5305 and 5302 in the same hub motor!

[Doctorbass]

Thanks Rick

[Doctorbass]

Hey Rick, I found the thread you alreadi contributed about custom winding : http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=3659&p=119686#p119686

you seems the most experiemnted man for that here!

By the way, i would really be interested to see some pics of your 5302/04 hub motor .. like the outside 18AWG wires, the axel you machined with your CNC.. and THE INTERIOR !!

I really want to try one.. I found two of that 3PDT relay for cheap that could work.. for every phase i need 3x DPDT relay so it can be also:

2x 3PDT !

now i got from 20$ on ebay 2x 3PDT so it's the same!..

115VAC 60A Sealed !!!!

for sure i will need to put them outside but that's not a problem for me if y use both side of the axel to pass the wires..


Doc.. happy

[rkosiorek]

well the motor came back home for christmas.

so i opened it up so i could take a couple of shots for DoctorBass. all that there really is to see is 12 separate leads. they are labeled:

Y1S - Yellow winding 1 start
Y2S - Yellow winding 2 start
Y1E - Yellow winding 1 end
Y2E - Yellow Winding 2 end

the same is repeated for the Green and Blue windings.

for 5302 mode all of the starts from all windings are connected together. then the yellow ends are connected together to the yellow supply from the controller. ditto for the Green and the Blue.

for 5304 mode Y1S, G1S and B1S are connected together. then Y1E connects to Y2S, and Y2E is connected to the yellow supply simmilarly for the Blue. B1E connencts to B2S. then the B2E is connected to the Blue supply, repeat for the Green.

anyway here are some pics of the wires comming out of the windings. at least i think that is what Doc asked for.

02_04001.jpg (74.51 KiB) Viewed 1821 times

02_04003.jpg (82.04 KiB) Viewed 1823 times

merry christmas Doctorbass

rick

[ZapPat]

You guys are making me drool with your big-ass multi-winding hub motors set into small diameter wheels!

This type of setup should give good efficiency too... if one can keep the throttle hand light that is!

[Doctorbass]

Wow.. Many thanks Rick.. you did a nice job!

After having re-reed your rely i found that i've missunderstod what you wrote about the Delta Star mode ratio...

this is not like 20% difference but more like between 100% or 173% speed difference when changing from star to delta!!

Examinating the expernal parts needed for this configuration.. like only one 3PDT relay!! instead of 2 like with serie/parallel, i think i'll go for the delta/triangle!

I discussed with some electrical maintenance senior at my job and they said thei use this configuration in star to delta their large 3phases motors and then they start them to requier less current. the gain ratio for speed is 73% if i remember?

[Tiberius]

I discussed with some electrical maintenance senior at my job and they said the use this configuration in star to delta their large 3phases motors and then they start them to require less current. the gain ratio for speed is 73% if i remember?

Doc, Rick,

I've converted a motor (3 phase AC induction) from star to delta. Its a standard change to operate off different voltages. I think the ratio change is square-root-three/2 = 1.732, which would be where the 73 % comes from. In the UK its done to switch between 230 V and 440 V supplies. I don't think the speed changes though.

My lathe has a complex winding system in the motor that allows 2:1 ratio speed switching from the same voltage

As Rick says, there's no point in having copper in the motor that isn't used, so star-delta switching is better than tapping into windings. On a DC motor, it would give you the equivalent of a 1:1.732 voltage change.

But all these things require complex switching. Why not instead look at a DC-DC converter to step up the battery volts. That could give you the equivalent of any voltage you like.

Also, AFAIUI, these changes are not the same as mechanical gearing after the motor. You can't make the motor more efficient at low speed by doing this, all you can do is push it to higher speed.

Nick

[Tiberius]

Guys,

I was thinking about this further, and I'm open to correction from someone who's done it on a BLDC, but....

Star-Delta switching introduces a 60 degree change in the phase relationship between the terminals and an individual winding (in addition to the 1.732 voltage change).

So, depending on the geometry of the motor, it is quite possible that the Hall sensors would need changing.

