×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Rotation by Lorentz or by Faraday?
2

Rotation by Lorentz or by Faraday?

Rotation by Lorentz or by Faraday?

(OP)
The following assumes the application of a constant DC current to rotors that have no rotational inertia to start with.

Some illustrations show the flux from the pole of a permanent magnet causing rotation by 'pulling' the laminate core of a coil wound electromagnet into alignment with the permanent magnet; before stopping. http://www.unicopter.com/1904_7.gif

However, other illustrations show the flux from the pole of a permanent magnet causing a portion of a coil of wires to transition across the full width of the flux; before stopping. http://www.unicopter.com/1904_8.gif

Are both of the above statements correct ~or~ where is my limited intelligence coming off the tracks.


Thanks,
 Dave

RE: Rotation by Lorentz or by Faraday?

Your first link shows an external rotor. That may be causing some confusion. The rotation would be the other direction if this were an internal rotor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

I think the first illustration is supposed to be a synchronous motor.  The magnets on the outside represent a rotating field without depicting either iron or copper.

The 2nd illustration illustrates interaction of current located within a magnetic field and you might qualitatively see that it results on force on the conductor in this case.

The exact question of exactly how / where the torque-producing force is one that is near and dear to my heart.  Not so simple as it may seem.  For machines with conductors located within iron slots, the normal torque-producing force acts primarily on the core (not the conductor), even though you can come pretty close to calculating the correct torque producing force using a force on conductor equation (F=q*v X B = L i x B) under the incorrect assumption that the conductor is located in the airgap flux.

I have written a short and long whitepaper on the subject, along with video and some other assorted stuff.
http://home.comcast.net/~electricpete1/torque_web/

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
Thanks gentlemen.

It's a small world. Peter, your short paper was read and printed out a while back. Then after entering the above posting, I took it out and read it again over coffee. Thank you for the informative and readable paper. Unfortunately, it does not appear to answer my weird question.

My query was not well presented. The following sketch includes the second sketch above (b), plus a third sketch (c). http://www.unicopter.com/1904_13.gif

Without considering the mechanical aspects of the 'motors', there are two electrical questions.

 ~ In sketch (b), will the conductor at location A move to location B before it stops due to lack of force?
 
 ~ In sketch (c), there is a row of adjacent and physically bonded conductors that are aligned normal to the flux path. All of these conductors have their current flowing in the same direction. In addition, there is no cycling of the current direction in these conductors.
Under these conditions, will an endless column of these of conductors be continuously moving by electromagnetic force from top to bottom of the sketch?


Dave

RE: Rotation by Lorentz or by Faraday?

Quote:

~ In sketch (b), will the conductor at location A move to location B before it stops due to lack of force?
I don't believe it is necessary to talk about motion for the question if the currents and fluxes are specified.  Instantaneous force is a function of instantaneous current and flux distribution and therefore the relevant electromagnetic results are captured by describing force as a function of position for the specified distributions.  (if we were trying to determine voltage and calculate an induced current, then motion would be important).    Also we assume the return path for this current is far away in a location of zero flux.  Then the force on the conductor would decrease as conductor moved from point A to point B and continue decreasing as it moves further in same direction (because flux density B is decreasing as we move from point A to point B and beyond... F = q V x B = Length * I x B)
 

Quote:

~ In sketch (c), there is a row of adjacent and physically bonded conductors that are aligned normal to the flux path. All of these conductors have their current flowing in the same direction. In addition, there is no cycling of the current direction in these conductors.
Under these conditions, will an endless column of these of conductors be continuously moving by electromagnetic force from top to bottom of the sketch?
You have to describe your physical problem better.  But if the conductors are rigidly linked together so that their forces are summed, then there is a constant force on the assembly as it moves downward through the gap.

* I think it can sometimes be misleading to try to talk about partial systems without considering the complete loops for current and flux.  Can you describe more completely the physical system you are evaluating?   I am also guessing this will be a better question for the magnetics forum.
 

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
Peter, the following link is to the proposed motor. It is intended for a 'backpack' coaxial helicopter. The physical system is to be an axial flux motor. It will consist of a central stator, plus a rotor on each side, which are to turn in opposite directions.

http://www.unicopter.com/1904.html

The primary question is if this concept motor can be driven in a phaseless or half-phase manner. OR, if it can even be driven at all.


Dave   

RE: Rotation by Lorentz or by Faraday?

It is not a type of motor I am familiar with.  You called it "axial flux", but it looks to me as if the copper winding is formed as a toroid such that flux would flow circumferentially within the stator and not interact with the rotor magnets.  

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Also, is this a hobby project?  If so, eng-tips is probably not the right place to ask. There may be hobby sites where you can get better answers.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Actually I think I understand now the reason for calling it axial.  The flux is supposed to flow axially from the magnets, past half of the stator winding, to the core in center of the stator winding, then back through shaft to rotor plate to opposite side of magnet.    It is a similar concept discussed here on the forum before here in this thread:
thread237-184398: brushless true dc motor?
(see my post 21 Apr 07 15:20 )

Sseems to me like it could in theory create a simple dc motor without need for mechanical commutation.  I don't think everyone in the forum agreed on that point.  Maybe I am missing something.
 

