Conclusive proof gearboxes are awesome.

[SplinterOz]

Hillhater- Its not using a lot of power to be ripping up the asphalt right from a start.

Its likely getting 20:1 or 30:1 leverage on its phase current to battery current used as its starting out. This is because its only loss is the same loss its having when its making this amount of torque in its peak efficiency range, only now it has even less losses, cause its not having all the additional core losses of turning while its making this torque.
.

In simple terms , are you saying the Tesla is making maximum torque from standstill, yet not pulling max amps from its battery ??

Yes that is what he is saying and many of us not only understand why but also have measured it.
Note that the speed of the motor is directly related to the Volts applied (Kv). So if the motor is slow then the volts across the motor must be low. Now unless the controller is very inefficient (which most are not) then the KW (electrical) put into the controller are the KW put to the motor (loose about 10% for efficiency).
Doing the maths....
Batt Voltage * Batt Amps = Motor Voltage * Motor Amps (* controller efficiency)

Now if the Motor Volts are lower (slow speed) then the Battery Volts then the Motor amps must be higher by the same amount.

Simple really.

 

[Hillhater]

....The Model S didn't need a gearbox to beat the Viper.

No, but it did need a dozy driver in the Viper. ! :wink:
The Tesla is a very impressive & desirable car, but this blind "fan boy" delusion that it is faster than a Viper or M5 does none of us any credit.
Youtube video's are not proof.!
There is enough independent test results available that show the top spec Tesla's best 1/4 is easily bettered by both the stock Viper and the M5.
And what happens beyond the 1/4 mile is probably not going to be mentioned

 

[Hillhater]

Batt Voltage * Batt Amps = Motor Voltage * Motor Amps (* controller efficiency)

Now if the Motor Volts are lower (slow speed) then the Battery Volts then the Motor amps must be higher by the same amount.

Simple really.

So at peak torque ( stall or low rpm) , you are implying there is very low amp draw from the battery ?

 

[John in CR]

Batt Voltage * Batt Amps = Motor Voltage * Motor Amps (* controller efficiency)

Now if the Motor Volts are lower (slow speed) then the Battery Volts then the Motor amps must be higher by the same amount.

Simple really.

So at peak torque ( stall or low rpm) , you are implying there is very low amp draw from the battery ?

Not very low, but quite significantly lower than peak battery current, which occurs well above 0 rpm.

 

[Nuts&Volts]

Batt Voltage * Batt Amps = Motor Voltage * Motor Amps (* controller efficiency)

Now if the Motor Volts are lower (slow speed) then the Battery Volts then the Motor amps must be higher by the same amount.

Simple really.

So at peak torque ( stall or low rpm) , you are implying there is very low amp draw from the battery ?

Yes
Power= Battery Voltage X Battery Amps = Motor Voltage (RPM) X Motor Amps (Torque)

Battery Voltages=constant ~100V; So Battery Current is directly proportional to power not torque

 

[liveforphysics]

Batt Voltage * Batt Amps = Motor Voltage * Motor Amps (* controller efficiency)

Now if the Motor Volts are lower (slow speed) then the Battery Volts then the Motor amps must be higher by the same amount.

Simple really.

So at peak torque ( stall or low rpm) , you are implying there is very low amp draw from the battery ?

Absolutely. My bicycle draws 20A off the battery from a Size6 Sevcon gen4 to be sending 660A phase current to the motor and making full torque at 0rpm. Yes, 20A to is what it draws from my battery to make over 500ft-lbs of torque to my rear wheel, while still geared to have a top speed higher than I wish to travel.

On a Zero, the 100% certain to get longer range method of modding the bike is to just put longer gearing on it. The motor spins slower, you have worse acceleration, but you're using less power to do it because the tax to make that torque (copper loss) is the same at all speeds, but core-loss just keeps climbing with RPM, and some core losses climb quite sharply. If you want less range, just gear it shorter, you're now able to draw more power sooner (because RPM's climb faster for a given change in speed), you accelerate quicker but have reduced range from all the increased losses involved with spinning faster.

I would be very surprised if a Tesla model S is even drawing 40A when you pin the throttle from a stop and it's generating full torque sending a thousand amps to the motor or whatever it's doing to make full torque. And a portion of that 40A is going into the slip loss than your PM motors don't even have to deal with.