Nick

[markcycle]

The Halls don’t need to change (according to Crystalyte) but the controller may not like switching on the fly as the controller will a instant phase jump and might shut down. The ratio between Wye and Delta is 1.73

The reason I know this is I plan on offering my Motorcycle hub motor in Wye and Delta configurations to adjust for different wheel sizes.

Happy Holidays
Mark

[kZs0lt]

Also, AFAIUI, these changes are not the same as mechanical gearing after the motor. You can't make the motor more efficient at low speed by doing this, all you can do is push it to higher speed.

Does this mean that the efficiency is the same whether star or delta, making the switch absolutely useless, unless you you want to change rim diameter or battery pack voltage? That would mean that you can stick to 5302 or delta for more speed, and switching to higher windig or star using SAME POWER(current x voltage at full throttle) would not give you more torque or better efficiency at low speed? Or at least the change would be so small that you couldn't feel it?
Those using 406/4011 or other dual winding hubs, what are your real experience?

[Doctorbass]

Also, AFAIUI, these changes are not the same as mechanical gearing after the motor. You can't make the motor more efficient at low speed by doing this, all you can do is push it to higher speed.

Does this mean that the efficiency is the same whether star or delta, making the switch absolutely useless, unless you you want to change rim diameter or battery pack voltage? That would mean that you can stick to 5302 or delta for more speed, and switching to higher windig or star using SAME POWER(current x voltage at full throttle) would not give you more torque or better efficiency at low speed? Or at least the change would be so small that you couldn't feel it?
Those using 406/4011 or other dual winding hubs, what are your real experience?

WYE give more torque Delta give more speed for the same power (watt)

I dont want to replace my 5305 with a 5302 when i need speed and to put it back when i need more torque... that have no sense.. The contactor to switch between these two winding mode is to allow me to have a motor taht for the same power can have the same power and speed caracteristics than a 5303 or a 5305 without the need to replace the hub motor every times i need to switch. This idea give me the same power rating capability for both mode and unlike the 408/411, i have no problem with power capability of this 400 serie (411 allow more power than the 408 mode carsu the amount of cooper is different in each mode)

also the 1.73 ratio will help me with that 5305 to reach higher speed without the need to keep a larger wheel diameter or higher voltage...

my configuration will be 100V wich is the max my controller can take with 20" diameter wheel (to get the max trust power as possible for a given torque)

so the dynamic of operation of my 5305 is i think the best setup possible for having monstrous torque and speed

At this day, there is absolutly no bicycle hub motor that is so powerfull as a 5305.. the reason: it work with more volt!

A for exemple with a 5202, to get 100A that will be fiddicult to reach over 60-65V.. at full throttle so it's 6500W max that you can put in

In the popposite a 5305 will need higher voltage to get 100A and will need around 90-100V at full throttle.. Thats 9000-10000W Max so this is a fact that the 5305 with longer winding allow more power than a 5302.

So with a 5305 on a 20" wheel at 10kW there is no possible configuration that can be as powerfull as this.. unless to use a kelly controller with 220A capability

Now this configuration have alot of trust and acceleration capability !!.. but it have a lack of top speed!....(around 62km/h) the only solution to keep these acceleration capability and to have the possibility to get higher speed is to play with the winding of the motor !!.. the Star(WYE) to Delta is not the most efficient solution cause it can allow a speed boost factor and a torque lost to around 1.73(without wind factor).. at speed of around 80-90kph many member here reported that the need of power at this speed is around 5000W.. so there is still enough margin to get the 105km/h i think.

The killacycle operate very similar as this method but this is more a question of serie/parallel motor than phase winding configuration wich offer a different factor of 2.0 instead of 1.73

Doc

Doc

[ZapPat]

Also, AFAIUI, these changes are not the same as mechanical gearing after the motor. You can't make the motor more efficient at low speed by doing this, all you can do is push it to higher speed.