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
A little more on the objective;

It is to apply electric propulsion to helicopters. Helicopters are extremely inefficient vehicles.
The 'backpack' configuration was selected for development since it offers the best payload to empty-weight ratio.

This 'strange' motor was conceived as a possible way to provide CW and CCW torque from a single motor. The question now becomes; Firstly, will it even work and secondly, will it work efficiently.

The idea (or hallucination) started with the consideration of a Rotor-Stator-Rotor axial flux motor and then the consideration of how to get the rotors turning in opposite directions. This initial concept is here; http://www.unicopter.com/1740.html.
________________________________________

Peter, I read your thread (21 Apr 07 15:20 ). It appears that my idea is a duplication of your earlier one. Did you ever come to a final answer?


The following drawing is an exploded version of a drawing on the web page given in a previous post. This shows the flux path.
this flux path differs slightly from that on the prototype.

http://www.unicopter.com/1904_14.gif

The flux flows from the magnets in one rotor, through the adjacent conductors, through a radial laminate core, through the return conductors, through the magnets in the other rotor, and then back via a very thin air gap to the originating magnets. The radial laminate in the core is intended to direct the flux from one side of the core (flattened doughnut) directly to its other side.  

The hope is that the opposite flow of electrons in the conductors will drive the permanent magnet rotors in opposite directions.
 

RE: Rotation by Lorentz or by Faraday?

I have not done my prototype yet, but I have a renewed inspiration and an idea how to make my construction job a little easier. Maybe this weekend (or maybe not).  I was mistaken to say mine is the same as yours.  Mine was radial flux. Yours is axial flux. There are still similarities in the approaches. If successful, they would both be dc motors working without commutation (if such a thing is even possible).   Yours has an interesting twist to take advantage of the opposite polarity of forces/torques produced on the two halves of the stator coil, since counter-rotating rotors is apparently something you can use.

I am pretty sure that if yours would work mine would work and vice versa.  

Here is what makes me suspect they have a chance of working.  If we draw the flux lines in the static condition, there clearly is a torque produced.  That much I am sure.  Next step... what would make that torque stop when the machine rotates? I can't think of anything.

Here is what makes me hesitate. The stumper question posed in the other thread: what is the mechanism for inducing voltage in the stator winding.  It is not a straightforward question to answer for textbook homopolar motor (using brushes) and it is not a straightforward question to answer for these motors (no brushes). I think to answer it one way or another would require careful application of reference frame theory converting B to E based on v.   I have not done that.  I think it would be easier and more productive to try to build it.

Have you tried yours?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Pete, I took the original sketch to be stationary coils and an external rotor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

You're probably right.  The annotation suggests some kind of mechanical commutation.  Not a very descriptive or useful figure.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
Peter, yes they were tried.

The first test, shown in this picture http://www.unicopter.com/1904_3.jpg, did not rotate either rotor. Perhaps this was due to; the center of the stator being a plywood disk, plus the other three air-gaps are quite large. In addition, the permanent magnets are only ceramic or ferrite.

The second test is shown in this picture http://www.unicopter.com/1904_10.gif. The + and - magnets only cover about 20-degrees of rotation. The U shaped single wire moved about 10-degrees when the current was turned on, and it moved about the same amount in the other direction when the terminals were switched.

This resulted in the following considerations;

 ~ If a CONSTANT dc current flowing through a wire cannot create a constant force; what about using PULSE Width Modulation to put a non-reversing current through the wire?
 ~ Might PWM be required anyway, since it may be the best means of varying the speed and/or torque?
__________________________

waross, you are correct.

It is Rotor-Stator-Rotor and the hope is that the electromagnetic activity at one side of the stator will not detrimentally effect the activity at the other side of the stator.

Dave
 

RE: Rotation by Lorentz or by Faraday?

Yes, the experts are not always right. But I believe that they are right much more often than is suspected.
A review of the basics of magnetism may be in order.
An energized coil will exert a magnetizing force. The strength of the magnetic field is roughly the magnetic force divided by the weighted length of the magnetic path. The weighing is based on the magnetic permeability of the materials that make up the path. One inch of air or plywood is equivalent to about 800 Feet of core grade iron. The textbook drawing of a wire in a very large air gap does work but if you want more force than a fraction of the torque of a d'Asonval meter movement you should spend a few hours reviewing the basics of magnetism and then seriously shortening your air gaps.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

Waross - Just to clarify - what is it that the experts say?  Is it written somewhere that a brushless dc motor cannot be built?  If it is written somewhere I'd like to read it?

On the other hand, I agree on is not built anywhere... that makes me suspicious there is a reason it is not built.

Alos, where is the large airgap you are talking about?  Here is the flux path:
http://www.unicopter.com/1904_14.gif
There is no doubt the designer of this device is aware of the need to minimize airgap distance since all airgaps are minimized.

 

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Quote (electricpete):

Waross - Just to clarify - what is it that the experts say?  Is it written somewhere that a brushless dc motor cannot be built?  If it is written somewhere I'd like to read it?

