Jump to content


Photo
- - - - -

Holden Time Attack Concept.


  • Please log in to reply
54 replies to this topic

#1 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 11 October 2018 - 22:04

Seems to have most of the features it should.

 

https://newatlas.com..._term=Read more

 

Video.

 

https://www.youtube....3&v=IAvvbaQ70YU



Advertisement

#2 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 11 October 2018 - 23:15

Cute. I'd really enjoy an article that compared all these silly show cars and renders with what actually emerged at the end as useful product.



#3 thegforcemaybewithyou

thegforcemaybewithyou
  • Member

  • 2,300 posts
  • Joined: April 12

Posted 12 October 2018 - 09:11

I must say I'm a fan of the active downforce fans, hopefully we see something similar on the 2021 F1 cars. Maybe not the rotating blades as can be seen in the video, but air streams blowing over defined surfaces, driven by a pressure reservoir that can be charged by the engine/MGUK/MGUH/battery.



#4 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 12 October 2018 - 22:33

I was briefly involved with a project to use fans on a roadworthy track car. The numbers are pretty cruel. It's OK with a single seater as the sidepods give you a better volume to play with.



#5 pierrre

pierrre
  • Member

  • 61 posts
  • Joined: June 17

Posted 15 October 2018 - 15:10

saw the cgi kangaroos...they look mighty impressed



#6 desmo

desmo
  • Tech Forum Host

  • 18,209 posts
  • Joined: January 00

Posted 15 October 2018 - 21:44

I like it, a halo supercar should have every bit of FIA-banned tech that can conceivably be put to use. 



#7 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 16 October 2018 - 07:50

That is one part that kinda dissapoints me with Koenigsegg. They are trying so so hard and winning in many areas. But their aero is so tradisional and simple they really lose out on performance by not having the same downforce as the others. They should have huge tunnels that the suspension arms poke through. ala 90s LMP style. like so:

https://upload.wikim...-Turbo_wing.jpg

 

http://www.car-revs-...ach-2014-32.jpg


Edited by MatsNorway, 16 October 2018 - 07:52.


#8 Kelpiecross

Kelpiecross
  • Member

  • 1,364 posts
  • Joined: October 10

Posted 16 October 2018 - 12:22

I realize that Holden's concept is just a design fantasy - even though it is interesting - I don't like it much. If we are talking about an all-out one-lap-attack car it is sadly lacking in imagination. What is needed is the one-lap equivalent of a TF dragster - extreme to the point of being ridiculous. Holden's concept also has odd things like four separate electric motors but apparently the "shifting" is handled by a 3-speed gearbox - one for each motor? Doesn't seem very practical or even make any sense. One big electric motor and one gearbox and two wheel drive (2wd is always faster than 4wd on a good surface) would seem lighter and faster to me. I don't think it would get anywhere near 1:29. Same with the fan-induced downforce - probably better when travelling very slowly or starting from rest but very heavy and complex compared to wings.
A current F1 car complying with current F1 rules probably could do 1:29 around Bathurst - they are usually about 30 secs faster than the V-8 Supercar lap time. And a car using F1 technology but being free of restrictive regulations could be quite a bit faster than 1:29. Remember this is a car designed to do one lap only and it doesn't need all the heavy power recovery gubbins or battery. The power could be two to three times what F1 cars are now - it only has to last about one minute in the same extreme way that a TF dragster's engine only has to last about 4 seconds or only a few hundred revolutions of the engine at full power. Wings and other downforce devices to be whatever is needed, tyres to be whatever size is best, driver aids like ABS, traction control, wings automatically adjusting to whatever is necessary, also acting as air brakes? The driver being advised when to brake etc. (you would need this approaching The Chase at 300MPH or so)- or maybe no driver at all?. And if you are a electric power enthusiast you could use a 3,000 HP motor in the car - could be better in fact for a one-lap car than with an ICE - as recharging problems would not apply.

In a similar manner to Holden's virtual calculations (and probably just as accurate) - my virtual lap time around Bathurst is estimated to be 59 secs. Like a TF dragster on the straights, faster than F1 in the corners is the general idea.

Edited by Kelpiecross, 16 October 2018 - 12:24.


