The Low Rpm Efficiency Myth - Busted?

[safe]

It took me a moment to realize what you were talking about...

Yes, in the "perfect scenario" where the "efficiency peak" and the "power peak" are one in the same then you will get the kinds of high efficiencies that you have pointed to. This is one of those "apples" verses "oranges" situations were we are talking about two tihngs at once. The PWM has NOTHING to do with this... this is all about matching the voltage to the rpms so as to hit the "sweet spot" of efficiency within the "normal" physics of the motor. That's why in the past I've stressed the idea that you need to align these two things and then the controller will automatically produce a very efficient overall powerband.

Back to the "myth" however...

The PWM "effect" simply gives you a little "extra" torque in areas that are nice to have... if you use a very high current limit YOUR LOSSES ARE STILL BAD. (PWM doesn't change the overall motor charactoristics that much, just a little "extra" in certain rpms)

The point here is that it's the motor configuration and the current limit that determine MOST of how your overall system behaves, not PWM.

The myth stands as a myth...

 

[safe]

Good job Joe! Even if safe is a little slow, others will read this topic and gain a lot of understanding.

Sorry, you don't realize that there are two separate variables:

1. Current Limit

2. PWM "Effects"

They BOTH contribute separately to the overall motor perfomance.

The "Heroic" PWM Controller Myth is still BUSTED!

 

[Leeps]

The pwm effect is just changing the input voltage to the motor, thats it its just that simple. It does not in any way shape or form change the motor characteristics nor does it give you extra torque. The motor characteristics are constants and as such stay that way. Torque is proportional to current and the current that a motor will draw can only be externally effected by voltage and shaft rpm. Even current limiting works in effect by lowering the voltage going to the motor in order to reduce current draw at the motor, there is a problem in the controller measures battery current and thus the dc-dc conversion doesnt allow the current limit to be a constant value, i suppose this is the torque boost you are speaking about, the same rules still apply the controller is droping voltage to try and keep the current in check and maintain efficiency.
I see things being very apple to apple or orange to orange depending on which fruit you prefer.
Joe

 

[safe]

Hopefully you guys haven't walked away before realizing the conclusion on this one. The PWM "effect" is totally unrelated to the choice of current limit for the controller. For a novice that knows nothing about this stuff they don't know the difference between the "current limit" and the PWM "effect" they just think "controller". So if they read this they will get the wrong idea.

It's true that a low current limit makes for a more efficient powerband and a higher one tends to give more power, but it also means that you lose that "sweet spot" of being in the "peak efficiency" of all the various voltages as you go up and down the rpms.

 

[xyster]

The "Heroic" PWM Controller Myth is still BUSTED!

I don't see that the myth ever existed.

Safe's focusing on whether or not controller characteristics can overcome motor inefficiencies. The answer, I agree and have agreed, is no. But I don't see Leeps has ever argued otherwise.

Leeps has been focusing on controller efficiencies at different voltages, and is correct in his assessment too that the controller is a very efficient voltage converter, especially when compared to not using a controller and instead using some extraneous load whether physical or electrical to moderate motor speed. I've not perceived Safe arguing otherwise.

You both win. Now hug and make up...

 

[safe]

...I suppose this is the torque boost you are speaking about

There is a REAL PWM "effect" where a little "extra" energy is passed through because of the momentum of the pulses. I've gone to great lengths to add this little "warp" into my calculations and compared my results to the 5304 Simulation and it all matches up. (and Fechter discussed this with me before too)

So maybe we agree...

Voltage, if perfectly matched to rpms, can produce great efficiencies in a motor. This is done by selecting a PWM controller with an appropriate current limit. The additional PWM "current warping effect" is actually a relatively small issue that mainly boosts low end torque with no change in efficiency. This was the "narrow" definition of what the "myth" was about.

We might be able to simply write this off as a misunderstanding...

 

[Leeps]

Well in the last sentence of your last post, it seems to me that your admitting to efficiency at low rpms. Is it true or are we going to keep this up
Joe
edit your last post went in when i was typing this. so yeah its inappropriate

 

[safe]

Well in the last sentence of your last post, it seems to me that your admitting to efficiency at low rpms.

It "can" be true if your current limit is configured correctly so as to "make" it true. But the PWM "current warping effect" has little to do with that. (there is an actual "effect" separate from current limiting)

The whole time we've been discussing ideas unrelated to each other...

 

[safe]

Sweet Spot Everywhere

Here's a 1200 Watt running so as to be at the "perfect" current limit so as to have the "ideal" powerband.

Now THAT'S sweet!!!

 

[Leeps]

The boost in torque that your noticing is really a current limit that isnt stable. The current limit is set for input current not output current. So as output voltage varies the current limit is varying also.

It should be noted that this only applies when your demanding a throttle setting that ends up demanding a current greater than the current limit. It does not fundamentally change controller function nor motor function, its better thought of as overriding your throttle input.

Ok as far as extra energy coming from the pulses, now your going into the deep end of the pool of electrical knowledge. Im glad that we agree up to this point, im willing to tackle the little extra pulses of energy also if you want to go at it and put an end to this debate.

Joe

 

[Leeps]

Ok well you got another post in while i was typing, but the last graph really just showed what i was saying.
im still up for the pulses of energy gig
by the way i thought the title of the thread was low rpm efficiency myth, not current warping myth or pwm effect myth, just that something doesnt feel right
Joe

 

[safe]

....I'm willing to tackle the little extra pulses of energy also if you want to go at it and put an end to this debate.

That was what the "myth" was centered around... the idea that the pulse effect allowed a little extra energy through than the voltage would normally suggest. We don't need to go too far into it becuase it's actually a pretty small influence, but since it shows up at low rpms it does help a great deal those people with fixed gear machines. However, that little "extra" boost doesn't change the overall truth that if the motor / current limit as a whole is inducing a bad efficiency situation the PWM "current warping and boosting effect" won't "miraculously" turn things around.

I think I've had enough typing for a while too... I think we are on the same page more or less now, so that's good.

(there was a moment when I was thinking "Oh my gosh, have I been wrong with this stuff for weeks now?" until I realized what was going on)

 

[safe]

I thought the title of the thread was low rpm efficiency myth, not current warping myth or pwm effect myth,

I take full credit for that mistake... terrible title on my part...

 

[Leeps]

Well you are right if the switching frequency is too low for a given motor inductance the current can rise to dangerous levels for the controller. In any case this would show itself at higher throttle levels, with a worse mismatch getting you into trouble at lower throttle levels. I seem to remember hearing older curtis motor controllers having this problem with axial flux motors. This effect is hardly desirable and is dangerous for the motor controller.
I had a thread about a new idea for a different approach to motor controllers that would match switching frequency to motor inductance, i have enough on my plate to work more on it.
I'm glad this is over
Joe

 

[fechter]

There you go, the controller should dynamically adjust it's on-time to maximize efficiency in the system. The iron and copper in the armature are acting like the inductor in a switching power supply. Those laminations are typically very inefficient at high frequencies, which can result in heating and loss. At the same time, you don't want the field to reach anywhere near steady state current, so the switching frequency can't be too low either.

I've seen some power supply circuits that measure the dI/dt to figure out the switching point.

I guess this idea should be a new thread.

 

[Leeps]

Fechter i had a thread on that. Remember my thread, new approach to brushed motor controller. Granted i didnt do anything with it, it was just an idea. And to boot my thread just went off on a tangent, maybe we should bring up the subject again, just doesnt seem like theres much to discuss on the subject.
Joe