Hi all.

Because there are discussions again and again, I wanted to do some objective examination of the problem I see with LFS tyre physics.

Here we go. I will prove my point. I won't use the XF GTI as some say it has not enough torque. I think it has enough, because there is enough wheelspin. (see clutch engaged ... difference of engine revs at 8000 and car speed)

Test paramters :

- I took the FZ50 (street version)
- Default Setup with 25% fuel
- No steering axis assigned to avoid steering during testing
- automatic gearchange activated to garantee the same shifting revs
- I used the car park with attached layout

Test Nr 1

Full throttle from start with TC on :

http://rapidshare.de/files/20129480/FTwithTC.wmv.html

2.83 at checkpoint 1
5.79 at finish line

Test Nr 2

Full throttle from start without TC on :

http://rapidshare.de/files/20129298/FTwithoutTC.wmv.html

2.81 at checkpoint 1
5.72 at finish line

Test Nr 3

Manual start. I tried to do a good start. No rev drop, but less wheelspin.

http://rapidshare.de/files/20129603/manualstart.wmv.html

2.81 at checkpoint 1
5.74 at finish line

____________________________________________

Test 1 and 2 are reproducable. Test 3 isn't, because every manual start is different.
Important is automatic gear change, because different revs at shifting result in different times.

With manual clutch and manual shifting a better time is possible!

Test Nr 4

Old one. XF GTI at acceleration lesson.
I managed a 7.09 with manual clutch and using a keyboard with full throttle.

http://rapidshare.de/files/19158049/709ft.wmv.html

Try to do much better with a good start. I bet a 7.05 is possible, but this isn't a realistic advantage against total full throttle wheelspin at start. It should be about 6.00 possible ...

Try it yourself if you don't believe me or show the videos.

Do a significantly better time with automatic gear change and show it in a video.

Layout attached.

Greetings
RIP

EDIT : 25% fuel used

Idea for further tests (I'm too lazy):

Experiment with different values for TC allowed slip and TC minimum speed. You could even make a longitudal force over slip ratio type of graph.

I played a bit with those, but all the results were even worse. Perhabs you can find a setting which is significantly better ... don't be lazy

I find drag times are best with TC set to 6% (although tbh didn't try going any higher, the tyres were bright green then and looked perfect).

Wheelspin when starting a FZ50 without TC :

@bob

I'll test 6% ...

I'm starting to think that the rest of the car may be real the problem.

Let's take a look at this first;
http://www.ismirdochlad.de/decel-slip.gif
(For the non-Germans, the graph shows slip ratio vs. friction coefficient of a tire)

The peak we see in the graph is also in LFS, you can verify this by looking at the longitudinal acceleration readout(F9) while launching with lots of wheel-spin. There's a spike in acceleration right when the tires regain grip.

If we were able to hold the tire at 20% slip (or wherever the peak is in LFS' model), we would see better acceleration. Problem is, you can't get near the peak without going over it. You can have wild wheel spin or virtually no slip at all, but nothing in between.

So either the edges of said peak are simply to steep, or the problem lies elsewhere. I suspect if it really was as simple as tweaking the slip vs. friction curve of the tire, this would have been fixed long ago.

My guess is that real cars behave differently because they have lots of slack and flex in the drivetrain. When you see a real car slightly chirping its tires, it's not a constant slip ratio like in LFS, but it changes as the engine rocks back and forth in its soft mounts, the suspension flexes and the driveshafts twist.



When you launch a car in LFS with spinning wheels and the tire goes from slip to grip, it STAYS that way unless you apply a lot more throttle to break it loose again.

When you launch a real car with spinning wheels and it goes from slip to grip, suddenly there's a lot more torque being applied to the engine because of the additional tire grip. That means the engine will move inside the flexible mounts. Since the engine moves relatively quickly, the maximum travel inside the mounts doesn't just depend on the torque being applied to the engine, but also on the engines momentum. That means it'll move quite a lot and then be pulled back towards its center location as soon as it runs out of kinetic energy. Moving back towards the center location will now in turn apply additional torque to the wheels and break them loose again.