#9 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 16 October 2018 - 22:14

2wd is always faster than 4wd on a good surface

 

Really?



#10 PeterScandlyn

PeterScandlyn
  • Member

  • 291 posts
  • Joined: October 17

Posted 17 October 2018 - 03:13

Amazing thing about all palaver being expended is it's a vehicle fantasized by that Australian badge engineering outfit, Holden.

 

Don't even build their own cars these days, let alone in their own country...... :well:  :)



#11 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 17 October 2018 - 05:44

To be fair they do have a styling studio and some sort of engineering/CAD office, oh and a nice little proving ground.



#12 GreenMachine

GreenMachine
  • Member

  • 1,361 posts
  • Joined: March 04

Posted 17 October 2018 - 09:19

Holden :confused:

 

Oh yes I remember, they used to make cars here didn't they - amazing how quickly the memories fade ...  when (if) the Camaro replaces the Holden in Supercars, there will be nothing to keep the memory refreshed.  I'd think the nameplate will be gone in a few years too.



#13 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 17 October 2018 - 09:49

Well they can't call them Peugeots, and Opel never got much traction here, Daewoo's a poisoned chalice, so Chevrolet or Buick? I think we'll be seeing Holden for a while yet. 



#14 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 17 October 2018 - 22:20

Don't get me started on the subject of Australia getting out of auto manufacturing.  :evil:



#15 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 17 October 2018 - 22:52

Yes, a rather short sighted step, but I guess we never made much money. Every new model I was involved with was a 'bet the company' deal. Having said that, it is always difficult to work out the true profitability of the local branch of a multinational, you can play games all day long with transfer pricing.



#16 Kelpiecross

Kelpiecross
  • Member

  • 1,364 posts
  • Joined: October 10

Posted 18 October 2018 - 11:22

Really?


Yes. Really.
Remember that the car I am proposing is of an F1 type on a dry, warm, smooth tar surface. Not a Land Rover/Landcruiser on wet grass. There has been much work done by most of the F1 builders and quite a few F1 cars actually built investigating the possible advantages of 4wd over 2wd. The unambiguous result appears to be that 2wd produces a faster lap time. There is plenty of information on the internet about these various trials. I suppose all the F1 designers have had the same experience as most of us of trying to get a Landcruiser or etc. moving on wet grass in 2wd without any apparent action, then putting it into 4wd and driving away with no wheel slip at all. The implication to most of us would be that 4wd would be great for an F1 car - but, oddly, it just doesn't seem to work that way.

Another odd thing is that nobody seems to really know just why 4wd is not actually better than 2wd.

So - in contrast to the Holden Electric Abomination - I could hardly propose 4wd for a F1-type of Lap Attack Car.

#17 Kelpiecross

Kelpiecross
  • Member

  • 1,364 posts
  • Joined: October 10

Posted 18 October 2018 - 11:25

Don't get me started on the subject of Australia getting out of auto manufacturing.  :evil:


You mean - getting out of all manufacturing - not just auto.

#18 GreenMachine

GreenMachine
  • Member

  • 1,361 posts
  • Joined: March 04

Posted 18 October 2018 - 20:59

Yes. Really.

... The implication to most of us would be that 4wd would be great for an F1 car - but, oddly, it just doesn't seem to work that way.

Another odd thing is that nobody seems to really know just why 4wd is not actually better than 2wd.


It is pretty obvious really, and I am sure the relevant quotes are there from in-period. Basically 4wd did provide advantages, just not enough to offset the weight penalty (there were probably other issues around complexity, weight distribution , cost, etc).

 

However that was yesterday's 4wd, today's may be a different set of trade-offs and a different outcome, but like autoboxes, six wheels, fan cars, I believe 4wd is not permitted under current rules.



#19 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 18 October 2018 - 22:38

Yes. Really.
Remember that the car I am proposing is of an F1 type on a dry, warm, smooth tar surface. Not a Land Rover/Landcruiser on wet grass. There has been much work done by most of the F1 builders and quite a few F1 cars actually built investigating the possible advantages of 4wd over 2wd. The unambiguous result appears to be that 2wd produces a faster lap time. There is plenty of information on the internet about these various trials. I suppose all the F1 designers have had the same experience as most of us of trying to get a Landcruiser or etc. moving on wet grass in 2wd without any apparent action, then putting it into 4wd and driving away with no wheel slip at all. The implication to most of us would be that 4wd would be great for an F1 car - but, oddly, it just doesn't seem to work that way.

