Race Sim Central
Rec.Autos.Simulators Archive

Car Physics - Engine and wheel (again)

11 posts by 6 people · 02 Jul 2001 09:51 UTC - 08 Jul 2001 16:44 UTC

Nick Bull 02 Jul 2001 09:51 UTC#175771
Hi all, Thanks for eveyone that helped out last week. I've been thinking about this over the weekend, and I'm still a bit confused. Maybe it's just the way I'm doing it..... I've got an engine. The engine produces a torque at a specific RPM. The Idle RPM is 850. The engine also has a flywheel that has inertia. When the engine is at idle (850rpm) and the throttle is at idle throttle (5%), then the internal torque of the engine equals the output torque of the engine (5% of the max torque at 850rpm) and hence no output torque is produced. If the throttle is pressed, then the output torque of the engine goes up, and the rpm of the engine is calculated by working out the acceleration of the flywheel. Now, if the engine is put into gear, then the inertia of the drivetrain and the drive wheels are added into the engine flywheel inertia (using the relevant equations). This seems to work pretty well, but it is only modelling the car if there is no reaction from the ground. The problem I have is modelling the external forces on the wheels that feed back to the engine. The original version I had derived the engine RPM from the angular velocity of the wheels which was derived from the speed of the vehicle. Now, I've got an engine that is it's own object. The engine can have an rpm when in neutral even if the vehicle isn't moving. So, the question is, how to work out how much negative torque the wheels apply on the engine. Cheers, Nick.
On 2 Jul 2001 02:51:22 -0700, [email hidden] (Nick Bull) wrote: >If the throttle is pressed, then the output torque of the engine goes >up, and the rpm of the engine is calculated by working out the >acceleration of the flywheel. Sounds good, inertia of engine itself is also included probably.
Quoted text (10 lines)>Now, if the engine is put into gear, then the inertia of the >drivetrain and the drive wheels are added into the engine flywheel >inertia (using the relevant equations). This seems to work pretty >well, but it is only modelling the car if there is no reaction from >the ground. > >The problem I have is modelling the external forces on the wheels that >feed back to the engine. The original version I had derived the >engine RPM from the angular velocity of the wheels which was derived >from the speed of the vehicle.
Glad you got that out of the way; the velocity of the vehicle has very little to do with the speed of the wheels, in racing at least. ;-) > Now, I've got an engine that is it's >own object. The engine can have an rpm when in neutral even if the >vehicle isn't moving. So, the question is, how to work out how much >negative torque the wheels apply on the engine. I think I explained quite a bit in a post before this one, hope you didn't miss that. About Pacejka, reaction forces, sliptrac.crv etc. In short, you just let the whole assembly rotate, using the total inertia (geared etc). Then, as the wheels begin to rotate faster than the ground is moving beneath them, a slip ratio appears. The slip ratio decides the reaction force of the surface (see the curves in RCVD), and you add that as negative force to the whole assembly. So now you have: acc=total_torque/total_inertia but you have to add the negative forces on the wheel to total_torque. You just got to have those curves in, or approximate by 2 or 3 lines: /\ / \___ / Something like that. From 0 to 100% (maximum traction), and once it starts slipping too much, it goes back to ~60% (~75% for braking). Ruud van Gaal, GPL Rank +53.25 Pencil art : http://www.marketgraph.nl/gallery/ Free car sim : http://www.marketgraph.nl/gallery/racer/
On 2 Jul 2001, Nick Bull wrote: > I've got an engine. The engine produces a torque at a specific RPM. > The Idle RPM is 850. The engine also has a flywheel that has inertia. > When the engine is at idle (850rpm) and the throttle is at idle > throttle (5%), then the internal torque of the engine equals the > output torque of the engine (5% of the max torque at 850rpm) and hence > no output torque is produced. When you get things right, then you should be able to slowly let out the clutch with the engine at idle, and start the car rolling (unless you have programmed in a very abrupt racing clutch that engages over a very small distance). This was one test we used during the development of Atari Hard Drivin' (1st arcade game with a real physics model, c.1989). -- Doug Douglas Milliken Milliken Research Associates, Inc.
Quoted text (11 lines)>When you get things right, then you should be able to slowly let out the >clutch with the engine at idle, and start the car rolling (unless you have >programmed in a very abrupt racing clutch that engages over a very small >distance). This was one test we used during the development of Atari Hard >Drivin' (1st arcade game with a real physics model, c.1989). > >-- Doug > Douglas Milliken > Milliken Research Associates, Inc. > >
