Here's my "question of the week":
Suppose I have a DC motor that I can either modify the strength or number of
magnets; or modify the number of windings. Given that
I can only use 12 VDC, what would be the best way to increase RPM - assuming
torque is not even an issue ?
Stronger or more magnets ? More windings? Less windings ? Remember that is has
to stay with a 12 vdc power source.
For a given torque the Power developed at the armature is WT where W
is the speed and T is the developed torque.
Is this a shunt,series or separately excited machine?
Anyway, to get more speed you need more power and the back emf E =WT.
Hence E must be increased.
However E is proportional to flux X W so that if the flux is increased
E will increase and hence the speed.
So you need more magnets! This will also increase the torque since it
is proportional to flux X armature current.
Simply a bigger motor (with larger magnetic field) has more speed
unless you increase the armature voltage which could damage the motor.
Did you believe that?
No? Well you are right. Here is teh true story.
V-E =Ir where r is armature resistance V is armature voltage (fixed)
and E is back emf.
Now E =const (sayk) X flux X W
hence speed W directly proportional to (V-Ir)/flux
Hence you need to REDUCE the flux to increase the speed. This is known
as field weakening and with a shunt connection you can do it easily by
adding a resistor in teh field. I suppose you can do something similar
if you have a separately excited field. This sounds counter intuative
but it is the case. It will increase the losses in teh field and
reduce the efficiency of course.
I thought I did...
If you have an existing motor the you need to run faster at a constant
input voltage, you can:
1. Reduce field strength as others have mentioned (but this is bad for
2. Reduce number of windings
Since you are talking about a PM motor the windings are armature windings,
so you would need to unwind the current armature windings and rewind with
fewer turns of thicker wire.
Ok, so I'm trying to figure out the theory that makes this work and I'm not
getting very far.
The field strength of the windings when the motor is stalled (not spinning)
it seems to me will be a function of the current times the number of turns
in the windings. And the stable state current will be limited only by the
coil resistance if you have a true constant voltage supply.
These factors combined with the strength of the permanent magnets would
directly define the stall torque (along with the physical size and shape of
the armature and magnets of course).
Now, from what I grasp in the other posts, when it starts to spin, a back
emf is generated in the armature which creates a voltage to offset the
supply voltage. So the faster it spins, the more back EMF, that reduces
the voltage across the windings and reduces the torque.
So for a constant torque load, the motor will reach a steady state RPM
value where the back emf reduces the voltage and hence the current, and
hence the torque until it equals the torque of the load. So the RPM of the
motor will be defined as the speed at which the back emf reduces the torque
to match the load.
So, if we reduce windings, what happens? The back emf is reduced so that
would imply a greater RPM value for the steady state. But the torque is
also reduced, so the means the motor will reach equilibrium at a lower RPM.
But coil resistance is also reduced if the windings are all series wound,
and that will increase current and increase torque.
So we have at least three factors at work here trying to push the steady
state RPM value in different directions. With out the exact math of how
these three factors work against each other, it's not at all obvious to me
which will win.
Same thing seems to apply for the other option of reducing the strength of
the PM. Back EMF is reduced, but so is torque. Oen effect would tend to
cause higher RPM steady state, and the other would cause a lower RPM steady
state. The question is which one wins. But I suspect the answer to that
depends on where alone the RPM torque curve the previous steady state was
Lets look at the end of the curve near the stall torque (0 RPM). If we
reduce magnet strength, we have reduced torque. Because the motor isn't
spinning, there is no back emf so that as no effect at this end of the
So, if the motor was operating very a very heavy torque load at a very slow
RPM, reducing the field strength could drop the stall torque to below the
torque of this load, at which point, the motor would simply stop spinning.
So at this end of the curve, where back EMF has very little effect,
reducing magnetic strength will reduce RPM as I see it.
But, if the motor was operating at the other end of the curve, near the no
load RPM end, the only torque on the motor is from the frictions in the
bearings, and the primary factor controlling RPM will be back EMF, not load
torque. So if you reduce the magnetic field in this case, I could believe
that the no load RPM would increase because the reduction in in back EMF
has a much larger effect than the reduction in torque. So that would push
the equilibrium point to a higher RPM.
But this only works as long as the torque is small compared to the back
emf. As the magnet gets weaker and weaker, the torque becomes a larger and
larger factor in determining the equilibrium point. Once the torque is as
much a factor as the back emf, decreasing the magnetic strength further
will simply lower the RPM, instead of increasing it. At least that what it
seems to me like what will happen here.
So, even without knowing the exact math that applies here, I'm pretty sure
that the answer as to whether you want to increase or decease windings, or
increase or decrease magnet strength will depend on what part of the torque
RPM curve the motor is operating in. Changing either will change the shape
of the curve - my guess is that it mostly changes the slope of the curve -
making one end go higher and the other go lower. So for a fixed torque
load, the new RPM value will depend on which part of the curve the previous
equilibrium point was located.
So as far as I can tell, the question has no correct answer. It depends on
the motor and it's load. For a motor operating under heavy load, you have
to increase windings and increase magnet strength to make it run faster.
And for motors operating under a light load, you have to do the inverse.
Without knowing all the math I think that's the best answer I can deduce
based on how much I currently understand about PM DC motors.
Notice that they said reduce the number of turns AND use heavier wire.
The heavier wire will reduce the resistance and allow you to push more
current through the motor, giving you the same field strength that you
had with the higher turn count.
Other than missing the wire guage change you pretty much got it.
Yeah, that sounds good. I did miss that. If you increase the wire
thickness so that the reduced resistance gave you enough current gain to
get back the field strength lost by the fewer winding, you would still have
all the same torque but less back EMF over the entire curve giving you
increased RPM over the entire range.
However, as a question of whether it's possible, the size wire you would
have to switch to might make the windings too large to fit in the current
motor making it impossible to actually implement. :) But the theory sounds
like it might work. That's cool, I've learned a little more by this.
On Sep 10, 4:28 pm, email@example.com (Curt Welch) wrote:
I don't think thats right. Changing armature current only effects
torque. You would reduce armature resistance too and the back emf
would go up and the speed woudl not change. You must change the
armature voltage or the field -field is teh flux or magnetic strength.
As I understand it, because you have reduced the number of windings, back
EMF goes down. Back emf is not a function of resistance, it's only a
function of the number of windings cutting the field as the motor rotates
as I understand it.
The speed would increase because you have both increased torque, and
reduced back EMF while keeping the field strength the same. The increased
torque, acting against a reduced back emf, causes the motor to reach an
equilibrium point at a higher RPM. At least that seems logical to me.
On Wed, 10 Sep 2008 23:48:20 -0700, HardySpicer wrote:
The original question asked about increasing the speed
"assuming torque is not an issue"
If torque is not an issue you are running the motor in
a mode where its speed will be limited by BEMF, Reducing
the number or armature turns means the motor will have to
rotate faster to generate the ~12V of BEMF where motor
current (and therefore torque) are in equlibrium with
Think of it as making your 12V motor into a 9V motor but
still running it on 12V...
Show me the equations first. This up a bit down a bit stuff is for
Motor current and torque are in equilibrium? Torque is proportional to
Please show your equations to back up the claims. A steady-state model
Means you have no idea!! The design of electrical machines has been
well defined by simple equations for at least a century or more and yo
ucannot state your point. I take it you are not an electrical
That's good info - thanks!
But I'm asking a basic, theoretical question here. Let me put it a different way:
Given the same voltage supply; torque not being an issue; what results in
more RPM ?
Stronger magnets or more windings ? both ?
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.