On the other hand, I agree on is not built anywhere... that makes me suspicious there is a reason it is not built.
Just to clarify the first part of my quote, it is not argumentative... it is what it says: asking if this configuration is addressed somewhere.

As per 2nd part, I certainly agree that there is strong evidence to be inferred from the fact that motors like this have not been built.  I have zero doubt motor designers no boatloads more about their craft than me.  

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Also I put my linked file from my other thread onto an attachment of my post 25 Mar 10 12:03t since my ISP changed and link no longer works.

I have another way to think about my drawing that suggests why it might not work.  Replace the stator core with a permanent magnet which produces the same flux pattern.  In that case the stator-replicating permanent magnet would be toroidal magnetized in the circumferential direction (perhaps with an axial gap.).  Now looking at orientations of the stator-replicating permanent magnet and the rotor permanent magnet, we don't see any attraction created even in the static condition.   Hmm.  

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Quote (electricpete):

I have another way to think about my drawing that suggests why it might not work.  Replace the stator core with a permanent magnet which produces the same flux pattern.  In that case the stator-replicating permanent magnet would be toroidal magnetized in the circumferential direction (perhaps with an axial gap.).  Now looking at orientations of the stator-replicating permanent magnet and the rotor permanent magnet, we don't see any attraction created even in the static condition.   Hmm.   
My comparison above is not valid for analysing static torque in my drawing..... Replacing stator core with circumferentially-magnetized toroidal permanent magnet does not properly capture the effects of the high permeability of the  core which directs the flux from the rotor permanent magnets so that it passes by through only half of the stator winding loops.  In contrast, the flux from rotor permanent magnets would pass thru both sides of a stator-replicating toroidal permanet magnet.   
 

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Hi Pete;
My comments were based on a picture of a toroidial coil on a plywood core, and magnets a long way away from the flux which could be expected to be mostly inside the toroid. One of the advantages of a toroidial coil.
The second link in the first post may produce an order of magnitude less force than a d'Arsonval meter movement. Only one conductor instead of many and no return path in the flux field.
The first picture posted shows a sensor to switch the coils on and off. Switching on and off rather than reversing the current as is done in a brushless DC motor may give the same performance as a broken brushless DC motor.
If Intermesher wants rotation with a steady DC current look here:
http://en.wikipedia.org/wiki/Homopolar_motor
and here
http://en.wikipedia.org/wiki/Homopolar_generator

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

(OP)
The following is a questioning of the most rudimentary part of the large motor question.
 

Lorentz force shows that if a current carrying conductor is located in a flux path the conductor will move.

This sketch is the Lorentz force diagram in a slightly different format.
http://www.unicopter.com/1904_15.gif

It shows a pair of ring shaped permanent magnets, which are fixed in space (stator). A single wire conductor is located between the two magnets and this conductor is free to rotate about the axis that is common to the ring magnets.  In some un-described way a current is applied to this conductor.

The questions then become;
  A/ Will the conductor continuously rotate about the axis if the direct current is a constant one?
  B/ Will the conductor continuously rotate about the axis if the direct current is a pulsed one?
  C/ Will the conductor oscillate if the current is an alternating one?


Dave ????
 

RE: Rotation by Lorentz or by Faraday?

(OP)
Bill,

Thanks for the lead to homopolar motors.

It's too late to look deeply into tonight.(2:00 AM) :)However, it shows that it is possible. Perhaps today's super magnets will make it more attractive, particularly for a counterrotating application. Putting the coils in the stator also eliminates the slip rings.

It's worth looking a little further into. Thanks

Dave

RE: Rotation by Lorentz or by Faraday?

Quote:

The questions then become;
  A/ Will the conductor continuously rotate about the axis if the direct current is a constant one?..
  

As I see your most recent post is a simplification of the motor for purposes of clarifying the underlying principle.   It looks at the same problem in a different inertial reference frame.  As you no doubt know, the force is the same in any inertial reference frame. If it works on dc, the principle is the same as your earlier motor and the motor I proposed.  

Namely, the principles which suggest these motors will produce torque:
1 – We know there is force F=qVxB = Length I x B on a current carrying conductor in a magnetic field
2 – We think that that F=qVxB = Length I x B force still occurs when the conductor is moving relative to the magnets that produced the field (as long as current and field don't change).
3 – We think that at least part of the equal/opposite reaction force associated with that force on conductor occurs on the magnet which is located on the other side of the airgap.  This amounts to torque transfer accross the airgap.

Items 2 and 3 are subject to further discussion (are they correct?) but 2 and 3 must be true for these devices to work.  Even if 2 and 3 are correct, we still need another step of analysis to understand torque vs speed characteristics.

fwiw, I have done an F.E. analysis of a simplified/reduced version of my proposed motor in the condition (attached).   It shows that in the static condition of this simplified/reduced device there is a force on the conductors roughly predicted by F=qVxB, and that the equal/opposite reaction force occurs primarily on the permanent magnets. What it means is for this particular geometry in the static condition torque is produced.

