Thank you 👍 This makes me feel smarter. Next time I overcook the exit & die backwards in a wall I will use this knowledge to confidently blame the tyre model & excuse my own ineptitude.
@@rideroundandstuff I kinda sound like this.. I'm using an 'omni' ModMic that sounds better than their directional ones and a small mic amp thingy . The noise floor isn't great but its really more than fine and better than 9/10 gaming headsets in my opinion!
I listened to it on my phone's speaker yesterday and the amount of crackling was absurd. Now I listened to it again with my headphones and it's perfectly alright. Sorry. Funny though, because the sound quality of my phone's speaker is alright otherwise.
@@rideroundandstuff Probably my GENIUS being too much for the poor speaker to cope with. I get it all the time, my wife often doesn't hear what I'm saying either. ;-)
This is not always going to be true, some tires may not drop off at all, some a bit more.. And if your tire is melting because of the sliding, the drop might be considerable.. But its not the reason of 'breakaway' as seems to be often hinted by articles and books on 'how to drive'
@@NielsHeusinkveld I see, the drop off isn't because of the tyre loses grip, its that combining longitudinal and latitudinal slip causes the tyre to no longer have any MORE grip to provide. I can definitely feel this is sims like project cars where if your rear wheels spin you die, as compared to irl when 125cc TAG karting as long as you have quick wheel work and accurate and intelligent throttle movements, any reasonable slide is savable and a certain amount of sliding is fast.
The thumbnail is quite amusing. You're getting good at this RUclips lark. I can see PewDiePie quaking in his boots as his subs are lured away by the temptation of your spreadsheets.
Thank you for these videos Niels. Very interesting. It appears "easier" (I use that term loosely-of course it is all very complicated) to tweak empirical models from the physical models. As you explained in one of your earlier videos, you can isolate characteristics of the tyre model from other characteristics that detract from a "real-world" experience versus having to alter a physical model in dimension, stiffness etc which inherently affects all other characteristics . I look forward to seeing you take these out on track...😉
Also very well explainable with a g-g diagramm. When you are braking very hard and "move down" in the tire's traction circle the tire can take up less lateral force. The "width" at this specific point is slimmer than it was with no longitudinal forces. Very nice insights Niels, I hope more videos are coming! :)
You see that it follows a friction circle, but what you don't see is the loss of lateral stiffness (initial steepness of the lateral curve) when you add braking slip ratio as I demonstrate here.
@@NielsHeusinkveld I've always thought it would be interesting to visualise the available grip with a dynamic traction ellipse of max available long. and lat. grip that changes size and shape and displays the current operating point at each time step. Always been too busy with the engines day job (cool t-shirt!) to see if the reality of this matches what's in my head.
Thanks you for making this! I kept trying to find more info on how slip ratio and slip angle affect each other while in a slide and couldn't find much on the subject anywhere else. Most of the information out there pertaining to tire grip while drifting only mentioned slip angle and would completely ignore slip ratio.
Good explanation, and I think it's really helped me understand why I can't seem to get on top of some sims. I've always felt that reducing my throttle doesn't help with grip the way it should.
Hi Niels. I remember discussing limited slip differential behavior with you back in the days when I worked on the GPL physics mods. I feel like there ought to be a way to relate lateral grip and longitudinal grip to the belt, side wall, and tread (or what passes for tread these days) distortions rather than to slip angle and slip ratio. That would get rid of the zero velocity problem and you would be able to park the car on a slope and still have the tire produce grip even though you had no slip angle. I am enjoying this series of videos, keep up the good work. Gene
Tire data / modelling reminds me a bit like the weather and the various computers models it has, sometimes they can get parameters close to what happens in real on the day and some days they can be off by a fair bit especially when dealing with severe weather as the atmosphere is so dynamic just like tires and the surface they are on. It's a never ending cycle to try and get better data / improve computer models, and like tire modelling weather models have improved a lot especially over the last 20 years or so with technology but they still have a fair way to go and even then they will probably never be able to be completely accurate with things being so dynamic in the real world.
Without any intended sarcasm, a great way to demonstrate practically is to drive a real car! I feel sometimes, other developers (both in commercial sims and in industry simulations) get a bit lost in the numbers and forget what a tyre actually feels like. A road tyre in particular demonstrates the behaviour really well. You can really feel the contact patch grip change as you add or remove long/lat load and/or a combination of the two 🙂 Very well explained with the dynamic spreadsheets though. Much easier to grasp than a full tyre slip plot!
Simple traction circle, max acc/brk or turning, not both at same time, entering a braking area, max brake, as soon as you need steering, roll off the brakes at the correct rate to allow max grip out of the tire all the way around the traction circle.Key is driving on the limit of both while doing it consistently (if your not trying to save tires/fuel/etc). You did miss the weight transfer part that is also apart of the traction circle, maybe make another video of that. Most new racers dont understand when weight transfer is good/bad for traction and to use that to their advantage, maybe you could go over that subject. I come from motorcycle road racing, much tougher job than cars (need to control 5 things at once vers 2 for cars). Thanks for the video.
