VOLTAGE, HIGH OR LOW: THE MATHS

[Doctorbass]

Here is an excellent explanation from guy and a fellow on E-S ( the New(2013+) 50kW Zero motor engineer) he is one of the people with the most knowledge about motor and controller. He helped Luke to tune the legendary death bike too.

https://evmc2.wordpress.com/2015/02/12/voltage-high-or-low-the-maths/

https://evmc2.wordpress.com/2015/02/10/voltage-high-or-low-explained/

Doc

 

[actionobject]

Doc,

Thank you for posting this. I am digesting it in little bites.

For our high powered ebikes, my very basic understanding, is that higher Voltage creates higher top speed, and higher Amps creates quicker acceleration.

Is that at all accurate?

 

[riba2233]

For the same motor, yes. But you can get higher speed with same voltage (but higher amperage) with using lower turn (higher speed) motor. So basically, you can use whatever voltage you want for the same results, same performance.

 

[teklektik]

actionobject-
Also see the ebikes.ca Hub Motor Tutorial - a good overview of the same core physics and some other general hub motor information.

 

[cycborg]

higher Voltage creates higher top speed,

Correct, phase voltage determines speed (field weakening aside), and phase voltage is limited to battery voltage.

and higher Amps creates quicker acceleration.

Phase current determines torque, and therefore acceleration. How this relates to the battery is more complicated than with voltage, however.

Say you have a 36 V battery with a maximum discharge rate of 40 A, and you want to double your low-speed torque, so you buy another identical battery.

- If you connect them in parallel, you have 36 V with a maximum discharge of 80 A, so now you can obviously double your phase current and double your torque. You have not changed your top speed, however.

- If you connect them in series, you now have 72 V with maximum discharge of 40 A. You have doubled your top speed, yay. But you have also doubled your low-speed torque, because the controller will able to transform the battery’s high voltage and lower current into lower voltage and higher current.

Of course, this assumes that your controller can handle these voltages and currents and that its current limits are set properly.

 

[actionobject]

Thx for the link and the explanation.

So how does one find the sweet spot for these two variables, for each specific application.

Does it depend on reducing the back EMF or optimizing for the cruise speed desired, or what?

for example between a 48V 20ah and a 96V 10ah. there are a lot of possible combinations that could work well (assuming the Motor and Controller can handle the high amperage and voltage)

Is it correct to assume that the higher amperage will have a greater possibility of thermal rollback?

How does one find the sweet spot. Which would be, for many of us, as fast and quick as possible!

 

[riba2233]

Sweetspot depends largely on available technology. It's easier to find bms for 48 V then for 96 V. But it's easier to find a more powerful motor and controller for 96 V. So 72 V could be sweetspot, it's safe enough, and you have enough available parts for that voltage.

 

[Doctorbass]

If you want speed but want also a simple battery with simple/cheaper BMS at low voltage , you can go with a moderate speed motor and use the filed weakening ( a kind of timing advance..)

in this case for most of the ride at normal speed you get great efficiency from your motor because throttle is high and PWM is maximized and once you need to go fast, YOU CAN!! but at a cost of efficiency drop.

Some reported riding at 100+km/h with a 18s pack with the use of the Adaptto that does field weakening

Doc

 

[Doctorbass]

Someone could build a 100kmh ebike with hub motor with only a 48V battery....

Power supply ( used as a charger) are easy and cheap to find
48V battery are easy to find
48V bms are very popular

You would only need to use 4 gauge wires for the battery and controller... :lol:

But as well ... this is in the "WHAT IS POSSIBLE" question... not in WHAT IS THE BEST...

Doc

 

[Doctorbass]

Usually in the WHAT IS THE BEST question, this include:

-Efficiency ( heat loss),
-Price
-Availlability of the conponents
-Built difficulty

In some case it is you controller that take all the heat (with fast wound motor...) and in some case it is your motor(with slow wound motor)...
But controller are easier to cool down than hub motor!!..

