Theoretical Resonant Motor

Sort of yes.

If a normal three phase motor is rewound to have all three phases set up as independant center tapped windings each respective phase can be set up with a pair of switching devices and a simple feedback circuit to make each operate as a simple push pull oscillator.

The spinning rotor will keep the three phases running at the correct average phase angles while the overall speed will only be limited by inductive losses or a specifically designed LC tank circuit setup incorporated into each respective phase oscillator circuit.

Given a DC power input a self running AC induction motor could be fabricated that has its speed controlled only by the limits of the LC tank circuits.

In theory anyway.

I think you may be referring to ferro-resonant circuits? Useful in some types of transformers. Have not seen it applied to motors.

http://en.wikipedia.org/wiki/Voltage_regulator#Constant-voltage_transformer

What is the intended goal? More efficient electric motors?

Resonant circuits do not produce any energy. They only store/accumulate existing energy in sometimes useful (and sometimes destructive) ways. A resonant motor may be beneficial at one narrow fixed frequency. Not good for applications requiring variable motor speed control.

"Not good for applications requiring variable motor speed control."

Why would it not be possible to use variable circuit admittance as a means of controlling torque? In hydraulic or magnetic coupling to an induction machine does it not follow that as the torque varies the angular velocity of the rotor would change?

Again ... "What is the intended goal?"

This is important if you don't want to waste a lot of time chasing the unattainable or are trying to reinvent something that already exists and works perfectly VERY well right now.

On paper first, try designing some of the high power variable inductors and capacitors required for this concept. If the large size and high cost are acceptable, continue to experiment and learn.

In single phase AC induction motors, capacitors are frequently used to phase shift one or more windings for starting torque and/or running torque. Alternate (higher inductance) windings are also used to provide the same functions. These motors operate optimally at one RPM dictated by the motor & winding geometry and the line frequency. The cost, size, and complexity of "variable" capacitors and/or inductors sized for these motors is typically unacceptable.

Three phase motors can be FULLY controlled (speed and torque) by very small, efficient, and reasonably priced Variable Frequency Drives (VFD). These small electronic drives synthesize continuously variable frequency three phase power signals at the PROPER voltage for the chosen speed (Volts/Hertz needs to be held approximately constant up to rated speed/HP).

I'm really interested in understanding the intended goal. What properties (cost, size, complexity, reliability, efficiency, etc.) of existing motor/drive technology is unacceptable and needs improvement?

Hi MJB,

You are absolutely correct in that conservation of energy applies. That is to say that one can not get out of the circuit more than one puts in. And that may also imply that it is pointless because converting energy in the tank circuit into rotational (mechanical) energy may dampen the tank circuit too severly.

I'm trying to think outside the box a little bit, mostly to understand what might be the reasons why this is not in any literature. I'm not likely to be the first one to toss around such a concept. I don't think conventional controls would be very useful.

What I'm kicking around is a concept. The other questions I asked in the original post was intended to turn loose some thought processes. What I'm thinking about is whether or not it might be possible to have a low RPM drive at full (rated speed)torque by breaking some of the conventions without breaking rules of physics.

In a purely reactive circuit, no work can be produced. And a tank circuit would be just that except for parasitic losses. To produce rotation, I'm going to have to cause some sort of phase shift so I have a "Real" component and thereby produce some torque. The question really is could that be controlled in such a way as to produce a low RPM motion at high torque? I wouldn't think a conventional motor would work simply because the core would be saturated too quickly (before I reached resonance). Consequently, a new core would be required. When was the last time a really new type of motor was introduced anyway?

There is nothing wrong with thinking outside the box. Some new positive things have been created that way. Just don't forget that the box is there for a reason. Inside the box is where you find the stuff we already "know" works well.

"...whether or not it might be possible to have a low RPM drive at full (rated speed)torque" ... I'll try to explain why this is readily available now.

It is easy to drive a DCPM (Direct Current Permanent Magnet) motor down to about 10% of its rated speed using a PWM DC supply (or a variable phase rectified AC supply) and some armature current feedback for good speed control accuracy. Accurate speed control below 10% rated speed is possible if an additional speed measuring feedback signal is used (e.g. a tachometer or an optical rotor speed/position sensor). Full rated torque (sometimes MORE torque) is typically available from zero to full rated speed. I've built a few of these drives and they performed as expected.

For 3-phase induction motors, it is easy to drive these motors down to 1 Hz (1/60 of full rated speed) with a VFD. Full rated torque (sometimes MORE torque) is available from zero to full rated speed (RPM). I have NOT built VFDs, but have purchased new ones for work use and acquired used/surplus ones for personal home use. The performance seems amazing the first time you see it. I've used VFDs many times for many years and am still impressed how well they control 3-phase motors.