Also we know that any dc motor of the type we discuss must include generator action.  If you rotate it backwards it should produce dc voltage.  If you rotate it forward under load, it should produce a voltage opposite in polarity to the applied voltage which permits conservation of energy (I*E_induced = Pmechanical including mechanical losses).    I am still trying to figure out how that would occur.  As a start, in my motor diagram, the flux from a single magnet linked to a single partial arc coil changes over time which could possibly provide the mechanism for induced voltage.   But one would think the device should work even if the rotor was uniformly loaded with magnets, in which case it is very difficult to see any mechanism for change in linked stator flux over time.  Without that piece of the puzzle, it can't work.

Quote:

The questions then become;
.....
  B/ Will the conductor continuously rotate about the axis if the direct current is a pulsed one?
  C/ Will the conductor oscillate if the current is an alternating one?
As I see it, if these devices as built do not work on dc, then they will not work on ac or pulsed dc.   To take advantage of any of these, you would need more than one phase in your stator winding.    

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Quote (electricpete):

If you rotate it backwards it should produce dc voltage...
Correction -  I should have said polarity of induced voltage will reverse depenpding on direction of rotation and motor/generator action depends on comparison of terminal voltage to induced voltage.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Here's another typo, Pete.

Quote:

If you rotate it forward under load, it should produce a voltage opposite in polarity to the applied voltage
Induced voltage or back EMF is in the same direction as the applied voltage. As the motor is overdriven by an overhauling load or other factor, the back EMF or generated voltage increases. At the null point of changeover from motoring to generating, the applied voltage and the back EMF or generated voltages are equal and no current flows. As the machine speed is further increased, the generated voltage exceeds the applied voltage. At this point a reverse current may flow from the higher voltage source (the machine, ie; motor cum generator) to the lower voltage source IF the external circuit allows reverse currents.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

I guess "opposite polarity" or "same polarity" depends on the particular circuit your are picturing.  Here is the circuit I am picturing: draw a loop including the source voltage and the induced voltage and traverse that loop in the direction of current flow during motor operation.... then the induced voltage has a polarity opposite the applied voltage (if you disagree I can post a picture).   On a more practical level, the applied voltage is in a direction to increase current flow and the induced voltage is in a direction to decrease current flow (during motor operation).  But I personally don't care what you call it.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

The polarity question is not important and I know Bill understands it better than anyone else here.

I just wanted try out to see if I could actually draw it here in the forum using  Takoma 10-point font. Looks ok in my preview window so here goes:

DC Motor Drawings:

1st drawing highlights the series aspect of voltages.... same polarity with respect to the current loop:

╔====  -Vsource+ ======== R ====== +Vinduced- ============╗
║                                                                                                          ║
╚== =============================================╝

2nd drawing highlights the parallel aspect of voltage sources:  opposite polarity with respect to common node at bottom:

╔==================== R =========================╗
║                                                                                                           ║
+                                                                                                            +
Vsource                                                                                                    Vinduced
-                                                                                                            -
║                                                                                                           ║
╚== =============================================╝


The interesting thing: you might be inclined to change your terminology based on which drawing you look at, but they're of course the same circuit.  

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Whoops the terms opposite/same  were of course reversed in my labels. Should've been:

Quote:

1st drawing highlights the series aspect of voltages.... opposite polarity with respect to the current loop:

2nd drawing highlights the parallel aspect of voltage sources:  same polarity with respect to common node at bottom:

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Just put a voltmeter on the machine terminals and an ammeter on one of the lines. As the motor is over driven from full load to full generation, the voltage will increase slightly and the current will fall to zero at the transition point and then rise to maximum in the other direction.
The back EMF will rise more than the terminal voltage but it is awkward to measure directly.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

Hi Pete;
Try this to get your end characters to line up.
[tt]MonoSpaced Text[/tt]

The current changes direction depending on generating or motoring mode.

Try a 12 volt battery in parallel with a 10 volt battery.
The 12 volt battery is the source and the 10 Volt battery is the back EMF of the motor. Be sure to add some conductor resistance and an equivalent resistance for the commutating poles and armature resistance.
Now change the 10 volt battery for a 12 volt battery. The back EMF has risen 2 volts to 12 volts. equal voltages and no current.
Now change the motor equivalent battery for a 14 volt battery. You will have full current in the opposite direction. This agrees with both your sketches.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

As I overheard on a jobsite (actually about someone younger than me);
"He's old enough to remember DC."
We don't see the big DC machines on new construction that we used to.
I used to tell my students, sometimes the only difference between a motor and a generator is a couple of volts or a couple of RPM.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

Bill - thanks for all the continued tutorials.  However there's nothing new in there for me I'm afraid.   You'll recall this whole discussion started when you said I had a typo regarding polarity.  I think I have addressed that and I have clarified my comments applied to motoring mode.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

I was reviewing Electromagnetic Field Theory" (1979) by Zahn and found the following statement on page 431:

Quote (Zahn):

It is impossible to design a commutatorless dc machine.  Although the speed voltage alone can have a dc average, it will be canceled by the transformer electromotive force due to the time rate of change of magnetic flux through the loop. The total terminal voltage will always have a zero time average.
I have to admit I don't totally understand that (like a lot of stuff in that book), but Zahn ought to know.
 

