Gears used in this electric supercar...please explain.

[modern_messiah]

Not sure if people are aware of this but here is a quite sexy looking electric supercar doing it's thing: http://youtu.be/2jIcWBe2_4w

Now the thing that gets me is (and admittedly I never have had the best grasp on gears in electric vehicles - but this is something I am attempting to fix) why/how are they using gears? I note the time between each shift is quite small. This is obvious given the fast RPM build up electric motors tend to have. The best I can think of is roughly as such (this is borrowed and summarised from elsewhere):

With an electric motor the torque is constant, from 0rpm and up, for as long as you can provide the same current to the motor. Power builds like a gas engine so RPM is an integral part of power regardless of the type of motor/engine you are using. The main difference however is that electric motors are more efficient at higher rpm whereas ICEs are generally the opposite. So for efficient cruising in an electric car you want to have the motor run around the mid point or slightly above in the RPM range. So you could theoretically have a gear box setup where the first gear is the required ratio for the best launch, and a few more gears that are each designed to keep the motor ticking over at its most efficient RPM for a given speed.

Now my own thinking leads me to believe that you could possibly also use gears to allow you to use a high torque, low speed motor in a car as well. This sort of motor would give you excellent low speed control and power, and then via the use of gearing you could 'up' the motor RPM to allow for greater road speed.

Or I'm full of sh*t. I'm still a newb to a very large portion of EV/electric motor theory, so please forgive me if I'm talking out my bum.

The insight of people with a greater deal of knowledge than me would be very interesting to read.

Cheers.

- Matt

 

[jonescg]

I've always thought that a three speed gearbox on an electric motorbike would be a formidable weapon. If we get good drive from 1/5th of the max rpm all the way to the top, there isn't much need for a reduction. So a close ratio box with a few cogs would work fine, and take up less room / weight?

 

[Hillhater]

No matter how much torque a motor produces, a gearbox will multiply that torque at the wheels.
More torque = more acceleration
Then of course changing to a higher gear will allow a higher top speed and more efficient cruising.

 

[modern_messiah]

If we get good drive from 1/5th of the max rpm all the way to the top, there isn't much need for a reduction. So a close ratio box with a few cogs would work fine, and take up less room / weight?

Excuse my ignorance, but can you elaborate? Do you mean no gears that reduce the ratio - only ones that increase it for higher top end speed? This assumes the motor produces enough torque for good take off and acceleration.

No matter how much torque a motor produces, a gearbox will multiply that torque at the wheels.
More torque = more acceleration
Then of course changing to a higher gear will allow a higher top speed and more efficient cruising.

So how would you go about applying a gear box to maximize this?

 

[MikeB]

With an electric motor the torque is constant, from 0rpm and up, for as long as you can provide the same current to the motor.

I'm not an expert, but I've followed several discussions on the issue.

The second part of your sentence is one key issue: "as long as you can provide enough current to the motor". My scooter has weaker acceleration at very slow speeds because it's running into the max current limits imposed by the controller and motor. High current means higher heating and electrical load, so you need heftier components. There are times when allocating weight/cost to a gearbox is a better trade-off than allocating weight/cost to heftier components.

The efficiency curve of an electric motor is also not flat, you can get significant reductions in efficiency at very low speeds. Again, a gearbox can be used to keep the electric motor closer to it's optimal efficiency range, and thus you get better range out of a given set of batteries. Again, it's a trade-off between gearbox cost/weight and battery cost/weight.

So, it's really a function of what characteristics you are trying to optimize for, and what components you can use. Sometimes a gearbox is a good idea, sometimes it's better to throw that money and weight into your other components. And it's also a function of availability: an e-vehicle has different needs from it's transmission than a ICE vehicle, so finding an optimal gearbox for a good price is difficult.

 

[Pure]

I agree with pretty much everything said, but wanted to add..

Damn that's the first electric car that I've heard that actually sounded sexy!

 

[REdiculous]

Look at what Tesla Motors did with the Roadster. It had a 2 speed but apparently it wasn't reliable enough so they locked it in second. They were originally aiming for 4sec 0-60mph but locking the transmission in 2nd resulted in 0-60mph in 5.7sec....not much difference.

Later they moved to a single-speed and got the 0-60 time down to 3.9sec. They improved motor cooling, tweaked drive electronics and improved the motor's rpm limit to keep the top speed and improved torque at the same time.