Another odd thing is that nobody seems to really know just why 4wd is not actually better than 2wd.

So - in contrast to the Holden Electric Abomination - I could hardly propose 4wd for a F1-type of Lap Attack Car.

 

Ok now I see where you got that idea. I wouldn't propose 4WD for an F1 car (even if the rules permitted 4WD, there are still many other rules that work against its adoption). I would propose 4WD for an electric time attack car (and yes the ultimate time attack car would be electric).

 

4WD benefits:

 - higher acceleration at traction-limited speeds (which is most speeds in a lightweight 1000kW 2WD car)

 - Higher cornering grip and even higher with 4 wheel torque control, where each tyre can operate at the same angle on the potato plot.

 - Front wheel regen' braking permits a reduction in battery pack weight.



Advertisement

#20 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 19 October 2018 - 01:35

Yes, a rather short sighted step, but I guess we never made much money. Every new model I was involved with was a 'bet the company' deal. Having said that, it is always difficult to work out the true profitability of the local branch of a multinational, you can play games all day long with transfer pricing.

Oh I believe there is a lot more at stake than company profits and government subsidies. Why is (almost?) every car maker around the world, government subsidised?



#21 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 19 October 2018 - 05:02

Because the engineering eco system that grows up around  an auto plant is the foundation for a modern manufacturing economy. IMO. You can't just rent a workshop, buy some crates of parts and start building aircraft that people want to buy. You can do exactly that with cars, using CKD kits. The barriers to entry are therefore fairly low. You aren't going to make big $$ assembling CKD, but as you take baby steps towards local sourcing the eco system grows and away you go. Sure it takes 20 years.



#22 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 19 October 2018 - 21:14

(and yes the ultimate time attack car would be electric).

Oh hell no. Too heavy.

 

You can cram some wings on a "Formula ford" tub made of carbon, shoehorn a 1.5L bike engine or whatever custom of same capacity and get 80s turbo power if you are a car manufacturer. Electric is not touching that kind of power to weight ratio any time soon.

 

Or you take Gordon Murray seriously and you could consider a turbine going full tilt all the time, providing downforce, trust and wheelpower and use a clutch ala. Top fuel or koenigsegg Regera to control it.

 

We are talking 4-500kg and 1000hp+


Edited by MatsNorway, 19 October 2018 - 21:17.


#23 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 20 October 2018 - 03:07

Mats they are not heavy at all. The battery is the heavy part of an EV and time attack cars don't need a big battery. EV topology permits lightweight 4WD (no transmission, final drives, etc) and individual wheel torque control. The Holden engineers got it mostly right.

 

Have a look at the quickest cars in formula student, Pikes Peak etc.

 

eg Formula Student 150kg 0-100 km/hr 1.5s in a car that is designed to go around corners. (and heavily regulated on general safety, electrical safety, max voltage, max power etc)


Edited by gruntguru, 20 October 2018 - 03:09.


#24 Kelpiecross

Kelpiecross
  • Member

  • 1,364 posts
  • Joined: October 10

Posted 20 October 2018 - 05:28

Ok now I see where you got that idea. I wouldn't propose 4WD for an F1 car (even if the rules permitted 4WD, there are still many other rules that work against its adoption). I would propose 4WD for an electric time attack car (and yes the ultimate time attack car would be electric).
 
4WD benefits:
 - higher acceleration at traction-limited speeds (which is most speeds in a lightweight 1000kW 2WD car)
 - Higher cornering grip and even higher with 4 wheel torque control, where each tyre can operate at the same angle on the potato plot.
 - Front wheel regen' braking permits a reduction in battery pack weight.


