The Low Rpm Efficiency Myth - Busted?

[safe]

There is a myth floating around that says that the "Buck Converter" which is a PWM Controller does such a "magical" job in passing extra current to the low rpms that it can counter act the efficiency losses that low rpms always suffer. In this thread I hope we can once and for all thrash out all the misconceptions and confusions that make this myth so pervasive.

The "Heroic" PWM Controller Myth will soon be BUSTED!

[safe]

First we look at the chart of a regular 750 Watt Unite motor that runs at 36 Volts without using a controller.

[safe]

Second we add a 40 Amp PWM Controller BUT DO NOT YET apply the benefits of the "current warping" that the controller can achieve.

[safe]

Finally we add the PWM "Current Warping" Effect. ("Buck Converter") These results were compared to the 5304 Simulation program and they matched very closely. Notice how the power output has a slight bend to it in the low rpm area.

[safe]

Summary

Okay, so what can we say about this myth?

Well first of all we can "understand" why this myth exists. There is an increase in torque that exists due to the "extra" current that manages to pass through the PWM Controller and this does make the bike "feel" more powerful down in the low rpms because it actually IS STRONGER than if the effect didn't exist.

However, efficiency has to do with "how well" power is used and not it's gross quantity. While more energy is being passed through the controller when you "bog down" in the low rpms you are also LOSING just as much of it as you would expect. POWER (torque) is improved but the current DRAIN to achieve the same "effect" does not.

Efficiency is ALWAYS best at the "efficiency peak" and despite the extra "torque" you find down low you still "pay" a higher price for that torque.

Low rpm torque is "more expensive" than peak efficiency rpm torque!

[xyster]

There is a myth floating around that says that the "Buck Converter" which is a PWM Controller does such a "magical" job in passing extra current to the low rpms that it can counter act the efficiency losses that low rpms always suffer...

Where do you get these "myths"? I think you just make them up to stir the proverbial hornet's nest. There is no pervasive "myth" in re controller efficiency I'm aware of. Controllers are widely understood to be less than 100% efficient, hence most people understand nothing magical is going on. Just simple math that:

Amps In X Voltage In = (Amps Out X Averaged Voltage Out) + Efficiency losses which turn to heat.

so:

Amps In X Volts In = Amps Out X Average Voltage Out + heat

Note: It's been years since I took college physics w/calculus...the "averaged" voltage out may be an integral function and not an average.

The fact that motors are less efficient at low rpms is independent of controller or no controller, hence any "controller myth" as it applies to motor inefficiency is a red herring anyway.

[safe]

The fact that motors are less efficient at low rpms is independent of controller or no controller, hence any "controller myth" as it applies to motor inefficiency is a red herring anyway.

You don't subscribe to the myth, so it doesn't apply to you personally, however there are those that are still "struggling" with this concept.

On your bike you've simply said:

"Hey, reality is reality, I'm going to use a big battery and a big motor and not decieve myself into anything else. And my way works and it's more reliable anyway."

In a way I do respect that... you are not fooling yourself with your designs... they are "RAM Tough"... HEMI Powered Tough.... they work... solid as a rock...

[xyster]

"Hey, reality is reality, I'm going to use a big battery and a big motor and not decieve myself into anything else. And my way works and it's more reliable anyway."

Bingo!

[Leeps]

I really really think that this was aimed at me. I never ever suggested as much.
There are two things that i have to say. The controller itself is magnificently efficient. Synchronous controllers that omit the flyback diode and use another mosfet thats switched inversely with the main power mosfet are about as efficient as your going to get.
On the motor i never claimed anything with efficiency. The simple fact that safe posted graphs with current and torque shooting up at the bottom of the scale, and then noticing that this doesnt look right and saying a current warping of 50% seems right, shows me that more studying needs to be done on the subject.
Just ask me about permanent magnet motor efficiency, ill gladly explain it to you.
Joe

[safe]

Just ask me about permanent magnet motor efficiency

Well I guess this thread is addressed to you.

So do you or do you not agree that the curve for efficiency follows the "general" shape of the one presented?

Let's start there first...

All the dyno charts of real world (and simulated) motors fit the physics as defined, so if you could present some kind of "argument" that supports the idea that at, say, 25% of the maximum rpm that you can "somehow" end up with comparable efficiency to the "peak efficiency" value then I'd VERY MUCH like to see it!!! (a miracle!)

Xyster seems to simply "agree" that while you can get a little bit "extra" torque because of the PWM effect it's not like you get it for "free". That "extra" torque "subtracts" from the efficiency at the same rate that it "adds" because you are INCREASING the current (through the effect) and therefore increasing the amount of energy DRAINED at the same time. (which counts as a negative)

Do you see that?

You get the "extra" current (which comes as a welcomed surprise) but you have to "pay" for that current because it's still draining the battery. The overall efficiency doesn't change. You would KNOW THIS if you played with the numbers a little. If you "foolishly" try to add the effect into the efficiency (my first inclination) you get "non-sensical" results where the efficiency can EXCEED 100%!!! (look ma, we invented the perpetual motion machine!)

