Old wives tails and Back EMF - 9th wonder of the world? - RC Groups
Mar 04, 2005, 02:43 AM
Permanently Banned

# Old wives tails and Back EMF - 9th wonder of the world?

Mabey its not the 9th wonder of the world, but Back EMF is certainly interesting!

Ever hear that old wives tail about "runaway" motors? It goes like this - "Dont run your motor with no load on it - it could over rev - the RPM's could "runaway" and the motor will blow up!!!"

This is totaly untrue and Back EMF is the reason.

Ever wonder why you can put 25 volts to a motor with an internal resistance of .025 ohms and NOT pull 1000 amps?

Back EMF is why.

At the most basic level a brushless motor is a coil of wire and a moveing magnetic field.

It is important to remember that this is also what makes up a generator/alternator.

The ESC sends a pulse of current thru the coil of wire which generates its own magnetic field. This field attracts or repels the fixed magnets, causeing them to rotate.

The generator/alternator aspect of electric motors is one of the keys to how they work. When that pulse of current causes the magnets to rotate, those same coils of wire begin to act like a generator (actually an alternator is more accurate).

The same coil of wire that starts the rotation, also generates a voltage IN THE OPOSITE DIRECTION to the original applied voltage.

This is what is called BACK EMF. EMF stands for Electro Motive Force = voltage. The "Back" emf is opposite to the applied or "forward" voltage. Way back in the good old days we called it "Counter" Electro Motive Force.

It (Back EMF) is a key to how all modern ESC's work. It is also what determines exactly how fast a motor will spin - its Kv. Back EMF also determines how much current a motor will draw under any given load.

Note that the Back EMF is opposing the original applied voltage. As the motor begins to speed up, the back EMF increases. The faster the motor spins, the higher the back EMF. The motor will continue to speed up untill the point that the back emf ALMOST equals the original, or forward, voltage. At that point the RPM CANNOT go any higher.

The two voltages are never quite equal. The difference betwen the "Forward" voltage and the Back EMF is what makes the motor go and it is this small difference that determines how many amps your motor pulls.

Take a typical motor like a Hacker C50. It has a winding resistance of roughly .025 ohms. Using Ohms law, if you applied 25 volts to a load with only .025 ohms, you should get a current flow of 1000 amps!

Back EMF is why, in the above example, you can put 25 volts across a .025 ohm motor winding and only draw, for example, 30 amps instead of 1000 amps. In this example, the Back EMF is exactly 24.25 volts. The voltage difference is .75 volts.

Remember Ohms law - E=I*R - .75volts = 30amps * .025ohms

If the motor (for some mysterious reason) trys to spin faster, the back EMF will begin to increase. The voltage difference would then be reduced (or mabey reversed) and the motor would, in effect, have the breaks applied.

If the motor begins to spin slower (if a load is applied - like a prop) the back EMF will go down as soon as the rpm drops.

This creates a bigger voltage difference and so allows more forward current to flow, increasing rpm and power draw, untill a new balance is reached.

This is why an electric motor CANNOT "run away" in RPM. The laws of nature will not allow it.

All electric motors have basically a built in govonor.

Back EMF is also why electric motors draw more current as the load increases. Put a bigger prop on and it will spin slower.

This means less Back EMF which means MORE forward current and more power use.

The back EMF is what determines the motors Kv.

More powerfull magnets, or more turns of wire will both lower the Kv. Weak magnets or fewer turns of wire mean higher Kv. I used to think this had to be wrong - more turns of wire should produce a stronger magnetic field and so it should spin FASTER not slower - but things actually work just the oposite way.

A very strong magnet means more back EMF voltage is generated per rpm. More coils of wire (more "turns") also means higher back emf per rpm. MOre turns means more back EMF, which means lower current flow and therefore lower RPM/volt = lower Kv.

Fewer turns means Less back EMF means higher current and higher Kv.

Thats why motors with fewer turns of wire or weaker magnets have higher Kv. Cheep motors are always hi Kv because they have cheep/weak magnets and dont pack as much wire onto the stators.