"Higher cornering grip and even higher with 4 wheel torque control" ? Again you claim this apparently mainly because you just "know" that it's true.
Fundamentally I think 4wd cornering grip is less than 2wd/rwd for the same reason that front wheel drive/2wd only is less than 2wd/rwd - and that is that the front wheels are
at their absolute limit resisting sideways forces in a corner - asking the front tyres to then work even harder in supplying propulsive force is just not going to work - and any attempt to do so will just reduce cornering force. The only way "torque control" is going to help is if it cuts all driving power to the front wheels during cornering - which sort-of negates the purpose of 4wd. This effect became apparent in the testing of the '70s F1 4wd prototypes when they found that progressively reducing the drive to the front wheels resulted in proportionately faster cornering - until the fastest cornering speeds were with no drive to the front at all. In fact some cars had a clutch to disconnect the front drive completely in corners.

I have nothing basically against "electric" for "time attack" if it will do the job weight and power wise. (I think it's ridiculous for general road use because of the recharging problems etc.) Maybe a capacitor would be better for TA than a battery? If this proposed TA car is absolutely "anything goes" possibly something like a hydrogen peroxide/permanganate turbine would be best? These motors can be thousands of BHP for really only a handful of kilos. And about as safe as flying a Me163 - but that is not the point here where the only rule would be - "must be wheel-driven".

Higher grip at traction-limited speeds? I would imagine that all TA laps would be of the "flying" variety not "standing start". The 4wd (and the downforce fans) would probably be better for "standing start" but I would think that after the first hundred metres or so the speeds would such that wings/2wd would have the advantage for traction.

And I suspect you would need a lot more than 1000kW - the '80s turbo cars were more than this.

Edited by Kelpiecross, 20 October 2018 - 05:32.


#25 NotAPineapple

NotAPineapple
  • Member

  • 165 posts
  • Joined: July 13

Posted 20 October 2018 - 12:24

4WD its self won't give you more cornering grip, but if it can include torque vectoring it definately can while simultaneuously giving better corner exit traction.



#26 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 20 October 2018 - 14:55

Mats they are not heavy at all. The battery is the heavy part of an EV and time attack cars don't need a big battery. EV topology permits lightweight 4WD (no transmission, final drives, etc) and individual wheel torque control. The Holden engineers got it mostly right.

 

Have a look at the quickest cars in formula student, Pikes Peak etc.

 

eg Formula Student 150kg 0-100 km/hr 1.5s in a car that is designed to go around corners. (and heavily regulated on general safety, electrical safety, max voltage, max power etc)

 

The VW EV was that fast up Pikes Peak is because it is the first proper manufacturer effort to go as fast as possible.

 

The peugeot was a quickie for the team. Tubeframe and spare parts they had around. And the car has the shape of a hatchback. The VW barely beat the "conservatively" tuned 3.5L engine in the Peugeot. They could do a rerun with the same peugeot, have a experienced driver like Dumas, up the power in it, run alcohol perhaps. and boom they beat that time again.

 

Problem is that tradisional motorsport and their engines have been so tamed for so many years people forget what kind of monsters we could have built if it was not for the dangers of the drivers.

 

And there is a reason VW did not let the Porsche LMP time attack car run at FOS. 

 

For an all out effort there is no reason a manufacturer could not go 400-500kg, 4WD (and it would have to be 4WD to be competitive) and 1000hp-1500hp+ in a IC car. That is just not happening with electric yet. And i know very well the potential a electric car has. They are the fastest RC cars you can get basically. Not even Nitro can touch them over a single lap.

 

For an Electric motor how light can it be pr kW?

 

the KERS was 80kw and 20kg? that is 4kw/kg so if you want 1000kW you get a 250kg motor? assume 200kg?

 

Either way you need a ESC and battery that is probably equal to the motor in weight surely. I believe that is not far off reality for RC cars. How much in weight for the rest of the emty car?  The motor itself does not really need a drivetrain. It could be slapped directly over the driveshafts and be running as a spool to save weight.

 

When you approach the extreme ends of something it becomes bizarre. Just like in math when you are near a asymtote. Top fuel runs 0-500km`t in less than 4 seconds and does it in first gear. An all out time attack car would probably be wildly different than most people imagine.


Edited by MatsNorway, 20 October 2018 - 15:25.


#27 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 20 October 2018 - 16:34

having electric motors on the front wheels would enable better energy recovery... so 4wd does make sense



#28 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 20 October 2018 - 16:37

Problem is that tradisional motorsport and their engines have been so tamed for so many years people forget what kind of monsters we could have built if it was not for the dangers of the drivers.

