**Important** reality check on motor, voltage, current etc.

Joined
Oct 28, 2008
Messages
14,539
Location
Manhattan Beach, CA, USA
I spent a lot of time making this for you guys.

It explains a critical element of putting together a battery/motor/controller system.

The phase current and battery current relationship, and how and why phase currents are multiplied.

Why does this matter to you? Because battery current is never seen by the motor, only phase current, and phase current is what determines torque output of a motor, as well as the heating the FETs see.

Hopefully this will clear a little fog away for some folks.


pwmcontroller.png
 
Excellent work Luke!

That certainly helps me understand the principals a bit better :)

burtie
 
Burtie said:
Excellent work Luke!

That certainly helps me understand the principals a bit better :)

burtie


Thank you my friend. :)
 
LFP nice write up. I have one comment. I’m not sure if I have this right.

You state that a motor with 2 turns will have half of the PWM “on time” of a 4 turn motor. I think you mean that since the KV constant for the 2 turn is 2x the KV of the 4 turn to produce the same speed, the throttle would be cut down to get ½ the PWM on time, if the operator wanted to maintain speed. I guess your other thought may be that the controller would reach battery current limit, but that might not be true.

If the operator stayed with WOT the PWM would stay the same 2 or 4 turn, if the Current limit would support it. The same voltage used with 2 turns would act with the higher KV to give twice the speed.

When the 4 turn setup was used at low speed, its higher KT would give more torque than the 2 turn by a factor of 2.

The power drawn from the battery would probably be the same in both cases, limited by current limit. The power dissipated in the controller at 100% duty cycle would be the same voltage, with current differences because of inductance and resistance differences . The way that power was applied to the wheels would change. 4 turns High torque at low speeds. 2 turns low torque at low speeds and the inverse.
 
I understand your general message, but this part seems misleading or misworded.

"anytime PWM is used to regulate battery currrent, the average phase voltage drops, and the average phase current increases to balance the power equation"

If the average phase voltage drops, how does the average phase current increase?

Ignoring BEMF calculations (It doesn't affect the general principle), it's simple Ohm's law. If you reduce voltage to a resistor (The phase resistance), you decrease current. What would make a motor the exception?

Another simple rule is that anytime the motor slows down in the same environment/situation, the lower the torque the motor outputs (It takes lower torque to sustain a lower speed). Since there's lower motor torque, you can deduce the average phase current is decreased.

I think everyone needs a meter that shows them phase current stats. Perhaps direct observation would be a better teacher than simple electrical laws.

BTW, on a separate tangent, I think that the "average voltage" is really just a model for understanding it. With an actual controller, I believe that the motor sees the full voltage of the battery during the ON part of the mosfet, but the motor's voltage is a fraction of that depending on the duty cycle due to inductance. Ok, now saying that, I guess that it's more than a model. :lol:
 
As one of those with the 2 turn per winding motors, I understand the issue. The controllers, however, have program limits on the phase current as well as the battery current limit. I accept that I probably have give up low end torque compared to a higher turn count motor by limiting the phase current, but I get higher top speed in exchange. I wonder if the width of the pulse changes to adhere to the phase current limits. If not then which current limit dominates, or does the battery current get limited when either the the battery or phase limit is reached?

I have 2 controllers with fresh off the charger voltage limits of 150v. With my current 20s packs, I've been up to just over 60mph. These controllers can handle a 37s pack, and I'll add a second motor to help overcome the additional wind resistance, along with address the aero of the bike a bit. This is really just for a few runs, and I don't want to risk blowing my controllers needlessly.

Do the math, what would you do, go for it, or stay at the low voltage?

John
 
It's not clear to me how a controller would affect a limit based on Phase current. If it found Phase Current too high and cut PWM duty cycle further it would increase Phase Current based on current multiplication. I think all it could do is shut down.

Don

John in CR said:
As one of those with the 2 turn per winding motors, I understand the issue. The controllers, however, have program limits on the phase current as well as the battery current limit. I accept that I probably have give up low end torque compared to a higher turn count motor by limiting the phase current, but I get higher top speed in exchange. I wonder if the width of the pulse changes to adhere to the phase current limits. If not then which current limit dominates, or does the battery current get limited when either the the battery or phase limit is reached?

