DIY Electric Car Forums banner

1 - 6 of 6 Posts

·
Registered
Joined
·
4 Posts
Discussion Starter #1
Hey guys,

I'm looking at designing a 3 phase motor controller for use in a car at some point in the future. I've got the prototype figured out, and it seems to work. However, I'm not sure what the best way to proceed is in relation to increasing or decreasing the phase period. I can do it, no problems, but I'm not sure what the best way to design it is.

If I was to do about the most straightforward solution, I'd treat the accelerator pedal as a linear potentiometer, and each point on that potentiometer would represent a phase period. However, this means that changing the potentiometer very quickly could result in the slip angle increasing too much, and cause the motor to stall.

In regards to AC motors, what function do you think the accelerator should have, and in a high abstraction sense, how would it work?

Thanks
 

·
Registered
Joined
·
4,891 Posts
I assume that you'll get some responses from people with vastly more knowledge of AC motor control than I have, and from people with infinitely more motor controller experience than I have. In the meantime...

First, are you talking about induction motors, or synchronous motors - you should be clear about that.
If induction motors, it is slip speed, not slip angle. If synchronous motors, there's no slip.

It's not clear to me what you mean by "phase period". Are you designing a sinusoidal power supply, or something providing a trapezoidal waveform for a "brushless DC" motor?

I think that the driver's control should generally command torque (so full pedal is the motor's maximum torque at whatever speed the motor is turning). Since power is the product of torque and speed, that also means that full pedal is full power, and that's an equally valid approach. Even then, as you already recognize that's a target for the control system, not directly the controller output. In basic control system design the torque would be the setpoint... you do have some understanding of control systems, right?
 

·
Registered
Joined
·
4 Posts
Discussion Starter #3
I assume that you'll get some responses from people with vastly more knowledge of AC motor control than I have, and from people with infinitely more motor controller experience than I have. In the meantime...

First, are you talking about induction motors, or synchronous motors - you should be clear about that.
If induction motors, it is slip speed, not slip angle. If synchronous motors, there's no slip.

It's not clear to me what you mean by "phase period". Are you designing a sinusoidal power supply, or something providing a trapezoidal waveform for a "brushless DC" motor?

I think that the driver's control should generally command torque (so full pedal is the motor's maximum torque at whatever speed the motor is turning). Since power is the product of torque and speed, that also means that full pedal is full power, and that's an equally valid approach. Even then, as you already recognize that's a target for the control system, not directly the controller output. In basic control system design the torque would be the setpoint... you do have some understanding of control systems, right?
Thank you for your response.

Let me try to clear some things up in my previous post. I'm talking about a 3 phase induction motor.

Perhaps slip angle was a little vague. It's something my professor talked briefly about. Supposedly, it's the angle at which the rotor is moving behind the magnetic field. Essentially the same sort of idea as slip speed.

When I said phase period, again, that was a bad way to put it. I meant the frequency at which the AC output from the controller is operating at. So, yeah, a sinusoidal motor controller.

As to what you said about torque, yes, that would be ideal. My problem is, that I'm not sure how to design an AC controller to change speed based on torque. Since AC motor speed is controlled by frequency, I don't know of a straightforward way to change the speed based on power to the motor. (short of doing some interesting programming on a micro-processor)
 

·
Registered
Joined
·
4,891 Posts
Let me try to clear some things up in my previous post. I'm talking about a 3 phase induction motor.
Thanks. :)

Perhaps slip angle was a little vague. It's something my professor talked briefly about. Supposedly, it's the angle at which the rotor is moving behind the magnetic field. Essentially the same sort of idea as slip speed.
Okay, but that's not how an induction motor works. The rotor does not rotate at the same speed as the stator's magnetic field (and just behind it by some lag angle); it rotates more slowly, with the resulting changing field through the rotor inducing the rotor field to react with the stator field. It really is a speed difference (which is why it is called "slip"), rather than an angle (which you might call "lag").