 

[grindz145]

I always say that I think we'll look back on the times when we didn't use gearboxes and thinking how archaic it was. We don't NEED gearboxes, but we will want them once the EVs get more and more refined. It will be a nice to have, and right now it is an unnecessary complexity.

Ask Thud what he thinks about gearboxes

 

[amberwolf]

Regarding the use of or lack of gearboxes in large EVs, and why they are or aren't used in particular ones, you might want to dig thru various build and discussion threads at DIY Electric Car forums as well. It is a question often brought up, as well as whether to use a standard or automatic.

 

[liveforphysics]

The answer is none of the above.

This is an off-the-shelf chassis with rear suspension setup for a 6spd sequential trans-axle. They bolted a motor up to it to make the conversion as simple as possible.

The time spent shifting is time you're not applying throttle. Almost every EV can benefit from a 2spd, though seldom is it worth the expense and complexity to folks. Some EVs with a huge speed range can benefit from a 3spd. This EV is suffering from it's 6spd being a sequential, which forces the driver to shift through each gear one at a time. If it were an H-pattern 6spd, he could simply shift 2nd to 4th to 6th or something similar (which is incidentally how I drive my civic racecar on the street, start in 2nd, go straight to 4th or 6th.)

 

[modern_messiah]

I agree with pretty much everything said, but wanted to add..

Damn that's the first electric car that I've heard that actually sounded sexy!

Agreed.

If it were an H-pattern 6spd, he could simply shift 2nd to 4th to 6th or something similar

So that is why he's shifting so quickly - to get to his desired gear. The only reason he pauses briefly between each shift is to minimise the time the car is not accelerating for (or rather stretching it out over a longer period). Hmm

I will have to go through the more detailed build threads because even if a conversion can benefit from a gear box I want to know about it. A car without manual transmission is unfortunately a bit boring to me :-(

 

[spinningmagnets]

If gears were NOT available in the world, it would be very expensive to create the infrastructure to produce them. As it is now, it is fairly easy and affordable (on a manufacturer scale) to specify a two or three-speed transmission. Thud did a great job of showing how a 2-speed could be made at the garage-level.

If a motor and system voltage is capable of providing your desired top-speed, then when the EV is in the lower half of its RPM range, it will be drawing more amps (creating heat), than when it approaches cruise. Of course, LFP is right about the sequential transmission in this case, as that is one of his areas of expertise. But FrankG converted a motorcycle over to electric while keeping the stock transmission, and he found it was quite beneficial to use two of the available gears (5-speed?).

The issue of whether to add or leave-out gears when using an electric motor, depends on your goals. For frequent stop-and-go driving (some E-bikes), much time is spend accelerating from zero. A two-speed would reduce motor and controller heat, and it would also ease the C-rate stress on the battery pack. Not an issue with LiPo, but there are still a LOT of LiFePO4 and LiMn packs being sold out there.

Non-hub drives, such as Miles, RWP, Ypedal, and GGoodrum, give the motor some gears, and they seem pretty happy with the low-heat and good range per aH of battery that they are getting. For the RAW PERFORMANCE and light weight of a race-bike, gears may not be worth the added bulk, weight, or complexity...it may be better to just have a LiPo pack with a big motor, and a simple heavy-duty + reliable single-stage chain (like Lukes race bikes?...)

 

[liveforphysics]

Motor heating is going to be exclusively a function of it's efficiency x input power. At the point of peak efficiency, it makes no difference whatever phase current may happen to be, it's going to yield the absolute best use of battery watt-hours to satisfy a mechanical power output need.

Peak efficiency always happens at the point at which copper resistive losses match all other combined motor system losses. This means for some motors peak efficiency happens quite soon and then tapers off with additional RPM, and the inverse is true for other motors.

 

[modern_messiah]

Please forgive me lack of education but...

For frequent stop-and-go driving (some E-bikes), much time is spend accelerating from zero. A two-speed would reduce motor and controller heat, and it would also ease the C-rate stress on the battery pack.

Would the 2 speed reduce heat and C-rate stress because you use a higher reduction to take off so the motor does not have to work as hard to get the vehicle moving (less amps drawn), and then switch to a lower reduction to get the performance up?

Peak efficiency always happens at the point at which copper resistive losses match all other combined motor system losses. This means for some motors peak efficiency happens quite soon and then tapers off with additional RPM, and the inverse is true for other motors.