 

[Arlo1]

Batt Voltage * Batt Amps = Motor Voltage * Motor Amps (* controller efficiency)

Now if the Motor Volts are lower (slow speed) then the Battery Volts then the Motor amps must be higher by the same amount.

Simple really.

So at peak torque ( stall or low rpm) , you are implying there is very low amp draw from the battery ?

Not very low, but quite significantly lower than peak battery current, which occurs well above 0 rpm.

John you will find its even lower with a torque throttle and a more advanced controller

 

[Punx0r]

It's definitely an interesting debate.

Does anyone know the towing capacity for the Model S? I can't find it on Google.

OK, it's not a typical car you'd use for towing a caravan etc.

I'm figuring an average family car can pull something around it's own weight up a 25% grade at less than 20mph (steepest hill you're likely to find, using first gear).

A model S would NOT have any trouble pulling anything reasonable for any other car can pull, it would just have it's range decreased obviously by the amount of added energy you're requiring to be dragging a trailer through the wind and accelerating it's mass when you take off etc.

That brings up something extra odd to the table on this topic as well. The Tesla can't be making much power when you're taking off from a stop (because motor RPM is near 0). Mathematically it can't be making much more than say 30hp when you're below say 5mph with it. This might seem like a huge disadvantage vs a gasoline engine with a clutch in respect to launching the car from a stop. HOWEVER! The gasoline car with a clutch, as I've learned from dyno'ing them plenty of times and watching the power I put out starting the dyno rollers from a stop or nearly stopped and slipping the clutch to be in the powerband and launching the car, you don't put down jack-squat for HP with a gas car when it's nearly stopped either.

I should have clarified: I wasn't asking if the Tesla had enough torque, just wondering if the motor could cope with a heavy, low speed load without overheating, or excessively draining the battery. This kind of relates the "inefficient speed" argument. If you can tow your buddy home over the Rocky mountains without issue then I'm convinced it does quite nicely without a transmission.

BTW, the Model S gear ratio is 9.73:1, so with 300ft.lb of motor torque we have around 3000ft.lb at the wheels at launch. That's a heck of lot for a 300HP car. Considering the typical 4:1 first gear ratio on a 300HP ICE car it would be making only 1200 ft.lb. Obviously adjusting for wheel radius gives us the more meaningful thrust figure, but car tyres don't vary that much in size...

The point regarding ICE power at low wheel speeds is very interesting. I wonder where an automatic transmission with its torque converter fits in. They apparently can multiply the torque in such situations, although I'm not sure by what typical ratio. So they ought to have an advantage off the line compared to a manual cog-swapper, but IME they don't.

 

[John in CR]

John you will find its even lower with a torque throttle and a more advanced controller

I believe it, though I probably won't get to find it unless the CA3 does a good enough job imitating it, at least until Sevcon gets some competition to exert downward price pressure. Lebowski thinks his phase current sensing may not work with my 6 phase motors. On the bright side I finally tracked down some bargain sine controllers, and I'm hoping my test units ship next week.

 

[John in CR]

I should have clarified: I wasn't asking if the Tesla had enough torque, just wondering if the motor could cope with a heavy, low speed load without overheating, or excessively draining the battery. This kind of relates the "inefficient speed" argument. If you can tow your buddy home over the Rocky mountains without issue then I'm convinced it does quite nicely without a transmission.

Might as well ask it to operate as a tow truck too. Aren't you setting the sights a bit high? I mean what luxury sedan can pass that test?

 

[liveforphysics]

I should have clarified: I wasn't asking if the Tesla had enough torque, just wondering if the motor could cope with a heavy, low speed load without overheating, or excessively draining the battery. This kind of relates the "inefficient speed" argument.

What you have stated here, is the absolute core of the fallacy. When a motor is making full torque at 0rpm, it's using less power and making LESS heat, even though it's 0% efficient, than when it's making full torque at any non-0rpm, even though the non-0rpm will have a positive efficiency value.

What happens is you pay the exact same copper loss tax in the form of heat (winding resistance * phase current), and this copper loss heat is the TOTAL SUM of your motors losses at this point while it's not turning.

In other words, when you start at 0% efficiency and 0 power output at a stalled rotor (assuming your controller is one that actually reads phase current and sets a limit that it calls your 100% throttle peak torque), then the motor has LESS heat to deal with just sitting with a stalled rotor at peak torque and 0% efficiency than at ANY point where it has rotation on the shaft (and hence non-zero efficiency), because full torque WITH any rotation is both the exact same copper loss, but now it also adds various core losses.

