Faster motor winding or slower the eternal discussion!

[Doctorbass]

I know that many will say faster motor winding is better and heat less for the same generated output power and i think it make sense.
There is hundreds of various thread about that on E-S.

But there is something else we should add to the calculations: the ability to the controller to drive the crazy phase amp required.

Many time during i have used my faster winfing motor like the 5303(kv of 12) and 5302(kv of 18) i blew my controllers. This never happened when i used my 5304 or 5305 wicth are slower motor.

This is a matter of phase amp required for the desired torque..

Faster motor require more phase amp for teh same torque but produce less heat for the same output power.

Problem is that:
- Faster motor run cool but controller run HOT !
- Slower motor run hot but controller run COOL !

I have read on the Adaptto forum that for every 100A phase current there is a 10V spike generated when the mosfet switch ON and OFF.

From my understanding This mean that with mosfet of 100v rating, if you set 100A phase current on the controller you should not use battery above 90V.

what i see i sthat if you want to use faster motor (appear to be more efficient) than you need a high amp and high volt controller

A 400A phase controller woudl require about 40V margin between the mosfet voltage rating and teh battery used.. ex the 4110 would require to not be used with battery above 60V to get the 400A...

Let's continu this discussion with your opinion guys...

Doc

 

[gogo]

I know that many will say faster motor winding is better and heat less for the same generated output power…

Who said that?

 

[friendly1uk]

I know that many will say faster motor winding is better and heat less for the same generated output power and i think it make sense.
There is hundreds of various thread about that on E-S.

That depends who you are, but I find it inaccurate.
Most of us when polled want to do ~18mph, which a slow wind will do. A faster wind will be less efficient, wasting power as heat.

Your talking about the minority that ride at full speed everywhere. A high full speed.

Where people have a target speed, the slowest motor possible is likely the right one. Giving best torque and efficiency to do that job.


I'm sure your thread is simply not aimed at the average user. I'm just posting so Mr Average is not confused. I look forward to zipping my mouth and learning something now

 

[liveforphysics]

I know that many will say faster motor winding is better and heat less for the same generated output power and i think it make sense.
There is hundreds of various thread about that on E-S.

Many fools may say such things, but we've gone through this so deeply now in the threads KF started that I think most everyone has got it now and doesn't believe the nonsense anymore.

Copper fill % in the slot is all that matters.

 

[cwah]

I was thinking that faster motor would need more amps thus generate more heat due to loss.

That's why slower motor and higher voltage make sense, so amps is lower and loss is also lower.

Isn't the reason why our electricity line are 220V? To minimise loss?

 

[MrDude_1]

Clearly the best choice is the one where both the motor and the controller are more than capable of running within the desired power output.

So the problem is the human that always wants the maximum power output.

 

[flathill]

DrBass. You should already know this from your car audio days...

It's like running a 1-ohm subwoofer driver on an amp designed to drive a 16-ohm driver and wondering why it got hot and blew up.

It's like running a 16-ohm driver on a amp that can handle 1-ohm loads and you wonder why it is not getting as loud as with your other 1-ohm driver, or as loud as the other amp with a higher rail voltage designed to drive a 16-ohm load.

The driver and amplifier and rail voltage need to be selected to work together.

The motor and the controller and battery voltage need to be selected to work together

 

[teslanv]

Here is what I believe Doctorbass is getting at:

Let's say you have a DD Hub motor with different winding options. - For Instance the MXUS 45mm Motor.

If you assume the same % of copper fill (same total number of strands), then which winding is easier on a given controller such as an Xie Chang (Infineon) 12X4110 Controller?

Motor A has 20 strands in parallel and 3 turns. (60 strands of copper fill) The Kv is about 12 RPM per volt and Kt is 0.796 Nm/A
Motor B has 10 strands in parallel and 6 turns. (60 Strands of copper fill) The Kv is about 6 RPM per volt and Kt is 1.592 Nm/A

To produce a given amount of torque at a given speed, Motor "A" wants to use 1/2 the voltage and twice the phase current of Motor B. Conversely, Motor B wants to use twice the voltage and 1/2 the phase current of motor A.