Apply a certain voltage to a motor it will turn at a certain rpm, based on that applied voltage times the Kv. The rpm can only increase if the voltage increases. It can only go down if the voltage goes down - assuming the load is constant.

That old wives tail is just an old wives tail - brushless motors CANNOT "run away" if the voltage is constant.

oops! Have I said that already?

Back EMF is also why motors and ESC's are destroyed if a motor gets stalled. The stall current of a motor can be orders of magnitude higher than its full power current IF IT IS NOT ROTATING! When the motor stops spinning it also stops generating the back EMF and that allows the forward current to follow Ohms Law.

It is also why ESC's and motors get so hot if they fail to start.

Ever had a motor go into the jerky back and forth twitching instead of starting? Check the ESC and motor and you will find them both getting VERY HOT VERY FAST. The ESC is applying full voltage pulses and the motor is not providing enough back EMF to reduce the current flow. This means very hi current flows and that means lots of heat.

Never continue to try to start a motor that is not spinning. Find out why.

Earlier I said a motor could not and would not "run away" in rpm given a fixed voltage.

There are actually two failure modes for motors that can cause a dramatic RPM increase. It is still NOT a "run away" situation tho.

If you ever have a motor that suddenly starts to act like it has a much higher Kv than it should, one of the two following things has probably happened (assuming the same voltage is being applied and the same load).

1) If you over heat the magnets in a motor they will loose some of thier strength. This will reduce the back EMF generated, thus INCREASING the motors efective Kv. The motor will draw more current and turn at a higher RPM. The effeincy will go way down tho and it will also run a lot hotter than before. This extra heat kills the magnets even more which causes higher Kv and more current and more heat and higher Kv and more current and more heat.....melt down.

2) It is possible for the windings to partially short internally. If this happens just so without causeing a total short - it can effectively reduce the number of turns the motor has. This again reduces the back EMF, increases current flow and rpm and heat.

Mode 2 is pretty rare tho. A short typicalls only happens in one "leg" of the three phases and it will cause a very unbalanced load condition. The motor will usually just not start and go ahead and self destruct or destry the ESC or both.

Back EMF also plays a key part in how ESC's work. Modern controllers detect the back EMF pulses and use them to determine the timeing of the power pulses - the "advance" settings. They also use this to control motor direction by detectinng the phase of the pulses.

Back EMF can also destroy a controller. If you suddenly reduce throttle, but the motor continues to spin at a hi rpm (free wheeling prop with no break for example), the back emf can suddenly go very hi. This can sometimes produce excessive reverse voltage conditions on the FETS and cause them to blow. Older controllers were very subject to this problem.

Ever noticed that sensorless motors dont like to run at very low "idle" speeds? This is because at low rpm levels they dont generate enough back EMF for the controller to detect and relyably sync with inorder to keep the motor running.

Sensored motors on the other hand - like some fo the older Aevox motors, would run at super low rpms. the sensors told the ESC exactly how fast the motor was spinning and how to time the power pulses.

Start up time and very low rpm running are just about the toughest things for motors and controllers. If it does not go well, both can be destroyed very quickly.

Low rpm conditions are not the same as 'no load' and are also not the same as partial throttle.

Hope you found this interesting

Larry

P.S. Now - STOP spreading that old wives tail!

PPS - I have ofcourse over simplified several things here and left others out and totally ignored losses and inefficiencies and PWM etc etc. I may even have gotten something 'bass ackwards'.

