3
\$\begingroup\$

In a BLDC motor the stator always looks like a gear and the coil always is wrapped around each gear like this:

BLDC Stator

I'm assuming it will be less efficient to do it like this because nobody is making BLDC motors like it, but I don't know why it will be less efficient:

Why not use this for a BLDC stator?

In a brushless DC motor, why must the stator be gear-shaped with teeth, instead of using stacked iron (or steel) rings with spaced copper wire wrapped around them?

Is there any kind of analysis or software that can help me visualize how it will be less efficient?

Edit:

This is my research about this kind of design.

Here is an outrunner design, and the 4 magnets are in cyan color (modeled by me):

enter image description here enter image description here

I only find one thing challenging, and that is how the rotor and stator are supposed to be kept together so rotation will happen. I also find it difficult to understand how the ball bearings will get into the stator, because the ball bearings will definitely touch the windings.

Normal design (from a youtube video):

enter image description here

\$\endgroup\$
8
  • 4
    \$\begingroup\$ The efficiency of the second figure is zero. It isn't a motor at all. You need the magnetic field to extend outside to interact with the magnets. \$\endgroup\$ Commented Apr 15 at 14:35
  • 1
    \$\begingroup\$ dsa - Hi, Where did the images come from? To comply with the site rule on referencing, details of the original source of any copied / adapted material must be provided by you, next to each copied / adapted item. If the original source is online, please edit the question & add the webpage/PDF/video name & its link (URL) (e.g. website name + webpage title + URL). (I doubt it applies here, but if the source is offline (e.g. printed book / private intranet) then add source details "to the best of your ability" e.g. title, authors, page, edition etc.) TY \$\endgroup\$ Commented Apr 15 at 14:41
  • 4
    \$\begingroup\$ There are free finite-element magnetics analysis packages like FEMM, but you should learn the fundamentals of electromagnetics first. \$\endgroup\$ Commented Apr 15 at 14:50
  • \$\begingroup\$ There are ways to do a non-gear-shaped rotor, but they aren't like you've drawn here. The coils still need to be perpendicular to the radii. The rotor in the image is called a salient-pole rotor, a term you might be able to use to look up more details. \$\endgroup\$ Commented Apr 15 at 15:42
  • 1
    \$\begingroup\$ @TimWilliams - Yes, well-formed question, but lack of research. I didn't downvote (I rarely do). I wrote a quick comment because I didn't have time to write a decent answer. Andy basically said the same thing (not that he needed my help). \$\endgroup\$ Commented Apr 15 at 21:04

4 Answers 4

Reset to default
2
\$\begingroup\$

Your idea of "tangential" stator electromagnets can work with the following modification. It can also be wound on a traditional toroid winder. However, you are stuck with needing "teeth":

modded stator

The fundamental issue is that you got to get the flux out of your ring design to perform work. The way your design is currently configured, you magnet flux is trapped in a high-permeance core - great for transformer design (where you want to contain the "magnetic energy") but poor for motor designs (where you want to "squirt out" the "magnetic energy").

So as a compromise to your design - you can get a reasonably performing motor with the stator coils oriented in the tangential direction (as opposed to the traditional radial-flux designs and the modern axial-flux designs) but you're going to need the flux nubs (teeth) to "spray" the magnetic flux out of the ring. And this is precisely why you will never see a pure, closed ring stator motor in practice.

\$\endgroup\$
8
\$\begingroup\$

In the same way that the universal answer to

"Why does the speedrun do..?"

is "because it's faster", most anything of industrial importance has the answer "because it's cheaper / more efficient".

That doesn't mean it's impossible, and indeed there are examples. Even famous historical ones.

Consider Tesla's "Egg of Columbus" demo:

enter image description here

We see a (probably laminated iron) toroid, with windings on a few sectors, and a solid metal rotor (the egg) in the middle.

When the windings are energized alternately, a rotating magnetic field is created, and the usual induction, phase shift and torque system is set up. This works equally well for a permanent magnet (which then rotates synchronously to the field, when lock is maintained as it spins up).

The key that other answers/comments (at time of writing) are missing is how the windings are energized. In this case, we have:

enter image description here

during one phase, and

enter image description here

during the next (1/4 phase), and so on alternately (the first again but opposite polarity, then this in opposite polarity).

This does not use the iron ring as a toroidal inductor, but [ab]uses it as a sort of bar magnet with a hole in it. The top-down view looks like so:

enter image description here

Notice windings are energized in opposing direction, forcing magnetic field out of the core and into space, where otherwise leakage inductance would be present if this were a plain old transformer.

In this setup, the core doesn't do a whole lot, actually; since most of the magnetic field line length goes through air, the effective permeability is fairly low (single digits, 2-8 say), even if the iron's relative permeability is some thousands.

The real question is, how you intend to couple this stator to the rotor. A loose coupling, as here, gives very poor effect indeed. The stator gets rather hot while the egg takes some minutes to spin up -- pitiful torque, passable as a demonstration but hardly an industrial marvel. A close-fitting rotor (typically composed of laminated iron with a "squirrel cage" conductor framed around it) intercepts much more of the field lines, giving good power coupling and efficiency, and you have the basic induction motor; or use a permanent magnet (typically strontium ferrite or NdFeB supermagnets embedded in a solid or laminated iron "pole piece") for the synchronous equivalent.

