Yes, I've been imagining this too since 25 years ago now, also thinking how it would not need turning wheels, just power differential like paddling a boat. Not sure if this part can be done though or if it would break under the pressure... I should have studied engineering to find out.
This is how tanks steer! And also (to some extent) why using them on roads ruins the roads. If you tried in a traditionally wheeled car then all of the drive shafts would snap though!
I used to think that the treads actually ruin the road because it changes to contact surface on which you apply the power differential from a line (on a tire) to a rectangle(the tread), i.e. it tries to rotate a flat surface which can't be rotated, so somethings got to give. Arguably a tire wouldn't have the same problem.
No turning wheels means no parallel parking. Zero turning radius would be cool, but even with complete differential steering that would scrub the tires a bit much. I don't think you could practically remove steering tires entirely. I think you're still stuck with a rack and pinion, ball joints, tie rods, strut bearings, etc so even with AWD with independent electric motors on all wheels differential steering isn't really that useful. On a performance car, it'll help with handling around turns. I don't see that becoming commonplace on average cars because of the added cost.
In addition to the sls this idea is deployed to some extent in IC cars as well: electronic limited slip differentials can send continuously varying torque to each wheel, see the Porsche 918 spyder (combo electric and IC) or the Nissan GT-R which has a famously odd driving characteristic as a result of the computer controlled torque vectoring.
The article mentions it, but this is very similar to the setup in the Honda Accord Hybrids. Many spec sheets list them as "CVT" or "e-CVT" but those are all wrong -- there's no normal or CVT transmission at all. Instead the car is motivated by an electric motor that can close a clutch plate (in the Koenigsegg it's a hydraulic system probably to handle the greater HP and torque) and engage the ICE engine. When the ICE isn't being used to help out the electric motor it runs basically whenever it wants to and charges the batteries.
A prius works in the same way, correct? In some ways this is a simpler system than conventional ICE-only drivetrains, I think this should have been pushed downmarket even faster than it has been. For people without the ability to charge (e.g. in Switzerland the vast majority lives in rentals without charging ports in the garage) this seems like the best solution so far.
The Prius system is actually much more complicated involving a set of planetary gears to couple torque in and out of the drivetrain. The Koenigsegg system doesn't use the ICE below 30mph (because that would stall the engine at its fixed ratio) where the Prius can.
A modified Koenigsegg with ratios adjusted for cruise at the local speed limit (and a lot less power!) sounds like it would work quite well.
The Prius eCVT is just like a traditional planetary gearbox, but instead of having a complex set of auxiliary gears on clutches which need to be closed in pairs to vary the speed of the planet gears, thus the output shaft, it uses electric motors to spin the planetary gears at exactly the speed needed to obtain the desired output speed.
The Prius CVT is absolutely genius in its simplicity.
TLDR: At low speeds, such a car is basically an EV with a petrol power generator and some power control circuitry which chooses to alternate between charging the battery or powering the electric motor directly. Under heavy acceleration and after a certain speed threshold (enough for the petrol engine to utilise the fixed final drive ratio), a clutch is engaged and then mechanical power is sent directly to the motor, then to the wheels.
I want to see the dyno chart. Either there are some serious power spikes/valleys with this setup, or they are nerfing (ie reducing) power at various points to smooth things. There is no way direct-drive IC engine makes anywhere near similar power at 100kph as at 200 (double the rmp). This may be fast, but a transmission would make it much faster.
(A modern trans, at this power/price, would only be 200lbs at most. Less than carrying a passenger.)
Why add 200 lbs to a performance car? You may reduce acceleration rates (including braking and cornering), which is kind of the point in a car like this.
The clutch is a hydraulic coupling, meaning that it acts pretty much like a torque converter. They claim that it stays "locked" most of the time (at constant speed), but the amount of slippage during heavy acceleration must be considerable. Looks like far from an ideal setup in terms of drivetrain efficiency.
Locked means a regular, slip-free friction clutch housed inside the torque converter. Unless that clutch slips (which would be a mechanical failure), then there will be no slippage during heavy acceleration.