It's possible that the professor was referring to an angle in coordinate system other than the physical rotation of the motor, or referring to the angle between current and torque vectors (rather than stator field and rotor). If so, you need to catch up to the fundamental material on which this representation is based, before you can implement anything using it.

When I said phase period, again, that was a bad way to put it. I meant the frequency at which the AC output from the controller is operating at. So, yeah, a sinusoidal motor controller.
Okay, so you've got a handle on what you need to control :), but terminology is important to communication.

As to what you said about torque, yes, that would be ideal. My problem is, that I'm not sure how to design an AC controller to change speed based on torque. Since AC motor speed is controlled by frequency, I don't know of a straightforward way to change the speed based on power to the motor.
It sounds like you're starting from the very beginning, which makes me think that your prototype is essentially a power section without a usable controller. Induction motor control has been studied, and products built to control it, for decades; maybe it's time to read a textbook. Seriously - you're at the same point as about a million people before you, so why not read what has already been written (either in printed form as a textbook or online as prepared educational series) for exactly your purpose?

You could ask in this forum for assistance with specific points that you are having trouble understanding, or specific design choices that you are making, rather than expecting anyone to provide a custom online course for something that is already available.

For a start, I suggest a look at the Wikipedia article for Vector control (motor). My guess is that it won't immediately make sense, but as always in Wikipedia there are the references for source material.

... (short of doing some interesting programming on a micro-processor)
Even DC motor controllers are routinely built with microprocessor control; it seems reasonable to expect that any practical and fully functional induction motor controller will be microprocessor-based. There are very long discussions in this forum of controller design and construction which I haven't read (because I'm not building one), but my guess is that if you read them you'll find they are a mix of power section design, analog driver design, and programming of a microprocessor controller. I could be mistaken...

But even if you are willing to program something as a controller, I don't see a point in starting yet because you haven't established what the program needs to do. There's fundamental understanding and design needed before they're any point in coding.
 

·
Registered
Joined
·
4 Posts
Discussion Starter #5
It sounds like you're starting from the very beginning, which makes me think that your prototype is essentially a power section without a usable controller.
Well, it's not usable in the sense I want. It's essentially a transistor driver chip, and the frequency is controlled by the resistance between two control pins. So, it's just using a potentiometer.

Induction motor control has been studied, and products built to control it, for decades; maybe it's time to read a textbook. Seriously - you're at the same point as about a million people before you, so why not read what has already been written (either in printed form as a textbook or online as prepared educational series) for exactly your purpose?
That's a good idea. I didn't even know that direct torque control in AC was a thing. I should have guessed that it was, because it would be highly critical in certain applications. I've looked into few textbooks, and I'll see what I can learn there. Thanks for the suggestion. :)
 

·
Registered
Joined
·
3,141 Posts
Microchip has quite a few application notes and other materials on ACIM fundamentals as well as advanced concepts. Here is one that seems pretty good:

http://ww1.microchip.com/downloads/en/AppNotes/00001660B.pdf

Links to source code, devices, etc.:

http://www.microchip.com/wwwAppNotes/AppNotes.aspx?appnote=en567897

An overview:

http://www.microchip.com/design-centers/motor-control-and-drive/motor-types/ac-induction-motor

I attended a seminar at Microchip's MASTERs conference in 2004, and it showed how calculations and concepts for induction motors is made easier by using the rotor as a frame of reference. From that viewpoint, the magnetic field rotates around it at the slip speed, which remains within a fairly narrow range over a wider range of RPM.

Even a simple open loop trapezoidal waveform control needs to implement a V/f concept, which might be provided by a waveform with a fixed width and variable spacing for speed control. Sinusoidal waveform by means of PWM is probably ideal, but it might be worthwhile to try this simple approach as a benchmark. Torque control might be as simple as adjusting the pulse width and monitoring current, possibly even just DC supply current.
 
1 - 6 of 6 Posts
Top