Is there an electric motor type which gives torque at low current

[Infraviolet]

Ok, the title isn't long enough to properly ask what this thread is about so...

Is there any type of electric motor which, unlike the usual scenario, can give a large holding torque with minimal current requirements and then consumes power primarily as a consequence of providing higher speeds?

Particularly because when you consider a motor which is holding a position, it is not doing any work over time, it is just holding against a torque load (say a pulley or something), so at the level of an overview of the system, torque at low speed should not require a power input. As soon as you start moving against a torque you must consume power, but at standstill the power a motor consumes simply to apply a torque is simply being wasted as heat.

To consider some motor types which given counterexamples to the type for which I am wondering about the existence of:

A DC brushed motor has torque, current and speed relationships which mean, powered from a constant DC voltage, it draws the least current when not loaded and spinning at maximum speed. Increasing the torque load it is moving slows it down and increases the current required, right up until stall, where you get maximum current at zero speed.

A stepper motor is set with a given level of current, often fixed by typical chopper driver units with microswitches but this could be varied according to commands, which gives a given level of torque. Running faster requires a greater voltage in the motor's coils to overcome back-emf, until some high speed threshold beyond which a given power supply can no longer supply enough power to provide both the desired current level and necessary voltage at the same time. The motor, unlike a DC one with constant voltage driving,is consuming power power at higher speeds, if you consider a fixed current level, but... this is still a situation where having a higher level of current in the first place often takes more power than achieving higher speeds but with lower currents. In the motor itself, holding torque requires just as much current as torque at greater speeds, in the system of the motor and driver the power requirement rises with speed for a given level of torque, but at low speeds the level of torque required is having more influence usually than the speed.

A brushless permanent magnet synchronous motor, whether driven with AC waveforms or with DC type six step commutation, can have controlled levels of current supplied to its coils so as to control the torque produced. Considering a situation where the driver system is closed loop and knows the rotor angle, so can always energise the phases in the right fashion as to provide the maximum torque per amp... The current required, at a zero or low speed, is in proportion to the level of torque being generated by the motor. Where you've got buck-converter style driving circuits providing lower voltage higher currents in the motor, from a higher voltage lower current supply, your power requirements, for the higher voltage supply supplying the driver circuit, for a given level of torque, increase with speed as back-emf rises. Like the stepper though, and afterall a stepper and BLDC driven can be pretty similar in principle, at a given speed, or at standstill, producing a greater torque requires a greater current in the coils and a greater amount of power from the supply. Again, the amount of extra power from the supply needed so as to produce more torque is often higher than the amount needed to simply provide higher speed against zero or minimal load torque.

AC induction motors have unstable regions at low speeds, so struggle with providing torque at a standstill. This doesn't sound like the sort of scenario where a motor could be at its most efficient at zero to low speeds.

Lets ignore reduction gearboxes, as they simply shift the "problem" and let you get higher torques and lower speeds at their output whilst not changing the principle that greater torque requires more motor power, even when that torque is applied at standstill.

Is there then, any electric motor type where, unlike the examples above, you're able to hold at a standstill "for free", where the power consumes is in proportion to torque*speed, so you'd be looking at situations where a high torque at low speed, up to whatever level the system can physically deliver, needs only as much power as a running at high speed with a low torque?

This idea of a motor which gives torque "for free" (not actual zero power usage, there are always losses somewhere from every system, but "free" as in insignificant when compared to operation at speed) but then consumes power as greater output speeds are demanded would also include motor types which consume power according to torque*speed^n, so increasing the speed demands extra power more quicky than increasing the torque does. That is to say, motor types that, opposite to a simple brushed DC motor, could become inefficient at speed, but would be very efficient to provide torque at standstill.

Does such a concept exist?
Thanks

[Andy Chee]

I'm having a hard time understanding your concept.

Can you illustrate your concept in the form of a torque/RPM/power graph?  Can you overlay your concept over the top of a typical permanent magnet graph, so a direct comparison can be made.

[K5_489]

So....you're basically trying to come up with a way to violate the laws of physics?   

