* * * MOTOR UNIVERSITY (Lab 101) * * *

[eP]

Everything makes sense except the 100% duty cycle part.

<snip>
Look at a typical 36 Volt 750 Watt Unite Motor performance chart... (it's possible that the numbers don't fit right, but the idea is sound... if you understand his train of thought :? )

Power(In), Power(Out) and Heat are in Watts times the scale.

(and a final note, if you overvolt the motor it changes everything, so the only way to get to 2500 Watts for peak power is to overvolt, which means all the charts need to be modified to view that)

In which way you are going to dissipate heat generated by 50A current if the motor is rated for little over 20A for the sake of low ability to heat dissipation ?
In other way its nominal power out will be much higher than rated 750W what we can see at your chart.

I see you are going to burn next one.

 

[safe]

In which way you are going to dissipate heat generated by 50A current if the motor is rated for little over 20A for the sake of low ability to heat dissipation ?

In other way its nominal power out will be much higher than rated 750W what we can see at your chart.

Well, there's this idea of "forced air cooling" which Fechter has used and seems to increase the heat dissipation capacity. With all "R&D" of new ideas (or at least ideas that don't have published results) there is the potential of going past the limits of the technology. That's what we "do" as pioneers in this field... we try new things and see what "works". In fact, when one of the racing mechanics/designers was asked to describe his latest racing machine he said something like:

"Well, it's the result of a lot of effort and all I can say it that it 'works' to make our guy go fast."

...and that's how the name stuck. When they talk about "Work's Bikes" at the race track that's what started that phrase. So, yes, I will test, learn, think up new stuff, advance the machine in every way I can. If 50 Amps with air cooling is too much, then I back it down to 45 Amps, etc...

But the main idea is not the absolute value of the current allowable (which can only be know by real world track testing) but the relative benefits of using a "restricted" motor like one that uses Motor Current Limiting in the controller. Since the "Imperfect Rider" tends to favor the low rpms when they shift (human nature) that rider tends to produce more heat naturally and without thinking about it.

So hopefully by now you at least understand "the process" of how ideas get tested and developed. No one ever gained anything by not trying.

I'm from America... we love pioneers... :wink:

 

[safe]

I was asking: "What changes (specifically) did you make when you altered the graph?"

i.e. from (?)percent time spent at peak power to (?)percent time spent at peak power...

All I did was change the values of the percentage that the motor stays in any particular rpm. By bunching the "Bell Curve" approximation to the middle you amplify the benefits of the lower current limit of the standard controller. The data was what you would call "ad hoc" and other than the fact that all the slices added up to 100% it was mostly just a lot of:

=BO67+$A$65 (look to cell to left and then add a constant.. repeat...)

 

[TylerDurden]

I see...

So, would you say that the Devil's chart shows running @ peak 80% of the time, or more?

 

[safe]

So, would you say that the Devil's chart shows running @ peak 80% of the time, or more?

Yes, something like 80% - 90% of the time is at peak in the Devils Advocate percentage chart. (which is just about never the case in the real world)

A hub motor that is designed for a top speed of 20 mph will quickly pass through the power peak and stabilize in the high efficiency area... but at 20 mph.

A geared bike passes through the 20 mph mark and then upshifts and goes from 20 mph to 25 mph... upshifts... 25 mph to 30 mph... upshifts and on and on until you max out at 40 mph.

So it depends on your gear spacing, but generally speaking you tend to rev the motor a little past peak and then upshift which takes the rpms to a little before peak.

You also have to remember that the accelleration at the peak itself is the highest so you tend to get past the peak rather quickly and go into the zone where both controllers are the same. (an area of good efficiency)

The way you ride makes a big difference...

I've noticed that if you run a gear lower and rev out the motor into the higher rpms before you do upshift that the heat is lower. When you try to upshift too early (like barely past peak) then you drop lower in the rpms and it still pulls about the same, but it creates more heat. So my "Imperfect Rider" behavior mode definitely (in the real world) produces more heat. Basically I've gotten good enough at knowing the "quirks" of the motor to now know how to exploit it's features, but a Motor Current Limited controller would tend to filter out "imperfections" in rider habits. Basically the Motor Current Limited controller does with technology what can already be done with subtle throttle control and gear selection. I'm trying to build a bike for cavemen:

 

[Malcolm]

But a bike for cavemen wouldn't have gears, would it?