The torque converter is most likely to handle a bit of slippage at low speeds, or just to provide a smoother activation than slamming a clutch. It's also supposedly a low-slip converter, unlike the normal "comfort" type.
I think it's literally just there to allow the engine to idle when the car's stationary. Otherwise it'd stall since it's direct drive gearing.
With a redline at 8250rpm at 249mph, that gives you ~600rpm (which many V8s can idle comfortably at) at 20mph and ~1500rpm at 45mph. I'd expect it to lock somewhere around 45mph and be effectively a rigid connection above that.
Edit: From TFA the hydraulic coupling starts to engage around 30mph, so not a bad guess. :)
A torque converter a type of liquid transmission. At low speeds it converts most of the slippage to torque to allow the IC engine to deliver power without stalling. A hydraulic coupling just slips and the extra power ends up as heat.
I think the point in this car is that they have enough torque to spin the tires at any of those speeds where they've had to "nerf it" to make things smooth. Once you're at that point might as well take advantage of it to simplify. A transmission wouldn't add anything except maybe some efficiency which they don't care much about given the application.
It basically sounds like the car's "stuck in 5th" with electric boost to make up the shortfall in torque when the motor's at low RPM. They're not getting maximum engine power out of it at 100km/h, you're right there, but they don't need to because the electric motors make up for it when accelerating.
There shouldn't be any spikes, really. Much less so than with a transmission, which has peaks for every gear.
The car would basically have the following stages:
1. All electric. ICE is just powering a generator.
2. 50km/h+: ICE clutches in at 1000 RPM. Not much power provided.
3. 100km/h+: ICE starts delivers at 2000 RPM.
4. 250km/h+: Peak torque (assuming sweet spot of around 5000 RPM, just a guess))
5. 400km/h+: Redline at 8250 RPM, nothing left.
You'll probably have two torque peaks: Initial electrical acceleration and its drop-off, and ICE peak and drop-off. Depending on how well-tuned this is, it might be smooth between those two peaks.
A trans, on the other hand, would have lows, peaks and shift losses between every gear change, and that's combined with a higher permanent loss.
However, arguably, you could have made a more impressive car while saving weight: By removing the ICE. You'd probably lose a bit of top speed, but make up for it in acceleration. 400km/h is useless.
Probably a fine setup for optimising straight-line acceleration and 0-400 times, with questionable track performance. A conventional high-performance car is able to make use of the engine torque at any time by shifting to the appropriate gear and keeping the engine in a narrow rev-range. This, on the other hand relies solely on the electric motors to fill that gap before the torque peak at, say, 250 km/h.
A conventional high-performance car will struggle with its lacking acceleration and narrow optimal rev-range. Every shift is a performance penalty, and the transmission itself is inefficient. It's only really acceptable for straight-line 0-400, where the shift range is only gone through once, and even then it's often too much, with people preferring two-speed transmissions for drag. For a real race, it becomes extremely wasteful.
The only benefit of a conventional setup is the ability to handle extreme top-speeds that direct-drive will have difficulty with. However, in most races you're quite far from 400km/h, making this benefit useless. Instead, you constantly need immediate acceleration and power from variable speeds, which an ICE with transmission will always struggle with.
While this particular car is not that impressive performance-wise, it's tuned for racing, not top-speed. An purely electric racer will run circles around it, with their only downside being tendencies to battery overheating for certain races. On that note, see VW's recent pike's peak record: https://electrek.co/2018/06/21/vw-all-electric-race-car-fast....
A conventional rig has spikes for each gear, one for each ratio as the engine revs through its power band. Sticking to only one gear ratio only consolidates those spikes into one massive mountain of a spike.
This would be true for if there was only one type of motor involved.
For this car, there will likely be two spikes very far apart, with ICE picking up as electric is dropping off, effectively smoothing out the valley between the two peaks.
Unlike combustion engines, electric motors have much higher peak power output compared to steady state maximum power. (ie. they are limited by cooling, both of the motors and controller). Typically the pulse power for 1 second might be 10x the steady state output!