A masterpiece :-) I drive it whenever I see it, as well as the sequel, Race Drivin'. Todd Wasson --- Performance Simulations Drag Racing and Top Speed Prediction Software http://PerformanceSimulations.Com
| | >>When you get things right, then you should be able to slowly let out the | >>clutch with the engine at idle, and start the car rolling (unless youhave | >>programmed in a very abrupt racing clutch that engages over a very small | >>distance). This was one test we used during the development of AtariHard | >>Drivin' (1st arcade game with a real physics model, c.1989). | > | > A masterpiece :-) I drive it whenever I see it, as well as the sequel,Race
Quoted text (7 lines)| >Drivin'. | | Any chance the source is opened up? ;-) | This way we could check out the clutch functions and such that are | often not in our sims. :) | |
Ouuuuh that WOULD be nice <g> Ash
On 03 Jul 2001 21:52:44 GMT, [email hidden] (J. Todd Wasson) wrote:
Quoted text (8 lines)>>When you get things right, then you should be able to slowly let out the >>clutch with the engine at idle, and start the car rolling (unless you have >>programmed in a very abrupt racing clutch that engages over a very small >>distance). This was one test we used during the development of Atari Hard >>Drivin' (1st arcade game with a real physics model, c.1989). > > A masterpiece :-) I drive it whenever I see it, as well as the sequel, Race >Drivin'.
Any chance the source is opened up? ;-) This way we could check out the clutch functions and such that are often not in our sims. :) Ruud van Gaal, GPL Rank +53.25 Pencil art : http://www.marketgraph.nl/gallery/ Free car sim : http://www.marketgraph.nl/gallery/racer/
Afaik, Atari Coin-Op still owns it (does anyone know who owns Atari these days?). A modified version of the Race Drivin' model was used (so I hear) in San Francisco Rush, an arcade game from a few years ago. I don't think that there is any chance of the source being opened up in the near future. That's not the Atari style...<grin>...the development building I used to visit had separate electronic locks with credit-card-sized inductive-keys for every separate lab and project. I had to borrow my friend's "key" to use the washroom. -- Doug On Thu, 5 Jul 2001, Ruud van Gaal wrote:
Quoted text (21 lines)> On 03 Jul 2001 21:52:44 GMT, [email hidden] (J. Todd Wasson) wrote: > > >>When you get things right, then you should be able to slowly let out the > >>clutch with the engine at idle, and start the car rolling (unless you have > >>programmed in a very abrupt racing clutch that engages over a very small > >>distance). This was one test we used during the development of Atari Hard > >>Drivin' (1st arcade game with a real physics model, c.1989). > > > > A masterpiece :-) I drive it whenever I see it, as well as the sequel, Race > >Drivin'. > > Any chance the source is opened up? ;-) > This way we could check out the clutch functions and such that are > often not in our sims. :) > > > Ruud van Gaal, GPL Rank +53.25 > Pencil art : http://www.marketgraph.nl/gallery/ > Free car sim : http://www.marketgraph.nl/gallery/racer/ > >
On Sat, 7 Jul 2001 06:02:16 GMT, Doug Milliken <[email hidden]> wrote:
Quoted text (8 lines)>Afaik, Atari Coin-Op still owns it (does anyone know who owns Atari these >days?). A modified version of the Race Drivin' model was used (so I hear) >in San Francisco Rush, an arcade game from a few years ago. I don't think >that there is any chance of the source being opened up in the near future. >That's not the Atari style...<grin>...the development building I used to >visit had separate electronic locks with credit-card-sized inductive-keys >for every separate lab and project. I had to borrow my friend's "key" to >use the washroom.
Geez, that source code must have been REALLY good! ;-) A bit over the top. A shame though that the company is so closed up. But good to hear The Source was used even a few years ago; that probably means the code was ahead of its time. Ruud van Gaal, GPL Rank +53.25 Pencil art : http://www.marketgraph.nl/gallery/ Free car sim : http://www.marketgraph.nl/gallery/racer/
On 06 Jul 2001 21:58:18 GMT, [email hidden] (J. Todd Wasson) wrote: >>Any chance the source is opened up? ;-) >>This way we could check out the clutch functions and such that are >>often not in our sims. :) >> > > That would be great, but I doubt it :-) What's your clutch problem? Not really a problem, although I was thinking of how to slowly add torque from the engine to the drivetrain. A first hack would be to do something like: torque_entering_differential=torque_engine*clutch(0..1) acc_engine=(torque_engine+torque_feedback_from_differential)/ (inertia_engine+inertia_drivetrain*clutch) (or (1-clutch) for the drivetrain inertia) I should draw a picture though, as the engine rpm is directly derived from the incoming differential axis in my sim. So the differential is really the central place. The problem is then that I now (still not working though, but it worked before) separate the engine rpm entirely when the clutch is disengaged (engine runs free), and