Does this mean that the efficiency is the same whether star or delta, making the switch absolutely useless, unless you you want to change rim diameter or battery pack voltage? That would mean that you can stick to 5302 or delta for more speed, and switching to higher windig or star using SAME POWER(current x voltage at full throttle) would not give you more torque or better efficiency at low speed? Or at least the change would be so small that you couldn't feel it?

I've been wondering about the efficiency issue myself, because in Doc's case for example it would be simpler for him to just hardwire his motor into a Delta, and drop the wye. Theoreticaly this Delta/100V setup would be just as efficient if Doc were to instead boost his battery pack voltage by 1.7X (to 170V!) and keep the Wye, right? Since this is a ridiculously high voltage, lets compare the two winding options with a 100V pack like Doc has, but using the controller's duty cycle to compensate for the windings. So lets look at a situation where the motor is soaking up 10,000W in delta, then in wye:

Delta @ 59% Duty cycle = Wye @ 100% Duty cycle
Delta @ 59% Duty cycle: V_motor = 59V, I_motor = 170A
Wye @ 100% Duty cycle: V_motor = 100V, I_motor = 100A

This means that the FETs in the controller are seeing much higher peak currents if left permanently in the Delta from a standstill, instead of starting in a Wye. So the motor itself may not gain much efficiency from this, but the controller has much lower current demands on it, and thus also is working more efficiently. And if we also consider switching losses in the controller, keeping the duty cycle of the controller at 100% for as much time as possible also eliminates these losses.

I guess the reduced controller requirements mostly explains why the killacycle people decided to use a similar series/parallel switch. Apparently, the Tesla gang also did something similar after destroying their two-speed mecanical gearbox from excessive torque!

BTW Doc, I believe that the Kelly controllers are rated by motor current and not battery current, so take this into account when reading that 220A rating. I guess they use motor phase current limiting instead of the basic battery bus current limiting that we get on all our cheap ebike controllers. This is a better way to do current limiting for sure, but it also means that the advertised curent rating is higher than an equivalent ebike controller rated by battery current, when running at less than 100% duty cycle. For example, your 220A controller will top out at 110A (battery) when the controller's duty cycle is at 50%, either because of throttle demand or if the controller is doing current limiting. That's still insane for an ebike anyways mind you...


Now, here's a question for some knowledgeable motor guys out there:
What kind of efficiency gain do you get when going from a one speed system to a two speed system with a 1.7/1 ratio (when talking about mecanical gears of course)? Any links to good ressources about this are welcome.

Pat

[rkosiorek]

it definitetly improves drivability and performance. it must also improve efficiency but i don't know how much.

in the 5302/5304 the motor pulls very well from a start in 5304 (series) mode. as your speed increases the rate of acceleration decreases. now if i switch to 5302 mode the Kv of the motor is doubled and i can continue accelerating.

i have no scientific evidence of an efficiency difference. but there must be some. pushing my trike at 25kph the throttle position is considerably lower in 5302 mode and i can still accelerate. in 5304 mode i would be pretty much maxed out for throttle.

rick

[John in CR]

... for example with a 5202, to get 100A that will be difficult to reach over 60-65V.. at full throttle so it's 6500W max that you can put in

In the popposite a 5305 will need higher voltage to get 100A and will need around 90-100V at full throttle.. Thats 9000-10000W Max so this is a fact that the 5305 with longer winding allow more power than a 5302.

I've read through this thread a number of times, but I keep coming back to the thought that the 5202 is the more powerful motor. To me its shorter windings and effectively greater wire gauge from more parallel strands is capable of handling greater current flow due to the lower resistance. The part I really don't understand is why the 5202 would be limited to 60-65V, while 90-100V is possible with the 5305. Why would the voltage limits be different? I thought these BLDC motors didn't really have voltage limits other than structural limitations of the maximum rpms they can withstand. That would mean the one that can handle the most current is the most powerful.

John

[ZapPat]

... for example with a 5202, to get 100A that will be difficult to reach over 60-65V.. at full throttle so it's 6500W max that you can put in

In the popposite a 5305 will need higher voltage to get 100A and will need around 90-100V at full throttle.. Thats 9000-10000W Max so this is a fact that the 5305 with longer winding allow more power than a 5302.