So then you step back and realize:

"Oh, now I get it... you have to balance the equation and subtract from the other side."

I think you've managed to "forget to balance" the equations and allowed yourself to focus on the "extra" current and ignore the fact that the "extra" current counts as a negative in efficiency.

Get it now?

The PWM Controller "effect" is to "shift" a little extra torque to the lower rpms, but it doesn't effect the efficiency curve because the equations always need to balance.

[Leeps]

Before we go any further, let me ask you a question. This will have great bearing on my approach to this conversation. If you had a battery that can supply for example 200 amps, is it necessary for the motor to consume said 200 amps. Now in the same breath the fact that there is a dc-dc conversion taking place, and thus more current is available to the motor with less current being taken from the battery is it necessary for the motor to also consume this current, in your opinion?
Joe

[xyster]

Xyster seems to simply "agree" that while you can get a little bit "extra" torque because of the PWM effect it's not like you get it for "free". That "extra" torque "subtracts" from the efficiency at the same rate that it "adds" because you are INCREASING the current (through the effect) and therefore increasing the amount of energy DRAINED at the same time. (which counts as a negative)

Safe, you're connecting two separate dots (extra torque available and controller efficiency losses) to suggest I agree with your contention. I never connected those dots. But instead of clarifying my position, I'll sit on the sidelines on this one. I could stand to learn more first. So let the debate continue!

[safe]

...more current is available to the motor with less current being taken from the battery is it necessary for the motor to also consume this current, in your opinion?

I think the electrical unit we need to discuss is the "Joule":

From wikipedia:

The work required to move an electric charge of one coulomb through an electrical potential difference of one volt, or one coulomb volt.

The work done to produce power of one watt continuously for one second, or one watt second.

From what I see in the dyno charts and by running the calculations the quantity of energy (in Joules) that can be transmitted because of the PWM effect will INCREASE for a given lower voltage (low rpm) so the result is more "work" in the form of watts. This is the PWM "effect".

Efficiency needs to take this "extra" current (in Joules) and subtract it as a LOSS in the equation because the battery must drain extra units of energy in the form of "Joules".

The net "effect" is to get more "torque" but the efficiency is not affected. (because it all balances out) The "torque" still is "paid for" normally.

As to this idea about "less current being taken" from the battery... well... if the battery has a certain number of Joules of energy you can extract them at any rate that you choose. You CANNOT magically expand the "pool" of Joules because of the PWM "effect". All the "effect" does is changes the location in the powerband where Joules of energy are sent to the motor... that's it... the motor still obeys all the laws of physics about efficiency.

If it were possible to create "extra" Joules then you would have:

A perpetual motion machine... (producing "free" energy)

At the bare minimum conceed that the laws of conservation of energy need to be preserved... you can't "invent" energy from nothing.

[safe]

We have to remember that motor efficiency is something that is about the physical design of the motor itself. Every motor has it's "preferred speed" to be run at and when you get way outside of the "preferred speed" the efficiency drops off. The optimal efficiency occurs at a point where the power output is dropping off with increasing rpms. The "peak power" is always at a lower rpm than the "peak efficiency" and it's natural curve is a parabola that peaks at the middle rpms. Low end torque is in an area of EXTREMELY low efficiency. (often 50% or lower... well below the 80% peak)

[Leeps]

Sadly you completely ignored my question.
Joules really is not what we should be talking about, we can talk in units of energy if you like, it will only complicate things why add the dimension of time. Instantenous power transfer is all we need to look at the rest will be obvious.
In any case, you mention that the quantity of energy transmitted increases at a lower duty cycle. All i can say is justify it.
its really simple input voltage*input current=output voltage*output current*controller efficiency
Controllers are roughly 95% efficient and that is taken care of. Left side and right side must balance, the input voltage is fixed, input current can go up and down. The output voltage is proportional to throttle. Now if the output voltage is less than input voltage guess what the input current mustnt equal output current. There is no overunity nothing going on here.
please answer my original question, as i said it is pertinent to the manner in which i have this discussion
Joe

[Leeps]

You got a post in while i was typing.
Suggesting that peak efficiency occurs at a point where power output drops off with increasing rpms, is also suggesting that peak efficiency is occuring around the power peak. Im sure this is not what you were trying to say, please be more specific on this.
and as before please answer my original question.
Joe
edit i re read your original post, and noticed that you said optimal efficiency and not peak efficiency in this case all is well i do agree. Im sure we will touch on motor efficiency much more deeply later in this discussion.

[xyster]

Because I'm truly interested in this discussion and its efficient progress, for this thread only I've appointed myself impartial sideline umpire.

With that I ask:

Please specifically answer each others' questions at the first available opportunity.

Thank you for your support.

[safe]

In any case, you mention that the quantity of energy transmitted increases at a lower duty cycle. All I can say is justify it.

Think logically for a second...

If there is say 1000 "units" of energy in the battery and "normally" at a given low rpm you are able to extract 20 at a time. Now because of the controller you are instead extracting 25 or 30 "units" even though the voltage as seen by the motor "seems" the same then you are still using more "units".