John
 
It's my understanding that is how the phase current is limited, by reducing the limit of multiplication. The stock controllers that came with the motors are a bit overly restricted in terms of the phase current limits. To limit the phase current it must either limit the battery side too when the phase limit is reached or widen the pulse.
 
donob08 said:
It's not clear to me how a controller would affect a limit based on Phase current. If it found Phase Current too high and cut PWM duty cycle further it would increase Phase Current based on current multiplication. I think all it could do is shut down.


Controllers I'm familiar with simply switch the FETs off if the current is exceeded, which is sensed very quickly on a cycle-by-cycle basis. The phase current will start to fall (it is maintained for a while by the motor inductance causing current to flow through the commutation diodes).

Over a few cycles, this has the effect of shortening the ON portion of the waveform, so reducing the duty cycle as you said. This reduces the average output voltage, so by Ohm's law the phase current falls. Not the other way around.
 
swbluto said:
I understand your general message, but this part seems misleading or misworded.

"anytime PWM is used to regulate battery currrent, the average phase voltage drops, and the average phase current increases to balance the power equation"

If the average phase voltage drops, how does the average phase current increase?

Ignoring BEMF calculations (It doesn't affect the general principle), it's simple Ohm's law. If you reduce voltage to a resistor (The phase resistance), you decrease current. What would make a motor the exception?

Another simple rule is that anytime the motor slows down in the same environment/situation, the lower the torque the motor outputs (It takes lower torque to sustain a lower speed). Since there's lower motor torque, you can deduce the average phase current is decreased.

I think everyone needs a meter that shows them phase current stats. Perhaps direct observation would be a better teacher than simple electrical laws.

BTW, on a separate tangent, I think that the "average voltage" is really just a model for understanding it. With an actual controller, I believe that the motor sees the full voltage of the battery during the ON part of the mosfet, but the motor's voltage is a fraction of that depending on the duty cycle due to inductance. Ok, now saying that, I guess that it's more than a model. :lol:

After thinking about your statement, I think I know what you're referring to. You're referring to a "full throttle" simulation graph or some such graph. If your speed decreases at "full throttle", then yes, your phase current increases but it only increases because the physical load increases i.e., when you go from cruising on flatland to climbing a hill. It's not increasing simply because your motor voltage is decreasing. If you reduce throttle, you get a "partial throttle" graph in which case your phase currents decrease.

This was something that took me some time to wrap my head around in the past. (And makes me think... if I'm among the very few who "gets it" , and those who don't are closer in the level of understanding of n00bs, would n00bs be more likely to be misled, being more able to understand the less-complete understanding?)
 
EVan said:
donob08 said:
It's not clear to me how a controller would affect a limit based on Phase current. If it found Phase Current too high and cut PWM duty cycle further it would increase Phase Current based on current multiplication. I think all it could do is shut down.


Controllers I'm familiar with simply switch the FETs off if the current is exceeded, which is sensed very quickly on a cycle-by-cycle basis. The phase current will start to fall (it is maintained for a while by the motor inductance causing current to flow through the commutation diodes).

Over a few cycles, this has the effect of shortening the ON portion of the waveform, so reducing the duty cycle as you said. This reduces the average output voltage, so by Ohm's law the phase current falls. Not the other way around.

Evan,
He and LFP are both saying the power in stays the same, so the reduced average voltage out means current out must go up accordingly. I don't believe this is what necessarily happens in actual use. If it hasn't reached the phase current limit yet, then yes, but once it has reached the phase current limit then either incoming current has to get limited, or the duty cycle increased to increase voltage out. Phase current limits is a programming option in these controllers for a reason.

John
 
John in CR said:
EVan said:
Controllers I'm familiar with simply switch the FETs off if the current is exceeded, which is sensed very quickly on a cycle-by-cycle basis. The phase current will start to fall (it is maintained for a while by the motor inductance causing current to flow through the commutation diodes).

Over a few cycles, this has the effect of shortening the ON portion of the waveform, so reducing the duty cycle as you said. This reduces the average output voltage, so by Ohm's law the phase current falls. Not the other way around.