I've read through this thread a number of times, but I keep coming back to the thought that the 5202 is the more powerful motor. To me its shorter windings and effectively greater wire gauge from more parallel strands is capable of handling greater current flow due to the lower resistance. The part I really don't understand is why the 5202 would be limited to 60-65V, while 90-100V is possible with the 5305. Why would the voltage limits be different? I thought these BLDC motors didn't really have voltage limits other than structural limitations of the maximum rpms they can withstand. That would mean the one that can handle the most current is the most powerful.

John

John, check ebikes.ca hubmotor info page to understand the winding lenght/thickness vs voltage/current relationship. In a nutshell, a winding two times thinner, but twice as long will give you the same torque, but with twice the voltage and half the current. This being said, I realise that I recommended to methods to change from a wye winding to a delta for a longer high torque startup... but I just now wonder if it might not make for lower initial torque if his controller is already hitting it's motor phase current limit. If this is so, I guess I can see better why we would want to have a delta/wye switch after all! And of course let's not forget the series/parallel possibility for the battery pack which will do about the same thing, but with a ratio or 2 instead of 1.7.

Following this logic, a permanent delta would give as much low speed torque as a wye only if the controller's phase (motor) current limit is not being hit. For these big motors I think this might well happen, even with method's large kelly controller. So the best thing would be to start out wye at low speeds and switch to delta once the controller stops doing current limiting into the motor. I wish Justin's simulator would have a "delta/wye" winding option in advanced mode, and especially a "controller phase current" limit option (instead of just the battery current limit as now). I don't think either of these addons would be too hard to do, but like everything they take some time, and I'm sure Justin is a busy man!

[John in CR]

ZapPat,

I read Justin's page again, and I understand that same motor with different windings but with the same amount of copper, produces the same power given the same power input (One just turns faster and the other turns slower but with more torque). I also understand the benefits Doc is going for with the switchable windings. The part I don't understand is regarding maximum power. It seems these motors don't have a voltage limit, at least not a cheaply achievable one, but they definitely have a current limit. Won't the one with the shorter thicker windings be able to handle more current? Then assuming you're running the same voltage, then the one that can handle more amperage is capable of producing more power. It won't have the same low end torque without wasting batteries, but it will push through more wind, up at the top end, won't it?

John

[rkosiorek]

ZapPat,

Won't the one with the shorter thicker windings be able to handle more current? Then assuming you're running the same voltage, then the one that can handle more amperage is capable of producing more power. It won't have the same low end torque without wasting batteries, but it will push through more wind, up at the top end, won't it?

John

sure if it were not a motor and just a lump of wire with straight plain DC going through it. that would be a static thing. it would be much like a resistor then. but wait, it is a MOTOR. windings which change their position within the strong magnetic influence of the permanent magnets. not only that but the controller feeds pulses of DC through the windings so the magnetic force created by the windings is constantlly expanding and collapsing. another complication is that the polarity of these pulses alternates directions, changing the direction of these magnetic fields. nothing static about it. a lot of different things interacting and influencing each other. sounds sort of like AC dosen't it? it should because it is.

the windings increase in resistance with RPM because as the motor turns it also acts as an alternator. this generated voltage acts in reverse and opposes the voltage going into it. its called Back Electromotive Force (BEMF). BEMF is proportional to the nummber of turns. lower number of turns = lower BEMF = less significance at high rpm = higher rpm. balancing that lower turns also means a weaker magnetic field so less force is generated for each pulse = less force at low rpm.

the motor makes up for this less force per pulse because at high rpm it has a lot more pulses. power is the total of the energy of all those pulses. that energy or power is the force per pulse (X) the number of pulses in a given time period.so a high turns count motor produces higher force per pulse but fewer pulses. the total energy is the same as the identical motor with fewer turns at higher rpm. i.e. higher FORCE (X) fewer pulses will be equal to a higher number of PULSES of lower force.