If the energy "units" don't come from the battery then where do they come from? You can't "invent" energy out of nothing.

The only plausible theory would be that the "normal" physics that define a motor "change" when the voltage and amps are "shifted" and what would normally be a TERRIBLE efficiency situation becomes slightly less so. If this is the line of logic to pursue then it's more about the MOTOR side and not the BATTERY side.

Do you want to go down this road?

With such an argument the "magical energy" that is "created" is actually "losses" that are being "recaptured".

[safe]

Suggesting that peak efficiency occurs at a point where power output drops off with increasing rpms, is also suggesting that peak efficiency is occuring around the power peak. Im sure this is not what you were trying to say.

My point was that going from maximum rpms downwards the "milestones" are:

1. Maximum (No Load) RPM

2. Peak Efficiency RPM

3. Peak Power RPM

4. Low RPM (where the PWM Effect matters)

[Leeps]

At this point i demand an answer to first question with the analogy of the battery and motor controller. Ive repeated myself it does matter in how i answer my questions, i promise to touch on all of your questions afterwards.
As far as the milestones of efficiency, thank you for clarifying.
Joe

[safe]

If you had a battery that can supply for example 200 amps, is it necessary for the motor to consume said 200 amps. Now in the same breath the fact that there is a dc-dc conversion taking place, and thus more current is available to the motor with less current being taken from the battery is it necessary for the motor to also consume this current, in your opinion?

"...and thus more current is available to the motor with less current being taken from the battery"

When you say "current taken" that has to be thought of in terms like Watts or Joules. Obviously the highest voltage is in the battery and so high voltage / low current produces a certain amount of energy "units". Since the "chopper" is taking this "energetic" pulse and pushing it very fast to the motor the "momentum" of the pulse is faster than if it were being "motivated" by a low voltage. So I understand the "momentum" idea.

What seems to be the case is that the pulses hit the motor "as if" they were at "full charge" and "full momentum" and the motor reacts to it in such a way that for the microsecond that the pulse passes through it behaves as if it were running at full throttle.

I do believe that I understand the idea of "momentum" of the pulse and the benefits of this... that possibly there are efficiency benefits that can be achieved through the pulse concept.

But on the flip side...

Think of the efficiency "hole" you are digging yourself out of?

The physical design of the motor just plain doesn't like being run at low rpms and naturally you would expect 50% LOSSES or more. So let's say that there's a few percentage points of improvement on the "pulse" effectiveness... so what? You still are hovering around 50% LOSSES... which sucks...

To get anywhere we need to think about the motor and how the motor might have some strange qualities that allow something "structural" (the physical makeup of the motor itself) to be overcome through a "pulse" effect...

[safe]

This is VERY similiar to the idea of "Fat Pipes" on Two Stroke motorcycles. When you "tune" your pipes perfectly so that when the pulse from one exhaust stroke goes to the end of the pipe and then "resonates" back to the motor you get a little "standing wave" that can be exploited for extra power. You could get 50% more power that way. Of course, your fuel efficiency went down because all of a sudden you are using more fuel, but maybe the analogy breaks down at that point.

As I see it this PWM "effect" is like a movie with many little "frames" of energy that pass by very quickly. When the "pulse" makes it's way to the motor.. if it's timed right... it will give the most efficient energy transfer as is possible. Getting the "timing" to all match up perfectly requires that everything is tested and fiddled around with by the designers, so some controllers might work better on some motors than with others.

[xyster]

Safe appears unable or unwilling to specifically answer questions. Perhaps he's practicing to run for political office.

If you had a battery that can supply for example 200 amps, is it necessary for the motor to consume said 200 amps.

Answer: yes or no
Then launch into explanation of understanding. Then optionally offer a retort.

Now in the same breath the fact that there is a dc-dc conversion taking place, and thus more current is available to the motor with less current being taken from the battery is it necessary for the motor to also consume this current, in your opinion?

Answer: yes or no
Then launch into explanation of understanding. Then optionally offer a retort.

If yes/no questions can not be answered succinctly and swiftly, the debate goes off at least two different directions, hard to follow by others, and then all of us whether participants or not have zero chance of improving our understanding.

[safe]

Can the toilet lid be lifted up or down?

I realize there's a question, but I don't know where it leads to, so my response has been to try to get him to elaborate on where his idea leads. It's so much easier if he would just "spit it out" and say what theory he is working towards rather than being cryptic...

The only "fact" that I'm aware of is that no dyno that I've ever seen of an electric motor shows low rpms having high efficiency. The day I see "real world" data to support the concept is the day I start to give it credibility.

Hey... I live in the "Show Me State" (Missouri) so you have to figure I'm going to be skeptical...

[xyster]

Can the toilet lid be lifted up or down?

Is not a yes or no question. Yes/no questions are very simple and easy answer, they require either a "yes", or a "no", or an "I don't know". Yes/no questions are therefore very good for getting/keeping discussions between participants with widely varied views on track, as this discussion is straying far from.