Electric Motors – part 3
          Brian Muldermm

Magnets … how many, what size?
For a 9 pole stator, wound ABCABCABC, there are basically only two choices available — 6 or 12 magnets in the bell.   Just to recap, the bell is the rotating part that has the magnets attached.  For a motor required to give low torque and high RPM, then 6 magnets will be the choice.  If you want low RPM and good torque, then 12 magnets will be needed. Ironically though, you might well be using 12 magnets whichever choice you make, as there is the issue of ‘magnet coverage’ as well.

But how does the number of magnets determine the motor RPM and available torque you may ask?   In addition, those of you who are familiar with these motors will also have some knowledge of what some people call “electronic gearing”, which is also related to the number of magnets used.

Well, in an attempt to answer these questions for myself some time ago, I created a number of drawings in AutoCAD that showed how the current flows through the windings and creates the north and south poles at the stator teeth.  You can then see quite clearly how the ‘bell’ with its magnets moves in relation to the fields generated on the stator teeth.  There is even some nice software on the Internet that does the job via animation.  Now when I did this exercise, I modeled the drawings around the LRK motor due to a raging debate on RC groups.  So before we look at this,  lets get more familiar with the LRK motor.  You will then understand the drawings better.

The LRK motor was developed by three gentlemen named Lucas, Retzback and Kuhfuss.  In a nutshell, the LRK winding diagram is about trying to obtain the highest flux (maximum lines of magnetic force)  with a given amount of stator metal and magnetism.  The more flux created, the more torque becomes available, but the RPM is lower.   This does not mean that the LRK motor cannot run at high RPMs though.  They are quite capable of reaching high RPM figures, but speed controllers are normally the limiting factor, due to the high gearing ratio of the LRK motor which we will discuss in a little while.

To build a standard LRK motor, you will need a 12-pole stator.  You cannot use a 9-pole stator for these motors.  The next significant difference is the winding scheme.  You will notice that only half the teeth are wound . . . which actually makes life a lot easier for two reasons.  Firstly, only six teeth need to be wound, as opposed to 9 teeth for our CD rom motor, and secondly, the unwound teeth help create more space for the teeth that do need to be wound.  This is great when you have little space to put on a decent number of turns.

If you study the diagram, you would notice that for each phase, the winding direction for each of the two teeth are different.   The intention here is that as current flows through the coil on the first tooth, it will create a magnetic field of say, north polarity, and then a south field on the opposite tooth.  This is important!  You have got to get the winding scheme right . . . but that rule goes for any motor.  Just make sure you study the winding diagram properly and wind accordingly.

At this point I would like to add my personal comment about the way the above diagram has been illustrated.  I have kept all my starting wires one after each other, which results in all the wire ends being together.  This is the natural way one would think when progressing from CD rom winding schemes to LRK.  The official diagrams on the internet however show it another way.

In reality, it does not matter at all how the wires are labeled from a winding point of view as they are all identical.  The problem however comes when people discuss how to connect the wires for a STAR configuration.  You see, it is not that obvious, as 99% of LRK motors would appear to be Delta wound.  So when somebody in the know says that for a STAR configuration, connect your start wires (or end wires) together for the star point connection,  which wires do you connect?  Well CD-rom-experienced people will do it according to my first drawing as this is the way they are accustomed to . . . and the problem is, the motor will run, but not very well.  The correct termination is every second wire being connected as per the pics which follow.

An additional twist to confuse people even more, is when they read the ABC notation, it can be described in two ways based on the above drawings.  My drawing would represent A-B-C-a-b-c   and the second one  A-b-C-a-B-c  (the – represents a missed tooth)

So depending on what format you are used to, mistakes can be made.  My conclusion is,  as I have already mentioned in the earlier articles, LOOK and read a drawing.

LRK Hookups


Now for those who love winding motors and feel that winding only 6 teeth is taking the fun out of it, there is a winding scheme that has all teeth wound.  This scheme is referred to as the Distributed LRK winding scheme.  It would appear this scheme is better suited to thinner stators though.  Of course, you could decide not to wind an LRK scheme and use the same scheme as a CD rom motor.  You will see what the major difference is in a little while.