Long story short, a real tire doesn't just go smoothly from 24% slip to 20% to 16%. The slack and flex of the drivetrain will make it grip and knock it loose over and over again several times per second (the frequency depends on how tight the drivetrain is), so it's basically cycling between 16% and 24% slip the whole time, which can have a significant impact on the transient phase between slip and grip.

My 0.02 €...

bal00 : Interessting point of view. I don't know what the problem is yet. I just analyse the result.

I am not sure if a good start really can produce better times, than with full throttle at all. Haven't experienced that yet.
If it does, it won't be much better than starting with full throttle. Not significantly at least.

@bob : I tested it with 6% TC and 10 m/s speed and the result was again worse than without TC : 5.78
Wheelspin has 5.72 ...

It's pretty much the same problem as trying to balance an object. You have to find THE perfect angle at which all forces cancel out each other.

It's the same with wheel-spin in LFS, unless you use JUST the right amount of throttle (which is also a moving target), you will quickly be on one side of the friction peak or the other.

Ever seen someone on a unicycle trying to stay on one spot? They don't just stand still and try to correct every little movement to one side. Going back and forth the whole time makes it a lot easier for them to keep their balance, and the alternating slip ratios are also what makes it easier to control wheel slip in real cars, since you don't have to hit that one magical throttle position.

That's a heck of a thought bal00

Is this what gives that intense vibration when accelerating in a RWD car (more drivetrain play...) with excessive power? One can feel the tires scraping against the ashphalt etc but there is invariably a longitudinal pulsating sensation under those conditions, your post explains why maybe?

Edit: I guess even wheelhop in FWDs right?

Quote from RIP2004 :

@bob : I tested it with 6% TC and 10 m/s speed and the result was again worse than without TC : 5.78
Wheelspin has 5.72 ...

Yeah I should have done more testing. Testing with BF1 I found 7.8% gives best drag times. 8% and one of the tyres goes red in forces view (but that could be caused by me steering with mouse from extermanl cam with forces view on...), 8.5% and both tyres go red and it feels very slidy under power.

With FZ50 the engine just bogs down too much when the TC kicks in for it to be useful (well, I haven't played with gearing enough yet...)

Quote from Ball Bearing Turbo :

Is this what gives that intense vibration when accelerating in a RWD car (more drivetrain play...) with excessive power? One can feel the tires scraping against the ashphalt etc but there is invariably a longitudinal pulsating sensation under those conditions, your post explains why maybe?

Edit: I guess even wheelhop in FWDs right?

Yes. Just think of the driveshaft as a big torsion spring to make it easier to visualize. Let's say you're in 1st gear and getting 1000lb-ft from the output shaft of the tranny. The tires still have grip and the torque "winds up" our driveshaft torsion spring (say 1 revolution). Now you give it more throttle, the tire breaks loose and because of the higher slip it can only transmit 800lb-ft onto the road (let's ignore the final drive). The result is that the driveshaft unwinds partially and makes the tire slip even more.
When the driveshaft is unwound to like 750lb-ft, the tire regains grip. Since the gearbox is putting out over 1000lb-ft and the tire does not slip anymore, it'll spend the first tenths of a second winding up the driveshaft again and due to that torque at the wheels increases relatively slow, until the driveshaft is wound up and the tire breaks loose again...

In the real world it's not so much the driveshaft that acts as a spring, but rather the engine/transmission mounts, the suspension bushings etc., but the effect is the same.

Quote from bal00 :

In the real world it's not so much the driveshaft that acts as a spring, but rather the engine/transmission mounts, the suspension bushings etc., but the effect is the same.

Hence here in NA, aftermarket motor mounts are available to stop wheel hop in SRT-4s...

rip... du muss auch immer was zu meckern haben, wa?

Quote from Screeny :

rip... du muss auch immer was zu meckern haben, wa?

Please write in English or if u just have to say what u just did...better not post!

I own a 2006 Corvette Z06. It has both traction control implemented via the ECU by cutting fuel off to one or more cylinders, and stability control implemented via individual wheel braking. In normal mode, both ECU and wheel braking are used to control wheel spin and eliminate yaw. During a 1st gear launch (redlines at 60mph), you can feel the wheels spinning a bit or sometimes more, but the computers quickly kick in and there's no hint of any yaw. The advantage of the wheel braking is that the wheel slip can be more accurately controlled, inspite of engine momentum.