[BrianHG]

Get a cheap DC motor where you can damage the bottom holding the brushes and rotate their orientation so that the distance between powering the field polarity with reference to the angle of the fixed magnets orientation is really short/close.  Your motor will be run at a slower speed, but it's output torque will increase dramatically.

Your new problem will be efficiency will drop if the motor's brush angle was previously tuned for maximum efficiency.  It manufacturer alignment was previously tuned for maximum RPM, your new alignment might end up being more efficient.

[langwadt]

superconducting windings?

[BrianHG]

Example, get a DC motor like this one with a plastic bottom holding the brushes:
Cheap 12v DC motor

Next remove the bottom, sand down the register keys holding the registration angle and place it back on the motor without fastening it back on so that it can be rotated to any angle while you power the motor.

Next, power the motor on your torque testing rig and rotate that bottom messing with the brush angle alignment monitoring your readings from your power supply and torque meter to see if you can tune the effect you want.


(They used to sell motors with a lever to adjust the brush angle to offer huge startup torques without having the need for mechanical gear box with different operation settings.  3-phase BLDC motors made this technology obsolete.)

If you want low current, then you will need to do my above instructions with a 48v or 120v DC motor, powering it at 12v with the proper angle set on the brushes for high torque, just before the short-circuit point.  Your DC motor will run really slow, like only 10-100rpm, but the torque will be as much as powering it normally at the full 48v or 120v.

[pqass]

A sprag clutch on the output shaft would limit the motor rotation to one direction only.

[Infraviolet]

Regarding the answers about changing the angle of a DC motor's brushes... aren't they already aligned such that the motor will generate a field in the windings at the angle (90 degress, electrical I suspect) which will give maximum torque? Any other angle would mean the motor would have its rotor generating magnetic fields at a more acuate angle to the field of the stator, and less torque would result? Also, would this not still be a system which drew more current at stall, it would just have a lower no-load speed, so if the motor isn't already optimally aligned and changing the alignment improves it it still doesn't change the behaviour?

Superconducting coils sound wonderful, but aside from the non-existence (yet) of a rom temperature superconductor, all superconductors have a current density limit, put too much current through them and they suddenly cease to superconduct due to the strength of magnetic field around the wire which that current generates. If I remember correctly that density limit is quite low for most materials? An ideal superconductor not subject to this current limit would perhaps fit the concept. If you built a 3 phase ac brushless motor from one you could set the desired currents running in the coils so as to produce a magnetic field 90 degrees (electrical) ahead of where the rotor was, and when the rotor was static you'd never need to overcome and back-emf, a current of any level could recirculate (so long as you had some inductance in the coils, which you will have ofcourse) forever without loss. A buck-converter style driver would effectively give "infinite" amps at "zero" volts in the coils, while drawing "zero" amps at whatever voltage you supplied it with. You'd only need to put extra energy in when having to overcome back-emf, and some energy input required whenever changing the relative ratios of the currents so as to change the angle their generated field was pointed in. Obviously infitiy and zero aren't going to occur in the real world like this, and there will be practical limits on how much current can flow (like the fields from the wires causing the wires to push each other around enough as to physically damage the motor), but over whatever the practical range was you'd get a situation where the motor's power draw was determined by the product of the speed and the torque, rather than dominated by how much torque was being required (even at low or zero speeds).

I was mainly wondering if one of the "other" (not typical DC brushed, stepper, PMSM/BLDC) motor types does display this kind of "reversed" behaviour? On a graph it would mean with speed increasing on the x axis, current on the y axis, current would follow a y=mx model, with positive m as a gradient. DC motors plot a y=-mx+c on that graph, with the stall current at x=0 and a minimised current when at the no-load speed (where y=0 nearly).

As for physics violations, how would this do so? Holding something static does not require work to be done, moving a distance against a force does. Maybe I've been stupid and missed something, but as power is proportional to torque*rotational_speed why shouldn't it be possible to apply very large torques with minimal power so long as the motion is slow enough, the power requirements only climbing as speed rises.

[IanB]

Ok, the title isn't long enough to properly ask what this thread is about so...

