Why does a DC series motor not reach a steady speed?

[cnh1995]

I have a confusion about the rotation of dc series motor..Suppose I put some load on it and turned it on..Due to no initial back emf, high current will flow in the armature (and field) and due to high starting torque, the motor will start rotating at say,1000 rpm. This will induce back emf which will reduce the current and the flux. Now, as the flux is weakened at the same 1000 rpm, back emf should also reduce leading to rise in current and flux. This will produce more torque and speed will increase, increasing back emf and reducing the current. If I'm right so far, this increase-decrease should go on forever and a steady speed should never be attained. But it's of course not the reality..Where am I screwing this up?? (I'm sure I am, but I just don't know where exactly..)

 

[Hesch]

I have a confusion about the rotation of dc series motor..Suppose I put some load on it and turned it on..Due to no initial back emf, high current will flow in the armature (and field) and due to high starting torque, the motor will start rotating at say,1000 rpm. This will induce back emf which will reduce the current and the flux. Now, as the flux is weakened at the same 1000 rpm, back emf should also reduce leading to rise in current and flux. This will produce more torque and speed will increase, increasing back emf and reducing the current. If I'm right so far, this increase-decrease should go on forever and a steady speed should never be attained. But it's of course not the reality..Where am I screwing this up?? (I'm sure I am, but I just don't know where exactly..)

If you load the motor by a constant torque, the speed, torque, flux, back-emf, current will find a balance. In practice the motor will probably not overshoot, thus not start an oscillation.

Of course if you have a motor with very high self-induction in the armature/rotor and very low inertia in the rotor/load, it could result in an overshoot, starting a dampened oscillation. But under the same circumstances, you can also bring a shunt motor into dampened oscillation

 

[jim hardy]

A series motor will accelerate until torque = friction and windage losses plus load. If unloaded it is likely to overspeed and self destruct. That's why you never connect one to its load by a belt.

here's a recent PF thread that's similar.
https://www.physicsforums.com/threa...c-motor-increase-when-flux-is-reduced.804006/

this image is posted there, of a series electric vehicle motor. It was in neutral and the accelerator got floored.
http://electricmr2ev.blogspot.com/2013/01/warp9-disaster-overspeed-motor.html

This occurred when the car was in neutral while in the garage, by mistake I carpet got stuck on the accelerator and the motor spun real fast probably 6K + at least it took about 2 seconds when the motor stopped

Centrifugal force throws the armature windings out into the air gap where they get chewed up.

I have a confusion about the rotation of dc series motor

i don't think you are the least bit confused.

old jim

 

[cnh1995]

If you load the motor by a constant torque, the speed, torque, flux, back-emf, current will find a balance. In practice the motor will probably not overshoot, thus not start an oscillation.

Of course if you have a motor with very high self-induction in the armature/rotor and very low inertia in the rotor/load, it could result in an overshoot, starting a dampened oscillation. But under the same circumstances, you can also bring a shunt motor into dampened oscillation

Is there any mechanism by which the torque,speed,flux,current will find balance? All the equations I have are for this balanced steady state. Is there one describing this process of balance?

 

[cnh1995]

A series motor will accelerate until torque = friction and windage losses plus load. If unloaded it is likely to overspeed and self destruct. That's why you never connect one to its load by a belt.

here's a recent PF thread that's similar.
https://www.physicsforums.com/threa...c-motor-increase-when-flux-is-reduced.804006/

this image is posted there, of a series electric vehicle motor. It was in neutral and the accelerator got floored.
http://electricmr2ev.blogspot.com/2013/01/warp9-disaster-overspeed-motor.html

Centrifugal force throws the armature windings out into the air gap where they get chewed up.

i don't think you are the least bit confused.

old jim

I just don't understand what's it that balances all the parameters and gives the motor a constant speed. If the field flux is constant
(shunt motor), I can understand the speed regulation in terms of armature current and back emf. But here, all parameters seem to be interdependent, so I don't know how the balance is established..I've made a huge mess..