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Check out homopolar motors, Pete. They do have sliding contacts of some type such as conductive bearings, brushes or a conductive liquid dip, but they do not have a commutator. Unfortunately they have little practical application due to their very low efficiency.
Now there's a challenge for someone; figure out how to improve the efficiency of a homopolar motor. I suspect that figuring out how to add some iron to the magnetic circuit may help.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

Thanks Bill.  But once again it is nothing new to me.   I mentioned that homopolar motors use brushes 24 Mar 10 23:23.  Zahn covers Homopolar motors on page 420.  He also makes the statement about impossibility of cummutatorless motors on pages 430-432.  From the context it is clear he is not considering homopolar motor in this category

http://ocw.mit.edu/NR/rdonlyres/resources/RES-6-002Spring-2008/Textbookcontents/chp06_text.pdf

If you don't like the terminology, feel free to take it up with him.  

The reason I quoted this is as it relates to the original post and the OP's followup questions related to his device (nothing to do with homopolar motors).  

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

To me, the Zahn quote is important because it makes me think it is less likely that "my device" and the OP's device would work.  But the exact reasons are strill tricky and I still can't understand completely why.

He highlights the subtleties of calculating induced voltage in similar situations.  A very good concise summary of approaches for caluclating induced voltages (more concise than Zahn imo) is on figure 9 the last page here:

http://www.physics.princeton.edu/~mcdonald/examples/EM/cohn_ee_68_441_49.pdf

I focus first on my device which I think is a little simpler.  It is clear there is flux cutting = motional induction on only half of each loop (the half inside the core).  There is also  transformer action, but to figure it out we need to know the return path for the flux.  Flux could either turn in direction of rotation or opposite direction of rotation once it enters on the core.  On the surface, it seems like we could control the polarity of that transformational component by our positioning of flux return paths relative to coils which would dictate whether the transformed component adds or subtracts.    But that is just a thought... could be way off base.

Now looking at the OP's device. One version of it is very similar to the homopolar motor, except that conductors are stranded / series instead of a solid conductor disk with terminals around OD and ID.  Zahn analysies  induced voltage in the homopolar motor page 423.  He uses equaiton (11) which is the Galilean transformation E = E' - v x B.   It applies when the contour moves compared to the conductor.  It is a fair analysis when the conductor is solid and contour remains inside the conductor. I'm not sure whether we can apply it when we have multiple conductors insulated from each other.

Let's set aside the question of induced voltage and just look at the force.  To me it seems pretty clear all of these devices produce force on conductor F = qVxB in static condition (anyone disagree?).  The question is then: if we hold the current constant, what is it that would happen when the rotor begins to move that would invalidate or cancel that force. It's tough for me to visualize what's going on that would prevent these devices from working.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Quote (electricpete):

I focus first on my device which I think is a little simpler.  It is clear there is flux cutting = motional induction on only half of each loop (the half inside the core).  There is also  transformer action, but to figure it out we need to know the return path for the flux.  Flux could either turn in direction of rotation or opposite direction of rotation once it enters on the core.  On the surface, it seems like we could control the polarity of that transformational component by our positioning of flux return paths relative to coils which would dictate whether the transformed component adds or subtracts.    But that is just a thought... could be way off base.
Actually, I could certainly make a case that tranformer-action emf is zero and motional emf while rotating is constant non-zero.   Considering that the rotor can be modeled relatively uniform with respect to angle (looks no different when we rotate it) the radial flux that it generates is uniform crossing the airgap.  Then even if the stator is not unfirom with respect to angle (and it cannot be due to the requirement for flux return path that doens't cross the coils), and if stator current is held constant, I think the flux linking stator coil should be uniform over time resulting zero transformer action voltage.    If true (it seems true at the moment), that supplies a plausible answer to the question I have been wondeirng since the other thread:  what is source of induced voltage.  (The answer would be the motional emf).  It makes the pendulum nugdge back a hair the other way towards believing it is possible this thing might work.

Assuming we control this thing as constant current source, it would be a constant torque device (torque ~ Radius*F where F = I L x B where I and B are constant).  For a constant torque device, power increases linearly with speed. That is convenient because induced voltage also increases linearly with speed, so energy would be conserved.    If it were supplied by constant voltage, it would act more like shunt dc motor.  That's how it might work anyway.... (or maybe not).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
Peter, after rereading the description of your radial flux motor concept, I agree that there is much commonality between your concept and my axial flux concept. A primary similarity is that the conducting wires are on the stator and the permanent magnets are on the rotor(s). This obviously has the advantage of eliminating the slip-ring.

Pursuing Bill's mentioning of the Homopolar motor, this web page came up 'Faraday Paradox' http://en.wikipedia.org/wiki/Faraday_paradox.

"The experiment proceeds in three steps:
1/ The magnet is held to prevent it from rotating, while the disc is spun on its axis. The result is that the galvanometer registers a direct current. The apparatus therefore acts as a generator, ....
2/ The disc is held stationary while the magnet is spun on its axis. The result is that the galvanometer registers no current.
3/ The disc and magnet are spun together. The galvanometer registers a current, as it did in step 1."

Further on it says; "There is no paradox or difficulty if one invokes the special theory of relativity."  
Now it is really becoming confusing.


