View Full Version : Torque

06-16-2005, 04:03 AM
Parallax has a Futaba servo.· The torque specs are:

3.40kg-cm/47 oz-in.· That I understand.

Their Unipolar Stepper has the following specs:


Could someone translate that for me?· How does the rotational power of the Unipolar compare to the servo?· Would the Unipolar turn anything the servo could turn?


06-16-2005, 04:16 AM

Most stepper motors have greater torque than a servo motor.· Basically you need to decide on how much torque you'll need.· Then price will become a factor, high torque servos can be quite expensive.


Dave Andreae

Tech Support
dandreae@parallax.com (mailto:dandreae@parallax.com)


06-16-2005, 07:20 AM
Newzed said...
Parallax has a Futaba servo.· The torque specs are:
·3.40kg-cm/47 oz-in.· That I understand.

Their Unipolar Stepper has the following specs:

Could someone translate that for me?· How does the rotational power of the Unipolar compare to the servo?· Would the Unipolar turn anything the servo could turn?

Sid, If my calculation and your data are correct, the stepper is about 0.27 kg-cm (at 200pps), or about 1/12th the torque of the servo.


Chris Savage
06-16-2005, 09:06 PM

·· It's a mixed bag...Some Stepper Motors have more torque, some do not.· You really need to compare the torque ratings to make a true comparison as each model can be different.· This also applies to servos.· There are heavy duty servos which easily out-torque many Stepper Motors.· Often with a Stepper Motor you can "gear-it-down" or add a drive system.

Chris Savage
Parallax Tech Support
csavage@parallax.com (mailto:csavage@parallax.com)

06-16-2005, 09:16 PM
Par, I should be getting the Micro Mill Monday or Tuesday.··Suppose I turn the little hand wheel until·the wheel handle·is at 9 o'clock and hang weights on it until the handle moves, and then suppose the handwheel moves with X ounces of weight.· If I give you the exact distance of the wheel handle to the center of the wheel and the weight at which it starts to move, could you tell me if the above stepper will do the job?



Bruce Bates
06-16-2005, 09:45 PM
Par -

Just to add to the message Chris just posted, what's not necesarily obvious to those who haven't opened up an R/C servo is that the DC motor inside is already heavily geared down. That's one of the reasons why some servo manufactuers sell metal replacement gears for the plastic ones often found inside. That replacement of metal for plastic will provide the maximum available torque over a longer period than the plastic gears might tend to.

Just as a sidenote there are many surplus gearhead DC brush motors that can be purchased quite reasonably; so too surplus geared stepper motors. Just ask if you need help sourcing them.


Bruce Bates

Loopy Byteloose
06-16-2005, 10:52 PM
As Bruce Bates points out...
It is not just about torque, it is about mechanical wear [the gears wear out faster than just direct drive bearings in a stepper]. Also, in some cases it is about 'brush noise' in DC gear head motors. Steppers can be choppy at certain rates, but they can provide precise holding power and position that is quite useful. While a gear head or servo may have better holding power, it is a bit more difficult to return to the exact same position in a open loop system. You have to add feedback.

G. Herzog in Taiwan

06-17-2005, 01:21 AM
Bruce Bates said...
Par -

Just to add to the message Chris just posted, what's not necesarily obvious to those who haven't opened up an R/C servo is that the DC motor inside is already heavily geared down.· ...
And, others have correctly pointed out that there's more to the issue of selecting a drive motor than simply its torque value(s), such as the material that the gears are made of, noise, tolerances, size, weight, ability to withstand wear and tear, ease of use, electrical power requirements, rpm, ·the application to which the device will be put, etc.

But, the original poster·(newzed)·provided torque data for two devices and asked which one has more "rotational power", and can the one·device·"turn anything" the other device can. I (loosely) assumed that "rotational power" meant "torque" for newzed's purposes; and, that it was fairly obvious that the one device newzed asked about, with only 1/12th the torque, could not·-unmodified- turn anything that the other (unmodified) device can turn.

I believe the torque rating given for the servo is the rating of the whole device (i.e., internal gearing included), not just of·the motor itself. So, the torque values given by newzed compare a geared-down servo to an ungeared-down stepper in this case.

Either device could have additional gearing added. But, newzed didn't ask about alternative gearing, nor about other brands/models of devices.



06-17-2005, 03:53 AM
I was· looking at only the Parallax Unipolar stepper at the moment.· That stepper takes 48 steps per rev which = .0393 inches of travel.· If I gear it down 10-1 then it will take 480 steps just to travel .0393 inches.· The thing I don't know is how many shaft rpm does the Unipolar make.· If its rotational speed is high enough that might be OK.· I found a unipolar at Digikey which has a holding torque of 189mN-m which is equal to about
26·oz-in.· That is about half the power of the servo but it is probably enough to do the job.· I'll know more when I get the mill tomorrow.