Thank you for this more indepth video Niels! I definitely learned something new and it was really interesting. The way you show how longitudinal slip impacts the lat' grip-curve, helped me understand better how trailbraking and turning should be transitioned between when driving. But thinking of trailbraking got me thinking about the weight transfer argument for why trailbraking works so well in racing. Which then had me wonder how much of an effect weight transfer has on the situation, you talk about in the video, where some sims are hard to control over the limit, possibly do to their overexaggurated drop off in tyre grip. How much of a role does weight transfer play in such a scenario? I don't have anywhere near your level of understanding of this topic, so hope you can further help me understand :) Thanks in advance
Thanks for this! It is very interesting and demonstrates the balance behind the limits. Funny, just last night I was using SimHub to create a dashboard for AMS2...and wanted to include a visual representation of tire slip. I'm sure there are other, better methods, but I ended up using the game's raw data for mTyreSlipSpeed and wrote so that if value is greater than 5, it'll activate the flag on the dashboard. More for fun than anything else.
There is one parameter missing: body mass transfer. Letting go of throttle raises the rear, decreasing load and thus grip drops off dramatically. All of the curves here assume static load. I personally agree, the slip doesn't grow and suddenly "break off".. Which is the feeling you often get, you get more grip, more grip, a bit of plateau and suddenly you are on ice.
Yes, I ignored that so we could focus on the combined slip at one load, rather than changing everything. Its best to not FULLY let go of the throttle, but just enough to restore lateral grip, while not transferring as much weight to the front. And common mistake, the rear rising doesn't mean its getting light. On stiff or weird suspensions, the rear can even sit LOWER when you lift off the throttle. The height of the suspension does not indicate weight transfer per se. Stiffer or softer suspensions also don't really make a difference as far as weight transfer is concerned.
This is really interesting - I’m over braking - never been able to diagnose what it was about my driving style that sometimes kills my grip on corner entry before. Got some experimenting to do
Many things can do this of course.. I frequently see people going on the brakes hard, and then coming off just as fast, removing load from the fronts and reducing the ability of the car to turn-in. The hardest part of braking is getting off the brakes and starting turning in.
The oversteer out of the corner is a bit of a tradeoff between the length of the car, the stiffness of the suspension and the load sensitivity of the tire too.
Very good explanation! Would be amazing if you can make a simplified explanation taking weight transfer into account. Because AFAIC it's the most important factor you left out here. And if some new sim racers are like me when I started they completely ignore it and spin without having a clue why :D
Abrupt full throttle lifting would move more weight on the front than 'just enough throttle lifting' so that you restore cornering grip without taking all acceleration away. So your best bet indeed is to avoid too much weight transfer to the front by figuring out how much or little to lift. Easy to underlift though and still spin out.. Tricky stuff this race car at limit business!
Excellent video, Niels- but I don't think you were qute clear on the slip angle (for a moment I though you were describing steering angle). It would be more efficient to explain it as the angle between where the tyre is pointing and where it's going (and lateral and longitudinal forces). E.g. throttle on oversteer with proper wheel drive (that's a good one, made me smile) and throttle on understeer with wrong wheel drive- even when the car is not sliding (sliding over/understeer), car will tend to turn larger/smaller radius curve when applying throttle (increasing slip will increase slip angle for given lateral load). It just gave me an idea- using lateral forces (line on your upper graph being horizontal, rather than vertical to represent the force required to keep turning the same radius corner)- gently increasing power will on RWD gently increas slip on the rear making the car want to turn on tighter radius which will be counteracted with unwinding the steering wheel. Suddenly mashing the throttle will make slip angle increase untill tyres can not produce enough grip and inertia will introduce yawing momentum (i.e. spin). Same explanation could be used on corner entry- easing of the reducing slip to allow tyres to generate increasing amounts of centripetal force as the wheel is turned. As an aside- the tyre model I have mentioned in earlier video maintained that the greatest lateral force was generated with a certain amount of slip (I seem to recal it was asymetric)... Sorry for the long post, and keep up the good work.
Hi Niels. Enjoying your videos. I sent you an email a few weeks ago regarding pedals, not sure who took it but They said they would forward it onto you. Hope your well :) Greg
Nice vieuw of the tyre grip Realy got mi mind gooing , interesting stuff 👌 Like the first curve when you dont mind the temp , but i like the secend more when your thinking with tyre temp. And even with weight transfer and temp the curve wil always be different but i like to think that the curve wil become more and more like the secend one when your tires are getting cooked. Anyway keep up the good work realy like your vids 💪👌
Hi Niels, thanks so much for making this video. Very well explained and helpful for insight into car handling behavior. I had one question: When you were describing the scenario with slowly adding more throttle coming out of a corner (with a V8 haha), are the two characteristic curves with lateral and longitudinal slip angle only describing the behavior of a single tire? In this case the rear tires if it was a RWD car. If that's true, the front tires in the same situation would have different lateral/longitudinal curves with virtually no longitudinal slip angle and maximum ability for lateral grip?
Wow that’s actually a really nice explanation of tyre slip, gg You mentioned how it depends a lot on the sim itself for this modelling, do you know which sims portray the tyre slip/grip better?
No idea! I was very relieved to see that AMS follows pretty much what industry standard models do. To any serious sim developer this information isn't rare or too hard to figure out. But how each sim does it is hard to tell without them telling us. :-)
very cool info thanks :D i think a 3d graph (or rather a surface plot colour coded depending on the y axis) would help a lot explaining this, if people didnt understand from this video :)
The biggest breakthrough with this will be getting data for sudden changes in steering input, such as quickly trying to catch a slide and it not being a quick snap into a wall.