WHAT IF:

The previous sketch 1904_15 was changed to http://www.unicopter.com/1904_16.gif
In the sketch, the commutators (plural) are the stator and the magnets become the rotors. These two magnets are now composed of many thin sectors, which all have their polarity in the same direction.

Might the 'laminated' PMs cause experiment 2/ to now operate as experiment 1/?


Dave

RE: Rotation by Lorentz or by Faraday?

I'll have to study that when I get a chance.

I am still interested in the commutatorless (and brushless) variety.  It should be pretty easy to set up a simple check for dc generator action spinning the rotor and seeing if I can generate a few millivolts change in meter reading.   I will do that and let you know results (this weekend or latest next weekend).   

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Looking at that wikipedia article, there is never voltage created unless the conductors spin... in which case there is a measured voltage in the non-spinning ref frame (given by E = v X B times number of turns).  That doesn't seem to bode well for my experiment.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
Peter, in the book http://ocw.mit.edu/NR/rdonlyres/resources/RES-6-002Spring-2008/Textbookcontents/chp06_text.pdf, which you previously mentioned, Fig 6-22 states "It is impossible to design a commutatorless dc machine.". This sketch and the paragraph above it appear to have come from a 1946 publication.

I don't think that transistors (and particularly power transistors, MOSFETs and IGBTs, etc.) existed back then.

Today's brushed DC motors are being replaces by BLDC motors. In addition, it is looking like Sensorless BLDC motors may start replacing the Hall Effect device, and, encoders in some applications.

Might it be possible to produce an operational 'unhomopolar' motor by pulsing the power and electronically reading the electrical activity?????

There have be 96 patent since 1977 that use the phrase "homopolar motor". I have not looked at any yet, but these inventors must be thinking of something.


Dave

RE: Rotation by Lorentz or by Faraday?

"Brushless DC motors are accepted by many in the trade as AC motors fallen prey to marketing jargon.

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

Quote:

"It is impossible to design a commutatorless dc machine.". This sketch and the paragraph above it appear to have come from a 1946 publication.

I don't think that transistors (and particularly power transistors, MOSFETs and IGBTs, etc.) existed back then.
I think the statement made in 1946 remains true, if we use the word "brushless" instead of commutatorless.    If we want to power the device with true dc where the dc lives in a stationary reference frame, then we get the required voltage induction only when conductors are rotating with respect to the stationary ref frame where the dc power supply lives, as shown by your wikipedia article.  So I think we can say the true brushless dc motor cannot exist.

Quote:

Today's brushed DC motors are being replaces by BLDC motors. In addition, it is looking like Sensorless BLDC motors may start replacing the Hall Effect device, and, encoders in some applications.
As Bill suggested the term BLDC seems a little bit of a marketing misnomer.    My understanding is typical BLDC resembles polyphase PM rotor synchronous motor, except instead of supplying it with sinusoidal ac, we supply it with switched and there can be a lot of intelligence built into the switching controls which perhaps resemble similar control flexibility in dc motors.  There are others on the forum that know a lot more about BLDC than me.

Quote:

Might it be possible to produce an operational 'unhomopolar' motor by pulsing the power and electronically reading the electrical activity?????
Again, I think it would be a challenge to try to apply your existing device as BLDC since you have a single winding rather than polyphase winding.   And if you made it a polyphase winding, I think it would make sense to swap your magnets to alternating polarities rather than same polarity.

I have to step back and ask the big picture question now....what is your main objective from this thread:
  • to modify the existing motor?
  • to learn about motors in general?
  • to build a motor from scratch?
  • to choose an available motor?
On the last two it usually helps to describe your application requirements and ask forum members for suggestions.  Maybe even start a new thread with your clarified objective.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Clarification:
"we supply it with switched"
    should have been
"we supply it with switched dc"

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

And by switched dc, I mean alternating polarity so there is an ac component.  

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)

Quote:

I think the statement made in 1946 remains true, if we use the word "brushless" instead of commutatorless.   
Perhaps true if the word 'commutator' can be applied to IGBTs.

I'm confused.
This diagram  http://www.unicopter.com/1904_8.gif is probably correct when it relates to a generator, where the Force is creating a Current.
However, could it be incorrect when it relates to a motor, where the Current is creating a Force. Consider that the sketch appears to represent a Homopolar motor. And, this will be particularly so if the PMs are considerably lengthened in the vertical direction. But, we are told that a Homopolar motor will does not work.

_________________________

My objective is the satisfaction that is derived from conceptualizing, researching and developing improvements in the field of vertical takeoff and landing (VTOL) aircraft. Home page http://www.unicopter.com

This thread is part of an attempt to see if it is possible to find a previously unconsidered or unobtainable means of providing a lightweight, powerful, slow turning, reliable and simple electric drive for rotorcraft. Asking a lot for sure. :)

Patenting or financial gain is not a personal objective. This note appears at the bottom of all pages that contain ideas, which do not conflict with the desires of participants in the specific idea.
"The above utility invention is openly and publicly disclosed on the Internet to negate an entity from patenting it, to the exclusion of all others whom may wish to use it. ~ Reference patent law 35 U.S.C. 102 A person shall be entitled to a patent unless - (a) the invention was known ... by others in this country, ..., before the invention thereof by the applicant for patent."


