlash. You can get gearboxes designed for servos that are very low in backlash and high in accuracy. Search for "servo gearboxes".

 

[Harmvd91]

but is it possible to mount these gearboxes on stepper motors?

 

[BrianE22]

For mounting, it should just be a matter of couplings and flange adapters. You might have to machine your own flange adapter.

 

[djs]

You might try RC type servos. Inexpensive and very accurate. One source you might check out in the US is Sparkfun (Google it).

 

[mikekilroy]

reality check time... you want to spend $100 for 30nm torque output at 0.1 degree accuracy.... speed at this point does not matter...

any precision planetary gearbox that can produce 30nm continuous output will be approx. 3-4" in diameter x 3-4" long.

any precision planetary gearbox of this torque with 6 arc min or less backlash will cost you approx $ 1500.00

NOW figure out whether you want servo or stepper.... you will likely want >100:1 ratio; any ratio over 10:1 will add approx. 30% more to the cost.

NOW pick your motor and add that to the gearbox price.

You will NOT get 30nm <.1 degree accuracy for $ 100.00 unless you steal it.

If this is for a client, walk away NOW.

http://www.KilroyWasHere<dot>com