As promised last month, I have started to put together an article explaining how you can go about building your own 3 phase motors.
Before we start to build a motor, you ought to understand some basic theory. Nothing too difficult to grasp, and it will help you understand why we do certain things. So lets start with the good old brushed motor.
The Brushed Motor
The basic layout for a brushed motor consists of a pair of stationary
permanent magnets positioned on the inside of the outer can. One
magnet will be a North Pole and the other South.
The rotating armature part consists of a soft iron core (material that will not hold magnetism) that looks like a lot of hammerheads connected together in a circle. These hammerheads are used to wind the motor coils that create the changing magnetic poles. All the wire ends are then connected to a commutator which makes contact with two brushes that allow current to flow into the motor from the battery. As the armature rotates, the brushes pass current through the coils in various directions resulting in a changing magnetic field that repels and attracts the two permanent magnets causing the armature to spin.
The basic operation of the brushed motor is quite simple, but lets look a little bit further.
What about the speed of a motor? As you well know, if you want to make a motor go faster, you simple apply more voltage. But let’s take a step back. If you apply a given voltage, what stops the motor from simply spinning faster and faster? What makes it stop accelerating?
If you were to manually spin a motor shaft, the motor would act like a generator and produce current. When you connect a battery to a motor, current passes through the motor and causes it to spin. But as the motor spins faster and faster, so too does the motor itself produce current (referred to as back-emf) that fights the incoming current from the battery.
The torque a motor produces is directly proportional to the number of turns in the coils and the current flowing through them. So, as the motor accelerates, the opposing back-emf will result in less current from the battery flowing through the motor.
As the current drops, so too does the torque. You eventually get to a point where the torque produced cannot overcome the frictional and other losses in the rotating rotor and the motor stops accelerating.
Not too difficult to follow I hope. Later you will see how all of these factors come together when designing a homebuilt motor.
The Brushless Motor
As the name suggests, it is a motor without any friction-generating brushes. If we would like to convert the above example to a brushless equivalent, then we need to remove the outer perimeter magnets and replace them with an iron stator containing a number of coils. The outer magnets now move to the inside (suitably sized alternatives of course) and are attached to the rotating shaft. The outer stator connections are connected to an electronic controller that then switches the coils on and off in a specific pattern, resulting in a rotating magnet field. This field then attracts and repels the magnets fastened to the central shaft making it turn. The maximum RPM that can be achieved is still based on the same principle as the brushed motor. Pretty simple isn’t it?
To make the motor go faster, we simply speed up the coil switching sequence that results in a faster rotating magnetic field. The principle is pretty simple, however, the finer detail in how the electronics does all of this is not. There is another twist in all of this. Consider the brushed motor again. Higher voltage results in a higher RPM and the current drawn from the battery is, in simple terms, a function of the load attached to the motor shaft.
For a brushless motor though, the current supplied is somewhat independent of the speed. They are not linked like the brushed example. The controller has to do the job of monitoring the motor closely and supply just enough current to make the motor turn at any given RPM. Supplying too much current, means you are wasting battery power and gaining nothing in return. Too little current and the motor will not be able to maintain the specified RPM (the back-emf current will be greater than the supplied current) and the rotor will fall out of sequence with the rotating magnet field and the motor will stop.
There are two ways for the controller to operate — it can be set to spin a prop at (say) 5000 RPM, for which it would need to supply a certain current. If the pitch is increased (as in a helicopter main rotor) the controller would need to supply more current to maintain the 5000 RPM. The controller monitors the back- emf wave-form and adjusts the current needed by the motor.
The other approach (normally used in our controllers), works the other
way round. The controller supplies current to the motor and monitors
the back-emf to see how fast it can make it go for that given current.
The Inrunner vs Outrunner
The above example is what we refer to as an in-runner. The out-runner simply moves the rotating magnets to the outside and the stationary stator back to the inside, just like the basic configuration for the brushed motor already explained.
But why would we want to do this? What is to be gained from an outrunner over the inrunner?
Well, there is basically one main reason —
A properly designed out-runner will provide a lot more torque, resulting in the ability to swing a bigger propeller without the need for a gearbox! (It’s much the same as a long-stroke IC motor giving more torque because of the longer power stroke.)
Not all outrunners work well without a gearbox though. The smaller the diameter of the motor, the less torque will be available. The smaller 20 mm diameter CD rom stators, which are quite readily available, are pretty useless at swinging anything above a 5 or 6-inch propeller. Yes, they can be made to swing bigger props, but then the efficiency of the power system becomes bad . . . resulting in excess current being dissipated as heat in your motor. This heat can also kill the magnets and hence your motor. In order to get the most out of these smaller motors, a gearbox is required for bigger propellers, resulting in a pretty good power system, with efficiency approaching 90%.
The moment you move to the 24 mm CD rom stator size, bigger props can
be realized. Most motor builders still over-prop these motors though,
and fit something like a 9 or 10-inch propeller. The overall efficiency
then drops into the 50% range (at best), but the weight saving over a brushed
setup and gearbox still results in less current required to fly the plane.
The lower current levels and high Lipoly capacity makes the 50% figure
less of an issue at the end of the day. Nevertheless, if you could
gain 30%, you would see a similar percentage increase in flying time.
Stators & Stator Types
When you start looking for stators for motor building purposes, you will notice that they come in all sizes and pole counts. The pole count is the number of teeth available for winding coils on. For our purposes, a stator must have a number that is divisible by 3 in order that they can be used with commercial speed controllers. So this would mean that stators with 3, 6, 9, 12 etc are useable.
Yes . . . some clever people on the net have built motors with other
numbers of teeth, but that area is not worth pursuing for our purposes.
Most CD rom motors will have 9-pole stators, while hard-drive stators usually have 12 poles.
These two types are the most common, the latter pole count being the more sought after these days as it provides more options in winding and magnet configurations.
But what difference does it make to the motor at the end of the day you ask?
Well, this is where is gets interesting, not to mention confusing to some. For a given stator diameter, either version can be made to out- perform the other. It all depends on what the motor was designed to do. Now because CD rom type motors are the easiest to convert to aircraft use, they gained a lot of popularity. The 12 teeth stators were left to the more serious motor builders who acquired them from broken household appliances etc.
When some on-line stores started making 12 pole stators available, questions started to be raised as to which stator to use. Some experimentation has hinted that the 12-pole stator is the way to go, and I would tend to agree with that. Nevertheless, a 9-pole stator can be made to match the performance of the 12-pole stator!
In the brushed motor description, I referred to an “iron core”. Now this iron core is not just a chunk of iron, but rather a number of thin iron slices packed on top of each other, with each iron slice being insulated from the next. The reason for this is to reduce eddy current losses, but we will not be concerned with this as you do not really need to know much about it for now.
A CD rom stator will typically be about 4 to 5mm in thickness and contain a number of stator blades. Sometimes, these blades are encapsulated in some form of resin coating. These are great stators and significantly reduce the chance of shorts when winding your coils. For a simple motor, the 4 or 5mm thickness is all you need. For more power though, you could use two of these stators on top of each other.
The moment the stator depth increases, a number of factors relating to the design of the motor changes, one of which is the need for longer magnets. Normally, deeper stator stacks allow for a more powerful motor, but there are limits to how deep you can go.
Well then . . . How deep should a stator be for best performance? Articles that I have read would indicate that half the stator width is the biggest you ought to go. So for a 24 mm CD rom stator, that would equal 12 mm or about 3 stators packed together. What you must bear in mind though, as the stator depth increases, the balance of the motor becomes an issue. Due to the nature of the construction of these motors, the deeper the stator stack, the more critical the craftsmanship of the motor becomes. Normally the magnets available decide your stator thickness, but for a first-off motor, stick with a single CD rom stator.
Lastly, if you strip hard drive motors, you will notice that they are slightly thinner. Here you will need more that one stator stack to make a motor of suitable power. And trying to find two hard drives with the same type of motor is another story. Seems they come in all sizes and you may need to strip about 10 motors before you find two matching stators.
For lack of a better heading, there is another important issue that needs to be looked at in order to see if a stator is worthy of new life. The rotating bell (the outer housing containing the magnets) is connected to a shaft that must run through the centre of the stator.
Some stators will contain a bronze bushing and others a dual bearing assembly. When you open a CD rom motor and discover you have found a bearing motor, you have hit the motor jackpot!
Bearinged motors are pretty scarce as most of the newer CD rom drives no longer use them. A bearing assembly normally accepts a 3,0 mm shaft and is ideal for our purposes.
The bronze bushed stator is another story. I personally dislike
bronze bushes as they do not last too long and a bearing motor does run
more sweetly. There are two types of bronzed stator assembles.
The one type will be a stator that was designed for the possibility of
fitting bearings, and the other not. If the stator had the option
of bearings, then the bronze bush will be a large one that can be pressed
out and a bearing assembly inserted.
The other version will have a tiny bush that does not allow for bearings. All is not lost though. If there is enough metal in the center of the stator, you could turn out the required size hole on a lathe. You ought to have at least 2 mm of metal left from the hole edge to the stator teeth though. If not, go look for a better stator!
Oh yes . . . one other point you must look out for. You will need to wind a number of turns of thick wire onto each tooth. Some stators simply do not provide enough room to do so and are useless for our needs.
Yes, you can still make a motor out of them, but will not be able to
get much power out of them. As a reference, a small motor would need
about 15 to 30 turns of 0,4 mm wire around each tooth. If you are
not able to do that, continue looking for something better.
If you have a lathe available, you can build all types and sizes of motors. Then the whole world of motor building becomes very interesting and you are always on the prowl for discarded stators. So where do we look?
Firstly, any old or broken appliance around the house is worth looking into. From a broken drill or angle grinder to a sewing machine motor. Even check that old cake mixer before tossing it into the bin!
CD rom drives and hard-drives are the number one source of stators being used. Tearing apart a hard-drive though is not easy. You need the right tools to open them up, or they will do nothing more than frustrate you. Most drives tend to use those small star type screws and it may be worth purchasing a set of star bits for the job. Once you get the motor out, there is still the task of opening it up as they are ‘pressed’ together. I clamp mine in a vice and use a screwdriver and hammer, which provides the necessary persuasion!
Scrap yards are always worth a look. I have found bins of the stuff (mostly the wrong type of stator though) but you often tend to get lucky along the way.
Computer shop/Repair centers . . . most of these guys bin the unrepairable parts. I tend to tell them to hang onto them. I did find one chap who was into re-cycling and went to the trouble of stripping everything down and sorting them out in various boxes. Finding this sort of person is like finding a needle in a haystack though.
When you strip down your CD rom motor, you will find a circular magnet glued to the inside of the rotating can. They are a kind of strip magnet that has been magnetized in such a way that they create a number of North and South poles within the strip. If you hold a magnet in your hand and move it around the circumference of the magnet strip, you will be able to feel the attraction and repulsion at different points round the circumference.
There are two types of magnet strips commonly used. A black one and a grey one. The black variety is your typical ferrite magnet and of little use to anybody — they are too weak. The grey ones are stronger and can be used for low power CD rom motors. The simplest solution here would be to rewind the stator with thicker wire and re-assemble the motor parts. You have in effect created a more powerful motor that is capable of flying a small, lightweight RC aircraft. This is in fact how it all started on the web. The original thread on RC groups was called “one hour CD rom motor construction”. To get a flying motor, the only real work here is to cut away the un-wanted metal and make up some form of motor mount.
These quick conversions however are not really that powerful, and appear better suited to the higher RPM type requirements due to lack of available torque. The stock magnets are simply not powerful enough to be of any real use.
The next step in creating a better motor is to modify the rotor can for better magnets. The strip magnet is removed (various methods from hacking to leaving in various liquids) and replaced with 12 off 1 mm thick 5 mm square Neodymium magnets. All of a sudden, a good increase in performance was realised.
But there is even more to be gained by moving to stronger magnets. A stronger magnetic field will result in a higher back-emf generated for the same number of turns used with the weaker magnet. Simple magnetic principles again. The greater back-emf will have the effect of lowering your motor’s speed range . . . remember . . . as explained in the brushed motor principles. So in order to retrieve the lost RPM, we must reduce the back-emf being generated. Make sense? To do this, we reduce the number of turns on the tooth winding.
Aah . . . but fewer turns means more space is now available on each tooth. So what we do now is use thicker wire to fill the space between teeth for the specified number of turns required. Your end result is lower coil resistance due to the thicker wire. Or looking at it from a different angle, you can now get more current into your motor, thereby increasing the power output. All this just by using stronger magnets. So the trend today is to use the strongest magnets you can get your hands on.
As for sources of these magnets . . . well, I purchase mine abroad, but they are slowly becoming available locally.
Don’t let the problem of getting magnets bother you though. I do purchase from time to time, so a plan can be made.
How should magnets be magnetized?
People tend to forget about this. To be of any use, the magnets must be magnetized in the direction shown on the left, below.
Many applications use magnets with their axis as shown on the right
— these are of no use in our motors.
When stripping a hard-drive, the super strong magnets that drive the pickup arm are quickly spotted. They are not suitable as they are magnetised along their length something like N – S – N. Sticks well to the fridge though . . . and yes . . . be warned . . . keep them away from your bank/credit cards!
And whilst on the subject of warnings, do not go and cut Neodymium magnets to a specific size. I will admit that I have done this once before, using surgical gloves, but the chemicals / substances found in these magnets will harm us. Rather just purchase them to be safe.
Next month, we will look more closely at magnets and winding configurations.
In the mean time, see what you can find.
- - - o o O o o - - -