Cars (DATA REPLACEMENT) BMW M3 E30 Improved Physics by Arch

[Noctam]

Makes sense. The evolution of your work is great (I can't imagine the amount of time put into it), thanks so much for sharing!

 

[Billy Pilgrim]

Thank you fr continuing work on this.

Can I just overwrite with these new files? No need for a re-install from scratch?

 

[Kyuubeey]

Thank you fr continuing work on this.

Can I just overwrite with these new files? No need for a re-install from scratch?

Just make sure the folder names are correct.

 

[Leonardo Ratafiá]

but what has to do on how AC measures the ride height? if it's done correctly it should be the same, just 2 different ways to show the ride height.
but indeed there are no kunos cars or even other mods with such ride height figures

 

[Kyuubeey]

but what has to do on how AC measures the ride height? if it's done correctly it should be the same, just 2 different ways to show the ride height.

It doesn't do anything to the physics is what I mean. It doesn't matter what you put in, it won't change the car. It's just so the RULES section knows what height to reference. The actual height is +-5mm correct, assuming that BMW's schematics are correct.

I don't think it's right for the Sport Evo though because I probably didn't adjust it for the different wheels and tire diam.

BMW didn't report the alignment conditions, it's just referenced as a "standard" alignment condition. I believe it to be one occupant of 65 to 75kg and a 90% to 100% filled fuel tank, but I'm not sure. Either way the spec is +-10mm. Alignment conditions change from country to country so it's a bit difficult to ascertain them sometimes.

 

[mclarenf1papa]

@ForDownloading

Ride height in AC is just a number, it could be set to 600 meters and the car would handle no differently. Usually it's set to ground clearance instead of the way Arch set it, but it makes no difference besides the number itself.

 

[Tennisbiki]

First I thank you for your passion and work in this and other mod.

I am a relatively new to AC and even newer to digging into tuning and suspension.ini. Though I trained mechanical engineer, and had some success in AutoX locally. And I always thought I want to build a Locost type of car, but mid-engine. Anyway.

Recently, I started moving the suspension geometry and setting wheel rate and damping. During the last few days, I thought I tried to set a car's "ride height" to other cars, thinking there is some kind of "standard" from car to car. Now I realize that ride height is not one parameter determined from 1 reference point on one axle of a car to the absolute frame (world) but something else.

What I now wonder is that is it possible that the "ride height" of front axle and rear axle may be also not on the same "system".

Is the answer not likely or impossible?

 

[Kyuubeey]

First I thank you for your passion and work in this and other mod.

I am a relatively new to AC and even newer to digging into tuning and suspension.ini. Though I trained mechanical engineer, and had some success in AutoX locally. And I always thought I want to build a Locost type of car, but mid-engine. Anyway.

Recently, I started moving the suspension geometry and setting wheel rate and damping. During the last few days, I thought I tried to set a car's "ride height" to other cars, thinking there is some kind of "standard" from car to car. Now I realize that ride height is not one parameter determined from 1 reference point on one axle of a car to the absolute frame (world) but something else.

What I now wonder is that is it possible that the "ride height" of front axle and rear axle may be also not on the same "system".

Is the answer not likely or impossible?

The UI ride height is just referenced from the sprung CG location.

The actual ride height of the car, determine via ROD_LENGTH, depends on your static spring deflection. When "suspension travel" and "ROD_LENGTH" match, your car will load in at the design height.

You can find static spring deflection from the "telemetry app" that you can get by enabling dev apps in the assetto_corsa.ini inside of the directory inside of system -> cfg. It should be towards the left on the master slider and titled something like "suspension travel". The value is in millimeters. Convert it to meters for ROD_LENGTH input. You can also just make a spreadsheet and calculate it, but you might not know how to yet form the sheet. Later I will share a tool that includes solvers for all this alongside a basic physics documentation/guide.

 

[Noctam]

[...]Later I will share a tool that includes solvers for all this alongside a basic physics documentation/guide.

You got my attention!

 

[Tennisbiki]

This is a newbie question about the geometry for the trailing arm suspension on the E30. My understanding is that since the whole arm-wheel unit turn about one axis, the chassis mount point for the WBCAR_STEER just need to be on this axis or simply one of the WBCAR (swing arm) point.

But looking at this E30 plus other cars, such as older Porsches 911's, shows many cars has this STEER arm in .. what I consider incorrect places.

Am I missing something about this?

Just using CM Showroom looking at cars.. like 718 RS60, was that supposed to be like VW bugs front suspension? chassis mounting points are outside the car. Can't be that hard to model the real deal, can it?