 

[Lowell]

100% operational duty cycle, as in the motor could run at that power level forever without burning up since there is never more heat created than the housing can dissipate.

 

[safe]

But a bike for cavemen wouldn't have gears, would it?

Those sort of questions are what sent the poor caveman to the therapist... :wink:

 

[safe]

100% operational duty cycle, as in the motor could run at that power level forever without burning up since there is never more heat created than the housing can dissipate.

That's not a standard usage for "duty cycle". Normally duty cycle is supposed to mean the percentage of time that a PWM controller is in the "open" state. When the gate is open a lot you have a high percentage of the voltage passing through to the motor and it's said that you have a high percentage duty cycle number. In a hyper-simplified simulation with a fixed gear they are more or less related, but not on a geared bike.

The way you are using "duty cycle" you're thinking in terms of what percentage of time that "in operation" the rpms are at peak power or some other rpm. So you're creating a source of potential confusion. :?

What is the real name of the "operational duty cycle" if that phrase is incorrect?

Anyone?

The only way to create a "heat profile" for a motor is to figure out (in some manner like I've been doing recently) a statistical "average" of where the motor tends to get used in the powerband. Some rpms are seldom used, while others are used a great deal. This gets amplified with gears since you can shift the powerband around at will, but also can become pretty messed up because you have free will to shift at really stupid times.

What is the true name for the analysis I've been doing that creates a statistical profile of a motor and how it behaves with time?

 

[TylerDurden]

Load Rating?

The Unite MY1020's rated load shows an 78% eff.

It specs-out 960W P_in, to get 750W P_out. It can therefore dissipate only 210W in heat at any given time. If it could dissipate more, it would get rated higher power, but it can't.

On the flats, that motor can continuously push a racing bike* 30mph... 521W P_out x 80% eff. = 651W@48v = 13.5A. No problemo; since 651-521=130W to dissipate.

On the flats, that motor can continuously push a racing bike 40mph... 1144W P_out x 70% eff. = 1634W@48v = 34A. More than double cooling needed: 490W to shed in heat. :?

You wanna go 50A? OK, that gets you 46mph, but at even 70% eff. you need 2.4kw P_in... minus 1696W P_out = 726W to dissipate or 3.5 times the cooling of a stock setup. Not sure how to accomplish that. (Eff. will probably be much less.)

We have been just running flat. You wanna accelerate?

How about modest acceleration or a 5% slope? At 30mph, the motor will pull 37A or 1784W if it can muster 70% eff... again needing to shed 530W in heat. You could hit the limit of 50A with more acceleration or an 8.1% slope, but not for long; since 727W loss will need to dissipate. (That's at a very unlikely 70% eff.)

At 40MPH, forget about acceleration. For shizengiggles, pretend you could get 70% eff... even a 3% slope will pull 50A. :x


We haven't even thought about how many pounds of batteries it will take. :lol:



*80lb bike 160lb rider / http://www.kreuzotter.de/english/espeed.htm

 

[safe]

Well let's look at the chart again for a 36 Volt Unite motor... (assuming a standard 40 Amp controller)

The scale is 20x for the Power(in), Power(out) and Heat.

So if you look at the chart:

Peak power is 50 x 20 = 1000 Watts.

Peak efficiency power is 20 x 20 = 400 Watts.

Power input at current limit is 70 x 20 = 1400 Watts.

Worst heat at stall is 70 x 20 = 1400 Watts.

Heat at peak power is 20 x 20 = 400 Watts.

Heat at lows rpms that one might pull in a tall gear is 40 x 20 = 800 Watts.

Let's all agree on the data first...

Volts = ( Duty Cycle ) * ( Battery Voltage )
Amps = ( ( Volts ) - ( BackEmf ) ) / Resistance
Power(In) = Volts * Amps
Power(Out) = ( Volts - ( Resistance * Amps ) ) * ( Amps - NoLoad )
Efficiency = Power(Out) / Power(In)
Heat = ( ( Amps Applied ) ^ 2 ) * ( Motor Resistance )

 

[TylerDurden]

From Unite:

 

[safe]

You realize that's the 24 Volt version right?