there is also the issue of if you put a high enough voltage into the windings and stall the motor you can create a magnetic fields strong enough to demagnetise the magnets. that would be your absolute voltage limit. usually you want to stay well below that. it is also the flaw in permanent magnet motors.

so these motors are not universal. you need to get a motor that is designed to be happy at the RPM you intend to run. a lower turn count motor may deliver the max speed you desire, but you'll have to wait an hour to accelerate up to that speed, or it could stall and not turn at all at low speed with any kind of load at all. being able to switch Delta/Star or series/parallel windings will allow the best of both worlds. gives you a low speed high torque solution to get going as well as a high speed running solution.

rick

[rkosiorek]

the demagetising thing is a reality. i have a couple of Kolmorgen motors from a Currie/USPD that i have permantly demagnetised by running at 72V. they use older ceramic magnets and are much easier to demagnetise than the rare earth magnets NEO magnets that the X5 motors use.

they did get pretty hot. but not hot enough to trip the 105C thermal cutout i had attached to the windings. so i am sure it was the strenght of the magnetic field and not the heat that demagnetised these magnets.

rick

[ZapPat]

The only limit to the battery voltage would be the motor wire's electrical insulation, and I would guess that's gotta be pretty high.

OK, now for the short/fat (5202 style) winding vs the long/slim (5204 style) winding difference.

The short/fat coil will produce the same torque as the long/slim coil, but at twice the current and double the voltage. When thinking about the electric/magnetic relationship of each, one just uses a single turn of double current conductor, and the other uses two turns of single current conductor, both giving the same resulting torque. They also each have the same net copper. It's kinda like switching two batteries between series/parallel connections - one gives twice the voltage, and the other can give twice the current, but both results have the same available power and energy.

I'm not a motor expert by any means and so my explanations might not be great...

Now the fun part (for me) is to figure out how the controller comes into this, since it plays a crucial part in the performance outcome of the setup. At low speed conditions (acceleration), the controller's current limiting behavior will limit your torque, and specifically it's phase current limit (same as motor current) because current is directly proportional to torque for a given motor. What I don't know is the torque vs current relationship differences between a delta and wye. Would it be the same thing as having a 52[02] vs an imaginary 520[3.4]? I mean, is it like having a 1:1.7 turns ratio difference between the Delta:Wye or not? If so, then the Wye would produce 1.7 times more torque than the Delta at startup so long as the controller is operating in it's phase current limiting region. With a controller that has a high enough phase current rating this torque loss wouldn't happen, but you need one that is rated at a 1.7X higher phase current rating than with the Wye... more expensive!

Pat

[John in CR]

I'm still trying to get to the bottom of Doc's explanation for why he says the 5305 can put out more power than the 5302. Maybe Rick explained it, just in terms not simple enough for my lack of knowledge. Is it because of BEMF, and that's what limits the current, instead of only the controller and battery limiting current like I was thinking?

Here's why I ask. I have some motors similar in size to an X5 at 12kg. They're wound fast and with the 60V controllers from the factory I go over 40mph on the flats. The no load speed is 950rpms at 60V and I'm running a 20" tire. Small grades and strong headwinds slow me down some, but as grades approach 10% it really bogs down. They never get more than slightly warm unless I allow it to really bog down on a long steep hill. My thought is that my controller is limiting the motor, and a higher current controller will deliver the amps needed to maintain speed under higher load and keep the motor operating in it's more efficient rpm range, so it still doesn't run hot. I have the speed I want, now I want better performance on hills and faster takeoffs. If some kind of dual winding switch like is being done for Doc is the answer, then I'm game for that too, or do I just need a beefier controller, or should I just add solder to the shunt first? I hesitate on the shunt mod, since the controllers already run warmer than the motors.

John

[Doctorbass]

John, Here is the explanation:

(all numbers are exemple.. not absolute value!)