 

[Kyuubeey]

This is a newbie question about the geometry for the trailing arm suspension on the E30. My understanding is that since the whole arm-wheel unit turn about one axis, the chassis mount point for the WBCAR_STEER just need to be on this axis or simply one of the WBCAR (swing arm) point.

But looking at this E30 plus other cars, such as older Porsches 911's, shows many cars has this STEER arm in .. what I consider incorrect places.

Am I missing something about this?

You are technically correct. Many good software such as SHARK will produce correct curves with an arrangement like that. If only life was that simple in real-time simulations.

However kinematic solving methods differ, and I believe that the welding of the CAR points coupled with the trailing nature of the assembly causes some error in how AC solves the toe deflection. Everything else appears to be very correct however. Log curves from your geometry and just find the unique toe-arm arrangement to match the toe curve. A bit of a pain in the ass because most software will not solve it in the same way, but oh well.

 

[Tennisbiki]

Thanks for your explanation.

What I understand is that sometime unrealistic suspension geometry are used to approximate the target suspension motion.

Sorry off topic: But personally I cannot stand having any suspension arm wider than the width of the car. How can an A arm be mount to outside the vehicle? I am going to try my hand fixing the double torsion front suspension and see if AC would take it. Time to search for a VW bug or 356 and see if other mod get that right.

 

[Kyuubeey]

Thanks for your explanation.

What I understand is that sometime unrealistic suspension geometry are used to approximate the target suspension motion.

Sorry off topic: But personally I cannot stand having any suspension arm wider than the width of the car. How can an A arm be mount to outside the vehicle? I am going to try my hand fixing the double torsion front suspension and see if AC would take it. Time to search for a VW bug or 356 and see if other mod get that right.

It's not unrealistic geometry if it produces intended results. Many of those old front suspensions just have an infinitely long virtual swing arm that doesn't swing, but translates vertically. There is no way to replicate that behavior by replicating the geometric centers of the assembly. The wheel-side points on many multilink suspensions when made with DWB are well outside of the dimensions of the wheel, because that is where one of the virtual axis pivots is.

 

[Stereo]

This is a newbie question about the geometry for the trailing arm suspension on the E30. My understanding is that since the whole arm-wheel unit turn about one axis, the chassis mount point for the WBCAR_STEER just need to be on this axis or simply one of the WBCAR (swing arm) point.

But looking at this E30 plus other cars, such as older Porsches 911's, shows many cars has this STEER arm in .. what I consider incorrect places.

The problem with just putting the WBCAR_STEER on that axis is the way setup toe is handled - it moves the chassis mount point laterally, keeping same length toe arm. So setup moves it slightly away from the axis unless it's a pure trailing arm design. The farther out along the axis you have it, the less this difference makes to the dynamic toe, so very long steer arms work better at keeping alignment correct.

 

[Kyuubeey]

The problem with just putting the WBCAR_STEER on that axis is the way setup toe is handled - it moves the chassis mount point laterally, keeping same length toe arm. So setup moves it slightly away from the axis unless it's a pure trailing arm design. The farther out along the axis you have it, the less this difference makes to the dynamic toe, so very long steer arms work better at keeping alignment correct.

I forgot about that. Any toe adjustment in setup will effectively throw the toe-arm position off whack. Do you reckon that is the reason there is some "solver error"? To me it makes more sense why only toe is off.

Now when I take a look, all my trailing geometries are built from an angled camber and toe value, a quite significant toe value as well, so it will definitely affect it a bit.

 

[Dr. Death]

Strange. I cannot replace the stock E30 of yours using the KS E30 sound because the engine just doesn't rev enough to appear in CM.

 

[Błażej Wianecki]

Hi. I have to say that the latest update has dulled down the chassis quite a bit. I has lost so much of that adjustability that had made the previous versions so good imo.
Henry Catchpole in his carfection review said that the e30 almost felt like a well setup fwd car, which at first is hard to wrap your head around. But it totally made sense while driving your earlier renditions. Now there is very little to none lift off oversteer. Maybe a bit more in sport evolution but the car is still very sluggish to rotate.
The behavior of real car can be observed very well here in this video in some of the drive by shots. And from the cabin around 6:40 mark. The rear end seems properly reactive on abrupt lift off.

Similar thing can be said about most of the updated cars and especially gt86 which used to dance on throttle just beautifully with very subtle and manageable rotation on throttle.

 

[Leonardo Ratafiá]

don't worry, he's gonna come back with a new update anytime and it will change again

 

[Kyuubeey]

Would be a bit funny if I did a telemetry comparison and we found out that the new version has more LOO than previous ones.

Anyway, it's fine, I'm not going to be updating any of these mods anymore, at least after one more update to a few. Old versions also exist so just use those. Fix it yourself if you want.