Evan,
He and LFP are both saying the power in stays the same, so the reduced average voltage out means current out must go up accordingly. I don't believe this is what necessarily happens in actual use. If it hasn't reached the phase current limit yet, then yes, but once it has reached the phase current limit then either incoming current has to get limited, or the duty cycle increased to increase voltage out. Phase current limits is a programming option in these controllers for a reason.

John

His original statement:

If it found Phase Current too high and cut PWM duty cycle further it would increase Phase Current based on current multiplication. I think all it could do is shut down.

Yes, if power going in is the same and the duty cycle decreases, phase current would be higher. But, if nothing else changes, why would the power going in be the same?

If the duty cycle decreases, assuming everything else remains the same(I.e., you're not all of sudden climbing Everest when the duty cycle decreases) the power going in decreases because the motor voltage decreases. Less voltage = less current = less power. Remember, P = IV.

Btw, phase current limiting is a pretty good idea. Assuming the phase current, and the switching losses, don't exceed the thermal design of the controller, then it should be impossible to blow the controller. Also, that's assuming the "controller's algorithm" isn't 'stupid'. Stupid controllers = short lives. Hey, people with lower IQs have lower lifespans on average, so it works just about everywhere. :)
 
swbluto said:
If the average phase voltage drops, how does the average phase current increase?

Ignoring BEMF calculations (It doesn't affect the general principle), it's simple Ohm's law. If you reduce voltage to a resistor (The phase resistance), you decrease current. What would make a motor the exception?
The fact that it's not a simple resistive load........?
 
John in CR said:
Evan,
He and LFP are both saying the power in stays the same, so the reduced average voltage out means current out must go up accordingly.


Homerun for John!


Power IN needs to match power OUT. PWM drops the average voltage across the coils, power needs to balance, so current goes UP by the same amount voltage dropped.


The system works just like a buck-type power supply. The motor winding is the inductor. Example of a bucking power supply would be starting at 100v and 20amps IN (2000w), and your supply switches at 50% duty cycle to buck the 100v IN into 50V out, but we've still got 2000w of power, so the current becomes 40amps to balance the equation.
 
Miles said:
swbluto said:
If the average phase voltage drops, how does the average phase current increase?

Ignoring BEMF calculations (It doesn't affect the general principle), it's simple Ohm's law. If you reduce voltage to a resistor (The phase resistance), you decrease current. What would make a motor the exception?
The fact that it's not a simple resistive load........

Miles, are you ever going to provide a specific counterpoint?

Yes, it's not JUST a resistive load. It's inductive, it has BEMF and it's rpm is affected by the physical load upon it.

Let's take that one at a time.

Inductive - this just shapes the current curve and causing it to ramp up. What's the deal here? The average current would be less than the ohmic predictions. How much less depends on the electric RPM, and the L/R ratio. From a phase current perspective, less current would be less harmful, so no problems there.

BEMF - this is what causes increased phase currents when the motor slows down. Lower BEMF = higher effective applied voltage.

Physical load - this is what causes the motor to slow down when the physical load increases.

So, in summary, if you keep the physical load the same, the bemf is the same and inductivity decreases the current below the ohmic prediction. So, the ohmic prediction still heavily applies. We're not exactly concerned about decreased phase currents...

Now, yes, if you're increasing the physical load, then obviously phase current will increase. This is true whether you're decreasing duty cycle or not (Decreasing duty cycle has a decreasing "influence" on the phase current, while increasing physical load has an "increasing influence"). This is true even during full throttle. This is just true everywhere.
 
swbluto said:
Let's take that one at a time.

Inductive - this just shapes the current curve and causing it to ramp up. What's the deal here? The average current would be less than the ohmic predictions. How much less depends on the electric RPM, and the L/R ratio. From a phase current perspective, less current would be less harmful, so no problems there.


Current ramps up delayed, because energy is stored in a field... then after the FET is switched off, the inductor releases the stored energy by maintaining the current... keep connecting more dots... what effect does this delay rise, then maintain effect have on current...


If you ever end up understanding this SWbluto, I should get an honorary degree in teaching.
 