Dave

RE: Rotation by Lorentz or by Faraday?

I don't have much more to contribute, but I'll comment where something sounds possibly a little off... not being picky, just repeating back for understanding.

Quote:

Perhaps true if the word 'commutator' can be applied to IGBTs.
Imo the BLDC motor is not a dc motor.  As shown in wikipedia, for a true dc motor, the dc supply needs to be in a different reference frame than the conductors in order for the counter emf to be generated. That is not the case for the BLDC motor where we the supply is connected to the stator.  imo the BLDC uses electronic switches to create ac... it does not substitute switches for the commutator function on a dc motor. Maybe that's the same thing you were saying.

Quote:

This diagram  http://www.unicopter.com/1904_8.gif is probably correct when it relates to a generator, where the Force is creating a Current.
However, could it be incorrect when it relates to a motor, where the Current is creating a Force
Both a motor and a generator need to have motor and generator action.  i.e. they both need to create voltage and torque.  It just may be a counter voltage (for motor) or a counter torque (for generator).

Quote:

But, we are told that a Homopolar motor will does not work
Some homopolar motors work.  i.e. Zahn page 420 or Bill's link. In all cases brushes are required since these are dc devices.

But the homopolar motor I tried (and I think you tried) without brushes won't work because there is no mechanism for generator action.

Now there is one point of confusion maybe you have that I cannot address.  Our homopolar designs put a conductor in a field. So we know there is a force on the conductor.  We also presume at least some of the counter-force will occur across the arigap on the PM of the rotor  (and I believe I proved it with F.E. for my design).  So we think  our devices have the capability to produce tangential force and torque in the static condition.  Yet we know they can't function as a motor because they can't fulfill the required generator action (required to satisfy conservation of energy).  How do we reconcile the device apparently produces a torque in static condition but cannot act as a motor? Seems like only two possibilities:
1 - We were wrong to conclude there is a torque in the static condition (?)
2 - The torque is there in static condition but somehow disappear when the rotor begins to rotate even with current held constant (?)
Beats me.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

If I look back at my F.E., I only analysed a 2-magnet, 2-partial-arc coil. For that particular geoemetry, I only proved it could generate torque with the rotor in one particular position.  But looking at that analysed 2-magnet 2-partial arc coil, it's easy to imagine that in other positions it might stop producing torque or generate opposite torque.   And if I added magnets uniformly around the periphery circumference of the rotor, it might act differently.  I'm guessing item 1 above might be the more likely logical error (these devices don't actually produce torque in static condition).   

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

I find the sketch in  
http://www.unicopter.com/1904_8.gif
to be less useful than it may be. The sketch neglects the location of the return conductor of the current carrying wire. When the wire moves, does it enclose more or less lines of flux? Many homo-polar designs will both generate and motor.
 

Bill
--------------------
"Why not the best?"
Jimmy Carter

RE: Rotation by Lorentz or by Faraday?

Continuing my earlier line of thought... I am now sure that the brushless dc design that I posted will not produce torque even in the static condition.  Assume that the rotor has a lot of tiny magnets... so many that they can be considered axially uniform.  

Now we calculate torque as Te =  d/dtheta (Wm)
where
Te = electrical torque
theta is angle of rotor with respect to stator
Wm is stored magnetic energy
we choose to hold current constant (alternative choice is flux linkage constant... results in sign reversal).

From angular symmetry of the rotor, we can see that moving the rotor a small angle dtheta will not change the stored energy at all.  At first we may think that we still have force on conductor but equal/opposite force on stator core. But another experiment where we move only stator conductor (not stator core) results in same Te = 0 so there is in fact no force even on the stator conductors.  

It seems a paradox F = L I x B does not apply.  But I believe it is correct.  Sticking with the energy approach, if move a current carrying conductor strip within an airgap, there is a change in stored energy which occurs at the leading edge of the conductor strip and at the trailing edge of the conductor strip, but no change in energy at the middle of the conductor strip (see the Long Version page 32 for rough picture).  It is the shifting of the flux rising/trailing edge associated with flux from conductor that creates that change in energy. If we have stator conductors all the way around the circumefernce of the airgap, there is no rising/trailing edge (*) and so no change in energy.  If we have stator conductors only part way around, there is an equal/opposite rising/trailing edge.  It can create torque when it interacts with poles, but we have nothing resembling a pole on these machines.  (* it raises yet another question, what if we apply the same logic to the conductor disk of a textbook homopolar motor which we know does produce torque... a question for another day... I believe homopolar motor has some real complexities).

I have kind of hogged the thread talking through a subject that is of interest to me.  I apologize for that.  I will be glad to continue to talk throught these things  or whatever you want in this thread...

However some people may have tuned out on this thread (my fault) and you may miss some good input on your real question of finding a good motor for your application. May I suggest that you post a new thread with title something like "slow lightweight motor?". I'm sure you will get some new suggestions and ideas.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Correction:

Quote:

so many that they can be considered axially uniform.  
should have been

Quote:

so many that the rotor can be considered to have angular symmetry.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Quote (electricpete):

If we have stator conductors only part way around, there is an equal/opposite rising/trailing edge
That was wrong. The relevant thing is that in the motor I have drawn all conductors within the airgap have the same polarity.  (there is no alternation of polarity).  That is what prevents the rising/trailing edge in radial flux profile that we're looking for to create a torque.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)

Quote:

Peter said;
I will be glad to continue to talk throught these things...
This is what I would like if you, Bill, and perhaps others, are willing.

