Stuka Stunt Control Line Forum
Archive, 2000–2021 · recovered from the Internet Archive
Forums › Stuka Stunt Main Forum

A question on electric speed controlers

Stuka Stunt Main Forum · 13 of 13 known posts recovered

steve1935 · Feb 05, 2007 09:51 PM

#0 source
Is it possible to increase tthe power to the electric motor when the plane changes attitude as in the 4-2-4 power change.
Steve

KDOHERTY · Feb 05, 2007 11:20 PM

#1 source
Steve,

Although it is within the realm of being doable, it has not shown itself to be necessary outside of the power modulation that the ESC's Governor Mode already provides. As we move forward with our research, we will be looking at improving the system and this is one area that we will almost certainly look at. It is not a trivial undertaking but well within our capabilities if it were to prove beneficial. To say that a properly sized electric motor/battery in an appropriate airframe is not lacking in power is a major understatement. The system I am flying will turn a 14x6 prop fully within governor mode (will not fall off) with power to spare.

Kim.

Brett Buck · Feb 06, 2007 12:41 AM

#2 source
>Is it possible to increase tthe power to the electric motor
>when the plane changes attitude as in the 4-2-4 power change.

Yes, although you don't want it to be based on attitude, but on load.

Kim may be turn out to be correct that you may not necessarily need it if you get a big enough system. That's how a lot of people are running their 4-strokes. I would sure be tempted to try it that way, if I was going to do it.

Brett

Alan Hahn · Feb 06, 2007 11:04 AM

#3 source
Steve,
In actual fact, simply raising the nose will automatically increase the power in an electric system. This is due to the "fact" that the maximum power output of an electric motor will occur at 1/2 the rpm of the no-load value. So for example, say you have a motor with a "kv" of 1000 rpm/volt. If that is hooked up to a 10V battery with no prop, it will spin at 10*1000=10000 rpm. Therefore the maximum power output is at 5000 rpm (1/2of the no-load value). I note that the power and torque are both zero at the no-load rpm. Maximum torque is at zero rpm. Since power = torque * rpm, it will peak right at 1/2 the no-load value. All this ignores friction and hysteresis losses in the motor, but the basic numbers are not too bad.

So if you prop the motor to run level flight at 8500 rpm, this is above the maximum power rpm at 5000. When you pull the nose is, the prop load will increase (the rpm will fall) and torque and power to the prop will increase. I note that the current (amps) from the battery will also increase, as well as resistance losses (heat) in the battery, ESC, and motor.

I am guessing this isn't exactly what you meant. If you use a ESC which has a governor mode, and you set up the intial throttle value less than max, then as the rpm tries to drop, the ESC will try to increase the effective voltage (and thus current and power) to the motor, up to the maximum the battery can supply. This is more power than the non-governor case.

Even this may not be what you mean. In the Fox 35 for example, when it comes out of a 4 stroke, the rpms actually increase (I believe although I haven't actually looked carefully enough), so this may be more effective at keeping the airplane velocity from dropping too much in a climb. I have heard people are thinking about coming up with a way to implement this in an ESC, but I think it might be a little tricky if all you have to go by is the motor current draw. If you also have some way to input the plane attitude, maybe that can be used.

At some point you could program in the whole flight power needs (now if we cann just get rid of the weakest link--the pilot!), and then I would have a real question about how "fair" that is. Also and why not a adjustable pitch prop!
Alan

cfrizell · Feb 06, 2007 02:30 PM

#4 source
Ummmmm

Brushless (3 phase) motors and controllers are usually set up to be "constant speed" devices. The motor will hold the set speed until the current limit point is reached (if there is a current limit function) and then fall linearly off from that speed if the controller has the ability to track down in frequency when safe load is exceeded.

Conversely, when the load comes off the motor it will NOT speed up like an IC motor. (Well, a little bit, maybe 5% or the slip value). Again depending on the controller it is possible to actually put a braking force on the prop if it is driven over set speed (synchronous speed)

The marvel of electronics is that almost any desired characteristic can be obtained within limits; constant speed, constant torque or constant power. The simplest and most usual is a constant speed setup.

Charles

Alan Hahn · Feb 07, 2007 10:55 AM

#5 source
Charles,
Not all ESC have governor mode. Remember the throttle pulse width modulation which sets the voltage to the motor, not the current or rpm. The current and rpm depend upon the load from the prop. So in this "normal" mode, rpm is not kept constant. Now governor mode will do just that (try to keep rpm constant within the limit of your maximum battery voltage), but you need to set it in the ESC's which support it.

Note that the pulse width modulation (which is the throttle control and brushed ESC's have it too by definition) is different that the switching a brushless controller does to supply the coils witht the correct phase of the current to keep the motor spinning.
Alan

cfrizell · Feb 07, 2007 01:15 PM

#6 source
AC motors (and DC too) are interesting.

As you increase the frequency (speed) you increase the voltage because what limits the current is saturation of the core(s). Just like a transformer. So usually increasing speed also increases power.