 

And there is a reason VW did not let the Porsche LMP time attack car run at FOS. 

 

For an all out effort there is no reason a manufacturer could not go 400-500kg, 4WD (and it would have to be 4WD to be competitive) and 1000hp-1500hp+ in a IC car. That is just not happening with electric yet. And i know very well the potential a electric car has. They are the fastest RC cars you can get basically. Not even Nitro can touch them over a single lap.

 

For an Electric motor how light can it be pr kW?

 

the KERS was 80kw and 20kg? that is 4kw/kg so if you want 1000kW you get a 250kg motor? assume 200kg?

 

Either way you need a ESC and battery that is probably equal to the motor in weight surely. I believe that is not far off reality for RC cars. How much in weight for the rest of the emty car?  The motor itself does not really need a drivetrain. It could be slapped directly over the driveshafts and be running as a spool to save weight.

 

When you approach the extreme ends of something it becomes bizarre. Just like in math when you are near a asymtote. Top fuel runs 0-500km`t in less than 4 seconds and does it in first gear. An all out time attack car would probably be wildly different than most people imagine.

 

I love the pikes peak peugeot, hell I work for them, but nothing conservative in that engine.. it had something like 900 HP at the wheels at altitude.. lovely car by the way, done in a few months..

Interesting thing about VW was that they were really conservative on the power.. opting for lower power so that they could have less weight of the battery pack..

As for the motors, there is a lot of advantages of running one motor per wheel, be it only 2 wd or 4wd



#29 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 20 October 2018 - 19:37

It is conservative by modern standards. For starters Mullen and others have had 800-1000hp engines at the peak in the past. Granted none was as drivable as the Peugoet. 

 

A Naturally aspirated engine of same size can make more at sea level. you lose 20% at 4000m i believe is typically claimed. That is a good pointer to what kind of power an all out Turbo engine should make for a event like pikes peak when it comes from a big company.

 

F1 engines with less than halv the volume are pumping out 700-800hp now? And they are built with limitations like fuel flow limitations as well as fuel flow ramping with rpm, bore and stroke limitations. On basically pump gas. Granted the power to weight ratio of the engine is really important too and i am not sure how light the peugeot is.. Could be lighter than a modern F1 engine.

 

For comparison Eklund claimed 800hp at altitude in his old saab. a 2L i believe (which i to this day find doubtful buit spectacular if true) that was the 1990s.. his record in open? stood until fully paved roads btw. 

 

There could be advantages to running one pr wheel. but you need double the controllers and in anything extreme you do not want double anything. extra mounts, extra cables, extra cooling pumps. Cables alone is probably  5-10kg if not more.

 

We are talking about a very extreme teoretical car, that has more in common with drag racing, Can Am and the wildest F1 cars ever from all the differen eras combined.

 

As for VW going less weight and less power, it makes perfect sense. As cornering speeds go up so does the importance of low weight. And the more power you have, the less important it becomes. Schumacher getting second place in 5th gear is a good example of that.  That will never happen in WTCC..


Edited by MatsNorway, 20 October 2018 - 19:43.


#30 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 21 October 2018 - 07:52

ok, but two motors with total power of one single one will have practically the same thermal and current requirements. Ok, twin controllers but total weight of the rest should be similar. What twin motors give you is the possibility of torque vectoring, which is a big deal. Also, you dont have to run a diff, which is one thing that will offset all weight increases in the rest of the system, possibly being lighter overall.



#31 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 21 October 2018 - 09:03

A conventional open diff is a horribly inefficient device, although trying to get numbers is tricky. At full load, 85% has been measured, and at part load, say around town, 60%.

 

As such two motors per axle is a no brainer.



#32 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 21 October 2018 - 14:55

Greg, I know this is OT but are there some solid numbers on transmission losses, or better yet, driveline losses for various types ofdriveline layouts? i.e. crank to wheel or crank to road loss.. ?



#33 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 21 October 2018 - 16:24

This is not a competition in efficiency but speed. If the opposite was true no one would spew most of the nitromethane out of the exhaust for downforce.

 

As for gear loses.

