Discussion of motor characteristics

WillAdams
Posts: 8488
Joined: Mon Apr 09, 2012 6:11 pm
Location: Pennsylvania --- south of the Turnpike, East of US-15
Contact:

Discussion of motor characteristics

Post by WillAdams » Wed May 21, 2014 11:06 am

Came across an interesting thread on Sawmill Creek: http://www.sawmillcreek.org/showthread. ... r-question

Distilled it down to the following --- seem reasonable?

==Characteristics== (available at: http://www.shapeoko.com/wiki/index.php/ ... cteristics )

* Higher torque means that a motor can manage to overcome greater resistance without loosing steps.
* Stepper motors produce more power at lower speeds
* Power supply requirements --- Take the total amperage of all motors. Minimum power supply should be at minimum 70% of that number.
* (For larger machines?) 2--3 turns of the motor for each inch of travel is one suggested guideline
* Micro-stepping --- Torque drops off 8--15% for each level (1/2, 1/4, 1/8, etc) of additional micro-stepping.
* Step resolutions higher than 1000, typically place unnecessary limits on the machine.
* Torque is usually specified as stall torque --- Higher quality motors have lower inductance and maintain a larger percentage of their stall torque at higher speeds. It is possible that a cheaper stepper may be rated for a higher torque but not perform as well as a higher quality stepper rated for less torque.[http://www.sawmillcreek.org/showthread. ... ost2268920]

Anything which should be expanded upon? Corrected?
Shapeoko 3XL #0006 w/Makita RT0701 Router w/0.125″ and ¼″ Elaire precision collets
Nomad 883 Pro #596 (bamboo)

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Discussion of motor characteristics

Post by cvoinescu » Wed May 21, 2014 1:05 pm

WillAdams wrote:* Power supply requirements --- Take the total amperage of all motors. Minimum power supply should be at minimum 70% of that number.
Not exactly true. Higher supply voltage decreases current requirement and lower voltage increases it. Motor current is not the whole story, either. You'll get away with four 1.7 A NEMA17 motors on a 24 V 5 A supply, but you'll need more current from a 12 V supply. If you switch to small 1.7 A NEMA23 motors, the 24 V 5 A supply won't be enough; you'll need 36 V 5 A or 24 V 7.5 A, as a ballpark.
WillAdams wrote:* Micro-stepping --- Torque drops off 8--15% for each level (1/2, 1/4, 1/8, etc) of additional micro-stepping.
Patently false. We've been there, talked this to death -- it's a common misconception based on unstated assumptions that make a huge difference in how torque would be measured. In brief, microstepping does not reduce torque at all, but it does not increase the precision with which the actual position of the motor follows the commanded position. Microstepping increases the resolution to which the motor can be commanded, but, in absolute terms, the actual position still deviates about one full step from the commanded position under full load, regardless of microstepping. The statement is only true if we measure the torque for a deviation of one microstep, which is only relevant in some situations.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

WillAdams
Posts: 8488
Joined: Mon Apr 09, 2012 6:11 pm
Location: Pennsylvania --- south of the Turnpike, East of US-15
Contact:

Re: Discussion of motor characteristics

Post by WillAdams » Wed May 21, 2014 1:17 pm

Thanks!

I'll wrap my mind around this stuff yet.

Changed the former to:

* Power supply requirements --- 70% of the total amperage is one guideline, but this interacts w/ the voltage of the power-supply as discussed above.[http://www.shapeoko.com/forum/viewtopic ... 03&p=26352]

deleted the latter since it's discussed farther up on that page --- let me know if there was some better discussion which I missed which should be added.
Shapeoko 3XL #0006 w/Makita RT0701 Router w/0.125″ and ¼″ Elaire precision collets
Nomad 883 Pro #596 (bamboo)

tm america
Posts: 84
Joined: Tue Feb 04, 2014 5:58 pm

Re: Discussion of motor characteristics

Post by tm america » Mon Nov 03, 2014 12:44 am

I would like to add my finding using different motors and electronics.. I had been building my machines using a tb6560 with 300oz nema 23 motors using a 24 volt psu and was limited to around 320 ipm in x and y but only able to get around 17 ipm in z without loosing steps .Then i started using ramps 1.4 with a 12 volt psu and 72 oz nema 17 motors.. and power and feedrate skyrocketed.. i am able to run up to 700ipm on the x and y before loosing steps and i can run 300 ipm on the z axis without loosing steps.. but i ussually limit it down to around 600 ipm for x and y and 150 for the z ..This way i have some buffer ..Took me some time to figure out how this could be possible smaller motors being more powerful than a motor with almost 5 times the holding power?Seems to me the answer is in the pulse time since electricity travels at a set rate it takes longer for the pulse to get to the other side of the winding on the nema 23 than is does on the nema 17 .so the motor has a full circuit for a longer time with the smaller winding .i would never believe this unless i seen it myself.i had the nema 23 on the machine only able to run 15 ipm on the z.Then i switched the motor to the nema 17 and speeds went up to 300 ipm....
For most people they probably never need the speed of a nema 17 doing milling so the extra holding power of a nema 23 might be better but for 3d printing and cnc plasma the nema 17s work much better for the app
also i would like to say i have tried pretty much ever belt-pulley combo and used different microsteps.. and i find that using nema 23 motors micro stepping ussually takes torque away as the pulses aren't long enough for the power to complete the path thru the long winding and complete the circuit but on nema 17 motors it has the opposite effect .more pulses mean more time the motor is under power per inch...It is all about having the right combo..