Is there any type of electric motor which, unlike the usual scenario, can give a large holding torque with minimal current requirements and then consumes power primarily as a consequence of providing higher speeds?

Particularly because when you consider a motor which is holding a position, it is not doing any work over time, it is just holding against a torque load (say a pulley or something), so at the level of an overview of the system, torque at low speed should not require a power input. As soon as you start moving against a torque you must consume power, but at standstill the power a motor consumes simply to apply a torque is simply being wasted as heat.

Yes, this is trivial.

Just put a worm gear on the output of the motor. Worm gears cannot be driven backwards. So the motor will hold any position you set it at with power off, and will only move the output shaft if you apply power to it.

Furthermore, with sufficiently high gearing, the motor can provide infinite torque with infinitesimal input power.

From here you can figure things out with basic electro-mechanics.

[ArdWar]

Short a (permanently excited) motor winding and you get massive torque for free.

The huge holding current you often find in real applications is simply the result of rather unrealistic expectation, where motor are expected to hold an approximately zero angular position instead of approximately zero angular velocity.

[Andy Chee]

Incidentally, I have a project for a "rope tensioner" where a rope winch drum is maintained at a a constant torque by a stalled motor, but still allows for rope to feed in and out.  So this topic is of interest to me.

[bdunham7]

Is there then, any electric motor type where, unlike the examples above, you're able to hold at a standstill "for free", where the power consumes is in proportion to torque*speed, so you'd be looking at situations where a high torque at low speed, up to whatever level the system can physically deliver, needs only as much power as a running at high speed with a low torque?

As you've stated, the zero-speed high-torque case won't be "free", but in general many electronically commutated motors of various designs can do pretty much what you are asking for.  Take almost any EV drivetrain for example.  The key is in the electronic drive and control system.  To illustrate how and why, if you wanted to do this with a regular cheap brushed DC motor you'd use a variable buck converter since that type of motor needs high current to produce torque--but not very much voltage at low speeds.  The more sophisticated EV does the same thing, just in a more efficient and sophisticated way.

[Infraviolet]

I found this article:
https://things-in-motion.blogspot.com/2019/03/basic-bldc-pmsm-efficiency-and-power.html
a pretty good description about loses which cause motor effiency to be imperfect.

What is really striking is how, for a BLDC/PMSM the I^2R losses (losses primarily affected by the torque being demanded) have far more influence than the core losses (those determined by speed).

I guess one way of phrasing my question would be to ask whether there exists a type of motor for which the I^2R losses are relatively minimal compared to the core losses instead? I can see how a superconducting BLDC/PMSM would fit that description, if R=0 then I^2R will be pretty low (R won't be actually 0 with superconducting wires as the "mosfet" (or equivalent) switching systems will still have some losses which will look like resistance) , but is there any alternative design of motor which achieves the same and already exists?

[Andy Chee]

Electrically speaking, rather than a stalled motor, perhaps one could reframe the question as a transformer with the secondary short circuited.

In this case, the question would be; what core material, winding topology, drive frequency, or construction, would give the most short circuit current with lowest heating.

In the transformer world, this might be LLC/ZVS.  So is there an equivalent control methodology in the electromechanical world?

[IanB]

I guess one way of phrasing my question would be to ask whether there exists a type of motor for which the I^2R losses are relatively minimal compared to the core losses instead?

You have ignored my answer, which is the perfect one. A drive which holds its position at unlimited torque with no power supplied has no losses and therefore has perfect efficiency.

If you insist, for whatever reason, to stay in the electrical domain, then you construct a motor which is the electrical analog of the worm gear. You make the motor have a very large diameter armature, with a very large number of poles. Torque is proportional to the armature diameter, and having a large number of poles (hundreds of them) is the electrical analog of a high gear ratio.

[Ground_Loop]

Servo systems will hold position with substantial torque, but you will also have winding losses.  If I need to hold position for long periods I use a friction brake which use no energy when engaged.

[bdunham7]

So is there an equivalent control methodology in the electromechanical world?

Yes.

[SteveThackery]

An ultrasonic motor might be another option. They have exceptionally high holding torque without consuming any power at all. They will rotate with extremely high torque well below 1 rps. They are compact.