Once you understand that I already understand, you should graduate from the level of competitiveness measurements to getting a degree in truth acknowledgment.
 
swbluto said:
Once you understand that I already understand, you should graduate from the level of competitiveness measurements to getting a degree in truth acknowledgment.


OMG... you've gotta save these posts for LOL value to be re-read for the time when you understand how phase current multiplication works.


I guess we should start at a very basic circuit, like a bucking LED driver in a flashlight that uses a FET and an inductor. Maybe it runs a pair of lithium cells in series, so it takes in 7.4v at 300mA, it switches the FET at around 40% duty cycle through the inductor, and ends up outputting something like 3v at 750mA to the LED. Are you capable of grasping this concept so far?
 
LFP

..A friend came up with a good explanation:

..If you are traveling at your systems design speed V = Battery Voltage / KV. You are not using a lot of battery current. You are voltage limited, NOT power limited. That battery would supply more but it hasn't the push. V Batt - BEMF = little bit.

..So if you go from 4 turns to 2 the BEMF goes to 1/2, the current goes to 2 times a little bit and you are back in business and push toward 2X speed. There is then enough voltage to oppose BEMF. Right before and right after the change in turns the PWM cycle will be 100%.

Don

PS This is really about your first note on the value of changing turn counts. I think changing has value. The PWM duty cycle need not be changed when turns change. As far as the other debate between you and your friend, I'll have to contact my other friend the Chaplin.
 
Let's talk about how this issue occurs in real world use. eg Taking off from a stop once I get past about 1/3 throttle acceleration, sounds, etc. all seem the same whether it's 1/2 throttle or WOT, so isn't that all at full duty, not PWM chopping? If I use slightly less throttle, then I do hear some noise that seems to come from the motor. A cross between a click and a groan is the best I can describe it. That's the only time I hear that noise, which I take to be the audible result of PWM under load.

I don't expect to tune my HV controllers to high current, so either very easy acceleration using low power or maximum acceleration is easy, and I already avoid that middle ground that the motor warns me about.

The instance where my 100v100a Methods controller failed was while going up a hill that was gradually increasing in grade, but traffic was slowing me to less than full throttle. With PWM in play, I guess the phase current skyrocketed with the high power demand of the hill, but at less than full duty. If that is indeed where the risk lies then I can make a point to avoid it. If I'm on the right track here, I can't really think of another time I'd be at high load and less than full duty unless I tried cruising along at a 60-70 mph that was somehow partial throttle using a single motor. With the dual motor I don't think the load would be high enough, and the only way I'd get the dual motor under high load will be a top speed attempt, so WOT.

What am I missing?

John
 
I missed reading all of the in betweens.

..I think.....

..For the controller: power in from battery = useful power to motor. So in that way the controller is zero sum.

..But in addition there is power that comes to the controller from the motor when the controller tries to cut off current for PWM. That power is generated in the motor. It is sort of like instantaneous regeneration. The motors inductance enhanced by the motion of the wheel sends “new” energy to the controller. That's the energy that heats up controllers at low duty cycle. That goes through the other FET that isn't driving.
 
liveforphysics said:
swbluto said:
Once you understand that I already understand, you should graduate from the level of competitiveness measurements to getting a degree in truth acknowledgment.


OMG... you've gotta save these posts for LOL value to be re-read for the time when you understand how phase current multiplication works.


I guess we should start at a very basic circuit, like a bucking LED driver in a flashlight that uses a FET and an inductor. Maybe it runs a pair of lithium cells in series, so it takes in 7.4v at 300mA, it switches the FET at around 40% duty cycle through the inductor, and ends up outputting something like 3v at 750mA to the LED. Are you capable of grasping this concept so far?

Wow, you know what.. if it isn't obvious by now, I created the simulator. See the link below. It's not like someone had handed me the equations and I created a wrapper for it. I understand the law of "conservation of power"/energy (Power is just the time rate of energy), and I further understand the law of induction. Yes, I know that a collapsing magnetic field induces a voltage value on the coil via Lenz's Law and that PWM allows one to control the ratio of collapse/field-growth to create an average voltage of your choosing. Once you get that average voltage, you calculate average current, after subtracting the bemf from the motor voltage, via ohm's law. When you calculate the power via the motor current and motor voltage, minus the bemf (And diode voltage for more accuracy), it's *gasp* equal to the power of the input battery. I_1*V_1 = I_2*V_2. when V_2 goes down (The motor's average voltage), I_2 goes up to compensate.