That is an essential part that I kept out to simplify it. Obviously when braking, tire loads on the front increase while the rears decrease, affecting the available front and rear grip quite a lot as well.
In the first half you describe slight grip loss. In real racing that is the “tires screaming” in a corner. That chirping squealing sound. The fastest drivers IRL “do not torture the tires.” Meaning not much noise from the tires thus the least amount of slip and the most grip. Also curious on sim racing games how do you calculate the coefficient of friction for each track? Also each track subsurface like painted rumble strips, hard run off areas and no race line that would be dirty from dirt and rubber debris? In a lot of sim racing the fastest drivers are 2 wheels off track. In the sims are each tire’s slip figured on each type of surface with different coefficient of friction?
Very late reply.. In AMS, there is a big text file that has grip multipliers for different surfaces, say 0.5 for grass, 0.9 for concrete, 1.02 for the racing line, 0.95 for curbs etc. I don't know how other sims do this but I would imagine it is similar!
Good explanation, but is there a way to incorporate a friction circle into that Excel sheet? Would make the combined slip easier to understand, and I would really like to see how it works with various slip angles.
I have it in the sheet but that looks at grip, not at slip. So you can't really see if a point on the circle is the result of silly slip angles or ratios. Perhaps I'll show it in a future video.
Thank you for your nice videos explaining vehicle and tyre dynamics in a simple language. Can you explain, for the following videos, singularity of the vehicle and tyre models when the vehicle speed becomes low? Definition of slip angle becomes ill defined when vehicle forward velocity goes very small, which causes problems in my simulations. I would like to learn how driving sims tackle this issue
AMS takes the lateral velocity to compute slip angle and when velocity nears 0 you get issues yes. I heard it does 'something simple but clever' near standstill but I have never looked into it. By default the 'low speed' behavior is gone at around 40km/h which I often change to a lower speed so I never really encounter it. Until AMS 'parallel parking' edition is released.. Probably you get some divide by 0 errors or big / weird numbers near 0..
So if I understand this correctly, break away is a function (not sure if that's correct nomenclature) of the torque band of the engine as opposed to the tire grip dropping off?
When its 'on power oversteer'. If you're mid corner zero throttle or brakes, and you're sliding, the 'feel' is still probably not really 'drop off' but more a 'sudden loss of expected grip increase'
Niels Heusinkveld Surely must also depend on which wheels are being driven? RWD, FWD or AWD, shouldn’t break away in the same way, adding downforce also complicates this further. I feel that the more information you give us, as a calculus problem it’s too difficult to calculate in real time. My guess is that’s the real crux of the problem as there has to be latency with the return signal as the circuit is feedback not feedforward, and as that’s digital there would logically be a quantization limit dictated by the cpu core clock rate, Im thinking its qubits or bust plus an inordinate amount of tire data from the manufacturer. Wondering if there might even be some tensor calculus involved as well, effing nightmare if you ask me. We’ve only just got circuit modelled compression that will run in real-time and that’s a much simpler piece of code, even valves are simpler to model than this.
From wiki tensor calculus page Tensor calculus has many real-life applications in physics and engineering, including elasticity, continuum mechanics...
Is the way you determine the lat and long components just by doing some simple vector mathematics? Breaking the overall force into its perpendicular components.
From the little I know this works fine for pure and combined slip. Of course you'd have to add some parameters for load sensitivity, peak slip that moves, temperature, and perhaps a dropoff as I belief by default the brush model just saturates.. But its a bit out of my comfort zone, I'm not aware of the pro's and cons enough.
Niels, maybe you can help me, i have this issue regarding tire grip, in all my games my cars doesn’t seem to have the same grip level into corners the other people online, i’m always 2 sec slower in all games compared to other people. I’m always at the limit of the car, i mean the car just starts to understeer when others seems to “stick” to the radius of the curve. Do you might have an idea where this might come from? I changed the gear, went from Fanatec to Simucube 2, this helped a little, changed the pedals, went from Fanatec to your pedals the sprint edition but still i can’t figure out from where this issue might come from!