I don't know how efficient they are - that would require some research.

Disadvantages are that they won't go very fast, they are rare and expensive. Also they need custom controllers, although the drive is actually very simple to implement, so that's not really an issue.

[Xena E]

Ok, the title isn't long enough to properly ask what this thread is about so...

Is there any type of electric motor which, unlike the usual scenario, can give a large holding torque with minimal current requirements and then consumes power primarily as a consequence of providing higher speeds?

Brake motors exist.
electromagnetic actuator holds off the shaft mounted disc brake whilst the motor is being powered, and locks them when unpowered.

Simples!

X

[macboy]

I agree with the worm gear idea, it does nearly exactly what you say you want. I am curious why you have ignored this suggestion?

[BeBuLamar]

A motor when it's running at zero speed would need a lot of current and little voltage. The power will be less than when it run at high speed but the current is high and not low. Without the current you don't have torque.

[JustMeHere]

Geared

[Infraviolet]

Yes, gears are the practical solution, though with enough gearing you're unable to have the kinds of fancy direct drive (springiness, backdrive when you choose to reduce motor power...) tricks you can do when gearing isn't present. I was mainly asking this thread as a theoretical exploration though, just wondering if there wa some class of motor which, in its natural ungeared state has power demands largely controleld by speed rather than torque.

I guess the closest thing would be a motor, it would seem of either DC brushed, brushless BLDC/PMSM or stepper, where a constant current rather than constant voltage is used in the motor's coils. That way, for the same torque level, your buck-converter style driver does draw more power at the higher voltage supply when your motor turns faster. Which is somewhat opposite to the constant voltage brushed-DC behaviour of maximum power draw at stall.

I guess, though, short of the sort of huge pole count, wide diameter system (undoubetdly pretty heavy in mass of copper) which IanB mentioned as the electrical analogue of reduction gearing, that electrical motors, of all types simply don't naturally give a particularly pleasant Newton-Metres per Amp ratio at their direct shaft?

[IanB]

You have the same problem with motors that you have with your muscles.

If you hold out a heavy weight with your arm without moving, your arm is not doing any work as the weight is stationary. However, your muscles are exerting a lot of effort to keep the weight up, and are consuming energy in doing so.

There is probably a law of the universe that declares this must be so. Without some kind of mechanical lock, resisting a force is going to take energy.

[T3sl4co1l]

I don't know that it's a standard type, but, I suppose you could build something like a pre-biased stepper motor.  That is, one winding instead of being around an iron core, is wound on a permanent magnet; energizing the coil is necessary to counter the magnet and allow normal stepping.

Probably these aren't very practical, because cogging is generally considered an undesirable property of a motor.  Steppers already have significant cogging of course, but much much more when energized.  Replacing a winding with a magnet would allow that holding force to persist without applied power, assuming you only need full-step holding of course (static fractional stepping isn't going to happen here).

Mechanical solutions, like worm drive or (solenoid or otherwise) clutch (brake), or just, enough holding friction and a brute force motor to overcome it, are preferable.

Related, there isn't a direct equivalent for the title, but torque at low RPM describes a series-wound motor.  This has a quadratic current-torque curve (more current means more field to push against), and an inverse torque-RPM curve (since EMF reduces current and thus field strength).  Perhaps at 0 RPM (and thus little applied voltage), the torque curve could be steep enough to be useful here; but it would be even nicer if the curve were just high all the way across (i.e. linear, with the same steep slope as the quadratic has at rated current) -- which describes an ordinary PMDC motor.

Or, if we took a PMDC motor and magnetized the rotor, it can cog in place (a "360° step" motor), but then we have to spend additional power to counter the magnet, whether by applying excess current during part of the rotational cycle or putting a counter-field winding on it (connected with slip rings) -- either way, we inevitably lose winding and/or iron cross-section on the rotor, reducing the power-efficiency curve.

Since a clutch/brake can do its job with very little displacement (or indeed be bistable, costing asymptotically down to zero average power), it's a superior solution, avoiding the compromise of a purely electromagnetic system.

Tim