Likewise, for making a given torque, the least power/energy you consume to make that peak torque is ALWAYS at 0rpm, and then the power used goes up from there, BEMF being the biggest driving force, but this is OK, because the increase in voltage(and hence power) needed to drive the motor at a non-zero RPM.

For a given torque, 0rpm is your least battery energy costly way to make that torque, as well as the lowest amount of heating in your motor for that given torque, despite 100% of the power you're dumping into that motor being wasted as heat to generate a torque (because with no rotation you're at 0% efficiency), it's still lower heat in the motor than any speed making that same torque WITH shaft rotation.

These concepts are not intuitive at all to see if you're just looking at an efficiency plot blinding without actually breaking down the losses in a motor and looking at what's really happening. Almost everyone gets this concept wrong, I don't hold it against you guys, and I appreciate the good debate.


I think of all the posts in this whole thread, Chalo's was perhaps the only one that was perfectly factually correct and rational.

 

[John in CR]

I understand what you're saying Luke, but I can't reconcile it with results using my motors with the series?parallel switching on the windings. In low speed (series), the motor generates less heat with slightly better acceleration than in high speed mode. I know it's not a gearing change, but the BEMF does go up quicker, so shouldn't losses, or is this apple and oranges because it's essentially a differently wound motor and more efficient while still at the low rpms, enough to offset the greater battery current?

edit- or is it the result of non-phase current limiting controllers?

 

[John in CR]

That analysis also doesn't consider time. With the lower gearing of a true gear reduction you'll either have the greater losses for less time or lower current for the same time.

 

[Hillhater]

I think i found a graph that shows what Luke has been saying about low battery current at low rpm on an AC drive.

 

[liveforphysics]

Precisely.

Full motor torque for the cost of ~24A of copper loss on a 600A system.

This is why its a huge fallacy when you guys talk about needing more battery to do direct drive etc.

 

[Hillhater]

Yep, crossed wires there Luke,.. i still had in mind the Dc brushed motors ( Warp, etc) from when we were discussing in the drag race angle, since most of the EV drag guys are using brushed motors.

Luke, any idea why the torque and power curves die at 3000rpm, whilst the amps are still increasing ??
..or is that just a setting on the controller ?

 

[Arlo1]

Yep, crossed wires there Luke,.. i still had in mind the Dc brushed motors ( Warp, etc) from when we were discussing in the drag race angle, since most of the EV drag guys are using brushed motors.

Luke, any idea why the torque and power curves die at 3000rpm, whilst the amps are still increasing ??
..or is that just a setting on the controller ?

I'm studying AC induction motors right now because I just scored one for free. Might use it for a CAR conversion.

The RPM is controlled by the frequency when there is no load on the motor.
The Torque is limited by the amount of phase amps you can feed the motor.
The HP is dependent on the amount of voltage on the Battery side combined with the torque curve.
Higher voltage will make the torque extend to a higher rpm making more HP!

One thing I need to point out is AC induction motors have a flat efficiency curve for a range of rpm. I found a graph that showed the same efficiency from ~3000 rpm -6000rpm
Its realy cool. Although it seems AC induction is not as power dense as PMAC (BLDC) its still able to make crazy power and efficiency when done right!

 

[Hillhater]

Its realy cool. Although it seems AC induction is not as power dense as PMAC (BLDC) its still able to make crazy power and efficiency when done right!

Yes, there is a reason why the drag guys are still using brushed DC..
They pick a 200lb, 50 kW motor frame, mod it, feed it 400v and 2000 + amps ...and bingo ! ..800kW and stump pulling torque.
...and bolt on another one if you need more speed !
Efficiency is not high on their list .

I guess its not quite that easy with AC.

 

[Arlo1]

Its realy cool. Although it seems AC induction is not as power dense as PMAC (BLDC) its still able to make crazy power and efficiency when done right!

Yes, there is a reason why the drag guys are still using brushed DC..
They pick a 200lb, 50 kW motor frame, mod it, feed it 400v and 2000 + amps ...and bingo ! ..800kW and stump pulling torque.
...and bolt on another one if you need more speed !
Efficiency is not high on their list .

I guess its not quite that easy with AC.

Controllers are harder/more expensive to build for hi HP AC motors

 

[Hillhater]

Tesla power/torque /rpm graph ( Roadster motor )
Shame they dont show the battery amps !
And again there is a sudden drop off of torque at 4750 rpm ??

