On Wednesday 09 December 2009, Andy Pugh wrote: >> My thoughts drift toward the automotive engine control. Electronic >> valves, spark timing, fuel injection timing etc... I think this would >> require much faster response than a path control for an automobile. > >Not easy to use for your own purposes though. The hardware is >certainly capable (typically a fairly high-spec embedded PowerPC) the >platforms are basically completely closed. > >> The current state of engine control seems to be the evolution of the >> slow, early processors with lookup tables. > >Indeed, many thousands of them, interacting with each other in complex >ways. Good for the job at hand, and incorporating hundreds of PID >controllers (and more elaborate variants where P, I and D are >themselves the output of cascaded lookup tables.) > >> An update every 36 degrees of crankshaft rotation and every 18 >> degrees of (virtual) camshaft rotation at 6000 rpm. > >Typically things are calculated every mS, or scheduled on a particular >tooth of the crank encoder. > >> Some of the new generation diesel motors have no camshaft. The valves >> are solenoid actuated. Granted, they run at approximately 2000 rpm so the >> timing requirements of the solenoid cycles are not as severe. > >I am looking forwards to the day that tech gets to automotive engines. >With a turbo diesel, once you have boost, you can switch to 2-stroke. >I am not sure if anyone realises that. > A better question is when the first such engine ships and works, who is gonna get caught with their pants at half mast. Entirely doable with solenoid valve actuation at lower rpms. Not practical for cam driven valves though.
Another consideration would be the heat loading on the piston centers and how much actual overflush of the air could be arranged since the poppet valves areas are considerably smaller than a side of the cylinder port for a loop scavenged design. In the go-karts of yore, I have seen the McCollughs turning nearly 20k rpms, with ignition purely by the magneto running with the points turned off once they get to 7k or more rpms. One guy selling Mac's claimed his 6.1 cid engine could make 28 hp on alky, and this was 40+ years ago! But one kid figured he knew it all, bought one of them & fed it pump gas. It went, for about 1 heat race which got it into the main that night, but 3 laps into the main it went away and when we picked up the pieces, the piece of piston I found was burnt away down to the wrist pin. I had seen nitro do that, but that's the first time I ever saw gasolene do it. >> Some of the high end automotive gasoline motors do not require a starter >> motor. They sense where the motor is, inject fuel into the correct >> cylinder and fire the spark plug to get the motor moving. > >Really? That's news to me, but identical to how out 1916 Dennis fire >engine (9.12 litres, 4 cylinders) is started. Do you have a reference? >The engineering/science students who look after the fire engine would >love that. That sounds like there is very little 'new' under the sun. OTOH,has the starter motor that can turn a 2+ litre per cylinder engine over and reliably start it ever been designed? Andy, shoot some video of that & put it on utube, I think that would be super cool, and would put the designers/engineers of today on notice that in case they are thinking of a patent, by golly just look at the prior art, now nearly 100 years old if that 1916 is the year it was built. >> There is testing with injecting gasoline into a diesel motor under load. >> They get the best power and mileage with an 85% gas/15% diesel mixture. Chuckle, sorta reminds me of the time the truck driver pulled the wrong caps at the gas station & dropped 2700 gals of super diesel into the regular tanks. I had to run, with the Big Max on the trailer, up to Colorado & work on a failed microwave relay, so I 'gassed' up, and realized I was in deep do- do about 5 miles north of Farmington NM cuz it was sounding like a diesel about ready to lift the heads plumb off on the hills.. So when it got too noisy, I'd switch tanks to pull the hills & switch back when the pull was easy, in a 2 tank equipt 77 bronco. And everytime I went by a gas pump, I borrowed a funnel and filled the front tank back up with ethyl. By the time I got back I'd pretty well flushed the diesel & that 302 was pulling 25% better than it had pulled new! I always figured we bounced all the carbon out of it, (or carboned it up and raised the compression ratio) and of course in those days the lead in the ethyl probably wrote a ~30~ to the cat convertor but we didn't have a quarter to call anybody who cared in those days. The extra power eventually went away but it took about 10k miles to do it. >It has often occurred to me to wonder: If you were designing the >optimum petrochemical fuel, would it be petrol, or diesel, or, far >more likely, neither? Probably today it would whatever they could get the most profit out of since the engine can be made to burn old socks if its required. :) >(My day job? I am one of the guys who programs those automotive ECUs, >for the european arm of the Big Three manufacturer likely to make an >actual profit this year) Good. Now tongue in cheek, tell me how to get rid of the throttle lags in a drive by wire throttle equipt 2002 2 litre Jetta. That SOB is gonna kill me cuz it won't run till about 2 seconds after you've taken your left foot off the brake pedal. Once it decides to run, it will light the tires up VERY nicely. If I could find a mechanical throttle conversion it would be done tomorrow. -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) The NRA is offering FREE Associate memberships to anyone who wants them. <https://www.nrahq.org/nrabonus/accept-membership.asp> Should I get locked in the PRINCICAL'S OFFICE today -- or have a VASECTOMY?? ------------------------------------------------------------------------------ Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