That's not what I'm arguing. Your statement of that fact is true. What I'm arguing is that you when you're in real life and you decrease the duty cycle, you slow down, and your phase currents go down. It seems like some have been led to believe from your statement that they actually go up. Maybe that's not what you're actually saying, but that's what the wording conveys and it's what some have been led to believe.
 
John

..No, 1/2 throttle is not WOT. 1/2 throttle is 1/2 duty cycle. There is chopping. Duty cycle and throttle voltage/position have a 1:1 relationship.

..That middle ground you speak of is because of the loading not the throttle position. I think you have the sense right. If the motor is a high power situation, meaning high current, and then you cut the throttle the motor will throw lots of energy back at the controller. Because its been using lots of current.

..The controller won't like it.

..High speed can mean high current just to overcome air resistance. So High torque requirements at partial throttle should be avoided. Full throttle will tax the battery. But partial throttle will tax = heat the controller.

Don


John in CR said:
Let's talk about how this issue occurs in real world use. eg Taking off from a stop once I get past about 1/3 throttle acceleration, sounds, etc. all seem the same whether it's 1/2 throttle or WOT, so isn't that all at full duty, not PWM chopping? If I use slightly less throttle, then I do hear some noise that seems to come from the motor. A cross between a click and a groan is the best I can describe it. That's the only time I hear that noise, which I take to be the audible result of PWM under load.

I don't expect to tune my HV controllers to high current, so either very easy acceleration using low power or maximum acceleration is easy, and I already avoid that middle ground that the motor warns me about.

The instance where my 100v100a Methods controller failed was while going up a hill that was gradually increasing in grade, but traffic was slowing me to less than full throttle. With PWM in play, I guess the phase current skyrocketed with the high power demand of the hill, but at less than full duty. If that is indeed where the risk lies then I can make a point to avoid it. If I'm on the right track here, I can't really think of another time I'd be at high load and less than full duty unless I tried cruising along at a 60-70 mph that was somehow partial throttle using a single motor. With the dual motor I don't think the load would be high enough, and the only way I'd get the dual motor under high load will be a top speed attempt, so WOT.

What am I missing?

John
 
John in CR said:
The instance where my 100v100a Methods controller failed was while going up a hill that was gradually increasing in grade, but traffic was slowing me to less than full throttle. With PWM in play, I guess the phase current skyrocketed with the high power demand of the hill, but at less than full duty. If that is indeed where the risk lies then I can make a point to avoid it. If I'm on the right track here, I can't really think of another time I'd be at high load and less than full duty unless I tried cruising along at a 60-70 mph that was somehow partial throttle using a single motor. With the dual motor I don't think the load would be high enough, and the only way I'd get the dual motor under high load will be a top speed attempt, so WOT.

John

As the physical load increased, i.e., the hill, your phase current would've sky-rocketed whether you were going full throttle or partial throttle. If you're at full throttle, the phase current will be even more than at partial throttle BUT the mosfets endure switching losses while PWMing during partial throttle, so if you're not at low enough partial throttle, the mosfets will endure more instantaneous heat than at full throttle.

If your phase currents are high enough, your controller will blow even at full throttle. There's a current limit to mosfets and full throttle doesn't protect you from that. This is why it's pretty wise to limit the phase current if your setup is at risk of sustaining high phase currents. On a hub motor, most of this extra "phase current" is just heat for the motor since the motor's magnets can start to saturate at high phase currents.
 
Don,

That doesn't seem correct, because that's the same as saying the duty cycle is determined only by throttle position. I believe the controllers are smarter than that. While accelerating there are lots of points in throttle position that result in the same acceleration as WOT, at least until you approach the speed related to the lower throttle position where acceleration decreases.
If that was the case then my motor would run significantly hotter from these partial throttle accelerations to speed, because the phase currents would be so much higher, but that simply isn't what happens. Maximum acceleration, whether full throttle or partial throttle is used, must be full duty.

John
 
Back
Top