Since Motor A has twice as many strands of copper in parallel, it can handle exactly twice as much current as motor B, so to the Motor, Either scenario will result in the same waste heat generated.

However the Hypothessis is that Motor A would require phase wires that are twice as large and the controller would be working its MOSFETs twice as hard to generate the higher phase amps required to run motor A as compared to Motor B.

 

[riba2233]

That's why you need controller to match a motor

 

[liveforphysics]

I was thinking that faster motor would need more amps thus generate more heat due to loss.

That's why slower motor and higher voltage make sense, so amps is lower and loss is also lower.

Isn't the reason why our electricity line are 220V? To minimise loss?

Resistive loss heating increases at the square of current, yet winding resistance increases at the square of the number of turns (the wire gets both longer AND thinner to make the extra turns).

 

[flathill]

yes this is simple

every given controller has an optimum system voltage and optimum winding

every given system voltage has an optimum controller and optimum winding

every given winding has an optimum controller and optimum system voltage

the key is the desired torque/power vs rpm must be defined before you start, both peak and continous

if you have a motor on hand you design/select the controller and system voltage around it

if you have a controller on hand you design/select the motor and system voltage around it

if you have a system voltage on hand you design a motor and controller around it (this is your best option if you have the freedom)

 

[teslanv]

yes this is simple

Apparently no so much... :wink:

 

[Doctorbass]

DrBass. You should already know this from your car audio days...

It's like running a 1-ohm subwoofer driver on an amp designed to drive a 16-ohm driver and wondering why it got hot and blew up.

It's like running a 16-ohm driver on a amp that can handle 1-ohm loads and you wonder why it is not getting as loud as with your other 1-ohm driver, or as loud as the other amp with a higher rail voltage designed to drive a 16-ohm load.

The driver and amplifier and rail voltage need to be selected to work together.

The motor and the controller and battery voltage need to be selected to work together

Yeah that's correct I know that already flathill.. as well ! 8)

but your exemple would probably apply more to class D amplifier due to the PWM and "current multiplication" and voltage overshoot of the mosfet at high phase amp.

Doc

 

[Doctorbass]

It make sense to choose a controlelr and motor to match together .. as well !

But how would you determine if you select faster motor and lower voltage high amp controller or a slower motor with more volt and low amp controller...

That's what i'm suggesting to discuss about...

Luke i know you've always said it would be better to use lower voltage at higher phase amp ( witch correspond to using a faster winding motor i guess too)


Doc

 

[flathill]

the tricky part is only when you start to define the limits with regards to peak and continuous power

the next level of optimization makes them all limit around the same time

no use have a controller that can put out 500kw continuous if your motor can handle only 100kw continuous.

your battery may end up being the limiting factor if you cannot increase capacity.

 

[flathill]

DrBass. You should already know this from your car audio days...

It's like running a 1-ohm subwoofer driver on an amp designed to drive a 16-ohm driver and wondering why it got hot and blew up.

It's like running a 16-ohm driver on a amp that can handle 1-ohm loads and you wonder why it is not getting as loud as with your other 1-ohm driver, or as loud as the other amp with a higher rail voltage designed to drive a 16-ohm load.

The driver and amplifier and rail voltage need to be selected to work together.

The motor and the controller and battery voltage need to be selected to work together

Yeah that's correct I know that already flathill.. as well ! 8)

but your exemple would probably apply more to class D amplifier due to the PWM and "current multiplication" and voltage overshoot of the mosfet at high phase amp.

Doc

You can design a 1-ohm driver and 16-ohm driver to have almost exactly the same performance, as long as the amp can supply the current and voltage needed in both cases. This is true whether the amp is Class A, B, AB/G, AB/H, T, D, or whatever

 

[teslanv]

If we consider that most DIY ebike builders are going to use commonly available motors and controllers, this narrows the options a bit.

DD Hub motors and "Big Block" Inrunner motors come to mind, as do Greentime, Xie Chang, Kelly and Savboton controllers.