overdriving might happen here... you said you are at the limit of the car.. but there is no fixed limit of a car... for example when steering too much you reduce the amount of grip the tire can create because you are past the optimum slipangle... so you reduce the limit of the car... in that case steering less can give you more grip... or steering to fast.. the heavier a car is.. the lazyier it is.. because all the mass wants to go in the direction it is already traveling... if you suddenly turn the wheel very fast, the car has not enough time to shift the weight to the outside.... so it just scrubbs over the front... you can try that when going in a straight line.. if you turn slowly to x degree of rotation.. the car will follow.. if you turn very quick to x degree of rotation the car will scrubb some time before it turns or have a initial movement then scrubbs then turn (depending on the car and how fast you turn) ... light cars are more responsive because less mass that wants to keep the direction... so the heavier the car the more time you need to set it up for the turn.. like you are talking to it.. "there is a corner coming soon.. let us get all the mass shifted so we can travel around the corner" .. and there are many more things like how much you brake while trailbraking since you are "stealing" cornering grip by braking grip... so trailbraking is a delicate thing.. you want to have the weight shifted to the front and the time you gain by trailing your brake into the turn.. but you want to balance it so you don't robber all the grip away from turning... also you may don't want too much weight on the front to not overloading the tires... thats why fast people may brake a bit earlier but not as hard or releasing the brake a bit more or earlier to carry more speed through the turn.. because yes slow in fast out... but if you can get more speed through the turn you are even faster out because you already had a higher cornering speed as a base to accelrate from... and well more and more things.. but in short... try to go slower to get quicker... relaxe and set the car up for the turns... try not to force it to turn but let it anticipate the turn.. let it roll if needed (for example long turns).. to not steal any cornering grip by accelrating or braking... be smooth with your inputs... especially with heavier cars... on light cars like formula cars.. you can be quite aggressive with inputs while still have a smooth car.. sometimes it is even needed to be quite aggressive to keep the car smooth but fast... but the heavier the cars get like gt cars for example... the smoother the inputs needs to for them to be able to react to your inputs... otherwise the just don't do anything and want to go straight... so if you are asking how can I go faster when I am already at the limit... try to raise the limit of the car... with your inputs and driving style... the less forces it needs to struggle with the more headroom of limit the cars have... I had that in the past too... I raced mainly formula and lmp cars and such things... and then joined a gt3 league.. and I thought omg.. why I am so much slower... it wasn't the ability to not react fast enough.. quite the opposite... I was reacting faster then the car can do.. so I was driving over the limit of the car.. which sets the limit of the car even further down...(makes it slower) ... I was trying so hard to go even faster and faster and there was just a very marginal improvement in laptime if at all... till I got frustrated and thought well.. I don't even try to go fast anylonger.. I just roll around now like a grandma... and ding 2.5 seconds better laptime ?!?!? wtf ?! Then it clicked in my brain... so pace is all about trying to keep the limit of the car as high as possible... it's about telling it what to do in a calm way.. so it doesn't get upset and reject to work with you... the lighter and more engineered towards racing it gets the more you can expect it to do what you want fast... but the heavier it gets the more time you need to give it to do the work you asked it to do .. otherwise it says no and quits the job :) so if the car understeers it means the front can't handle what you asked it to do... so try to let it do less ..so it can do the job...
I'm afraid its just a driving skill issue. Plenty of world records have been set with entry level equipment.. Pretty hard to pinpoint what you're doing wrong but its probably a few things!
Niels Heusinkveld If it’s my self then it’s easier to fix, i will try with a different pc (i now have an intel and will try with amd, who knows)! Your pedals are great (i previously had the Fanatec V3’s) the sprints are miles away in term of quality of built and the smart control software is really helpful! Thanks for your support! I will keep you posted for my findings regarding my issue!
@@silviutudorescu8114 no equipment you change is going to make any difference, your problem is skill not equipment (certainly changing your CPU for another is gonna make zero difference, a CPU does maths, they all do the same maths, if they didn't programs wouldn't run, of course this is known or we wouldn't have computers). Honestly dude, it's hard to know where to start, I would strongly recommend a couple of rl track days; as you don't seem to understand how the the dynamics and weight distribution of a car effects how much overall(lat/long) grip a tyre has as the normal force exerted on that tyre changes. Feeling this in rl is way easier than in a sim, but the understanding helps sim driving generally. Car setup can play a substantial part in this; doing car setup and learning to drive at the same time is a tricky business. Try with someone else's quick setup for a particular track and then learn to drive that setup, as it's very likely setup to trail brake well, with correct toe/camber/caster and suspension for the track. Sorry for the brutality, but you'd be much better off not wasting more money on sim kit (when what you have already is superb) and spending that money on track days/rl instruction.
Tyrant Worm Thanks for the explanations on where i can look into more, i guess i have to study indeed more on car physics and how to apply that weight transfer on braking vs acceleration and cornering. It’s good to know that perhaps it’s not the gear!
Thank you 👍 This makes me feel smarter. Next time I overcook the exit & die backwards in a wall I will use this knowledge to confidently blame the tyre model & excuse my own ineptitude.
You don't do that already since day 1 of sim racing? WTF?
You should always listen to a man that uses Sennheiser headphones and a mod mic!
the sound quality is pretty bad though :D
@@rideroundandstuff how is it bad? It's no studio setup or what Ermin Hamidovic uses but he can be understood perfectly.
@@rideroundandstuff I kinda sound like this.. I'm using an 'omni' ModMic that sounds better than their directional ones and a small mic amp thingy . The noise floor isn't great but its really more than fine and better than 9/10 gaming headsets in my opinion!
I listened to it on my phone's speaker yesterday and the amount of crackling was absurd. Now I listened to it again with my headphones and it's perfectly alright. Sorry. Funny though, because the sound quality of my phone's speaker is alright otherwise.
@@rideroundandstuff Probably my GENIUS being too much for the poor speaker to cope with. I get it all the time, my wife often doesn't hear what I'm saying either. ;-)
The very low reduction of the tyregrip after it's optimum is an eye-opener. Good stuff Niels.
This is not always going to be true, some tires may not drop off at all, some a bit more.. And if your tire is melting because of the sliding, the drop might be considerable.. But its not the reason of 'breakaway' as seems to be often hinted by articles and books on 'how to drive'
@@NielsHeusinkveld I see, the drop off isn't because of the tyre loses grip, its that combining longitudinal and latitudinal slip causes the tyre to no longer have any MORE grip to provide. I can definitely feel this is sims like project cars where if your rear wheels spin you die, as compared to irl when 125cc TAG karting as long as you have quick wheel work and accurate and intelligent throttle movements, any reasonable slide is savable and a certain amount of sliding is fast.