The LRK motor appears to be hugely popular in Europe and I suspect they have been around a fair while longer than the CD rom motor.  LRK motors are normally built by the more serious motor builder as they do require a fair amount of lathe work.  Typically, 12-pole stators found and used are bigger than the 9-pole version which results in a bigger motor that is capable of a lot more power.  As an idea to what power levels can be achieved, some motor builder are already pushing the 2 kW limit.  Crazy amount of power!!!

The Rotating Field
Right, now that you are familiar with and LRK winding scheme, let’s look in a little more detail as to how the motor bell rotates.  The principle explained is the same for CD rom motors.

A speed controller has three wires that get connected to our motor.  Those three wires control how the current will flow in the motor.  By turning on and off FETs within the controller, current is passed through the motor in six specific directions and then repeated over and over.

Looking at the stator terminations, A, B and C, the current flow directions are as follows —
Sequence 1: from terminal A to terminal B
Sequence 2: form terminal C to terminal B
Sequence 3: from terminal C to terminal A
Sequence 4: from terminal B to terminal A
Sequence 5: from terminal B to terminal C
Sequence 6: from terminal A to terminal C

 —and back to Sequence 1.

We call this set of steps the commutation sequence.  Now, if we apply the sequence to our motor windings, we can see what magnetic poles are being generated around the stator.

Let’s look at a STAR winding termination.  It is easier to see and understand.

The red arrows can signify a north pole tooth and the blue arrow a south pole tooth.

So if we pass current into terminal A and let it exit at terminal B, current flows through tooth 1 and creates a magnetic north pole.  The current then travels to tooth 7 and through its winding creating a magnetic south pole due to the opposite winding.

The current now flows through the ‘Y’ hookup wire to tooth 3 and once again creates magnetic fields as mentioned.  The current finally exists at tooth 9 at our terminal B.

Now — by applying the entire sequence of steps, we can clearly see how the north and south poles fields are rotating around the stator.  You should also be aware that it has taken six controller cycles to generate one revolution of the magnetic field.







Right . . . let’s ‘assemble’ our motor and add a bell containing 14  magnets.  Once again, the Red colour magnet will signify north pole and blue South. We will also mark one of the magnets with a circle so we can see how the bell is rotating.

With sequence one being powered, the bell will snap to the first position shown. You can clearly see why.  The 14 magnets are spaced around the bell very nicely and have magnetic poles in the right position to be attracted to the stator teeth.

Next, the controller switches to sequence 2.  The bell now snaps to the best position in relation to the new magnetic field position.  Try and see if you can follow the sequence.

Now what is of interest here, is that the marked magnet has moved only 8,57 degrees for each sequence change.  This calculates to 42 steps per revolution.  Another way of looking at it, the magnetic field is rotating 7 times faster than the bell.  Make sense to you?   Well this is where the ‘gearing’ term comes into it.  You have in effect a motor that has a 7:1 gear ratio.  Strictly speaking, this is not the same thing as gearing a motor with a proper gearbox, but the principle is basically the same.  If you can make the motor turn even slower than the 7 times above, then you are effectively creating more torque and lowering your RPM/volt figure . . . just as you would achieve using a gearbox.

Now an LRK motor can use either 10 magnets or 14 magnets.  So what happens when we use 10 magnets?

If you carry out the exact same exercise above, you will see that fewer magnet poles results in bigger rotational steps.  In fact, 10 magnets results in 12 degrees per step . . . or a gearing ratio of 5:1    So torque goes down and RPM up.

Interesting stuff this!!

Lastly, as already mentioned, you do not have to wind an LRK scheme for a 12-pole stator.  Should you wish to use the standard CD rom sequence, ABCABCABCABC,  (note 4 teeth wound per phase now instead of the three for CD) you will need to use 8 or 16 magnets which result in gearing of  4 and 8  respectively.


So lets go back to our 9-pole CD rom motor now.   As mentioned already, you can use 6 or 12 magnets.  The 6-magnet setup will result in a gearing of 3:1 and the 12-magnet setup a gearing of 6:1.

