RE: [Algorithms] Car engine behaviour
Brought to you by:
vexxed72
From: Robert D. <bli...@go...> - 2006-04-27 07:44:38
|
LOL - not sure that applies to a "normal car" though ... I know for sure = my Vauxhall Astra 1.4 doesn't pull away in 3rd gear unless I'm pointing = down hill to help get me going :) Anyway, to the original questioner, you really need to decide how much = of a simulation you want before you get too carried away with all this stuff. You could spend months getting all the equations just right and getting everything running nice, or you could cheat, and just use a couple of = tables and some scalars for traction etc. Talking of traction, you didn't really mention it much so I will. If = you drive onto wet grass, you have very little traction and so you cannot = get very much drive from the wheels. But also, if you over do it and they = start to spin, then you lose pretty much all traction in all directions, so = you can't turn either. Remember, it's not just about pushing you forwards, = it's also about pushing you round corners. Rob > -----Original Message----- > From: gda...@li... = [mailto:gdalgorithms- > lis...@li...] On Behalf Of Tom Forsyth > Sent: 27 April 2006 08:16 > To: gda...@li... > Subject: RE: [Algorithms] Car engine behaviour >=20 > > If you do your gear ratio and power distribution > > from engine to the wheels right you should not > > be able to accelerate in gears higher than 3rd > > (with a normal car). >=20 > Er... what? I think you need to have a talk to Brian Wilson: >=20 >=20 > And comin' off the line when the light turns green > Well she blows 'em outta the water like you never seen > I get pushed out of shape and it's hard to steer > When I get rubber in all four gears >=20 >=20 >=20 > TomF. >=20 >=20 > -----Original Message----- > From: gda...@li... > [mailto:gda...@li...] On Behalf Of = david > pangerl > Sent: 26 April 2006 23:21 > To: gda...@li... > Subject: Re: [Algorithms] Car engine behaviour >=20 >=20 > Hi, >=20 > On more add :) > Be carefully when shifting gears; you need to adjust the engine RPM to = the > wheel(s) RPM with the new gear ration before applying the torque from = the > new gear otherwise your car will jump on gear shift. > I managed this with a simulated clutch. I release the engine = connection to > the wheels, shift the gear and in <clutch_time> interpolate the = current > engine RPM into wheel RPM * new gear ratio and connect the engine = again > (aka > release the clutch). Here is also a perfect place to put in a "shift = gear" > sound :) >=20 > If you do your gear ratio and power distribution from engine to the = wheels > right you should not be able to accelerate in gears higher than 3rd = (with > a > normal car). >=20 > Cheers, > DAVID >=20 > Tom Forsyth wrote: > To add to this - the torque of a normally-aspirated engine (without = turbo- > or super-charger) is moderately constant (pus or minus about 10%) > throughout > the usable rev range, which is typically around 2000rpm to 7000rpm > (revoloutions per minute) for most normal petrol road cars. If it's a > diesel, more like 1500 to 5000. Well-made lightweight sport engines = can > rev > higher up to numbers like 9000rpm, and the F1 cars rev to something = nuts > like 19000rpm. But still fairly flat torque all the way through. >=20 > Turbos add some wacky stuff to engine dynamics, and what they do = depends > massively on what size turbo relative to engine size, what the setup = is > and > suchlike. On most well-set-up road cars with turbos, they start = helping > around 3500-4000rpm (below that they actually steal a bit of power!), = and > it's basically a linear amount of torque boost up to redline. But it's = so > dependent on your setup that it's hard to generalise too much. >=20 > I believe superchargers are far more predictable, but whether it's a = fixed > amount of torque or rev-dependent, I'm not sure. Never owned a > supercharged > car. >=20 > The final aspect is gearing. Typically, each gear is a constant = multiplier > over the previous one. So if second is 1.5x of first, then third with = be > 1.5x over second, and fourth will be 1.5x over third, etc. This means = that > when you change up at maximum save revs ("redline"), you will always = land > at > the same lower rev speed, which ideally is the start of the power band = for > your engine. For example, whatever gear you are in, if you change up = at > 7000rpm, you will be at 4000rpm - or whatever the numbers are. >=20 > Racing drivers vary their gear