Question about Startup Power to an Already Turning Motor

[kostelacj]

It seems to be reasonably well known that it takes more energy to start a motor turning from a standstill with a load than it does to keep that same motor running with that same load. I think that is true anyway. Assuming it is, how does one calculate the amount of energy, and the attendant current spike(s) that will occur when applying power to an already spinning motor? And thus how can one allow for that in determining the the duration one can expect from a battery that starts at a particular state of charge?

Essentially, I want to be able to quantify the amount of battery saved or expended when I depress the throttle on my e-bike from a stand still or from a given rpm provided to the motor by my scrawny legs.

Any takers on this challenge?

Thanks, John
Bike-E recumbent e- Bike
GM 36V 750W in a 20" rear wheel
Powered by GM 36V 16Ah LiMn Pack

 

[dogman dan]

I'd do it the dumb carpenter way. Two identical rides in similar weather, but ride one pedal first and the other pedal later. Measure the ammount needed to recharge for each ride. I'm asuming you have no cycleanalyst.

 

[PhoenixOSU]

You could get a guess by calculating the motor current and multiplying by some number, 1.5 or 2 should get you a ballpark figure to start from. It can be quite a bit.

Current = power / voltage

 

[kostelacj]

Sadly, no cycle analyst. Yet. And no way to know how much I charged other than to stare at the charger until the little charge light goes green and that probably includes who knows what BMS activity. I am looking for some formulaic means now for design purposes. Eventually, I will get around to measuring it, but I would like to have something in my head, in the meantime. Besides the rocks some folks think are there, that is.

Thanks,
John

 

[Drunkskunk]

Its a more complex question than it sounds like it should be.
You need the formula for accelerating a mass to a given speed over a given time.
F = M X (V / T)
Force equils Mass times (Velocity devided by Time)
Force = newtons
Mass = kilograms
Simple enough. if you brake it up into set speeds or times, you can graph it.

the other part of that is were it gets complex. you need to know your motor's efficancy at a given speed, as as the force is the actual power, and doesn't include any losses. to calculate that.. well, the math behind calculating that is the subject of volumes of books. its far easier to measure it.

 

[Tiberius]

A trial would be the best way. Its not a simple calculation.

Its possible to calculate the kinetic energy you've created by pedalling before switching to motor power. That's simple enough and it won't be a great deal unless you are starting and stopping all the time. But that start up regime is where the motor is inefficient. In fact right at the start the efficiency is zero. So the actual saving in battery energy will be many times the kinetic energy you've put in.

Doing the full theory is of course possible, but a good rule of thumb could be to assume the battery saving is about 10 times the KE you put in. On that basis pedal first looks like a better strategy than regenerative braking.

Nick

 

[kostelacj]

I am excited by the responses received. After reading them and thinking through the math and reading some more web pages I have concluded that measuring it is what I am going to have to do. If I do that enough I might be able to come up with some good guesstimates based on my setup. If I could get some measurements on other setups I guess we'd could do some correlating and come up with a more generalized idea of the differences going on in the motor under those two scenarios. Does anyone have a tool that will capture the energy flow over time discretely? As opposed to just a summation like the various Wh meters and such. Would we have to PIC one up?

Thanks,
John
Bike E e- Recumbent ( BEe-R) (I think this could work for me.)
Driven by a GM 36V 750W 20" rear
Powered by a GM 36V 16Ah LiMn battery
Ridden by a MadMan

 

[dogman dan]

A lot cheaper than a cycleanalyst, is a Watts Up. This is a device you should be able to find for about 30 bucks including shipping, that is like an electric meter for one plug in the house. You can use it to measure how much power is used to charge a battery, and other things like finding out if the fridge is as bid a dog as they say it is, or finding out how much power your cable box is using when you aren't watching tv. I like real world tests myself, since I do have rocks in my head where math is done. I have found that the way I ride, which is to start pedaling after I reach about 5 mph causes a route that has a lot of stops to use about 1 ah more to get 15 miles home than a route with less stops. So about 20 stop signs can use an amp hour. It's pretty signifigant. Since I have the range in the battery, I don't worry about it, and prefer to stay in high gear, and let the motor get me to 5 mph. I definitely change the riding style if I need to get more range, and if riding slower in a lower gear, I can pedal start easily.

 

[Drunkskunk]

What you need is a Data logger. the Cycle Analyst latest version can be modified into a data logger, and some older versions can be updated and turned into a data logger as well. Justin (the creator) has a post about it somewhere on here.

A cheaper alternitive, but doesn't have the same power or speed capabilities is the E logger http://www.eagletreesystems.com/MicroPower/micro.htm

 

[fechter]

Another way to calculate it is by energy.
The kinetic energy of the bike is
E=1/2M x V^2

So to reach a certain velocity from zero, the energy is calculated as above. M = mass, V = velocity.
If you start at a certain speed, then the change in velocity is substituted for V. A reduction in the V^2 term will make a big difference in the energy needed.

 

[jdcburg]

Hi - I'm no expert but couldn't you get an inexpensive digital voltmeter and measure the voltage of a fully charged battery, do a set number of starts from scratch and remeasure the voltage, then repeat the experiment with running starts? The voltage drop seems like it should correlate to the power used. - jd

 

[Rassy]

Dogman wrote:

A lot cheaper than a cycleanalyst, is a Watts Up. This is a device you should be able to find for about 30 bucks including shipping, that is like an electric meter for one plug in the house.

Not to be picky, but to set the record straight, this is the Kill-A-Watt meter. Both are great meters, but the Watts Up costs about twice as much and can go between your battery and controller, charger and battery, or controller and motor to get basic amperage and voltage measurements.