When you have a current limit on a controller ...that limit will be reached easyer with a 5302 than a 5305 for the same power input cause the resistance of a 5302 is just lower than the 5305 right?..

so if a 5305 have 1 ohm between 2 phase wire with normal winding... a 5302 will have like 0.4 ohm.. so for the same voltage that you would operate them ..

let forget the inductive caracteristic and let see only as a resistive load. at 48V the 5302 would take I=V/R = 48/0.4 =120A/phase

on the other hand the 5305 with 1 ohm will take 48A (48/1=48)

so for 48V a 5305 take 48A and

a 5302 take 120A(out of limit.. and need to reduce the voltage)


you can see that on a different way:

let say your current limit is 50A on the controller

to reach 50A with the 5302 you will need V=RI = 0.4 x 50= 20V

to reach 50A with the 5305 you will need V=RI = 1 x 50= 50V

now that mean one motor, the 5302 can only accetp 20V with a limit of 50A and that's 1000W for 50A

but with the 5305 at 50A the voltage to reach that current is 50V.. that's 2500W

if we push it to the max limit value that crazy people like me can use: To reach the controller current limit, my 5305 need 100V.. and that give 100A

that's 10kW max


but if i try that with a 5302, to reach 100A i'll need 40V so the max power would be 4kW...


so for a fixed current the total power allowable to these hub motor is different!..

You need to examine that by taking the current limit and then to apply the corresponding voltage to reach it .

doc

[methods]

To me it seems like the most logical solution is to do what all the big dogs do. . . Throw money at the problem

In my experience so far the 5305 wont take much more than about 100A with my 88.8V lipo pack (testing done at 100V fresh off the charger).
It is only logical to me that the next best option is to put another 6S in line to make it a 109V lipo pack.

With 126V off the charger, the thrust at 40mph goes from just enough to maintain my speed (say 32lbs) to more like 65lbs

So now I have the same incredible torque I had before in the low end but I also now have that torque in the high end which gives me my high speed.
Faster than I ever want to ride the bike. . . Even for me 40mph is really where I draw the line of too dangerous.

So like with every sport, money is the name of the game.
Throw $700 at a controller that is rated for 18KW and throw another $200 at a 6S 10Ah 20C lipo pack.

but I know. . . I know. . . You guys are in it for the sport. The challenge...
I played the underdog for a long time with the Honda motors.
I understand the desire to do more with less. That is a noble cause do doubt.

So to my thinking it would be a given to maximize voltage, limit current, and then figure out how to solve the real problem. . .
What the hell do I do with all this heat

-methods

[John in CR]

Doc,

I think I understand, but you qualified it all with a current limit on the controller. What I'm looking for is, "Which is capable of more power?", assuming current and voltage aren't artificially limited by the controller or batteries.


Methods,

I think the best answer is to first minimize the creation of heat. I'm going to try some 2 motor alignments that aren't full time 2wd. Doc is trying the switchable windings. There's been talk by a few of cooling systems. I'm sure many are working on a better motor, but they'll be expensive and probably slow to become available. What else can we do without sacrificing performance?...Maybe a braking loaded spring or mechanical device to help bear some of the load of takeoffs...

John

[Doctorbass]

Doc,

I think I understand, but you qualified it all with a current limit on the controller. What I'm looking for is, "Which is capable of more power?", assuming current and voltage aren't artificially limited by the controller or batteries.


Methods,

I think the best answer is to first minimize the creation of heat. I'm going to try some 2 motor alignments that aren't full time 2wd. Doc is trying the switchable windings. There's been talk by a few of cooling systems. I'm sure many are working on a better motor, but they'll be expensive and probably slow to become available. What else can we do without sacrificing performance?...Maybe a braking loaded spring or mechanical device to help bear some of the load of takeoffs...

John

John, bout the "Which is capable of more power?",

I wrote the answer above.

We know that the power taken by a motor is determined bt the formula of

Voltage x Current = Power

Voltage =Resistor x Current

Voltage = motor load x current that can supply the controller

50A across a 1ohm resistor result to more power (2500W) than
50A across a resistor of 0.4ohm... (1000W)

The resistance of a 5302 is lower than a 5305. So current law can give the answer.