 

[motomoto]

So to conclude how gearboxes aren't awesome I think Luke should show some video of how a Zero motorcycle with no gearbox can
beats a 450 with a stupid clutch and gearbox in a drag race. I saw the video where Trevor was killing it at a hare scrambles and it looked
like he killed um at the start, but I would like to see what happens out of a gate. A quarter mile comparison with slick tires would
tell the whole picture.

 

[flathill]

what you can conclude is magnetic gearing is aws
99.9666%
all i know is i know no thing
I

 

[Chalo]

The knee in the torque curve is probably the point at which back EMF rises high enough to limit the motor's ability to accept as much current as the battery and controller can supply.

 

[toolman2]

You know what happens if you take the same motor, and you hook it to a transmission, and you start out in 6th gear vs starting in 1st gear right? If it has the torque to chug the load you've asking in 6th gear, then 6th gear will almost always be the lowest amount of energy to make that torque at the wheel. Once you start getting RPM, you have to pay for that torque with increased BEMF rather than merely paying the copper losses (which are often smaller than you might think), and the power you're drawing adds up very quickly as you accelerate.

umm, that is not correct im afraid luke..

i mean you know 1st gear will ALLWAYS make torque off the line more efficiently than 6th, surely?
in 6th you are right up in the death zone for phase currents with the squared heat losses getting wildly out of hand, meanwhile the "cost" or "loss" of a few rpm would be minuscule in comparison, and theres the straight out geared lower torque advantage of 1st. makeing it piss easy on the motor with 4 times less ir2 losses by only having to provide half the torque.

-as this may well be key difference in thinking, and everyone is talking straight past each other, this one may need a direct answer

and yep, we get the amps per NM thing and this IS the std (and constant) price for torque, but the WATTS lost per NM output are squared, and the huge controller current multiplication spoken of is helpful, but can hide some quite large losses from the view of our CA's!

its not that you forgot that its actually winding resistance* phase current* phase current, not what you just wrote below?

What you have stated here, is the absolute core of the fallacy. When a motor is making full torque at 0rpm, it's using less power and making LESS heat, even though it's 0% efficient, than when it's making full torque at any non-0rpm, even though the non-0rpm will have a positive efficiency value.

What happens is you pay the exact same copper loss tax in the form of heat (winding resistance * phase current), and this copper loss heat is the TOTAL SUM of your motors losses at this point while it's not turning.

-again as others have said, we need some real numbers and loads to work with, and specs of some (say 10kg ish) bldc motors that we can actually buy, any further specs of the motors you have spoken of would be great luke, or we'll find some others and post them up, this has to help everyones understanding more than another 8 pages of..?

-and im fully clear that theres little need for gears on both the 14 000rpm tesla and the deathbike, cos the motor makes up about a third the weight of the whole bike

 

[liveforphysics]

Toolman- If you've got a motor and controller, and you're going to pin the throttle, you do get that anything with shorter gearing will be turning faster sooner and hence able to draw more power from the battery at any given speed where both bikes are identical motors and controllers right?

This is the mechanism in which you're able to accelerate faster, you're drawing more power off the battery sooner (from a start both draw identical current), but whatever is shorter geared will develop more motor RPM sooner and hence deliver better performance because its drawing and outputting a larger amount of power sooner. In fact, if one was geared 1:1 and the other 2:1, you would be able to draw the power off the battery at 25mph that the other motor isn't capable of drawing until 50mph, hence its not enabling the ability to draw the higher power levels until later speeds. This is why if you want to max your range, putting on the longest gearing possible that still enables the bike to get around, you grow range through not enabling the system to draw as much from the battery, your system outputs lower average power and uses lower average power (because its no longer even capable of drawing the same power because it cant get the motor fast enough to do it).

This is always true for a given motor and controller under the assumption the user is going to be operating the bike WOT when they wish to accelerate in either example. If in one example the rider only want to be 25% throttle or whatever while accelerating, then the math can get different and balancing core losses with copper loss and using lighter throttle can have different results, including the shorter geared bike drawing less power for a given amount of power output to power output, but IMHO, if the bike doesn't just flip over when you pin it WOT, nobody is going to do anything less than just pinning in WOT at every moment they want to accelerate, and in this case, you're ALWAYS drawing less power with the taller geared bike, as its not even possible for it to draw the same power because its RPM is lower.