That means, for the typical use case of accelerating from 0 to 250 mph, they can afford to use the electric engines in 'peak' mode rather than steady state mode, then let them cool down when the IC engine is engaged at higher speeds.
Attach the system to a dyno and require it to run at a particular speed for an extended period of time and it's going to look very disappointing though.
Is accelerating from 0 to 250 a typical use case? What about steady speeds, or repeated acceleration between speeds? As i said, they will have nerfed this thing. It would be a better car by adding a tramsmission. Performance should not be abandoned just to make a point.
The ICE in this is a 5.0L, twin-turbocharged monster that produces 1100hp and 950ft.lbs of torque. The electric motors really just need to carry this thing to about 30mph before the RPMs are high enough for the ICE to propel this thing with some authority.
900ft.lbs of torque at 3,000 RPMS is still 515 hp.
The combined HP of the car is actually around 1,800, but they've overlapped the curves to peak at 1,500, you never get the full combined power of both ICE and electric, in order to make it smoother.
Thanks! The TLDR seems to be that there's an input turbine and an output turbine. (I probably should've seen that coming.)
The modern replacement, torque converters, seem to be essentially the same but with some magical modification of fluid flow in the middle: https://en.wikipedia.org/wiki/Torque_converter
It's a torque converter, like on a conventional automatic gearbox. Just obviously built to handle significantly more power and better cooled to handle the large amount of slipping required for low-speed driving.
It doesn't actually take much of a torque converter to handle 1000+hp.
A stock style converter from a C6 or TH400 (half century old designs) should do the trick as long as you can cool them. Having a lockup converter that can deal with a 1000+hp is a whole different story.
>It also has a "battery drain mode." Basically, when you are ending your journey, you will tell the Regera where you will be parking the car, at home or a charging station. The Regera will then make sure that the battery is fully drained before it gets to the end point, which means it'll be ready to accept a charge. Very smart. It has a fully electric mode of about 22 miles.
I thought this was the worst thing you could do to lithium cells?!
"Fully drained" for the system probably doesn't mean "fully drained" at the cell level, since (as you say) there is a point below which draining cells damages them. The reported "0%" line will be some way above that.
On second thoughts, I'm guessing that what this ACTUALLY does is make sure that the petrol engine doesn't waste fuel recharging the battery unnecessarily at the end of the trip when it's going to get plugged in.
I like that idea - it means if you are doing less than 22 miles you can run it as a plug in hybrid and not use fuel. I used to have an old high performance sports and the fuel use for going down the shops etc was annoying.
I think it's pretty telling as to what sort of person I am, that I initially read the title as "How the 1,500 Hit Point Koenigsegg Regara" rather than "How the 1,500 Horse Power Koenigsegg Regara".
I believe the rotary engine will be relegated to range-extending electricity generation for an otherwise pure EV in Mazda's setup. Much like the BMW i3 has a small generator option for longer range.
65 comments
[ 0.21 ms ] story [ 118 ms ] threadA complete real time tweaking of per wheel outputs based on live slip/velocity/momentum for perfect AWD.
Full AWD Torque vectoring.
Rimac Concept One (the old model):
- Acceleration: https://www.youtube.com/watch?v=eT7KKxoAvvk
- Presentation at Stanford: https://www.youtube.com/watch?v=1FsrgJ66wUw
Rimac Concept Two (the new model, they plan to build ~150 of them):
- Geneva motors show: https://www.youtube.com/watch?v=XvjvYzzBh-k
- Company website: http://www.rimac-automobili.com/en/
I would like to see a serious off-road EV.
This review from 2014 has a great explanation: https://www.thetruthaboutcars.com/2014/02/review-2014-honda-...
A modified Koenigsegg with ratios adjusted for cruise at the local speed limit (and a lot less power!) sounds like it would work quite well.
https://www.youtube.com/watch?v=TRSQEgLk_2M&index=40&list=PL...