 

[flathill]

What DrBass really wants to know is what is the optimal motor to use with the Adaptoo Max-E

Pick your top speed and work backwards given the constraints of the controller to find your motor while taking inductance into account.

 

[Doctorbass]

flathill, not only the Max-E.. but all controller and all motors.

here is couple of my personal expériences:

with a fast Wind motor ex: kv of 12.3 ( ex: my Giant with the 5303 and 113kmh with the 125v 220A Kelly i had really nice powerband.. not peaky, just flat but still great torque from the start and power up to abot 106kmh... but it blew few times my controller.

My 5302( the real controller killer.. !) with a kv of 18 worked really short period before to blow all my controllers after one or 2 runs. The torque was too low for me and i was uder impression that the powerband was like spread on a too wide range up to 150kmh+ but had not enough torque to reach that speed and was only 109kmh... The controller match for that controller wold require insane phase amp and high volt margin above the battery max voltage.

My 5304 and 5305 (kv of 9 and about 7) never blew any controller worked for years with great service and awsome torqure and fun but i had to keep an eye on the temp sensor all the time...


I think that high speed winding ( high kv) require NOT ONLY high phase amp rating but also HIGH VOLTAGE to keep a safe margin from the battery voltage due to the high volt spike. Fast motor require extreme volt and extreme amp so extreme $$$ to work great without any problem...
Look like high kv motor are dedicated for ultra high power controller only... but... maybe i'm wrong..

Doc

 

[liveforphysics]

It make sense to choose a controlelr and motor to match together .. as well !

But how would you determine if you select faster motor and lower voltage high amp controller or a slower motor with more volt and low amp controller...

That's what i'm suggesting to discuss about...

Luke i know you've always said it would be better to use lower voltage at higher phase amp ( witch correspond to using a faster winding motor i guess too)


Doc

The choice is easy. As you know the battery doesn't care, it can be configured in any arrangement without penalty, and the motor doesn't care, it can be wound for any Kv without penalty, you're left looking at the controller, and hence whatever switches power most efficiently.

This means something like 20s if your goal is ultimate high performance, as it's about the minimum safe margin for using 100V MOSFETs, which are currently the power density switching maxima (with the exception of a couple very costly freak parts in the 650V range that offer a tiny power switching advantage, but the BMS cost and life-safety risks would be insanely pointless for a controller that is like 10% smaller.)

Motor = Voltage agnostic
Battery = Voltage agnostic (except they get cheaper and safer and less failure prone and corrosion prone as the BMS has less channels)
Controller = Limited by available Silicon performance in whatever the latest MOSFET dice can offer, right now the highest power transfer and/or most efficient controller is going to be using 100V MOSFETs, because they are currently leading the industry.

 

[flathill]

I'm guessing your high Kv winding also has lower inductance

maximum current ripple occurs when the back emf is equal to half the system voltage, and is inversely proportional to the motor inductance and the controller switching frequency

This does not mean higher inductance is always better. Your inductance could be so high that current can't reach steady state at high rpm. Again the controller and motor must be matched.

 

[cwah]

I noticed my leaf bike motor was blowing all my controllers. I needed to get an 18 fet for barely 48V35A to make sure they don't blow.

Would be good to know if there are fets that are so good they never blow with high windings motor

 

[teslanv]

Controller = Limited by available Silicon performance in whatever the latest MOSFET dice can offer, right now the highest power transfer and/or most efficient controller is going to be using 100V MOSFETs, because they are currently leading the industry.

Is there a better MOSFET than the common IRFB4110 for the TO-220 package?

 

[teslanv]

Perhaps the TI CSD19536KCS ?

http://endless-sphere.com/forums/viewtopic.php?f=30&t=58311

 

[Doctorbass]

Allex recently hit 160A batt current and 400A phase on 20s and 13.2kW on the MXUS 3000 4T and really great aceceleration!

Luke, Look like 20s battery is the key like you said many times!! the 100V mosfet seem to require 25-35V margin with high phase amp!

Doc