You know that for a given resistor (load) the voltage across it determine a current that cross it... just like a shunt..

So with the resistance that a 5302 have, the voltage that will be across it will be lower for a given current than for the 5305 that have a higher resistance.

The problem is that if you want to do 100V across a 5302 at full throttle.. normally the current would be like

100V/0.4ohm =250A...

and 100V x 250A would give 25000W.. But this is impossible!

If we use our exemple of the 50A limit controller, the equivalent voltage the controller give will be matched with the load and will by that way also decrease the current...

1: The power the motor have is the current x voltage that the controller send to the motor.

2: The voltage the battery have is not the equivalent voltage the motor will receive.. it depend on the load and the current limit of that controller

3: Power is Current x Voltage... and these two parameter also revelate the resistance (load) the motor have:

4: load = Voltage /current

5: 5302 = low resistsance 5305= higher resistance

6:power of the 5302 is limited by the maximum current the controler can output and with a lower resistance it is is just lower than with the 5305's higher resistance.

To resume.. what limit a motor power are lower resistance winding combined with lower voltage for a given current.

if you put 100V on a 5302 motor and set the current limit to 100A that can handle it, the output power will be less than the same setup on a 5305.

the reason?: the 5302 will reach the 100A limit at 40% throttle and will stop increasing power anyway if you try 100% throttle cause the internal PWM of the controller will limit the duty cycle of the output to sustain the 100A limit.. and the corresponding voltage at output will be like 40V

40V x 100A=4000W

But with a 5305, the 100A limit will be reached furter.. like 80-100% throttle so the corresponding pwm voltage to sustain the current limit at output would be like 80-100V

80V x 100A = 8000W



Doc

[John in CR]

Doc,

"Impossible", what if we add a liquid cooling system capable of rejecting 5kw of heat at 70°C? At the speeds involved I don't think that heat exchange will take too big a radiator. Can't a car radiator reject about 75kw of heat?

I think I understand your point now. With all else equal (controller and batteries) the 5305 is capable of higher power than the 5302.

Now lets say that under continuous real world use 100A is the limit for the 5305, which may be a true statement. If so, then couldn't a 5302 easily handle 150A? I don't understand the controller limits and the duty cycle stuff. Is the relationship linear? Then a 100V 150A controller would limit the 5302 to 60V instead of the 40V you had in your example, resulting in 9000W?

John

[methods]

It is worth noting that is is always easier to make power with voltage than it is with current.

10KW = V*I

V = 100V
I = 100A

Ok, 100A is about the most you can possibly run through "reasonable" cabling unless you want to run 4gauge down to your motor.

If we try:

V = 50V
I = 200A
W = 10KW

Now we have all sorts of trouble. If we even have 0.2ohms of resistance in our wire we see

200A * 0.2ohms = 40V !!!!!!!
That is 40V lost just to the wires!!!

The wires would have to be huge!
2gauge maybe.
The controller would be burning up, the connectors burning up, every part of the system supremely inefficient.

Now lets go the other way:

200V
50A
200A * 50V = same 10KW

Ah.....

Now we can use tiny little 10Gauge wires
Our loss through the same cabling above would be:

50A * 0.2ohm = only 10V
This 10V is only 5% of our available 200V
The above 40V loss would have been 80% of our available 50V

Now, we have not even broached the idea of the KV of the motor.

With the 50V 200A controller we will never be able to realize any speed
With the 200V 50A we would be able to realize a huge range of speed.

Just some thoughts.

These things are very important in the high power RC jet world
I have built jets that go 160mph and can sit on the front seat of my car
If I were to design for high current the weight of the wires would kill my performance
Designing with high voltage allows me to use tiny light wires.

Here is a 2.2KW tiny foam jet that I built
22V 100A RC jet

High Power RC.jpg (37.63 KiB) Viewed 5432 times

At the standard RC voltage of 3S 11.1V I would have had to run 200A

-methods