when the clutch is engaged (when it locks with the drivetrain; I have trouble remembering when you call the clutch 'engaged'; is it when you depress the clutch pedal, or when the clutch is locked with the drivetrain? :) ), I see the entire drivetrain as 1 big rotating assembly. In that last case I directly derive engine RPM from the differential 'input'. So there's no middle road, no slipping of the clutch when you slightly let go (and I don't have a pedal, so no easy way to test); it's engineRPM all the way in the engine, or all the way from the diff. Problem here is also that I would like to include support for 4WD cars sometime; someone's making a Subaru Impreza and that has 3 diffs IIRC; front, central and rear. Perhaps Gregor's formulae fit nicely, although I should define a nice way to define all those diffs in the car.ini files. Any idea what kind of pedals are nice to have as an aside from the normal wheel? (Alison Hine has separate pedals, and I don't yet support more than 1 controller, so it might be nice to test/support sometime). Ruud van Gaal, GPL Rank +53.25 Pencil art : http://www.marketgraph.nl/gallery/ Free car sim : http://www.marketgraph.nl/gallery/racer/
Hi Ruud, Ruud van Gaal wrote: > On 06 Jul 2001 21:58:18 GMT, [email hidden] (J. Todd Wasson) wrote: > > I should draw a picture though, as the engine rpm is directly derived > from the incoming differential axis in my sim. So the differential is > really the central place. I handle it as central as well :).
Quoted text (11 lines)> The problem is then that I now (still not working though, but it > worked before) separate the engine rpm entirely when the clutch is > disengaged (engine runs free), and when the clutch is engaged (when it > locks with the drivetrain; I have trouble remembering when you call > the clutch 'engaged'; is it when you depress the clutch pedal, or when > the clutch is locked with the drivetrain? :) ), I see the entire > drivetrain as 1 big rotating assembly. In that last case I directly > derive engine RPM from the differential 'input'. > So there's no middle road, no slipping of the clutch when you slightly > let go (and I don't have a pedal, so no easy way to test); it's > engineRPM all the way in the engine, or all the way from the diff.
The way I do it is similar to how I handle the locking wheels (and now the differential as well :) ). I have a flag that tells you whether the clutch is locked with the assembly or not. If the clutch is locked, then it's all considered as a single assembly. However, once the torque on the clutch exceeds how much the clutch can handle (which can vary with how much the clutch is actually applied), I set the flag to unlocked, and then treat the engine and the rest of the drivetrain as a separate rotating assemblies, but coupled with a frictional torque that the clutch produces (again, varies with clutch application) and points in the opposite direction to the slippage of the clutch. The clutch is again set to locked (with engine becoming a part of the assembly) when the direction of the clutch slippage reverses sign.
Quoted text (7 lines)> > Problem here is also that I would like to include support for 4WD cars > sometime; someone's making a Subaru Impreza and that has 3 diffs IIRC; > front, central and rear. Perhaps Gregor's formulae fit nicely, > although I should define a nice way to define all those diffs in the > car.ini files. >
Actually, I never properly thought about this. Since, however, the diff in my implementation is the highermost object in the rotating hierarchy, it wouldn't work in my system. I think Chris is using a different hierarchy (but the same system of equations) with engine on top, and perhaps that way one can just couple a diff upon diff in a treelike structure, and it could work. I don't know for sure, though, and still have doubts. However, if you have a special mechanism (such as the one above) for handling locked wheels, then it all becomes really, really complex anyway as there are many combinations of locked yet interlinked wheels possible and one should separately handle them all. -Gregor

Sponsors

Affiliate Links

Sim Gaming Expo MOZA Racing Sim-Lab Fanatec TrakRacer Asetek Internet Privacy From NordVPN Dreamhost Capital One Credit Card Application Fanatical CDKeys Amazon Amazon UK Project Motor Racing iRacing Get your racing gloves, boots and more from Demon Tweeks. Enlist at Roberts Space Industries, developers of Star Citizen and Squadron 42


Back from the ashes since July, 2019. First created in 2001 with the merger of Legends Central (founded 1999) and simracing.dk.

A site by a sort of sim racer, for sim racers, about racing sims. News and information on both modern and historic sim racing software titles.

All products and licenses property of their respective owners.


Facebook Threads Bluesky Twitter Mastodon Instagram YouTube Tiktok Twitch Reddit Google News Apple News Subscribe via WordPress Flipboard eBay

Race Sim Central © 1999-2026 | Some links pay RSC a commission. Advertising does not equal endorsement.Cookie Policy