ratios so that certain corners are = taken > right in the power range for certain gears. But road cars generally = don't > do > this - they stick with a fairly even ratio between successive gears. = The > usual exception is top gear, which is often much "taller" than the = rest to > give good motorway fuel consumption figures. Also on 4x4s, first gear = is > often much shorter to give better traction on rough terrain, and in = things > like old Landrovers you pretty much ignore first when on the road and = pull > away in second. >=20 >=20 > Obviously this is all a gross generalisation, but if you're simulating = a > car > in any finer detail than this, then you need to go and find some real = car > mechanics and they'll tell you the truth, rather than this load of old > cobbler's. Obviously real car nuts will kill me for these gross > simplifications, but I've got a Seven and they don't, so stuff 'em :-) >=20 >=20 > TomF. >=20 >=20 > -----Original Message----- > From: gda...@li... > [mailto:gda...@li...] On > Behalf Of Jon Watte > Sent: 26 April 2006 17:37 > To: gda...@li... > Subject: Re: [Algorithms] Car engine behaviour >=20 >=20 > I'd like to add that when simulating, you can break this down > into a few > simple steps: >=20 > 1) Decide what velocity the user controls mean that the wheel should > spin at. (This is a cheat, as the throttle really just lets > more gas/air > mix into the engine...) >=20 > 2) Measure the spin of the wheels, and divide by the differential and > gearbox ratios to get the engine RPM. Look up the maximum > engine torque > based on this RPM and your engine model. >=20 > 3) Apply torque or counter-torque to get them spinning at > that speed, up > to the maximum torque for whatever the current RPM of the > engine is. If > you want to slow down, you may use a lower torque max, because the > engine isn't actively working against the rotation, just > rotating slower > through friction. >=20 > Note that this will work right even when the drive wheel(s) is/are > spinning, i e, you're doing a burnout. The wheels will spin > faster, the > RPM will go up. >=20 > The actual movement of the car comes from the friction model > between the > tire collision and the road, and is not directly a result of > the engine > spinning the wheel. >=20 > Cheers, >=20 > / h+ >=20 >=20 >=20 > Dan Glastonbury wrote: >=20 > On 4/27/06, *Madoc Evans* <tm...@ti... >=20 > <mailto:tm...@ti...>> wrote: >=20 > I'm implementing >=20 > some vehicle physics and I'm not >=20 > sure how to simulate the behaviour of an engine. I'm completely > ignorant in that I don't really understand the >=20 > difference between >=20 > horsepower and torque and how RPM affect engine power etc. >=20 >=20 >=20 > To paraphrase from www.overboost.com <http://www.overboost.com>: >=20 >=20 > ------------------------------------------------------- > Using Tomcat but need to do more? Need to support web > services, security? > Get stuff done quickly with pre-integrated technology to make > your job easier > Download IBM WebSphere Application Server v.1.0.1 based on > Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D120709&bid=3D263057& >=20 > dat=3D121642 > _______________________________________________ > GDAlgorithms-list mailing list > GDA...@li... > https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list > Archives: > http://sourceforge.net/mailarchive/forum.php?forum_id=3D6188 >=20 >=20 >=20 > ------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, = security? > Get stuff done quickly with pre-integrated technology to make your job > easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache = Geronimo > http://sel.as-us.falkag.net/sel?cmd=3Dk&kid=120709&bid&3057&dat=121642 > _______________________________________________ > GDAlgorithms-list mailing list > GDA...@li... > https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list > Archives: > http://sourceforge.net/mailarchive/forum.php?forum_ida88 >=20 >=20 >=20 >=20 >=20 >=20 > ------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, = security? > Get stuff done quickly with pre-integrated technology to make your job > easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache = Geronimo > http://sel.as-us.falkag.net/sel?cmd=3Dk&kid=120709&bid&3057&dat=121642 > _______________________________________________ > GDAlgorithms-list mailing list > GDA...@li... > https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list > Archives: > http://sourceforge.net/mailarchive/forum.php?forum_ida88 |