But what is also important on AC motors is the slip - that is to say the difference in rotational speed of the electric field and the rotor.

As EE's know, there are two basic types - synchronous motors with a fixed magnetic field provided by permanent magnets (electromagnets in large motors) and a rotating field. These operate with very little slip indeed, if any. That is to say, the rotor follows the field frequency exactly. Torque rises with slip angle and then suddenly falls off when the rotor is pulled out of synchronism.

Then there are squirrel-cage motors which form the vast majority of ordinary AC motors where the rotating field induces current in the rotor bars. These run with up to 10% slip. With these the torque increases as slip increases until "pullout" torque is reached at about 10% slip - then the turque falls very rapidy and the current rises to a high value, often about 6 times the normal full-load current.

Designing AC speed controllers is tricky because the field frequency must start from a low value as the rotor accelerates from rest, if maximum torque is required during acceleration (which it usually is) As some will remember model brushless controllers used to have sensor wires to tell where the rotor was (or, its speed). Now I believe its done very cunningly by sensing the voltage / current phase angle, thus doing away with the need for sensors. I have never designed this sort of controller.

In contrast, DC motors will run at a speed determined by the voltage, the current increases as the load increases. Torque is always directly proportional to current, and speed to voltage. They will slow down when loaded due to the resistive losses of the armature and brushes which effectively reduce the armature voltage. This applies to shunt-would and permanent magnet motors - there are also series wound (car starter motors|) and compound wound. It's a BIG subject!

Charles

Pat Mackenzie · Feb 07, 2007 08:50 PM

#9 source
>AC motors (and DC too) are interesting.
>
>As you increase the frequency (speed) you increase the voltage
>because what limits the current is saturation of the core(s).
>Just like a transformer. So usually increasing speed also
>increases power.
>
>But what is also important on AC motors is the slip - that is
>to say the difference in rotational speed of the electric
>field and the rotor.
>
>As EE's know, there are two basic types - synchronous motors
>with a fixed magnetic field provided by permanent magnets
>(electromagnets in large motors) and a rotating field. These
>operate with very little slip indeed, if any. That is to say,
>the rotor follows the field frequency exactly. Torque rises
>with slip angle and then suddenly falls off when the rotor is
>pulled out of synchronism.
>
>Then there are squirrel-cage motors which form the vast
>majority of ordinary AC motors where the rotating field
>induces current in the rotor bars. These run with up to 10%
>slip. With these the torque increases as slip increases until
>"pullout" torque is reached at about 10% slip - then
>the turque falls very rapidy and the current rises to a high
>value, often about 6 times the normal full-load current.
>
>Designing AC speed controllers is tricky because the field
>frequency must start from a low value as the rotor accelerates
>from rest, if maximum torque is required during acceleration
>(which it usually is) As some will remember model brushless
>controllers used to have sensor wires to tell where the rotor
>was (or, its speed). Now I believe its done very cunningly by
>sensing the voltage / current phase angle, thus doing away
>with the need for sensors. I have never designed this sort of
>controller.
>
>In contrast, DC motors will run at a speed determined by the
>voltage, the current increases as the load increases. Torque
>is always directly proportional to current, and speed to
>voltage. They will slow down when loaded due to the resistive
>losses of the armature and brushes which effectively reduce
>the armature voltage. This applies to shunt-would and
>permanent magnet motors - there are also series wound (car
>starter motors|) and compound wound. It's a BIG subject!
>
>Charles

FWIW, the motors used in models are brushless DC motors. They have the same speed/torque relationship as normal permanent magnet brushed DC motors.
The ESC is just taking the place of the mechanical commutator and external chopper speed control.
Earlier motors had position feedback to time the commutation. Modern motors sense the rotor position by looking at the "off" winding and syncing to it.
( 2 winds are powered at any one time. One is always off. As the motor rotates they take turns being the off winding.)

They are not AC induction motors or vector (phase) controlled AC servo motors.
The closest thing they are in AC would be a permanent magnet synchronous motor, but the drive circuits don't generate 3 phase sine waves.

In addition to the commutation and chopping (PWM) function, some speed controls add a governor mode to try to maintain a target rpm as load varies.
This is done by adjusting the PWM. Once you hit 100%, the motor just reverts to the standard load/speed curve.
So in order to use governor mode you have to "dial back " from full power to leave some headroom as the load increases or the battery voltage drops during flight.
Some are far better at maintaining rpm without undershooting and overshooting then others.

Pat MacKenzie

Brett Buck · Feb 07, 2007 02:20 PM

#7 source
> Now governor mode will do just that (try to
>keep rpm constant within the limit of your maximum battery
>voltage), but you need to set it in the ESC's which support
>it.

OF course, constant RPM is not the same as constant power.
The torque required to turn the prop at a given speed varies considerably with maneuvering. More torque*same RPM = more power.