 

[jim hardy]

But here, all parameters seem to be interdependent, so I don't know how the balance is established..I've made a huge mess..

Have you tried algebra?
The check on our mental process is to apply math and see if our mental model leads to an impossible conclusion.

The two simple equations for DC motor behavior are
(Counter-)EMF = K Φ RPM
T_(orque) = 7.04 K Φ I_a(rmature)

Let's try to get a single formula relating EMF, speed and torque.

________________________________________________________________________

T_(orque) = 7.04 K Φ I_a(rmature)
and for a series machine Φ is a pretty linear function of I_a
so lump the constants
T =7.04 k' I_a²
I_a = √(T/(7.04 k')

EMF = K Φ RPM
again Φ = k' I_a
EMF = k' I_a RPM
RPM = EMF/(k' I_a)
RPM = EMF / √(T/(7.04 k')
... lump some constants to make it look nicer and we have>> ta daaa--- (drum roll icon)
RPM = k'' EMF/√T
aha speed and torque have an inverse relationship as we'd expect

So - pick an applied voltage and a torque and you know RPM
what's limit of that as T_orque approaches zero ?

Speed of course won't make it to infinity the motor will fly apart as in that photo.That's why you noticed as a kid that Mom's vacuum cleaner motor speeds up when you put your hand over the suction port.
And it's why the vacuum cleaner motor has its impeller connected right to the shaft.
Old Chrysler automobile starters, which are series machines, had a small shunt field to prevent their self-destruction should the Bendix gear fail to engage.

So -
your mental model(if i understood you correctly) leads to a conclusion that agrees with multiple observations of real world phenomena.
Had it predicted instead the motor slows down under reduced torque we'd have to figure out why.
You should continue testing it in your leisure time.

But the truth is , you knew that all the time .
Have more faith ! Much of learning is discovering what we already know.

lastly - Please check my arithmetic - my "Senior Moments" are becoming laughably commonplace.

 

[Hesch]

Have you tried algebra?
The check on our mental process is to apply math and see if our mental model leads to an impossible conclusion.

Of course if you have a motor with very high self-induction in the armature/rotor and very low inertia in the rotor/load, it could result in an overshoot, starting a dampened oscillation. But under the same circumstances, you can also bring a shunt motor into dampened oscillation

cnh1995: You can make a model of a shunt-motor by Laplace transform:

K_m = motor constant [Nm/A] or [Vs].
R = rotor resistance [Ω]
L = self induction in rotor [H]
J = inertia of rotor and load [kg*m²]

The characteristic equation as for my model: L*J*s² + R*J*s + K_m² = 0. Finding roots in a 2. order equation you calculate some discriminand? ( right word? ): D = ±√( b² - 4ac ) and calculating a bit it leads to:

4*L*J*K_m² < (R*J)².

Otherwise the discriminand will be complex, and the motor will carry out a dampened oscillation, ( damping ratio < 1 ). It will make an overshoot.

 

[cnh1995]

cnh1995: You can make a model of a shunt-motor by Laplace transform:

K_m = motor constant [Nm/A] or [Vs].
R = rotor resistance [Ω]
L = self induction in rotor [H]
J = inertia of rotor and load [kg*m²]

The characteristic equation as for my model: L*J*s² + R*J*s + K_m² = 0. Finding roots in a 2. order equation you calculate some discriminand? ( right word? ): D = ±√( b² - 4ac ) and calculating a bit it leads to:

4*L*J*K_m² < (R*J)².

Otherwise the discriminand will be complex, and the motor will carry out a dampened oscillation, ( damping ratio < 1 ). It will make an overshoot.

Thanks a lot..I'm not fluent at LT but I have to be within next 25 days (for my end term exam)..I'll surely check out this mathematical model.. I'm sure it'll cement my understanding of motor..