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Discussion of motor characteristics

Post by cvoinescu » Mon Nov 03, 2014 1:45 am

I'm sorry, but your theory about the time it takes pulses to travel through the windings is wrong. Electricity would have to be millions of times slower for that to have a noticeable effect. And if NEMA 23 motors were really that bad, why is anyone still making them -- why do people still buy them?

A more likely cause is that, while the TB6560 is a good chip, the Chinese products using it are not designed properly (and there's also the possibility that the chip itself is counterfeit). Many have component values way outside the range recommended in the TB6560 datasheet, for no obvious reason -- they're not cheaper than the correct values, although they also skimp on some of the optocouplers, which can cause all sorts of problems.

Out of curiosity, what are the specs of those NEMA 23 motors?
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

tm america
Posts: 84
Joined: Tue Feb 04, 2014 5:58 pm

Re: Discussion of motor characteristics

Post by tm america » Wed Nov 12, 2014 5:49 pm

The motors were 1.8 step angle 300 oz 2.0 amp motors if i still had them here i would get the part number off them but the machine was sold and so was the motor i took off the z axis..So could you then tell me how a nema 17 can have more power and speed than a nema 23 in the same app same controller everything the same .Every one i have done this with same results one was using tb6560 the other used tb6600.I have yet to try it will the ramps 1.4 - pololu 4988 but i will on the next machine just to verify the results.On the x and y it didnt matter but on the z axis feedrates went for between 15 to 17 inch per minute to 175 to 300 inch per minute without loosing steps.And when i say same app i mean i unbolted the nema 23 and bolted on the nema 17 ..no other changes besides the coupler needed to fit the 5mm shaft. I find it hard to believe the tb6560 and tb6600 chips don't properly run a nema 23 as they are in use all over the world doing just that?..

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Discussion of motor characteristics

Post by cvoinescu » Wed Nov 12, 2014 6:38 pm

tm america wrote:So could you then tell me how a nema 17 can have more power and speed than a nema 23 in the same app same controller everything the same .
I can't, because it can't. I don't know what you're doing wrong, or what's wrong with the driver you're using, but a NEMA 23 motor is more powerful than a NEMA 17 under most reasonable circumstances.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

CastIrony
Posts: 650
Joined: Mon Jul 21, 2014 10:21 pm
Location: Portland, OR

Re: Discussion of motor characteristics

Post by CastIrony » Wed Nov 12, 2014 6:59 pm

cvoinescu wrote:Patently false. We've been there, talked this to death -- it's a common misconception based on unstated assumptions that make a huge difference in how torque would be measured. In brief, microstepping does not reduce torque at all, but it does not increase the precision with which the actual position of the motor follows the commanded position. Microstepping increases the resolution to which the motor can be commanded, but, in absolute terms, the actual position still deviates about one full step from the commanded position under full load, regardless of microstepping.
One full step corresponds to about 0.2mm on the X and Y axes with the default motors and pulleys, right? That seems like a high amount of error, even if it's only a worst-case.
Shapeoko 3 #192 - Complete! Needs limit switches.

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: Discussion of motor characteristics

Post by cvoinescu » Wed Nov 12, 2014 11:15 pm

CastIrony wrote:One full step corresponds to about 0.2mm on the X and Y axes with the default motors and pulleys, right? That seems like a high amount of error, even if it's only a worst-case.
That's correct. Remember, though, that's under full load, which, with the default motors and pulleys, is 7 kgf (69 N, 15.5 lbf). For a Shapeoko, that's a lot of force.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

tm america
Posts: 84
Joined: Tue Feb 04, 2014 5:58 pm

Re: Discussion of motor characteristics

Post by tm america » Fri Nov 14, 2014 8:55 pm

How could there be something wrong ..if i take one out and put the other in place and it moves faster and has more power-not loosing steps..?You can come by my show and i will show in person if you would like .. i didn't believe it either but facts are facts the eye doesn't lie..All i did was swap the nema 23 for the nema 17 and no other changes.same steps same settings same lead screw.there has to be a reason .. i agree bigger ussually means more power but it has to have a well tuned setup to function at its true potential ....

Post Reply