The Prius eCVT is just like a traditional planetary gearbox, but instead of having a complex set of auxiliary gears on clutches which need to be closed in pairs to vary the speed of the planet gears, thus the output shaft, it uses electric motors to spin the planetary gears at exactly the speed needed to obtain the desired output speed.
The Prius CVT is absolutely genius in its simplicity.
Here's a good video on how a traditional planetary automatic operates: https://www.youtube.com/watch?v=Ugao6jTyM7k
(A modern trans, at this power/price, would only be 200lbs at most. Less than carrying a passenger.)
The torque converter is most likely to handle a bit of slippage at low speeds, or just to provide a smoother activation than slamming a clutch. It's also supposedly a low-slip converter, unlike the normal "comfort" type.
With a redline at 8250rpm at 249mph, that gives you ~600rpm (which many V8s can idle comfortably at) at 20mph and ~1500rpm at 45mph. I'd expect it to lock somewhere around 45mph and be effectively a rigid connection above that.
Edit: From TFA the hydraulic coupling starts to engage around 30mph, so not a bad guess. :)
The car would basically have the following stages:
1. All electric. ICE is just powering a generator. 2. 50km/h+: ICE clutches in at 1000 RPM. Not much power provided. 3. 100km/h+: ICE starts delivers at 2000 RPM. 4. 250km/h+: Peak torque (assuming sweet spot of around 5000 RPM, just a guess)) 5. 400km/h+: Redline at 8250 RPM, nothing left.
You'll probably have two torque peaks: Initial electrical acceleration and its drop-off, and ICE peak and drop-off. Depending on how well-tuned this is, it might be smooth between those two peaks.
A trans, on the other hand, would have lows, peaks and shift losses between every gear change, and that's combined with a higher permanent loss.
However, arguably, you could have made a more impressive car while saving weight: By removing the ICE. You'd probably lose a bit of top speed, but make up for it in acceleration. 400km/h is useless.
Sometimes I hate HN comment syntax.
A conventional high-performance car will struggle with its lacking acceleration and narrow optimal rev-range. Every shift is a performance penalty, and the transmission itself is inefficient. It's only really acceptable for straight-line 0-400, where the shift range is only gone through once, and even then it's often too much, with people preferring two-speed transmissions for drag. For a real race, it becomes extremely wasteful.
The only benefit of a conventional setup is the ability to handle extreme top-speeds that direct-drive will have difficulty with. However, in most races you're quite far from 400km/h, making this benefit useless. Instead, you constantly need immediate acceleration and power from variable speeds, which an ICE with transmission will always struggle with.
While this particular car is not that impressive performance-wise, it's tuned for racing, not top-speed. An purely electric racer will run circles around it, with their only downside being tendencies to battery overheating for certain races. On that note, see VW's recent pike's peak record: https://electrek.co/2018/06/21/vw-all-electric-race-car-fast....
But yes, it can be massively improved. IIRC, this car is beat in acceleration by a Tesla Model S P100D, which isn't exactly a supercar.
For this car, there will likely be two spikes very far apart, with ICE picking up as electric is dropping off, effectively smoothing out the valley between the two peaks.
Unlike combustion engines, electric motors have much higher peak power output compared to steady state maximum power. (ie. they are limited by cooling, both of the motors and controller). Typically the pulse power for 1 second might be 10x the steady state output!
That means, for the typical use case of accelerating from 0 to 250 mph, they can afford to use the electric engines in 'peak' mode rather than steady state mode, then let them cool down when the IC engine is engaged at higher speeds.
Attach the system to a dyno and require it to run at a particular speed for an extended period of time and it's going to look very disappointing though.
900ft.lbs of torque at 3,000 RPMS is still 515 hp.
https://youtu.be/4jhB9Yx0hyY
The modern replacement, torque converters, seem to be essentially the same but with some magical modification of fluid flow in the middle: https://en.wikipedia.org/wiki/Torque_converter
A stock style converter from a C6 or TH400 (half century old designs) should do the trick as long as you can cool them. Having a lockup converter that can deal with a 1000+hp is a whole different story.
I thought this was the worst thing you could do to lithium cells?!