Now that we have covered the basics of winding a motor and seen how magnets relate to the design of the motor, you should now understand the following table.  The numbers going across the top of the table reflect the number of stator teeth and the numbers running down the left of the table, the number of magnet poles.

The colour codes represent the performance to be expected with the specific configuration.  Basically, there are only a few that work well.

The table was extracted from the Net, but no compiler was named — thanks anyway!

AacBBaCCb AAccBBaaCCbb

A = clockwise winding       a = counterclockwise winding      - = empty tooth

                                Thus -- AaBbCc -- means
phase/wire 1 :  wind tooth 1 CW, continue to tooth 2 and wind it CCW
phase/wire 2 :  wind tooth 3 CW, continue to tooth 4 and wind it CCW
phase/wire 3 :  wind tooth 5 CW, continue to tooth 6 and wind it CCW

Colours --- grey = does not work    blue = good combination
     white = it works ...    red = it works, but not very well

Other alternatives for 12-pole 18-teeth -- (no space in table)
AaBbCcAaBbCcAaBbCc   or   A-B-C-A-B-C-A-B-C

If you look at the 12-tooth stator column and work your way down to the 10 or 14 magnet row,  you will see A-b-C-a-B-c.   Remember, the dashes reflect the unwound teeth.

Magnet Coverage
To get maximum performance out of your motor, you should have an optimum magnet area inside your bell.  Just to make sure you understand what I am referring to — the drawing which follows has a motor using 4 mm wide magnets with a space of 5,36 mm between them.  This calculates to approximately 43% magnetic coverage.  Not good!!

For CD rom motors, this figure ought to be around 100% and for LRK motors, 80 odd percent seems to be ideal.  I myself have built motors with 50% magnet coverage that worked okay.

An improvement in power and efficiency was realised when additional magnets were added to achieve 100% coverage, so aim for the correct figures.

To achieve proper magnet coverage, you must purchase the correct width magnets . . . or for the above example, double up on the magnets. What we do here is glue another magnet next to the existing one.

This however is easier said than done.  What you want to do is glue two magnets next to each other that are of the same magnetic polarity to create in effect a wider magnet of a south or north pole.  Being Neodymium magnets, they repel each other in a big way.   With a bit of practice, you can get the hang of it quite quickly though.  Should you fail trying to do this, you can glue magnets of opposite polarity to make life easier for you.   There was a debate not so long ago about this very issue.  The argument was that if you double up your magnets with opposite polarity, the feedback signals to the controller are easier to read.

In my opinion, this is a non-issue, as controllers read the signals from either version of motor just fine.   At the end of the day though, motor bells using two magnets of same polarity next to each other do offer better performance.

One more point to consider.  The above example uses 10 magnets.  We do have the other option of using 14 magnets if the bell is for an LRK motor.   The additional 4 magnets then raises the magnetic coverage from 43 to 65%. …which may work well enough for intended application.   But wait there is more .. if we wind a conventional winding, we could use 16 magnets.   Magnet coverage goes up again.   So depending on what magnets you have available or that can be purchased for that matter, you have some room to play in so to speak.

Lastly, there is the issue of airgap, that being the distance between stator teeth and magnets.  As I have run out of time for this month’s issue, I will talk more about that next month.

Magnet positioning and Glueing
Positioning magnets within the bell is not easy.  You need to get them precisely positioned if you want your motor to run really smoothly. Magnets that are inaccurately positioned can result in jitter being created within the speed controller (uneven feedback) which in turn will have negative effects on your motor.   There are various tricks that have been tried to achieve accurate placement, but the best one in my opinion is to create a jig that does it for you.   Once again, you will have to wait for your next SE for that one.

As for gluing magnets into you bell, this is the easiest part of the whole assembly.  After positioning the magnets, simply run cyno into all the groves and edges of the bell.  It is as simple as that.   Whatever gap is left between magnets can be left as such or filled with some epoxy and microballoon mixture if wished.   For a bigger motor though, you ought to use epoxy for better for a better bond.  Magnets have been known to break free, so care must be taken.

- - - o o O o o - - -