There are attractive larger diameter, slow-turning, outrunner and inrunner, motors available. One example is this http://www.alliedmotion.com/Products/Series.aspx?p=10&s=1
This thread was specifically started to see if a phaseless motor was possible and to try and understand the correlation between the magnetic attraction shown here http://www.unicopter.com/1904_7.gif and the electromechanical activity shown here. http://www.unicopter.com/1904_8.gif

Quote:

Bill said;
Many homo-polar designs will both generate and motor.
Will you direct us to one or two of these motors? This may help in understanding the concern.
__________________

On the subject of the phaseless motor (homopolar motor). I wonder if magnetic reluctance in conjunction with Pulse Width Modulation or Pulse Amplitude Modulation might drive a drive a homopolar PM rotor?

Figure 6-1 on http://ocw.mit.edu/NR/rdonlyres/resources/RES-6-002Spring-2008/Textbookcontents/chp06_text.pdf shows the 'On' and 'Off' action of a switch. Perhaps, for my requirement at least, there is some way to direct the 'On' pulses to the CW rotor and direct the 'Off' pulses to the CCW rotor.


Dave   

RE: Rotation by Lorentz or by Faraday?

Quote:

" Many homo-polar designs will both generate and motor."

Will you direct us to one or two of these motors? This may help in understanding the concern.
The MIT / Zahn textbook link above page 421 gives description of homopolar generator.  If you read through, it says it can also operate as a motor.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

Quote (Intermesher):

On the subject of the phaseless motor (homopolar motor). I wonder if magnetic reluctance in conjunction with Pulse Width Modulation or Pulse Amplitude Modulation might drive a drive a homopolar PM rotor?
If you are referring to the textbook homopolar motor such as shown in Zahn (uses brushes and works as-is), you can use electronics to change the voltage.

If you are referring to my device and your device (no brushes and doesn't work), I don't think there is much that can be done with front-end electronics. It would be better to organize the rotor into poles (alternating magnet polarity) and the stator into separate poles and run it is a PM sync motor or BLDC motor.


Quote (electricpete 28 Mar 10 22:13):

It seems a paradox F = L I x B does not apply.  But I believe it is correct.  Sticking with the energy approach, if move a current carrying conductor strip within an airgap, there is a change in stored energy which occurs at the leading edge of the conductor strip and at the trailing edge of the conductor strip, but no change in energy at the middle of the conductor strip (see the Long Version page 32 for rough picture).  It is the shifting of the flux rising/trailing edge associated with flux from conductor that creates that change in energy. If we have stator conductors all the way around the circumefernce of the airgap, there is no rising/trailing edge (*) and so no change in energy.  If we have stator conductors only part way around, there is an equal/opposite rising/trailing edge.  It can create torque when it interacts with poles, but we have nothing resembling a pole on these machines.

Quote (electricpete 28 Mar 10 22:24 ):

That was wrong. The relevant thing is that in the motor I have drawn all conductors within the airgap have the same polarity.  (there is no alternation of polarity).  That is what prevents the rising/trailing edge in radial flux profile that we're looking for to create a torque.
I think the conclusion was right... there are (paradoxially) no net tangential force on conductors in my device even though those current carrying conductors are located in a radial flux field and right hand rule would predict such a force.   But the explanations above fall short.  One more try on the explanation.   Again Te = d/dtheta(Wm).   We proceed as in 12.2 of the long version (add together flux contributions of stator and rotor so we can differentiate to find torque))  There is one contribution of radial flux from the rotor PM's... easy to figure out – it is constant flux (does not change with theta).  The contribution to radial airgap flux from the stator conductors?... Zero!   We can show it with Ampere's law remembering that both conductor currents are in the same direction.   A simpler intuitive thought is to compare flux pattern for two conductors carrying current in same direction and opposite direction.   If currents flow opposite direction (as those shown in 12.2), then the flux lines encircle each conductor independently and we have flux lines flowing in the area between the conductors... these are the same flux lines that would cross the airgap when we put the conductors 180 apart in the airgap.  If currents flow in same direction, the flux lines tend to encircle the pair of conductors (no flux lines cutting the plane between conductors and no corresponding flux crossing the airgap).  We also know if we put an array of conductors all flowing the same direction to form a ring of current, the flux inside that ring (where the airgap is) is zero.  So if airgap flux from the stator conductors is zero, the energy approach predicts torque on those conductors will be zero.  It is a satisfying explanation for something that has been bugging me a long time.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

RE: Rotation by Lorentz or by Faraday?

(OP)
Peter, I read through the first part of your Long paper and skimmed through the latter equations. It is impressive.

This thread has been an interesting and informative one, but there is a need to return to the more practical.  Thank you Peter and Bill for your time and contributions.


Dave
 

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources