Electric motor types

I don't quite follow the above statement... A commutator appears to send d.c. current of alternating polarities and d. c. pulses of varying frequencies until it reaches operating speed through its mechanical arrangement with the brushes. The process appears to be quite different than what you see in AC induction Motors, for example.

Here's an animation,



What am I missing here?

 

"The commuter and brushes in a conventional DC motor convert the current going through the armature coils into AC."
Exactly that. Not an AC sine wave but still AC with commutator acting as both a current reversing switch and on/off switch at the correct time for each coil in the rotor.
There are no pulses of varying frequency.
A brushless "DC" motor needs electronic switching to replicate the commutator and gets a signal for the correct timing of switching and changing polarity either from a Hall effect sensor or from back EMF. It's beyond my knowledge to explain these!
EDIT: A brushless DC motor always has magnets on the rotor and coils on the stator.

 

AC can be of other types besides sinusoidal. For example, two other common wave shapes are square and triangular (sawtooth). The frequency and amplitude can change too. That means that AC does not have to be constant and symmetrical like the supply from a wall outlet.

 

Yes, but while it is common that the stator is forming the outside of the motor (housing) and the rotor is the part near to the motor axis, there are BLDC outrunners with stator coils attached to the axis and a bell shaped rotor on the outside (= main part of the housing). Main reason for this design is to gain more torque in relation to outer motor diameter.

 

Heimfried, yes, and very common in r/c model planes. That's what is in my DIY outboard.

 

Hello, guys!

To me, it seems like there's a difference between AC (sine, square, or triangular) running continuously through the coils of wire regardless of the angle of rotation of the Armature, compared to alternate polarity pulses of DC ( resemble switched rectified AC?) which depend upon the angle of rotation of the Armature. Seems like I remember at one point in AC motor development, that "Fly Away" brushes (or capacitor?) were used to get an AC induction motor going because of the low starting torque.. Of course all of that has changed with the newer designs, electronic commutation and 3-phase AC, Etc.

I'm guessing that the pulses of DC in a brushed motor start out at a low frequency as the motor starts from zero, (commutator slowly rotating) and increase as the motor reaches full speed, then change frequency slightly as the load conditions change?

Thanks for your comments!

 

Time for some "when I were a lad" reminiscing!

I was a technician in a pharmaceutical research place and had to find a means of stirring blood inside a vertical perspex tube for an oxygen dissociation experiment, I came up with a kind of early brushless motor. The stirrer was a "flea", a bar magnet encased in teflon so this was the rotor. I wound 6 coils and placed them around the tube radially and wired each opposing pair in series. So now I had a 3 phase motor and had to find a way to switch the d.c. power to the phases in sequence, then reverse the polarity of the power supply and sequence the three phases again to complete one rotation.

I had two cams on the shaft of a conventional mains motor driven from a variac (speed control); one cam had two lobes and switched power to each phase by operating three microswitches arranged over about 180 degrees, and the other cam had one lobe arranged to operate two microswitches together acting as a changeover switch to reverse the polarity for half a revolution. In operation it sounded like an antique tractor in need of tappet adjustment!

If you can understand my description, and can imagine replacing the mechanical nonsense with electronic switching you have an idea of how a synchronous (rotor stays aligned with the rotating magnetic field) brushless motor works.

 

-Will

 

As Gonzo started this thread to discuss different types of electric motors here's another type; Switched Reluctance Motors.
They have three phase stators and no permanent magnets or "squirrel cage" on the rotor, only soft iron laminations. I believe Dyson use them in their latest cordless vacuums.
https://en.wikipedia.org/wiki/Switched_reluctance_motor

 

Interesting topic. I can try to be as short as possible and as informative as possible and to stick as much as possible to graphical representations instead of long text. Mind that I will consider only key topics that I consider interesting and important for general understanding, otherwise we can be lost in details. Please, mind that I am not an expert in this topic, but it is field I often (re)visit and like.

For simplicity, the best is to start with brushed DC motors. So, here is the diagram showing all brushed DC motor types:



I will skip the electro-mechanical construction of each of the types since you can google it yourself anyway, because it is a lot of information and because we can be lost in details (of course, they all have electro mechanical commutators which you have already mentioned in the thread).

Let's see now the torque/speed characteristics, but also this characteristic vs the so called armature current (rotor current, but for the sake of simplicity we can also say that this is the DC current on the connection of the motor):



According to characteristics (as shown) and limitations of the motor type (google if interested) , you can select the proper one for the application. Nice feature of brushed DC motors is high torque at zero speed.

Let's discuss the control now. Mind the diagram showing torque vs current, and as you can see, the torque is proportional to the current. That means that if you want to control the torque of the motor, you need to control the DC current and if you want to measure the torque, you can simply measure the DC current (easy task).

If you do not want to control speed (RPM), you only need a single speed, electronic controller is not required and it is enough to connect the nominal voltage on the motor connections (of course, nominal voltage is given for "nominal" torque load). If you want to control RPM, then electronic controller is required.

Let's see what this controller should do. There are two options:

• Change the DC voltage to indirectly change the DC current and the torque. This is not efficient. The performance is also lost at low voltages (we would have to explore motor in more details to understand this, so I will skip. Simplified, the magnetic field is very weak if both voltage and current are low). So it is not done in this way today, but before existence of the hi-speed switching transistors (MOSFETs, IGBTs, SiCs, etc.), this was viable analog option.

• Keep the DC voltage constant at nominal value for maximum allowed torque load/DC current, but when we want to reduce torque we use PWM modulation on DC voltage (you can google PWM if not familiar with it). The advantages: a) PWM duty cycle is proportional to the DC current, so we really control the current (not voltage) and indirectly the torque; b) Even at lower torques/currents, we still provide impulses of high current generating strong magnetic field in the motor so original performance is better kept; c) electronic controller circuit (HW) is easy to implement (e.g. half bridge for single rotation direction, or H-bridge in case you need bi-directional rotation, you can google these circuits if interested); d) Algorithm is very simple and can be implemented with the simplest/slowest/cheapest microcontrollers or even with discrete electronic circuits.

Here is some general comparison of the types of these motors:


And yes, the commutator (and brushes) is the biggest flaw of these motors compared to other motors (it wears with use; it generates sparks, heat, losses and elctro-magnetic noise; it limits the possible maximum speed of the motor; not applicable in food industry due to dusting (though I am personally not sure if that is more dangerous compared to some of the ingredients in some of our food today, but still we should not eat it); in case of small DC motors, commutator is still cheap, but in case of big motors it is required more Fe, Cu and C to build it compared to "only Si" for electronic controller in case of AC motors).

This took me some time, even though I copy/pasted a lot from my personal reminder document, so I will give overview of AC motors a bit later.

 

Nidza, thanks for an excellent summary!

"Let's see what this controller should do. There are two options:

• Change the DC voltage to indirectly change the DC current and the torque. This is not efficient. The performance is also lost at low voltages (we would have to explore motor in more details to understand this, so I will skip. Simplified, the magnetic field is very weak if both voltage and current are low). So it is not done in this way today, but before existence of the hi-speed switching transistors (MOSFETs, IGBTs, SiCs, etc.), this was viable analog option."

My opinion is that using parallel to series switching can be slightly more electrically efficient (and has no EMP vulnerability) than when using transistors in some niche applications, like where only two speeds are required. The electrical drain is slightly less with parallel – series, speed – torque control, because there is no miniscule PWM drain from the electronic circuitry. Parallel setting is useful in limiting the stress and inefficiency from the huge start up influx current for devices repeatedly starting from zero at full load and operating almost entirely or entirely at their highest speed. I built my first PM motor electric bike many years ago using that principle, and it worked well for my particular low speed applications. In theory, you get some regenerative braking when dropping down from the series position into parallel with PM, which might put a little energy back into the batteries, especially on a long downhill brake using the parallel setting. Manipulating the load by using properly timed gear shifts can also be used as a crude speed control with PM, and also employed to produce regenerative braking energy, though it is not worth the trouble...

Hope this helps.

 

Any thoughts or inputs about Axial Flux Motors?

 

I presume that you are talking about something like on the following diagram:


Yes, that can work, and I respect that it fits to KISS rule. That works in bicycle application as you have described. The flaw for some other application(s) would be that you can only have discrete/steped values of torque/speed. I wanted to describe the way of control which can use the full capability (continual control) of the motor so that it can be compared to ICE engines, which is probably the key interest for many people here. Between full capability implementation and connecting the motor directly to DC voltage, there are many solutions with their own limitations, but as stated at the beginning of my post, I wanted to avoid going into all options and details because the post could then extend to "infinity". And I thought of this just as an introduction before switching to AC motors.

This axial motor represented in the video is synchronous type of motor and it was compared to induction (asynchronous) type of radial motor. It should be compared to synchronus radial motor for correct comparison. I think that the major mechanical flaw of this types of axial motors when used in car wheels, as shown on the video, is that basically they have to be very robust since they must survive direct hits to road bumps/holes, while the motors which are not in wheels are protected by the chassis schock absorbers. Therefore, I presume that it fits better to aircraft application, as shown and maybe boat application as well. Regarding performance, it would have to be analyzed on electro-magnetic level, which was partly done in the video, but compared with incorrect type of motor.

 

Axial flux? Larger diameter, shorter length, higher torque, lower speed. Radial flux? smaller diameter, higher speed etc.....
The YASA motors in that 'plane claim 20% better power density than typical axial flux motors, paper here:
http://www.mojaladja.com/upload/elmotor/Analysis of the Yokeless and Segmented Armature machine.pdf
Small axial flux motors can have ironless stators with the coils formed as a printed circuit. But small brushed motors can have ironless rotors... lots of choice....

 

I am currently busy to write long post and describe each type of AC motor yet, but here is the broader diagram showing what types of motors do exist:


Even though this picture probably does not cover all existing subtypes that do exist (and some subtypes are maybe even outdated), it is very good representation of main types and categorization. The deepest/lowest subcategories actually represent the type of implementation, but the same "technology". And since this is probably overwhelming to start with, I am giving simplifed picture with numbers showing which motor types to learn first and in which order for the best overall understanding of AC motors.


And at the same time when learning each of these motor types, it should be explained the way of control. In addition, it should be first explained when used only with AC network connection (historically), and what are limitations and why all other sub/subtypes came. Then it should be explained modern way of control with switching circuits (MOSFETS, IGBTs, etc.). And due to the modern way of control (which became possible with fast enough microcontrollers and fast high current switching elements), additional sub/subtypes came and are still coming due to current popularity of electrification. But, the electro-mechanical principle of all motors is almost unchanged for decades, it is just becoming better and better in design (better usage of electro-magnetic field, more efficient, less heat, radial/axial, ...).

When I have more time, I can write a bit about each of these types numbered on the diagram above. That would give you good understanding of any existing motor.