So basically if you were to use a gear box in an electric vehicle you would do so in order to maximise the work the motor does at it's peak efficiency, for example a higher cruise speed at the motors ideal RPM than if you were to use direct drive.

 

[spinningmagnets]

Would the 2-speed reduce heat and C-rate stress because you use a higher reduction to take off so the motor does not have to work as hard to get the vehicle moving (less amps drawn), and then switch to a lower reduction to get the performance up?

Yes, that is what I meant. For high performance, it may be possible for a given design to have a motor large enough that the motor is not the limiting factor, and only gets warm under normal use, but not hot. However, many people here want to maximize performance from a smaller motor, rather than upsize it. Lots of melted motors and controllers here at ES, some purely from heat alone.

Since electric motors have their full torque available from 1-RPM, I'm not sure if a 2 or 3-speed would help much in a performance design (a bigger motor with one speed may be better), LFP would be a better judge of that. I doubt a 2-speed would provide a much benefit in a light 20-MPH E-bike. I believe the one vehicle that most needs a 2-speed is a 70-MPH capable motorcycle, which is much heavier and usually wants more range from a very size-limited battery pack.

A 2-speed planetary could be built into the rear wheel of a non-hub E-motorcycle (or inline with the motor-shaft to reduce un-sprung weight), but I haven't seen one yet.

 

[amberwolf]

Motor heating is going to be exclusively a function of it's efficiency x input power. At the point of peak efficiency, it makes no difference whatever phase current may happen to be, it's going to yield the absolute best use of battery watt-hours to satisfy a mechanical power output need.

Peak efficiency always happens at the point at which copper resistive losses match all other combined motor system losses. This means for some motors peak efficiency happens quite soon and then tapers off with additional RPM, and the inverse is true for other motors.

So if you design the motor around it's intended use, or properly gear it to the speed range it will be used in, it should work fine without a multi-speed transmission?

But if you need a very wide range of speed, going up and down quite far in that range very often, especially to zero, does it still make sense to use only direct drive, and not a multi-speed transmission?


If you have a "random" motor that you don't know it's peak efficiency point (and either can't or aren't going to test it to find out), then to get the best out of it would it still make sense to use the multi-speed?

 

[modern_messiah]

Thanks for the info spinningmagnets. I guess what I'm trying to deduce here is why the supercar in the video in my original post is using a gear box.

We've determined that the sequential in use has wasted/useless gears but they clearly have a target gear in mind when shifting. So whether that gives them a higher top speed or not I'm unsure. Maybe they use one gear to maximise the acceleration out of corners and then switch to a lower reduction to hit the top speed as the first gear in use may severely limit the top end speed of the car, but aid the acceleration to a particular point. I say this because in the video he is clearly coming out of a corner when the in car footage starts, waits a few second and then switches up as fast as possible while moving in a straight line. Or maybe I'm reading into it too much?

Thoughts?

EDIT: And in response to amberwolf I guess the type of application you would use a multi-speed setup in would be...a track race car. Constant change of speed. Depending on the track of course but you catch my drift....pun not intended.

 

[Hillhater]

Any where you want to improve the performance of an EV,..acceleration, speed, efficiency ...will benefit from a range of gearing to optimize the torque and motor speed to suit the varying demands of daily driving.

 

[McDesign]

I've wondered this myself, and will be able to actually experiment.

I'm converting an old Accord to an EV, using the stock transaxle and clutch. I have a G-analyst, so I can easily graph acceleration.

I'll try acceleration runs with the car just in third gear (~1:1), and then compare that with runs rowing through the gears in the conventional fashion.

I'm interested to see what the reality is, especially without the advantage of modern millisecond-quick twin-shaft transmission shifting.

Forrest

 

[liveforphysics]

The reason is very simple and all ready covered.

The transaxle in that chassis has the suspension bolt right onto it. They made this as simple of a swap as possible just bolting a motor where the engine goes with as little re-work as needed.

 

[Hillhater]

The reason is very simple and all ready covered.

The transaxle in that chassis has the suspension bolt right onto it. They made this as simple of a swap as possible just bolting a motor where the engine goes with as little re-work as needed.

but... if it performed better using only one gear ( or two ) they would simply select a single ratio , or pair of ratios , and stick to them.
They have obviously tested and found that it performs better using all or most of the ratios.

 

[McDesign]

That's what I want to find out first-hand, with a car more like what we all might use!

Forrest