Nice chart detailing what ratios the numbers apply for. i have read 99% is obtainable for spur gears myself.

http://www.meadinfo....bevel-worm.html

 

It was found that the friction loss of the straight bevel gear is larger than that of the spiral bevel gear, and the gear friction loss ratio of the bevel gears decreases with the increase in the gear torque.

Furthermore, we calculated the gear friction loss and the churning loss of the bevel gear drives.

https://www.jstage.j...rticle/-char/en

 

So in a sense it is silly to punish RWD cars in WTCC as they have a bevel gear extra and thus loses out on ultimate power.

 

Porsche ran spools in the 962. It worked fine at the time.

 

And remember you get a motor where you normally would have a diffuser. four motors pr. corners might not be so cut an dry as you think. ofc. one could argue that running skirt and a fan for downforce is the way to go. so five motors then.. Interestingly no one is doing that at Pikes peak yet, despite the paved road and no rules. hmm. Perhaps that is the trick for the peugeot "evo"

 

For individual motors pr wheel. 

You would not run same size motor front and rear. Smaller motors up front. Sized to be right under the traction limit up to a speed where the rear is not traction limited.

Four motors is more than just four motors and controllers. You have four times the mounts for both the motor and the controllers. There will be a weight penalty of a few kg extra and mid corner speed is very important the more extreme the race car becomes. More speed, more downforce.



#34 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 21 October 2018 - 22:39

"Higher cornering grip and even higher with 4 wheel torque control" ? Again you claim this apparently mainly because you just "know" that it's true.
Fundamentally I think 4wd cornering grip is less than 2wd/rwd for the same reason that front wheel drive/2wd only is less than 2wd/rwd - and that is that the front wheels are
at their absolute limit resisting sideways forces in a corner - asking the front tyres to then work even harder in supplying propulsive force is just not going to work - and any attempt to do so will just reduce cornering force. 

 

Ultimate cornering grip is obtained when all four tyres are operating at their optimum slip angle - depends on the tyre but typically 5 - 10* for slicks. Under these conditions there are significant energy losses in the tyre and when added to the usual aero drag (lots for a high DF car), there is a also a requirement for the tyre (on the driven wheels) to provide a lot of forward thrust. This means that the drive tyres must operate at a point on the "traction circle" which is removed from "pure cornering" by some angle. For an open diff 2WD, both drive tyres will be providing the same amount of thrust however the inside tyre has much lower normal force (load) than the outside tyre. If the slip angles are the same for both tyres, the cornering force generated will be proportional to the normal force and therefore the outside tyre will have a higher cornering force. Consequently the inside tyre will be operating with a traction circle vector which is further rotated from pure cornering than the outside tyre. Optimum cornering force is obtained when all tyres are operating with cornering force AND thrust force proportional to the normal force. Clearly this is not the case for drive tyres unless some sophisticated torque sharing is applied.

 

Furthermore, even when this is present on a 2WD vehicle, the driven tyres are operating at a significant disadvantage to the non-driven tyres. Only a torque vectored 4WD can operate all 4 tyres at the optimum point on the traction circle during limit cornering. Interestingly, when this is achieved, all four tyres have their grip vector pointing at the same angle to the tyre centreline.

 

http://robotics.ee.u...ntrol-Brown.pdf  See Page 31 onwards.

 

"CONCLUSIONS As predicted, improvements were hard to observe during the test itself but from the data it can be seen that the torque vectoring system is effective. Understeer is still evident but the system was able to vector torque away from the inside wheel while cornering and increase outside wheel torque where it could be utilised. More test situations would show additional effectiveness, such as a chicane test or double-lane-change test. Even with the 2WD simplification of the algorithm, it is clear that a 17% increase in cornering speed for a given 6m radius is a valuable improvement in performance. The results of this test could be extrapolated to suggest improvement in handling would be observable when the system is tested on the 2013 AWD car."



#35 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 22 October 2018 - 02:06

17% compared to what? Open diff? teoretical model? Sorry for not reading the paper but if you have read it. That would be nice to hear.


Edited by MatsNorway, 22 October 2018 - 02:07.


#36 Kelpiecross

Kelpiecross
  • Member

  • 1,364 posts
  • Joined: October 10

Posted 22 October 2018 - 03:56

17% compared to what? Open diff? teoretical model? Sorry for not reading the paper but if you have read it. That would be nice to hear.