Thank you for having these insights! I like your work. Whenever you have more boring stuff to share .. here are enough people wanting to get bored :)
Careful what you wish for.... ;-)
How can we tag iRacing and Nicki Thiim ? :D
How can this guy be an engineer, he's holding the calipers the wrong way in the thumbnnail?!?
The thumbnail is quite amusing. You're getting good at this RUclips lark. I can see PewDiePie quaking in his boots as his subs are lured away by the temptation of your spreadsheets.
Maybe he's using a mirror
Thank you for these videos Niels. Very interesting. It appears "easier" (I use that term loosely-of course it is all very complicated) to tweak empirical models from the physical models. As you explained in one of your earlier videos, you can isolate characteristics of the tyre model from other characteristics that detract from a "real-world" experience versus having to alter a physical model in dimension, stiffness etc which inherently affects all other characteristics . I look forward to seeing you take these out on track...😉
Also very well explainable with a g-g diagramm. When you are braking very hard and "move down" in the tire's traction circle the tire can take up less lateral force. The "width" at this specific point is slimmer than it was with no longitudinal forces. Very nice insights Niels, I hope more videos are coming! :)
You see that it follows a friction circle, but what you don't see is the loss of lateral stiffness (initial steepness of the lateral curve) when you add braking slip ratio as I demonstrate here.
@@NielsHeusinkveld I've always thought it would be interesting to visualise the available grip with a dynamic traction ellipse of max available long. and lat. grip that changes size and shape and displays the current operating point at each time step. Always been too busy with the engines day job (cool t-shirt!) to see if the reality of this matches what's in my head.
Absolutely fabulous video. Thank you so much for this, I've been looking for information like this for years. Brilliant.
Thanks!
Thanks you for making this! I kept trying to find more info on how slip ratio and slip angle affect each other while in a slide and couldn't find much on the subject anywhere else. Most of the information out there pertaining to tire grip while drifting only mentioned slip angle and would completely ignore slip ratio.
Another great video, thanks! Still shocked how much slip is considered a small amount!
Good explanation, and I think it's really helped me understand why I can't seem to get on top of some sims. I've always felt that reducing my throttle doesn't help with grip the way it should.
That might also be because the weight of the car shifts forward when releasing the throttle.
Hi Niels. I remember discussing limited slip differential behavior with you back in the days when I worked on the GPL physics mods. I feel like there ought to be a way to relate lateral grip and longitudinal grip to the belt, side wall, and tread (or what passes for tread these days) distortions rather than to slip angle and slip ratio. That would get rid of the zero velocity problem and you would be able to park the car on a slope and still have the tire produce grip even though you had no slip angle. I am enjoying this series of videos, keep up the good work.
Gene
Excellent video Niels, thoroughly enjoy the physics.
Tire data / modelling reminds me a bit like the weather and the various computers models it has, sometimes they can get parameters close to what happens in real on the day and some days they can be off by a fair bit especially when dealing with severe weather as the atmosphere is so dynamic just like tires and the surface they are on. It's a never ending cycle to try and get better data / improve computer models, and like tire modelling weather models have improved a lot especially over the last 20 years or so with technology but they still have a fair way to go and even then they will probably never be able to be completely accurate with things being so dynamic in the real world.
Amazing video! Can't wait for even more in-depth knowledge!
Without any intended sarcasm, a great way to demonstrate practically is to drive a real car! I feel sometimes, other developers (both in commercial sims and in industry simulations) get a bit lost in the numbers and forget what a tyre actually feels like. A road tyre in particular demonstrates the behaviour really well. You can really feel the contact patch grip change as you add or remove long/lat load and/or a combination of the two 🙂
Very well explained with the dynamic spreadsheets though. Much easier to grasp than a full tyre slip plot!
Sometimes I wonder if some of these sim devs ever drive to work? Or do they take the bus? I imagine all pcars devs ride the bus.
Simple traction circle, max acc/brk or turning, not both at same time, entering a braking area, max brake, as soon as you need steering, roll off the brakes at the correct rate to allow max grip out of the tire all the way around the traction circle.Key is driving on the limit of both while doing it consistently (if your not trying to save tires/fuel/etc). You did miss the weight transfer part that is also apart of the traction circle, maybe make another video of that. Most new racers dont understand when weight transfer is good/bad for traction and to use that to their advantage, maybe you could go over that subject. I come from motorcycle road racing, much tougher job than cars (need to control 5 things at once vers 2 for cars). Thanks for the video.
Thank you for this more indepth video Niels! I definitely learned something new and it was really interesting. The way you show how longitudinal slip impacts the lat' grip-curve, helped me understand better how trailbraking and turning should be transitioned between when driving.
But thinking of trailbraking got me thinking about the weight transfer argument for why trailbraking works so well in racing. Which then had me wonder how much of an effect weight transfer has on the situation, you talk about in the video, where some sims are hard to control over the limit, possibly do to their overexaggurated drop off in tyre grip.
How much of a role does weight transfer play in such a scenario?
I don't have anywhere near your level of understanding of this topic, so hope you can further help me understand :)
Thanks in advance
Thanks for this! It is very interesting and demonstrates the balance behind the limits. Funny, just last night I was using SimHub to create a dashboard for AMS2...and wanted to include a visual representation of tire slip. I'm sure there are other, better methods, but I ended up using the game's raw data for mTyreSlipSpeed and wrote so that if value is greater than 5, it'll activate the flag on the dashboard. More for fun than anything else.