Brett

Alan Hahn · Feb 07, 2007 08:01 PM

edited#8 source
Brett,

I never said constant power, as a matter of fact, in all the cases I talked about, the electric sytem will supply more power, and more torque.

In the case where the voltage stays constant, the rpm will drop (the prop load increases), but the torque will rise (when the rpm drops, the back emf from the motor will drop, allowing more current to flow). As long as the original rpm is set above 1/2 the no-load value, the power supplied to the prop will increase.

In the second case (governor mode) , the average voltage supplied to the motor increases. The ESC ramps up the throttle to try and keep rpm constant within the limits of the battery itself. In the governor case since the rpm is held constant (and the back emf stays the same), more current will flow due to the higher voltage giving more torque, so again more power to the prop.

The thing to remember is that for a constant voltage, the torque linearly decreases from its max at 0 rpm (I am assuming here that the iron does not saturate). This decrease is simply due to the back emf from the motor decreasing the current flow from the battery. When back emf=battery voltage, thats just the no-load rpm--again ignoring any friction.
All the resistances ( internal battery resistance, ESC resistance, wire/connector resistance, and coil winding resistance limit the maximum current at zero rpm and set the torque curve for a fixed battery voltage. More battery volts mean a higher torque at any rpm, and a higher no-load value (nominally just kv*Vbattery ignoring friction and hysteresis losses).

Anyway since torque linearly decreases from zero rpm, and rpm itself increases linearly (obviously) from zero rpm, the power curve ~rpm*torque peaks out at 1/2 the no load rpm value. A higher battery voltage (or an ESC at a higher throttle setting) will lead to a power curve which peaks at a higher rpm (and higher power too) than the lower voltage case.

Anyway, the motor rpm is not a simple function of voltage (except for the no-load value), but depends both on the voltage and the loading.

added correction---I really mean eddy current losses in the iron, not hysteresis in the above (and anywhere else I used that term. Don't know why the neurons in my head grabbed it!
Alan

Brett Buck · Feb 07, 2007 10:16 PM

#10 source

>I never said constant power, as a matter of fact, in all the
>cases I talked about, the electric sytem will supply more
>power, and more torque.

Oh, I never suggested any differently, just though I would point it out.

Brett

Dave Moffitt · Feb 08, 2007 10:10 AM

#11 source
>Is it possible to increase tthe power to the electric motor
>when the plane changes attitude as in the 4-2-4 power change.

I am NOT a stunt guru, so please bear that in mind when you read what I say here....however I am an Electronics Tech-weenie....
What I believe we want from our motors is that they give a little more power when the plane pulls a bit more "G" ( thinks.. maybe a lot more "G" ) than level flight.
Surely this could not be too hard to do.....make a device like a car contact breaker.. in level flight it does nothing but when pulling some "G" it moves and cuts in a circuit to push more current to the motor.
Simple theory... but might take a lot of juggling with springs, distance travelled etc to achieve, but IF it could be done it would be very light.... would it be worth it?....
Not to me.... but to some... it would be a way of increasing power in high "G" manouvres without any external control input, which I believe is banned.
Dave Moffitt

KDOHERTY · Feb 08, 2007 10:44 AM

#12 source
>>Is it possible to increase tthe power to the electric
>motor
>>when the plane changes attitude as in the 4-2-4 power
>change.
>
>I am NOT a stunt guru, so please bear that in mind when you
>read what I say here....however I am an Electronics
>Tech-weenie....
>What I believe we want from our motors is that they give a
>little more power when the plane pulls a bit more
>"G" ( thinks.. maybe a lot more "G" ) than
>level flight.
>Surely this could not be too hard to do.....make a device like
>a car contact breaker.. in level flight it does nothing but
>when pulling some "G" it moves and cuts in a circuit
>to push more current to the motor.
>Simple theory... but might take a lot of juggling with
>springs, distance travelled etc to achieve, but IF it could be
>done it would be very light.... would it be worth it?....
>Not to me.... but to some... it would be a way of increasing
>power in high "G" manouvres without any external
>control input, which I believe is banned.


Dave,

I thought you were busy making aluminum ends for your arrow shafts !!

There are much simpler ways to do what you are suggesting using accelerometers. We certainly have the ability to do this right now but we are still busy dotting the "i"'s and crossing the "t"'s on the basic timer/processor control system. One step at a time.

We already have the ability to maintain a constant RPM regardless of the load on the motor through the ESC's Goveror Mode. This type of power control is permitted within the rules.

As an aside, there seems to be a feeling out there that electric power systems are lacking in power or do not run in a manner that a top level flyer would desire. If a model is lacking in power it is a result of having not done the necessary homework and testing to determine the appropriate components. There is nothing inherently lacking in a properly spec'd electric power train. As to the run characteristics, you will never experience a more uniform run if you could wish it so. The ESC deals instantly with the extra load in corners and the model just does not decelerate. It also does not wind up. Could we improve on it? Sure, and we will probably run a plane with an accelerometer this summer. Stay tuned!

Kim.