For racing purposes, peformance mode is used, ECU fuel cut off is disabled, but stability control remains enabled.

You can turn off both ECU and stability, but other than spinning the tires to show off, it's not used for serious racers. All the magazine testers run with stability enabled.

FIA F1 class doesn't allow stability (individual wheel braking) control, but many other racing classes (like Dutch Supercar Challenge) do allow it. This would be in a car similar to a BMW 320 STW (super touring class). I'm not sure which cars in LFS are similar to the super touring class cars.

Just want to add, damn Jeff, you had a Caterham, and now the new Z06? Hot stuff!!

Quote from Tweaker :

Just want to add, you had a Caterham, and now the new Z06?

I had to sell the Caterham. The guy that was doing the engine mods and maintaince is chronically ill, leaving one other person familiar with the specific mods for my Caterham that resides near San Francisco. I sold the Caterham to a person who lives near San Francisco.

Instead of owning a large home, we live in a medium sized, 2 story condominium we bought back in 1984, and was paid off a long time ago. My wife and I don't travel much, the kids are grown up and moved out, so now my status is married with pet. Since we don't spend money on kids, house, or travel, we spend it on cars every now and then. I usually just buy a car every 8 to 10 years, but the Caterham was spending 75% of it's time in the shop getting upgrades or fixes to the last set of upgrades, because the guy doing this could only do it part time. I ordered the car back in 2003, and sold it about 2 months ago, and the car only had 1800 miles on it (because it was in the shop so often, due to engine upgrade / prototyping). I had a chance to order a Z06 for a good price from a friend at a dealership, so I ordered the Z06 and sold the Caterham. I'm out some money, but I plan on keeping the Z06 for a long time, I got a 7 year warranty for it. I'm 54 years old, so I'm thinking it may be my last chance to own a really quick car. Paul Newman continued racing into his 70's but I don't know what my reactions and condition will be when I get that old (probably end up driving a golf cart). A pic of the Z06 with our home in the background (our place is the open garage door and to the left (single car garage)):



Uhm, did I forget to mention I also own a 2001 Suzuki Hayabusa (it's really quick)? 2001 was one of just 2 years where the side fairings have the falcon (hayabusa means falcon) head pattern. If you look at the picture, the large button is the "eye", and the smaller one lower down is the "nostril" of the beak.

Yesterday I tested maximum g-forces values like with the braking and as Scawen said it is possible to achieve more acceleration with not full throttle.

But I still think it is to hard to achieve. The area of best acceleration is far to small ... IRL it makes measurable differences in 1/4 mile races between full throttle and good throttle control. Even a short burnout with lots of wheelspin makes a bad time ... In LFS you can just achieve a better g value for 1/10th of a second or so and all in all you are faster away with full throttle.

So I don't know what and how much is wrong about LFS, but I am pretty sure that this is still an issue and it will kind of get out of attention if there are free starts somewhere in the future of LFS. Because then reaction time will do the difference ...

Try using the BF1 with full throttle start (no TC), full throttle (TC at 8%), or managed pull away (with TC). You'll see the times fall lots between each method.

Quote from bal00 :

I'm starting to think that the rest of the car may be real the problem.

Let's take a look at this first;
http://www.ismirdochlad.de/decel-slip.gif
(For the non-Germans, the graph shows slip ratio vs. friction coefficient of a tire)

The peak we see in the graph is also in LFS, you can verify this by looking at the longitudinal acceleration readout(F9) while launching with lots of wheel-spin. There's a spike in acceleration right when the tires regain grip.

If we were able to hold the tire at 20% slip (or wherever the peak is in LFS' model), we would see better acceleration. Problem is, you can't get near the peak without going over it. You can have wild wheel spin or virtually no slip at all, but nothing in between.

So either the edges of said peak are simply to steep, or the problem lies elsewhere. I suspect if it really was as simple as tweaking the slip vs. friction curve of the tire, this would have been fixed long ago.

My guess is that real cars behave differently because they have lots of slack and flex in the drivetrain. When you see a real car slightly chirping its tires, it's not a constant slip ratio like in LFS, but it changes as the engine rocks back and forth in its soft mounts, the suspension flexes and the driveshafts twist.