How to drive a car, the excel version :-D Great explanation, many thanks!
Thanks for the explanation. I'll put it to work.
It's hard to imagine (for me) the actual slip angle/steering angle without seeing it.
There is one parameter missing: body mass transfer. Letting go of throttle raises the rear, decreasing load and thus grip drops off dramatically. All of the curves here assume static load.
I personally agree, the slip doesn't grow and suddenly "break off".. Which is the feeling you often get, you get more grip, more grip, a bit of plateau and suddenly you are on ice.
Yes, I ignored that so we could focus on the combined slip at one load, rather than changing everything. Its best to not FULLY let go of the throttle, but just enough to restore lateral grip, while not transferring as much weight to the front. And common mistake, the rear rising doesn't mean its getting light. On stiff or weird suspensions, the rear can even sit LOWER when you lift off the throttle. The height of the suspension does not indicate weight transfer per se. Stiffer or softer suspensions also don't really make a difference as far as weight transfer is concerned.
This is really interesting - I’m over braking - never been able to diagnose what it was about my driving style that sometimes kills my grip on corner entry before. Got some experimenting to do
Many things can do this of course.. I frequently see people going on the brakes hard, and then coming off just as fast, removing load from the fronts and reducing the ability of the car to turn-in. The hardest part of braking is getting off the brakes and starting turning in.
The oversteer out of the corner is a bit of a tradeoff between the length of the car, the stiffness of the suspension and the load sensitivity of the tire too.
Very good explanation! Would be amazing if you can make a simplified explanation taking weight transfer into account. Because AFAIC it's the most important factor you left out here. And if some new sim racers are like me when I started they completely ignore it and spin without having a clue why :D
Abrupt full throttle lifting would move more weight on the front than 'just enough throttle lifting' so that you restore cornering grip without taking all acceleration away. So your best bet indeed is to avoid too much weight transfer to the front by figuring out how much or little to lift. Easy to underlift though and still spin out.. Tricky stuff this race car at limit business!
Excellent video, Niels- but I don't think you were qute clear on the slip angle (for a moment I though you were describing steering angle). It would be more efficient to explain it as the angle between where the tyre is pointing and where it's going (and lateral and longitudinal forces). E.g. throttle on oversteer with proper wheel drive (that's a good one, made me smile) and throttle on understeer with wrong wheel drive- even when the car is not sliding (sliding over/understeer), car will tend to turn larger/smaller radius curve when applying throttle (increasing slip will increase slip angle for given lateral load). It just gave me an idea- using lateral forces (line on your upper graph being horizontal, rather than vertical to represent the force required to keep turning the same radius corner)- gently increasing power will on RWD gently increas slip on the rear making the car want to turn on tighter radius which will be counteracted with unwinding the steering wheel. Suddenly mashing the throttle will make slip angle increase untill tyres can not produce enough grip and inertia will introduce yawing momentum (i.e. spin). Same explanation could be used on corner entry- easing of the reducing slip to allow tyres to generate increasing amounts of centripetal force as the wheel is turned. As an aside- the tyre model I have mentioned in earlier video maintained that the greatest lateral force was generated with a certain amount of slip (I seem to recal it was asymetric)... Sorry for the long post, and keep up the good work.
thank you, this explanation was very good. I’m dreaming now of you partnering up with 3Brown1Blue for the animations.
Oh that would be really cool. I'd love to see it!
Hi Niels.
Enjoying your videos. I sent you an email a few weeks ago regarding pedals, not sure who took it but They said they would forward it onto you.
Hope your well :)
Greg
Nice vieuw of the tyre grip
Realy got mi mind gooing , interesting stuff 👌
Like the first curve when you dont mind the temp , but i like the secend more when your thinking with tyre temp.
And even with weight transfer and temp the curve wil always be different but i like to think that the curve wil become more and more like the secend one when your tires are getting cooked.
Anyway keep up the good work realy like your vids 💪👌
Hi Niels, thanks so much for making this video. Very well explained and helpful for insight into car handling behavior. I had one question: When you were describing the scenario with slowly adding more throttle coming out of a corner (with a V8 haha), are the two characteristic curves with lateral and longitudinal slip angle only describing the behavior of a single tire? In this case the rear tires if it was a RWD car. If that's true, the front tires in the same situation would have different lateral/longitudinal curves with virtually no longitudinal slip angle and maximum ability for lateral grip?
Wow that’s actually a really nice explanation of tyre slip, gg
You mentioned how it depends a lot on the sim itself for this modelling, do you know which sims portray the tyre slip/grip better?
No idea! I was very relieved to see that AMS follows pretty much what industry standard models do. To any serious sim developer this information isn't rare or too hard to figure out. But how each sim does it is hard to tell without them telling us. :-)
very cool info thanks :D i think a 3d graph (or rather a surface plot colour coded depending on the y axis) would help a lot explaining this, if people didnt understand from this video :)
Thank you again, great help as allways.
More tyre videos!🤪🤣
The biggest breakthrough with this will be getting data for sudden changes in steering input, such as quickly trying to catch a slide and it not being a quick snap into a wall.
Another great explanation...and nice shirt. 😁
Thanks for these videos!
Fijne colleges dit! Mooi werk
Will you make a real physical tyre model (squishiness, slip angle, temperature) as you explained in one of your previous video?