The 36 Volt version is represented (very well) in the chart... but you have to plot the data in order to know the real values:

From Lesson 4.wks:

At 2814 rpms - 26.9 amps - Efficiency 78.2%

 

[safe]

Rated Load

The official "rated load" and the peak power and the peak efficiency are all different things. Here's the actual location of the "rated load" for the Unite 36 Volt 750 Watt motor. They basically just pick a spot in between the peak efficiency and the peak power and say "this is the spot".

What's interesting to observe is that the "rated load" heat is 163 Watts, but the worst case scenario for heat is 1400 Watts which is a full 8.8 times more heat. Even at a moderate load at 1407 rpms the heat production is 640 Watts which is a factor of 4 times the "rated load" heat.

So you have to know more than the "rated load" to really master heat management... :wink:

 

[TylerDurden]

They basically just pick a spot in between the peak efficiency and the peak power and say "this is the spot".

Whaddya bet they actually rate the motor at the highest point where it can run and not overheat?

You make it sound somewhat arbitrary.

 

[safe]

Whaddya bet they actually rate the motor at the highest point where it can run and not overheat?

You make it sound somewhat arbitrary.

At the "rated load" the Power(out) ends up being right around 759 Watts at 2800 rpm. The location of where the "rated load" exists in the overall powerband is somewhere between the 40 Amp current limited power peak and the efficiency peak (which is a constant). It's right "there" where you would expect it.

The "absolute" power peak is 1412 Watts at 1675 rpm on an unlimited controller.

The 40 Amp controller power peak is 1033 Watts at 2513 rpm.

The "rated load" is 759 Watts at 2800 rpm.

The efficiency peak is the same for all and is 463 Watts at 3049 rpm.

So when you look at it this way you see that their "rated load" is almost the same as the efficiency peak. Think of all the time that motors are asked to operate at half their efficiency peak? Think of how bad things can get if you give gears to an "Imperfect Rider" and let them run a too tall gear all day long?

It's scary huh?

 

[TylerDurden]

 

[xyster]

I like that color-coded graph, TD. How would it look if the motor was cooled by a blower?

 

[TylerDurden]

I like that color-coded graph, TD. How would it look if the motor was cooled by a blower?

The green zone moves up, but:

There will be a limit, since the design of the motor allows little airflow across the windings & rotor.

Methinks pancakes are better than cans in that regard.

 

[eP]

Wrong colors TD

Power(out) is always cool ! (green)

Power(heat) could be hot.
To reduce power(in) when motor becoming hot you don't need blind criple controller.

You could use cheap temperature sensor or even simple timer activated above given current limit.

This way you could get warning signal enough long before the limit will be reached.
Smart controller will be able to calculate for you how much time you still have before the limit will be reached and power_in limits will be activated.

 

[TylerDurden]

Wrong colors TD

Power(out) is always cool ! (green)

Power(heat) could be hot.
To reduce power(in) when motor becoming hot you don't need blind criple controller.

You could use cheap temperature sensor or even simple timer activated above given current limit.

This way you could get warning signal enough long before the limit will be reached.
Smart controller will be able to calculate for you how much time you still have before the limit will be reached and power_in limits will be activated.

Agreed.

Color was to illustrate time v. heat, by corrupting teacher's chart.

:wink:

 

[fechter]

I think the color bars need to be oriented 90 degrees to the way shown. Everything to the left of "rated power" should be in the red. At the extreme left side of the graph is where the motor is stalled and the power output is zero, but the heating is maximum.

Of course, that's just a guess since the axes of the graph are not labeled.

 

[TylerDurden]

Better?

 

[fechter]

Yes, that makes more sense to me now.

 

[Mathurin]

But that's for full throttle.

If your power out corresponds to where "brief use" is written - would that not be green? EG: Matt crawling along in grocery store with very low power usage / respectable eff - without his motor burning up.