For DC drives and VFDs, full motor torque is available from zero to full rated speed.

"I don't think conventional controls would be very useful." ... Conventional controls as described above are very useful for many applications. Can you define what problem you are trying to solve?

I suggest you find a way to "play" with a real DC motor & controller or a 3-phase motor & VFD. Once you see these systems in action you will understand how good existing motors & drives really are.

The reason it is not good for variable speed motors, is that variable speed requires a variable frequency of AC, and LC tank circuits have a naturally high Q, or 'selectivity, they resonate at one frequency, and other frequencies are 'dampened out,' and attenuated to zero.

LRC circuits have the same high-Q as LC circuits, resistors do not 'phase-shift' the current like resistors and capacitors do.

I'm with you on this one Joe. Like you I am outside my sandbox but its fun to look inside.

The idea of using circuit admittance (1/Z) to control the throughput power of induction machines has been an interest of mine for a very long time. Especially as how it may be applied to hybrid power control.

I believe the frequency of an induction machine approximates the number of poles times RPS or Np *(2π/w) where w is the angular velocity of rotor.

This makes the frequency quite low which may make the size of the inductance and capacitance quite problematic.

The advantage here is there would be no switching losses with near infinite variability of throughput power (ΔΓ*ω) That is; infinitely variable torque at any given ω within the limits of the power supply.

Also, I would not limit your control to variable capacitance. It might be possible to put a variable inductor in series with the motor or generator inductance, perhaps by using an adjustable relay to move the core of a coil in and out of the windings, combined with the simultaneous variation of plate area capacitance.

Another approach might be to manipulate the output frequency of the power source and feeding it through a fixed frequency circuit.

In equation it might look something like this for a parallel LC control circuit.

Np* (2π/w ) = 2π (√1/LC)

Np/w = √1/LC

(Np/w) ² = 1/LC

Where Np = the number of poles

w = angular velocity of the rotor

L = Circuit Inductance at resonance

C = Circuit Capacitance at resonance

In turn I believe the Z would look something like

Z= L / (C(Xl -Xc)) Where Xl - Xc = net reactance

In reference to the variable inductor, why not use a variable core that is controlled via series connected feedback from the load current, sort of a modified LVDT? Let the motor current drive the fixed winding, and use the variable secondary for feedback.

Just thinkin'......

Oh, I almost forgot. The new electronic ballasts use a variable Q to start and regulate a fluorescent lamp. Before it starts, the lamp has an impedance in the megohm range and it need upwards of 600 Volts. After it starts, the Q of the circuit changes so as to regulate the current and the lamp voltage is then a function of the length of the plasma with an effective impedance that looks resistive at about 200 to 300 ohms with a plasma voltage of about 10 volts per inch of plasma (at rated current). That is another resonant application!

Boy did I screw that one up !!!!

RPS = w/2π not the inverse.

I guess its not anyone's responsibility to fix my ignorance.

N_{p *} w/2π = 2π (√1/LC)

(N_{p *} w)² = 1/LC when the input frequency is at resonance with the rotor??????

Is that right?

Come on you guys; give this demented old man some help here.

This is a very interesting discussion. I would like to congratulate all in not turning it into a ridiculous topic. I have encountered several motor designs, when coupled with modern variable speed AC drives, produced apparent results with unbelievable efficiency that could not be obtained when the motors were connected across the line. I had a theory as to why, but never the time to prove it. The idea that a resonance might have been present due to the drive characteristics lights up my mind with a new possibility. Resonance?? Modern variable speed AC drives can sense and regulate Quadrature and In Phase currents separately in the form of pulses that are sent and the reflection wave sensed and a corrected pulse sent, many times per second. (This is not the cheap drives) This is a very crude description of these drives, but those who are not familiar will get the basic idea. Perhaps resonance could explain some unusual results that are sometimes obtained. Then again it may be simply the algorithm within the drive.

I have investigated failures caused by resonance, mainly on energizing or de-energizing circuits but have never thought about it on running circuits and as an idea to create an improvement in design. Interesting.

In a high Q resonant circuit, most of the energy is stored within the L and C. If your motor is required to do any work, this will look like a resistance (Power In -> Power Out) and will lower the Q. It's an interesting idea. If you didn't have a steady load on the motor but an intermittent load, I expect it would behave as if it had a heavy flywheel attached. Perhaps there could be an application where the weight of a flywheel would be a disadvantage.

At this point I'm worried about the dL/dt reaction as the rotor reacts with the stator. That variable Q circuit used in electronic ballast for fluorescent lighting is quite clever. See my post #11. I'm not proposing using a ballast, just tinkering with the concept for the sake of discussion. I can't be sure that I've considered this concept from all angles by myself. Just thinking out loud.