Also, changing the windings to a three-phase arrangement is a bit more efficient; Tesla's original invention used two phases at right angles, which is mathematically simpler but needs one more wire. With those modifications, and various other optimizations to structure, geometry and material, you have the modern "BLDC" motor component (the "BL" part; the "DC" part is another matter!).

\$\endgroup\$
6
  • 1
    \$\begingroup\$ The electrical losses of a 'toroidal transformer' are comparable to the efficiency of this motor approach(?) Means the 200hp coil input power would enable the copper egg transferring a 4-16hp torque(?) \$\endgroup\$ Commented Apr 17 at 16:51
  • 1
    \$\begingroup\$ The egg doesn't move usefully. Tesla might as well have placed a red herring inside the toroid instead of an egg \$\endgroup\$ Commented Apr 18 at 0:02
  • \$\begingroup\$ @Andyaka So it does move? This comment seems to contradict your answer. Perhaps there is more to cover on this subject than your answer currently discusses? Also, I indicated one way in which efficiency can be improved, indeed to the point of industrial value. \$\endgroup\$ Commented Apr 18 at 0:29
  • \$\begingroup\$ @TimWilliams I edited my post. Can it be said that Tesla's "Egg of Columbus" kind of design for BLDC motors won't support shafts and ball bearings? Or any ways to make it support it (especially the ball bearings)? \$\endgroup\$ Commented May 8 at 14:30
  • 1
    \$\begingroup\$ Regarding the edit, I'm not particularly concerned with mechanical details; you might ask separately on the Engineering Stack to satisfy that curiosity. (Major edits to a question, after answers have been made to the as-written version, are also discouraged. Which I see a moderator has already reversed, so there you are. The added material is still in the edit history if you wish to copy it elsewhere.) \$\endgroup\$ Commented May 8 at 14:53
2
+25
\$\begingroup\$

Please look at the design from our in-house magnetic modeling program showing an outer-rotor motor.enter image description here In this cross-section, the grey is the rotor, the red and blue are north and south magnets, the green is a lamination stack, and the brown is copper. Now take a look at the instantaneous flux and the "lines of flux": enter image description here You can see that the flux path has only a small air gap to cross between the stator teeth and the magnets. The path goes from the south magnet,through the rotor body, and back into the north magnet on the rotor side, through the stator teeth, and around the center of the stator. Some of the the lamination teeth are magnetized, producing torque as the field pulls the rotor magnets toward the energized teeth. The teeth are energized in sequence, with the energized pair of teeth "walking around" the stator as the rotor magnets try to keep up. The teeth can be wider at the point where they need to be near the magnets, further reducing reluctance.

Now consider your alternate design. It has the following drawbacks:

  • Because the wire must have a larger diameter than the core, the magnets cannot be close to your core. (You could make slots for the coils, which would then start to resemble teeth in the stator!)
  • The only way to generate a pole in the stator is to have two adjacent windings oppose each other magnetically, so it takes two windings to provide one flux path. enter image description here
  • There is no way to have two adjacent spaces between the windings have the same polarity, since the back side of one of your windings has the opposite polarity of the front. enter image description here

These are the main arguments - others are efficient use of space for copper and ease of manufacture.

\$\endgroup\$
6
  • \$\begingroup\$ (What is the significance of the white line through the copper in each gap?) With a stack of standard laminations, what if you place the windings not around each tooth, but around that ring of inner core/through each gap and the central opening? \$\endgroup\$ Commented Apr 24 at 2:42
  • \$\begingroup\$ These are the direction of the magnetic flux. In order to get an magnetic pole toward the magnet, the two coils must produce opposite polarity fields. If they are in the same direction, the flux path remains in the core and so do not attract the magnets. \$\endgroup\$ Commented Apr 24 at 22:40
  • \$\begingroup\$ The image presented first has areas described by "brown is copper". There are "concentric" white lines through these areas: What is their significance? \$\endgroup\$ Commented Apr 25 at 5:46
  • 1
    \$\begingroup\$ Sorry. This design is what is called a "two-layer top and bottom" wind. There are two phases in each slot which are placed one on top of the other, rather than side-by-side as in the questioner's first picture. The line schematically shows a boundary between the two phases. \$\endgroup\$ Commented Apr 25 at 12:48
  • \$\begingroup\$ What is the result with windings around that ring of inner core/through each gap and the central opening, every third powered one way, the other way, and not at all? \$\endgroup\$ Commented Apr 25 at 14:17
1
\$\begingroup\$

If you want a simple (but also correct and relevant) answer, ask yourself where would the magnetic poles be in your circular drawing. I see none hence, the rotor won't receive a differential magnetic field hence, it won't turn (not one bit).

Your circular "design" will produce no external magnetic fields that can be used to turn the rotor. All the magnetic field is locked within the toroid shaped grey-coloured mass of what appears to be the core.

Is there any kind of analysis or software that can help me visualize how it will be less efficient?

It's not just less-efficient; it has zero efficiency.

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Tim Williams answer is correct. Efficiency will not be great, but it won't be 0. \$\endgroup\$ Commented Apr 17 at 15:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.