When you launch a car in LFS with spinning wheels and the tire goes from slip to grip, it STAYS that way unless you apply a lot more throttle to break it loose again.

When you launch a real car with spinning wheels and it goes from slip to grip, suddenly there's a lot more torque being applied to the engine because of the additional tire grip. That means the engine will move inside the flexible mounts. Since the engine moves relatively quickly, the maximum travel inside the mounts doesn't just depend on the torque being applied to the engine, but also on the engines momentum. That means it'll move quite a lot and then be pulled back towards its center location as soon as it runs out of kinetic energy. Moving back towards the center location will now in turn apply additional torque to the wheels and break them loose again.


Long story short, a real tire doesn't just go smoothly from 24% slip to 20% to 16%. The slack and flex of the drivetrain will make it grip and knock it loose over and over again several times per second (the frequency depends on how tight the drivetrain is), so it's basically cycling between 16% and 24% slip the whole time, which can have a significant impact on the transient phase between slip and grip.

My 0.02 €...

very well said. However dont some of the "race versions" of cars have tighter motor mounts, and less slack in the driveline. Dosent tire pressure, wear on the engine, transmission, wear on the tires, motor mounts, chassis flex/warp, surface type, type of car overall, computer control throttle, (some newer cars) and engine managment, all play a role in howe a car launches. Only reason i ask is becasue, years ago i used to have a camaro, launching was tricky becasue of the type of tires i had. it seemed like it was either grip or no grip, But i have a newer cadillac now with computer controled throttle. So if i gas it, it gradually leans into the throttle, but someties i can a little bit of spin, then chirp, chipr chirp...

I think that It could be modeled in LFS by somehow replicating either the entire driveline slack. Im sure there is a way to replicate this

yeah if i had the money for a z06 i'd get it in that color looks absolutely stunning

Quote from Bob Smith :

Try using the BF1 with full throttle start (no TC), full throttle (TC at 8%), or managed pull away (with TC). You'll see the times fall lots between each method.

Sorry, have been on holiday for a while

Well you're right. They do. But it isn't as significant as in real life. You don't need a formula 1 car with TC ... some Touring car is enough.

The german DTM with similar cars like the FZ50 GTR :

The difference between a good start with little wheel spin and a start with a lot of wheelspin is over a second to 100 kph.

Good start : about 2,5 to 2.9s to 100 kph
bad start : 3,5 to 4s to 100 kph

And thats even just for a second of wheelspin by accident. If you would stay full throttle like everyone in LFS does the difference would be even worse.
Even with a short burnout and then less throttle you lose several positions in real life. In LFS its not even 1/4 of a car length between a full throttle starter and someone who tries to get a good start ...

So still something is wrong. It is not as wrong as I thought first, because the grip IS slightly better at some degree before total spinning, as you can see with the TC of the formula 1 car. But the difference is not significant enough ...

just map the clutch to the brake pedal if u only have 2 pedals.
for drag tests, accel and clutch is all u need. brake is not necessary.

Clutch won't help, as far as I can see with my own tests.... Clutch is actually really pointless to have in LFS. No advantage of using it in anyway (other than entertainment) in my opinion

Even with a manual clutch, there is still an auto clutch that interacts when you are at 0. You don't get any resistance of traction when starting off, nor have time to do a proper rate of easing in the clutch...... It's just a "drop n' go". Because the full on wheelspin gains momentum, speed and traction MUCH faster than you could ever properly or efficiently use the clutch. Kinda sucks.

I tried it with clutch too. You are free to try on the drag racing track. Its very hard to achieve even something like 0.02 seconds advantage on a 1/4 mile.

You can do it a little better, but the difference should be >1 second on the 1/4 mile and it isn't ...
It can't be that hard to get a little more traction. It's all about a little feel. And you can use the throttle very precisely in LFS. But it simply doesn't work like it does in real life.
People on a 1/4 mile event aren't robots. They aren't gods which have the perfect throttle amount on every second. But all of them get better results by trying not to spin the wheels. And in LFS it is damn hard even to achieve the smallest amount of time advantage against a full throttle all the time start ...