 Mar 04, 2005, 07:26 AM Registered User Bit long winded, but basically correct However running wiothout a prop CAN induce rather high RPM, as the RPM drops under load (the effective viltage teh motor 'sees' reduces by IRo). On cheap inefficient motors the volt and RPM droip under load is pretty high - 30-50%.
 Mar 04, 2005, 07:56 AM Crash Master The 'runaway DC motor' old-wive's tale is quite correct, it's just that it does not apply to a permanent magnet motor... It's intended for DC 'traction motors', or series wound motors. You see, the Back EMF, or generated voltage, is a function of the winding cutting through the magnetic field as the armature rotates. The stronger the field, the higher the voltage, thus the reason it HAS to turn faster or slower depending on the turns of the armature and the strength of the field. A shunt-wound motor is normally run with an exciter current on the field, producing the same type of constant field as a permanent magnet. The series-wound motor is a different animal. The field is wound to handle the armature current, and produce a similar field with this current level... However, at low loads, the field is very weak (Magnetic flux = ampere*turns), so the motor HAS to rotate very fast to generate the counter EMF... This can be high enough to produce sufficient centrifugal force for the armature to explode.
 Mar 04, 2005, 09:53 AM Registered User A very thorough dissertation!
 Mar 04, 2005, 10:49 AM Registered User Very informative and clear! JL
 Mar 04, 2005, 11:31 AM Registered User Thanks Larry for the informative post>
 Mar 04, 2005, 06:12 PM Registered User Just remember that the above only applies to permanent magnet motors, and in the general motor world these are quite rare. For 3 phase motors, induction motors are much more common and for dc motors, series or shunt wound motors are much more common. Steve
 Mar 04, 2005, 09:14 PM Permanently Banned Thanks for the positive comments! Gene and Steve - I dissagree - to some degree - mabey - with both of you. Perhaps we should make sure we are talking about the same things here first by defining the terms. When I say "run-away" rpm what I am refering to is having the rpm increase without limit or controll when the motor is run under no-load at a fixed voltage. "Over reving" is a different animal entirely. I define that as simply exceedng the maximum physicaly tollerable rpm. Any motor can be "over reved" by running it at an rpm where the rotor or bearings give out or burst. Just increase the voltage until that point is reached. My point is that no motor will "over rev" all on its own as long as the applied voltage times the Kv results in an RPM that is less than the physical max. the motor can handle. As far as I understand the theory - even series wound motors will NOT "run away" in rpm. They operate under the exact same principals as our brushless motors. Current flows are limited by back emf in the same exact maner and therefore these motors also have a "Kv" and rpm is controlled by voltage. The vast majority of drill motors are series wound. They dont "run away" when run with no drill bit installed. I have taken many apart over the years and run them in off the wall applications. They behave as they should Series wound motors may very well "over rev" when run at no load and too hi a voltage - but that is a differtent thing all together from a "run away" situation. Some other types - some induction and all syncronus motors for example - also follow these same principals, but the construction is such that the frequency of the input power (and number of "poles" and other construction details) is what determines rpm rather than voltage. If it wasnt for back emf - all electric motors - no matter what design - would draw so much current they would melt before they could get any work done. Ohms law will not be ignored It is the back EMF limiting the max current flow that also controlls and "governs" the rpm, preventing a "run away" situation. "over reving" is always possible, but "run away" never is. Another way to look at it - if a motor could "run away" in rpm - you would have a perpetual motion machine Larry
 Mar 04, 2005, 09:44 PM Registered User I think there may be a definition problem here. Motor "Run-Away" is not perpetual motion, nor is it unlimited rotation speed. It is simply "over reving" a motor far beyond it's tollerance causing motor destruction, fire, explosions, and the like. If you go back to the basic principals for motor control, you will find some motor control systems / motor designs that cannot successfully recover from a "run-away" event. Just because you understand the operation of one style or several styles of motors does not mean that you can make a blanket statement about all motors. When run-away occurs, the motor does stop. The problem is that it is usually on fire when it does. Of course the laws of electrical currents, EMF, and hysteresis all still apply. The problem is that when run away occurs(on large 500HP+ DC motors) the DC controller cannot stop the rotation, and the system burns up. I have been working in the mines a long time, and was on site five years ago when a large motor gear box shattered, and the motor flew apart in flames. The whole area had to be rebult because of the fire damage. You see if the main power had been removed in time, the motor may have been stopped before such damage occured, but the motor was operated remotely, and the circuit breakers tripped after the motor was already on fire.
 Mar 04, 2005, 10:36 PM Permanently Banned I am going to reverse myself and admit I was, for most all practical purposes, wrong about series wound motors. They do have an effective Kv that is, essentially, inversely proportional to load. That means they can pretty much "run away" under no load conditions. They are still self limiting, and back emf still applies - just not exactly as I described I was wrong. Darn! And I was doing so well this year too Larry Last edited by Larry3215; Mar 04, 2005 at 10:44 PM.
Mar 04, 2005, 10:44 PM
Permanently Banned
Quote:
 Originally Posted by acrobatorcrash I think there may be a definition problem here. Motor "Run-Away" is not perpetual motion, nor is it unlimited rotation speed. It is simply "over reving" a motor far beyond it's tollerance causing motor destruction, fire, explosions, and the like. If you go back to the basic principals for motor control, you will find some motor control systems / motor designs that cannot successfully recover from a "run-away" event. Just because you understand the operation of one style or several styles of motors does not mean that you can make a blanket statement about all motors. When run-away occurs, the motor does stop. The problem is that it is usually on fire when it does. Of course the laws of electrical currents, EMF, and hysteresis all still apply. The problem is that when run away occurs(on large 500HP+ DC motors) the DC controller cannot stop the rotation, and the system burns up. I have been working in the mines a long time, and was on site five years ago when a large motor gear box shattered, and the motor flew apart in flames. The whole area had to be rebult because of the fire damage. You see if the main power had been removed in time, the motor may have been stopped before such damage occured, but the motor was operated remotely, and the circuit breakers tripped after the motor was already on fire.