What he said.

I suspect this was written by the same bloke in the white coat in the "seamless gearbox explained" video.

The writer also confuses/misspells "sheer/shear" - he is clearly an idiot.

#37 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 22 October 2018 - 08:13

Owning a car with a active tq vectoring rear diff, and having being driven in a car with 4 wheel tq vectoring I have no doubt that a theoretical ultimate track car should have it.. :D



#38 MatsNorway

MatsNorway
  • Member

  • 2,720 posts
  • Joined: December 09

Posted 22 October 2018 - 15:43

That would depend entirely on the weight penalty. :)

 

(but yeah. It would with 99% likelyhood) . Speaking off diffs. How much does the tq vectoring weight? what kind of mechanisms are involved

 

Ford Focus RS module:

https://s3-eu-west-1...-module-rdm.jpg


Edited by MatsNorway, 22 October 2018 - 15:55.


#39 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 22 October 2018 - 18:03

kikiturbo2, I've only seen one paper on diff losses, it was an SAE paper and they measured the heat radiated on a dyno test (bizarre). A rule of thumb is 98% for a gear pair with parallel axes if you aren't being stupid, and a conventional diff is way stupid. The old Peugeot worm type ones were a much better idea, at least for efficiency.



Advertisement

#40 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 22 October 2018 - 18:28

That would depend entirely on the weight penalty. :)

 

(but yeah. It would with 99% likelyhood) . Speaking off diffs. How much does the tq vectoring weight? what kind of mechanisms are involved

 

Ford Focus RS module:

https://s3-eu-west-1...-module-rdm.jpg

 

Total weight penalty on a mitsubishi AYC diff, which is a true torque vectoring diff and not a diff with an active lock as most active diffs are, is close to zero, because the casing is alu instead of steel for the mitsubishis passive LSD, including the controller hydraulic pump with solenoids that drive it. Mind you, due to complexity it is not as strong. If one was to make an equally strong version best guess is a few kg difference..

It works by having an open diff inside plus a step up gearbox whose output is taken directly to the output shafts via two sets of clutch packs (one per side) with a hydraulically operated piston that activates it. If you log its activity you can see that it basically puts more pressure on the outside clutchpack depending on lateral G and steering input. If anything, steering input is the major factor in all mitsubishi active diff controlls (rear and active center diff which is only active lock), which means that the whole system is based on yaw or directional change instead of pure optimisation of cornering forces mid corner.

 

Focus RS diff is somewhat similar but it doenst have the open diff, so the clutchpacks do all the work and when it overheats it looses drive completely.



#41 kikiturbo2

kikiturbo2
  • Member

  • 801 posts
  • Joined: December 04

Posted 22 October 2018 - 18:30

kikiturbo2, I've only seen one paper on diff losses, it was an SAE paper and they measured the heat radiated on a dyno test (bizarre). A rule of thumb is 98% for a gear pair with parallel axes if you aren't being stupid, and a conventional diff is way stupid. The old Peugeot worm type ones were a much better idea, at least for efficiency.

 

Thanks, that would explain big losses in 4wd systems that have multiple diffs and transfer boxes. I have seen some tests where rolling road dyno was compared to normal engine dyno and the total losses were on the order of 20% for 4wd cars but that includes the tire loss...



#42 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 22 October 2018 - 19:38

The diffs are the beiggest contributor.When you set the meshing (in production) there's a horrible compromise. Ideally for fuel consumption you'd like perfect mesh at a fairly low torque. but that implies a poor mesh at high load. That means lots of heat. Diff oil has a very hard life as it is, and you'd end up burning the oil if you were towing up a hill if the mesh was set for the 'ideal' for fuel consumption



#43 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 22 October 2018 - 21:53

Interesting stuff Greg. Sounds like another argument for the typical FWD layout. (all shafts parallel)



#44 Kelpiecross

Kelpiecross
  • Member

  • 1,364 posts
  • Joined: October 10

Posted 23 October 2018 - 04:32

17% compared to what? Open diff? teoretical model? Sorry for not reading the paper but if you have read it. That would be nice to hear.