In case anyone cares, mN-m means milliNewtons per meter.· 1 mN-m =
.141 oz-in.


06-27-2005, 12:57 AM
If you are referring to a mini-mill from Harbor Freight or Homier or some such, then you will need a much larger stepper.

Considder that the mill will need some torque to move the wheel, and as you said, you could create a lever with a one inch radius, then hange a string to measure torque.

Expect something on the order of 200 ounce-inches. 200 ounces at one inch for actual cutting operations with the mill.

Motors with 24 steps per revolutions are not particularly strong and the power needed to make it move from one posistoin to the next is not efficently used for slow speeds. once the unit is running, the inertia or momentum increases the delivered power.

There are two general types of steppers, pressed can and Cast frame. the pressed can type found on printers and such are great for hobby stuff, but are woefully underpowered for machine tools.

the Cast frame have a general classification by phyusical size. NEMA National Electrical Manufacturs Assoc. classifies them by the bolt pattern. NEMA17 is 1.7 inches square and is found on the very early 5-1/4 floppies. very underpowered but neat.

These NEMA types come in two styles. round body and square body,
Rround motors are older and underpowered. the square body have considderably more power. You can find NEMA23 in 35 to about 150 oz-in the square body ones can go up to 600 oz-in or better. Square body NEMA17 can be found up to about 150 oz-in.

Slight note on termonology. When speaking machine tools, A DRIVER is the thing that sends the power to the motor the CONTROLLER that sends out the steps. In this case the STAMP is the controller and the GECKO is the driver.

Then there is the matter of driving them. for optimum performance, you need to use a Chopper driver. A chopper driver will deliver high voltage, between 5 and 25 times the motor nameplate, to the motor in a chopped pattern.

Trying to drive a motor at nameplate will deliver very poor performance, and you will need a much bigger motor.

If you want a nice set of motors and a nice driver, look at HobbyCNC.
AutomationDirect has nicer motors but they are bi-polar and the HobbyCNC board is uni-polar.

The Gecko's are $114.00 each, so getting another one would be cost effective and deliver much greater power.

IF you want more of a read, go to the www.geckodrive.com website and find the white paper on power supplies.

Even more can be found by googling Jones on steppers. But, Jones gets pretty deep.

Suffice it to say, the cost of making one is no savings as compared to buing one. (or three)


ps: I have a metal lathe so if you need a few pieces turned, let me know.

06-27-2005, 02:56 AM
The little crank handle on the handweel of my mill is .600 from the center of the wheel.· I hund weights on the crank handle when it was positioned at 9 o'clcok.· A weight of 44 oz. will make the handlewheel turn.· What size stepper do I need?

Dave, the stepper I have been working with is the one you sent me with the Gecko drive.· I have programmed everything to give me a travel of .100 for a degree input of 360.· The stepper actually turns about 2 1/2 times but I am interested only in how far the table travels - not how many degrees it took to get there.· I have substituted a 4x20 LCD for the debug screen and I'm using a 3x4 keypad for input to the Stamp.· The keypad is read by a 16F628 which serouts the number pressed to the Stamp.· Once programmed, the system is a stand-alone.· Haven't hooked it up to the mill yet - I'm waiting on a wheel adapter and a coupler to connect the stepper shaft to the handwheel.

The LCD displays the current position, then the menu.· After selection it prompts you for the number of degrees and gives you an option to accept or reject.· When the rotation is finished, the LCD displays the new current position and the distance traveled, left or right.· Each new position is stored in the Stamp EEPROM, which is read each time the system is booted up.· The position is displayed as·XXX for CW rotation and -XXX for CCW rotation.· Position is always referenced to a starting position of 0, so it you are at 360 degrees CW, then rotate 180 degrees CCW, the LCD displays a position ·of 180 degrees.

Sid Weaver
Do you have a Stamp Tester yet?


06-27-2005, 04:29 AM
Newzed, check out the CNCZone Forum...


06-27-2005, 11:41 AM
The motor is a NEMA23, single stack. that means that when you look at it, the squareish mounting plate is about 2.3 inches square and the body is about 2.3 inches long. the end caps being about the same thickness as the motor plates.

If you have the opportunity to gear it up, you can use a 3:1 gear reduction to increase power by a factor of three.

Makeing a very simple assumption that 44 ounces at about 1/2 inch would equal about 22 ounces at one inch. so, to move the handle with the same force, most motors would do that. Once you start cutting the forces will go up.

If you exceed the force, then the motor will miss steps, and you will loose the location. good reason for a switch to verify the home location one one end. Nothing (but maybe the part) will be damaged.

If you cut aluminum, you can test it by taking successivly deeper cuts until it misses some steps. then back of to 80% and use that as the limit.

If you are moving a no-contact load, like a PCB, it should be fine.

If you have problems, drop me a line and I'll loan you a double stack NEMA23 so you can see the difference. and if you wanna play, I can loan you a NEMA34 although that is physically a large motor.