Those are very valid points. Commercial and industrial motors are often run under load conditions and applied voltages such that they will self destruct if the load abruptly goes to zero - say if a shaft breaks etc.

Perhaps I should pull my comments back a bit and not make such broad statements without at least some qualifications attached

Brushless permanent magnet hobby motors cannot "run away" when run under no load conditions.

I hate it when my feet dont fit in my mouth

Larry
 Mar 04, 2005, 10:51 PM Registered User I agree with you Larry, Back EMF still applies, and is one of the contributors of the failure that occurs.
 Mar 05, 2005, 06:05 AM Registered User I think you have it finally. Series wound motors are not good to run offload, and any permanent magnet motor can be run at RPM past is point of destruction on a voltage that would be safe if it were loaded. Some brushless motor/esc combos seem to also like some mass on the motor to spin up properly. But thats a separte issue.
 Mar 05, 2005, 11:03 AM Crash Master The culprit in running a PM motor with variable loads is the resistance, and thus resulting drop in effective voltage. This 'stealing' of the voltage is the IR drop. It steals voltage from the inductance, and instead is on the resistance of the winding... The inductor voltage/current creates torque and results in a given speed, the resistor voltage/current creates heat. There is a vector relationship between these, so the math is not exactly linear, but for practical purposes we can say the resulting speed drop with load increase is directly proportional. Why does this happen? Why doesn't the motor HAVE to spin faster to make up for the resistive loss and resulting loss in back EMF? The fact is the voltage is effectively lower with higer current, because the resistance is stealing voltage... Thus the motor does not HAVE to turn as fast to generate this voltage
 Mar 05, 2005, 11:43 AM Registered User From my college days I seem to remember the shorthand for back emf was known as Lenz's Law, which states that all inductives effects are essentially suicidal - in other words the reactive effect always opposes the force that creates it. It also fits in with the difference between motors and generators - I'm sure many readers will be familiar with Fleming's left and right hand rules from their elementary school physics. I have recently had some difficulty persuading a corespondent both to this forum and one I haunt in the UK that resistance is nothing but a nuisance in our motors and that the ideal motor has zero resistance windings. As you rightly say, it's all down to our friend back emf. I seem to remember that traction motors (we used to have lots of electric trolley buses when I was a boy in the UK - like trams, but with normal wheels, which get their power from a pair of overhead wires) were always wound with the field in series with the armature because they have maximum torque at zero rpm. I was always told that they were liable to overspeed off load. A very good offering, Larry, thanks. Geoff