I have now ploughed my way through the paper a couple of times (it's pretty heavy going). Oddly enough he does apparently mean 17% faster compared to an open diff. So it seems that when in a tight corner if you restrict the power to the inside wheel by torque vectoring to prevent wheel spin and send more power to the outer wheel the car is 17% faster.
What the hell did he expect? This thesis really is pretty pointless. Us Oldtimers used to get that same effect with a Limited Slip Diff.

The paper really has no bearing on the contention that 2wd/rwd is faster in perfect conditions than 4wd.

The paper is strange in that it has quite a few spelling mistakes - surely a bit of proof reading would fix this. It even mentions "wheel sleep" - Mexican "wheel slip"?

#45 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 23 October 2018 - 06:34

I assume Mr Brown is an electrical student. If the forward velocity increases by 17% and the latacc increases by only 25% something is going on that a=v^2/r doesn't cover, unless they didn't care too much about r.

 

However, niggles with physics aside that's a pretty stellar improvement from a bit of code.



#46 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 24 October 2018 - 02:01

I have now ploughed my way through the paper a couple of times (it's pretty heavy going). Oddly enough he does apparently mean 17% faster compared to an open diff. So it seems that when in a tight corner if you restrict the power to the inside wheel by torque vectoring to prevent wheel spin and send more power to the outer wheel the car is 17% faster.
What the hell did he expect? This thesis really is pretty pointless. Us Oldtimers used to get that same effect with a Limited Slip Diff.

The paper really has no bearing on the contention that 2wd/rwd is faster in perfect conditions than 4wd.

The paper is strange in that it has quite a few spelling mistakes - surely a bit of proof reading would fix this. It even mentions "wheel sleep" - Mexican "wheel slip"?

 

You won't get the same result with a LSD. As I said before, ultimate cornering grip occurs when each tyre provides lateral grip in proportion to its vertical load AND longitudinal grip in proportion to its vertical load.

 

Spelling mistakes? - nothing strange about that these days.   ;)  I see a lot of poor spelling and grammar coming from some very clever students.


Edited by gruntguru, 24 October 2018 - 02:03.


#47 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 24 October 2018 - 09:59

Yeah, it's all about 5 friction circles (none of which are circular) - one for each tire and one for the car. The basic racing problem for the driver is how to sit on the cars friction circle, and the problem for the team is to optimise the 4 friction circles and all the other stuff.



#48 gruntguru

gruntguru
  • Member

  • 6,603 posts
  • Joined: January 09

Posted 24 October 2018 - 22:14

Greg, would you agree that max cornering is enabled when all four tyres have their force vector pointing in the same direction (relative to steer angle). The only assumption here that I am not sure of is the "shape" of the friction "circle" and whether that remains constant with varying normal force?



#49 Greg Locock

Greg Locock
  • Member

  • 5,550 posts
  • Joined: March 03

Posted 25 October 2018 - 06:46

Last bit first - that is a very thorny question. Although I have millions of dollars of test results on tires at my fingertips, they don't answer that question, because we never map the circle. The software i use fits an ellipse to the lat and long tire models.

 

Max latacc would be if the force vector is perpendicular to the path of the car, for all 4 wheels, but it also has to match the axle weights, otherwise you are spinning as well as turning. This goes back to trying to draw the force diagram for a drifting RWD car, where the traction force from the rear wheels is actually helping the latacc, and the front tires are also helping to accelerate inwards, even if they have opposite lock. Getting my head around that made me realise why many vehicle dynamics textbooks start with a discussion of coordinate systems (which I'd guess most people skip, including me).  



#50 thegforcemaybewithyou

thegforcemaybewithyou
  • Member

  • 2,300 posts
  • Joined: April 12

Posted 25 October 2018 - 11:37

So, if we could design a car that races on a circular track only and the fastest time for one flying lap determines the winner, how would the car look like if downforce is not allowed.

I'd start with a 50:50 weight distribution front to rear, a weight distribution of maybe 55:45 to the left (for a lefthand track) so that the dynamic lateral distr. is close to 50:50. Add individual steering to all four(or 6 or 8) wheels and power all wheels individually, too. Tyre size should be same for all wheels. The suspension is either active or designed in a way that the car is parallel to the ground during cornering. Aerodynamic drag should be minimised as high drag reduces the remaining tyre forces.