While the average ebike are working with 36V average, our extreme DIY ebike are more like 72V average

When you have experienced sustained power while accelerating (large powerband) this is a completly different experience than using a torque motor and accelerate VERY quickly but with speed limited to 60kmh.. ( when you feel that at 45km/h acceleration begin to drop alot) The first sensation is more BRUTAL and the second is more INTENSE !


Now this conduct to another great question..!! is it better to choose the trade off ? medium voltage ?? :lol:

Doc

Doc

 

[liveforphysics]

20s is the ideal cell string choice for high performance right now from perhaps 10hp to around 1,000hp.

Below 10hp I would run something in the 24-48v range just to further minimize BMS hassles.

 

[Doctorbass]

20s is the ideal cell string choice for high performance right now from perhaps 10hp to around 1,000hp.

Below 10hp I would run something in the 24-48v range just to further minimize BMS hassles.

Hmmm 20s.... 74V for 1000hp :lol:

746 kW of power on 74v...

hmmm.. well the thread subject is "THE MATH"

so let's do THE MATH :lol:

746000W / 74V = about 10 000amp ?

that's serious wire size Luke !!!

This make me think about this impressive comment i have found on ELECTRICAL ENGINEERING:

I worked on an electric locomotive starter years ago that ran the companion alternator in reverse as a motor to start the engine. We had OOO gauge wires running to the motor carrying 10,000A peak current (For a short time, RMS over a cranking cycle was much lower). When the motor started cranking the wires would slam back and forth into each other as the magnetic fields caused them to repel and attract. The CRT monitors in the building would all start to go wavy as well.

:lol:

So twist the throttle and see your car steel frame twisting around the wires!!!!

We are in the one tesla field range at few cm from the wires i thing right?

Doc

 

[liveforphysics]

10kA distributed between a number of practical to switch phases is quite reasonable for making 1,000hp.

 

[tobewankenobi]

Basically the voltage is not so interesting for the motor, it is usually possible to adept the winding to the voltage to get the desired performance.. the voltage and current is more interesting for the control, the capacitor, battery and cables to the motor.. you can have 10kA in a cable but it will be very thick and difficult to fit in your bike If you go above 48V however there is a risk of getting killed by the voltage, from what I've heard 48V is the limit for voltage that can kill you.. so if you make a bike with 72V or more there could be a tiny more risk.. higher voltage also means more expensive components.. lower voltage is usually cheaper.. if you would do 1kW with 12V you get probably arround 100A so the motor cables would probably be ~ 16mm2 which is thick but you can still bend them.. maybe the control will be a bit bigger thought due to the higher dc-current.. in the motor you would do parallel turns so no problem bending them..

 

[Punx0r]

If the future is large radius, relatively slow-spinning hub motors with lots of small, low-turn teeth, that 10kA could be one phase per tooth, with maybe 100 teeth and only 100A each. The electrical supply into the motor would need some thought to prevent the insulation of some many wires taking up an unnecessarily large amount of space, maybe square or flat copper insulated with varnish and stacked together. You could loose a phase or two and keep driving no problem...

 

[flathill]

There is no reason to stick to 3 phase motors other than cost efficiency

I think the optimal number of phases may be equal to the number of teeth on the stator. Each coil wound on each tooth will have its own switch (which may be built into the battery cell separator). It will have a self learning dsp control scheme running on a fpga which will communicate with each "phase" over the dc bus

During cruise conditions you can switch from a v8 to a v4 engine or whatever to save gas. Ultra fault tolerant perfect for flight

 

[actionobject]

Some reported riding at 100+km/h with a 18s pack with the use of the Adaptto that does field weakening

Doc

Yes Doc, that us what i ended up with 18s10p for my Adaptto explorations. built with the 25R 18650. Pretty excited about that!

20s is the ideal cell string choice for high performance right now from perhaps 10hp to around 1,000hp.

Below 10hp I would run something in the 24-48v range just to further minimize BMS hassles

.

Luke, 10hp is about 7500W, which is way more than i am running. IS it more efficient to keep things at lower voltage, say 48V and push more amps thru it?

Will this not create much greater heat?

Apparently many of the parameters we are dealing with are extracted from what is available to the consumer as opposed to what is optimal?