Great! Now let's add weight transfer factor to these curves! )
That is an essential part that I kept out to simplify it. Obviously when braking, tire loads on the front increase while the rears decrease, affecting the available front and rear grip quite a lot as well.
In the first half you describe slight grip loss. In real racing that is the “tires screaming” in a corner. That chirping squealing sound. The fastest drivers IRL “do not torture the tires.” Meaning not much noise from the tires thus the least amount of slip and the most grip. Also curious on sim racing games how do you calculate the coefficient of friction for each track? Also each track subsurface like painted rumble strips, hard run off areas and no race line that would be dirty from dirt and rubber debris? In a lot of sim racing the fastest drivers are 2 wheels off track. In the sims are each tire’s slip figured on each type of surface with different coefficient of friction?
Very late reply.. In AMS, there is a big text file that has grip multipliers for different surfaces, say 0.5 for grass, 0.9 for concrete, 1.02 for the racing line, 0.95 for curbs etc. I don't know how other sims do this but I would imagine it is similar!
Very nice video! Now how do we get you to go fix iRacing's tyre model :)
Fantastic
Thank you so much!!
I love this videos!
Game Stock Car DVD on the background!!!
Good explanation, but is there a way to incorporate a friction circle into that Excel sheet? Would make the combined slip easier to understand, and I would really like to see how it works with various slip angles.
I have it in the sheet but that looks at grip, not at slip. So you can't really see if a point on the circle is the result of silly slip angles or ratios. Perhaps I'll show it in a future video.
Niels Heusinkveld True, it wouldn't work on its own, just thought it would be nice to see it alongside these graphs.
Thank you for your nice videos explaining vehicle and tyre dynamics in a simple language. Can you explain, for the following videos, singularity of the vehicle and tyre models when the vehicle speed becomes low? Definition of slip angle becomes ill defined when vehicle forward velocity goes very small, which causes problems in my simulations. I would like to learn how driving sims tackle this issue
AMS takes the lateral velocity to compute slip angle and when velocity nears 0 you get issues yes. I heard it does 'something simple but clever' near standstill but I have never looked into it. By default the 'low speed' behavior is gone at around 40km/h which I often change to a lower speed so I never really encounter it. Until AMS 'parallel parking' edition is released.. Probably you get some divide by 0 errors or big / weird numbers near 0..
While watching the graphs change I kept having iRacing flashbacks :D
So if I understand this correctly, break away is a function (not sure if that's correct nomenclature) of the torque band of the engine as opposed to the tire grip dropping off?
When its 'on power oversteer'. If you're mid corner zero throttle or brakes, and you're sliding, the 'feel' is still probably not really 'drop off' but more a 'sudden loss of expected grip increase'
Niels Heusinkveld Surely must also depend on which wheels are being driven? RWD, FWD or AWD, shouldn’t break away in the same way, adding downforce also complicates this further. I feel that the more information you give us, as a calculus problem it’s too difficult to calculate in real time. My guess is that’s the real crux of the problem as there has to be latency with the return signal as the circuit is feedback not feedforward, and as that’s digital there would logically be a quantization limit dictated by the cpu core clock rate, Im thinking its qubits or bust plus an inordinate amount of tire data from the manufacturer. Wondering if there might even be some tensor calculus involved as well, effing nightmare if you ask me. We’ve only just got circuit modelled compression that will run in real-time and that’s a much simpler piece of code, even valves are simpler to model than this.
From wiki tensor calculus page
Tensor calculus has many real-life applications in physics and engineering, including elasticity, continuum mechanics...
In my opinion this is the bit that iracing has missed the most
"So we're going to look at some Breaking and Entering" (5:00). What exactly has 'tire slip and grip levels' got to do with watching Burglary?
I understand your disappointment. :-)
Engine break after sudden throttle lifting and how it snaps grip off.
Is the way you determine the lat and long components just by doing some simple vector mathematics? Breaking the overall force into its perpendicular components.
No it is quite a bit more complicated than that. You wouldn't get the initial slope changes visible here and in all combined slip data you can find.
Ah yep, my main focus is beyond the initial slope. Just trying to piece all the bits together as I figure it out. Videos are very helpful :)
Hey, where did you download the 1000hp V10 Kart from Automobilista ? i cant find the download anywhere
Its a silly car I made for Jimmy Broadbent and myself only, its not released and is unlikely to ever be. (you're not missing that much!)
How do you formulate the graph?
Hi Niels, what do you think of the brush tire model?
From the little I know this works fine for pure and combined slip. Of course you'd have to add some parameters for load sensitivity, peak slip that moves, temperature, and perhaps a dropoff as I belief by default the brush model just saturates.. But its a bit out of my comfort zone, I'm not aware of the pro's and cons enough.
Brake-away... you mean powerslide?