I've been told that a starting capacitor is used to "limit the current" and I honestly don't remember enough about the vector analysis to draw it out at the moment, but I don't think anyone has made a reference to the effective Q of the starting circuit for a motor with a starting capacitor. Again, just scratching my head a bit. (I hope it's not fleas!)

A starting capacitor is used to shift the phase of current in starter windings. In a single phase motor, the stator magnetic field does not rotate but just switches back and forth and there is no tendency for a stationary rotor to start turning. The starting windings with the current phase shifted by the starting capacitor generate a rotating stator field which provides starting torque. The starting winding is switched out with a centrifugal switch after the motor starts. In some designs, the capacitor circuit is enabled all the time and the capacitor is called a running capacitor.

Link to resonant electronic fluorescent ballast:

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=1620667&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D1620667

My understanding is that until the lamp "fires" it is a very high resistance and requires a high voltage to start. Once it fires, it is a low resistance and if the high voltage were maintained, the current flow would destroy the lamp.

A resonant circuit develops a high enough voltage to "fire" the tube which is a high enough resistance to not load down the resonant circuit. Once it fires, it's low resistance loads the resonant circuit (lowers the Q), so the proper lower voltage is then applied to the tube.

There seems to be some confusion here, with some comments referring to DC motors and Stepper motors. Since you are asking about making the stator coils part of a resonant circuit, you are obvious referring to an AC motor, either synchronous or induction. At this point I'm worried about the dL/dt reaction as the rotor reacts with the stator.As far as I know, there is not much change in the stator inductance due to the position of the rotor in an induction motor. So in this case, I don't believe dL/dt is a problem.
In a synchronous motor, the impedance of the stator coils can be inductive or capacitive, controlled by the excitation of the rotor. This effect is used for power factor correction in power systems.
See: http://en.wikipedia.org/wiki/Synchronous_condenser
I found a patent description which is very similar to what you are proposing:
https://www.google.com/patents/US7034498

You wrote:-

I came across an interesting article that stated that, "resonant circuits are only used in RF applications" which may or may not be the case.

That is not the case, whoever wrote that did not know what he was talking about. I am using one in a "Dog Scarer" that I am developing for the Postal workers here, for example.....Many other "things" use resonance to improve efficiency for example......its a common way to do things, usually with Oscillators for example.

(I am loath to say "and thats that" because I bet 50 other people will come forward with many other "things" using resonance in an LC circuit.....!) I simply do not know exactly "everywhere" it is used......

Resonance can also be a problem, for example with stepper motors, that can actually reverse the direction, when it is not intended to be reversed.....due to resonance.

Some electronic filters use resonance........

I have to agree with you. I was surprised to see that statement in print. Today, I remembered the electronic ballast which I wrote about in some posts above. And your point about problems with stepper motors has jogged my memory. Consequently, the stepper motor design is very much the wrong way to go.

I think I remember how a stepper motor is constructed. If I remember correctly, it is driven with a signal that looks a lot like the output of an encoder with quadrature. Definitely not the answer to the question.

" What I'm thinking about is whether or not it might be possible to have a low RPM drive at full (rated speed)torque by breaking some of the conventions without breaking rules of physics."

I have worked with many, many types of motor controls over the last 40 yrs and con tell you that the drives using PWM modulation will do everything you are talking about.

I'm not aware of your level of knowledge on electric motors, so bear with me.

An electric motor has max torque at stall and curves downward as the rpm's increase.

Let me point you to a good website for this topic. I'm sure there are others who can point out other sites as well.

http://www.reliance.com/prodserv/motgen/b7097_2.htm

It discusses the rotor/stator relationship as a "transformer". An AC induction motor changes the input frequency from interaction between the stator and rotor currents due to the amount of slip present.

Also, there are drives out there that model the motor+system mass and maximize speed and performance without having the resonance issues.

How this helps. Keep being curious about things.

Thanks Bill, I gave you a GA point for your information. I am aware of the PWM drive and have even built a very low power version using a Basic Stamp board just to play around. It works very nicely! In general, I don't feel like I get enough opportunity to arrive at that intuitive understanding that you must have. And this exercise is most likely pointless considering everything else that is available.

All I really wanted was a small powerful low RPM motor that has some mussel without having to get into fancy control circuits or gear reduction devices. Perhaps this thread is done, but I do appreciate your input. Thanks for the link!

You wrote:-

All I really wanted was a small powerful low RPM motor that has some mussel without having to get into fancy control circuits or gear reduction devices.

A common and simple way to achieve this is a DC motor and a PWM circuit.....some small ones use a 555 and a transistor, others use a PIC or similar and a transistor or two......there are MANY more ways to do this......

Is this what you are looking for?

https://www.google.com/patents/US7034498