Niels, maybe you can help me, i have this issue regarding tire grip, in all my games my cars doesn’t seem to have the same grip level into corners the other people online, i’m always 2 sec slower in all games compared to other people. I’m always at the limit of the car, i mean the car just starts to understeer when others seems to “stick” to the radius of the curve. Do you might have an idea where this might come from? I changed the gear, went from Fanatec to Simucube 2, this helped a little, changed the pedals, went from Fanatec to your pedals the sprint edition but still i can’t figure out from where this issue might come from!
overdriving might happen here... you said you are at the limit of the car.. but there is no fixed limit of a car... for example when steering too much you reduce the amount of grip the tire can create because you are past the optimum slipangle... so you reduce the limit of the car...
in that case steering less can give you more grip... or steering to fast.. the heavier a car is.. the lazyier it is.. because all the mass wants to go in the direction it is already traveling... if you suddenly turn the wheel very fast, the car has not enough time to shift the weight to the outside.... so it just scrubbs over the front... you can try that when going in a straight line.. if you turn slowly to x degree of rotation.. the car will follow.. if you turn very quick to x degree of rotation the car will scrubb some time before it turns or have a initial movement then scrubbs then turn (depending on the car and how fast you turn) ... light cars are more responsive because less mass that wants to keep the direction... so the heavier the car the more time you need to set it up for the turn.. like you are talking to it.. "there is a corner coming soon.. let us get all the mass shifted so we can travel around the corner" ..
and there are many more things like how much you brake while trailbraking since you are "stealing" cornering grip by braking grip... so trailbraking is a delicate thing.. you want to have the weight shifted to the front and the time you gain by trailing your brake into the turn.. but you want to balance it so you don't robber all the grip away from turning... also you may don't want too much weight on the front to not overloading the tires...
thats why fast people may brake a bit earlier but not as hard or releasing the brake a bit more or earlier to carry more speed through the turn.. because yes slow in fast out... but if you can get more speed through the turn you are even faster out because you already had a higher cornering speed as a base to accelrate from...
and well more and more things.. but in short...
try to go slower to get quicker... relaxe and set the car up for the turns... try not to force it to turn but let it anticipate the turn.. let it roll if needed (for example long turns).. to not steal any cornering grip by accelrating or braking...
be smooth with your inputs... especially with heavier cars...
on light cars like formula cars.. you can be quite aggressive with inputs while still have a smooth car.. sometimes it is even needed to be quite aggressive to keep the car smooth but fast... but the heavier the cars get like gt cars for example... the smoother the inputs needs to for them to be able to react to your inputs... otherwise the just don't do anything and want to go straight...
so if you are asking how can I go faster when I am already at the limit... try to raise the limit of the car... with your inputs and driving style... the less forces it needs to struggle with the more headroom of limit the cars have...
I had that in the past too... I raced mainly formula and lmp cars and such things... and then joined a gt3 league.. and I thought omg.. why I am so much slower... it wasn't the ability to not react fast enough.. quite the opposite... I was reacting faster then the car can do.. so I was driving over the limit of the car.. which sets the limit of the car even further down...(makes it slower) ...
I was trying so hard to go even faster and faster and there was just a very marginal improvement in laptime if at all... till I got frustrated and thought well.. I don't even try to go fast anylonger.. I just roll around now like a grandma... and ding 2.5 seconds better laptime ?!?!? wtf ?!
Then it clicked in my brain...
so pace is all about trying to keep the limit of the car as high as possible... it's about telling it what to do in a calm way.. so it doesn't get upset and reject to work with you...
the lighter and more engineered towards racing it gets the more you can expect it to do what you want fast... but the heavier it gets the more time you need to give it to do the work you asked it to do .. otherwise it says no and quits the job :)
so if the car understeers it means the front can't handle what you asked it to do... so try to let it do less ..so it can do the job...
I'm afraid its just a driving skill issue. Plenty of world records have been set with entry level equipment.. Pretty hard to pinpoint what you're doing wrong but its probably a few things!
Niels Heusinkveld If it’s my self then it’s easier to fix, i will try with a different pc (i now have an intel and will try with amd, who knows)!
Your pedals are great (i previously had the Fanatec V3’s) the sprints are miles away in term of quality of built and the smart control software is really helpful!
Thanks for your support!
I will keep you posted for my findings regarding my issue!
@@silviutudorescu8114 no equipment you change is going to make any difference, your problem is skill not equipment (certainly changing your CPU for another is gonna make zero difference, a CPU does maths, they all do the same maths, if they didn't programs wouldn't run, of course this is known or we wouldn't have computers).
Honestly dude, it's hard to know where to start, I would strongly recommend a couple of rl track days; as you don't seem to understand how the the dynamics and weight distribution of a car effects how much overall(lat/long) grip a tyre has as the normal force exerted on that tyre changes. Feeling this in rl is way easier than in a sim, but the understanding helps sim driving generally. Car setup can play a substantial part in this; doing car setup and learning to drive at the same time is a tricky business. Try with someone else's quick setup for a particular track and then learn to drive that setup, as it's very likely setup to trail brake well, with correct toe/camber/caster and suspension for the track.
Sorry for the brutality, but you'd be much better off not wasting more money on sim kit (when what you have already is superb) and spending that money on track days/rl instruction.
Tyrant Worm Thanks for the explanations on where i can look into more, i guess i have to study indeed more on car physics and how to apply that weight transfer on braking vs acceleration and cornering. It’s good to know that perhaps it’s not the gear!
"How hard can it be?"
Cosworth :)
So is there any evidence that real modern tyres have a large drop off?
Not that I have seen, but the lack of all evidence can also mean they DO drop off to some degree..
Accelerometer data.
CAN I GET SPEERDSHIT
??
That's how I feel most of the time too, don't worry! :-)
Niels Heusinkveld Sorry it was bugged for me, couldn’t watch it for some reason, will watch it now