Scientists Claim Major Leap in Engine Design 775
An anonymous reader writes "Purdue researchers say they have made a major advance in the design of the internal combustion engine, one that could seriously boost fuel efficiency and cut emissions. A key portion involves building intake and exhaust valves that are no longer driven by mechanisms connected to the pistons, a departure from the way car engines have worked since they were commercialized more than a century ago. 'The concept, known as variable valve actuation, would enable significant improvements in conventional gasoline and diesel engines used in cars and trucks and for applications such as generators, he said. The technique also enables the introduction of an advanced method called homogeneous charge compression ignition, or HCCI, which would allow the United States to drastically reduce its dependence on foreign oil and the production of harmful exhaust emissions. The homogeneous charge compression ignition technique would make it possible to improve the efficiency of gasoline engines by 15 percent to 20 percent, making them as efficient as diesel engines while nearly eliminating smog-generating nitrogen oxides, Shaver said.'"
Related story (Score:4, Informative)
So it's glorified Variable Valve Timing, then (Score:5, Informative)
http://en.wikipedia.org/wiki/Variable_valve_timin
According to Wikipedia, VVT has existed since the 1960's. The only improvement I can see (and that's from reading between the lines) is that they've developed a means of controlling it more precisely.
Been done (Score:3, Informative)
Why does this warrant such a sensational article? Racing engines have been actuating valves pneumatically for years and exhaust gas recirculation has been around forever too. Hell, Honda's old CVCC used a similar exhaust technique back in the 70's.
Pretty Low I Would Say ... What Motive Is There? (Score:5, Informative)
It's not just Purdue working on this, nor is it cutting edge. The idea of variable valve actuation [wikipedia.org] has been around for a while [delphi.com] as well as HCCI, which has some problems that are yet to be overcome. One of the notable ones that I recall is simple power. As the Wikipedia article notes, in a gasoline engine, you increase the fule/air charge to increase power. In a diesel engine, you just inject more fuel. In an HCCI engine, it's tough because "many of the viable control strategies for HCCI require thermal preheating of the charge which reduces the density and hence the mass of the air/fuel charge in the combustion chamber, reducing power. These factors makes increasing the power in HCCI inherently challenging."
For more info, the Wikipedia page has some great references:
- Research, publications at Lund University [vok.lth.se]
- Research at Chalmers University of Technology [chalmers.se]
- Research at Stanford University [stanford.edu]
- Research, publications at University of Wisconsin, Madison [wisc.edu]
- Research at University of California, Berkeley [berkeley.edu]
So, it's got a lot of benefits but a few trade offs that need to be addressed first. Honestly, why would Ford/GM buy this out and kill it when they could just develop the technology themselves and integrate it into their vehicles like Hitachi's research [hitachi.co.jp]? I mean, just because technology changes doesn't mean they should kill it instead of changing with it, right?Re:I RTFA yesterday when I saw it on the Firehose (Score:5, Informative)
The cam/valves are really the last mechanical part of the loop. The fuel/air mixtures are now fully controlled by the ECU, and can change on the fly to adjust for altitude, temperature, manifold pressure (turbo and supercharged systems), and the octane of the fuel. As I mentioned above, the spark systems are now fully controlled by a computer, and advance or retard the cylinder ignition, sometimes in conjunction with the fuel curve, to best burn the fuel/air mixture. Being able to dynamically change the valve timing, opening, closing, overlap, duration opens up even more possibilities for tuning and timing an engine.
Re:I RTFA yesterday when I saw it on the Firehose (Score:4, Informative)
The other "development" you mention is that adding water to the combustion process allows a higher compression ratio to be used without the risk of preignition or knock. This results in more (and more even) combustion pressure, meaning more torque everywhere from the same amount of gas. This method is very popular among drag racers, and is sometimes used with air/water injection into a turbo- or supercharger's plenum to keep the pressurized air from becoming uselessly hot before compression in the motor.
Rotary valves are a much older development. They have no history of producing more power or reliability (or even efficiency) than traditional valves. Of course, there is no reason they should. Both of the above techniques combined might double the efficiency of the internal combustion engine.
Re:I RTFA yesterday when I saw it on the Firehose (Score:5, Informative)
I don't know if you've ever had to compress a valve spring, but they're pretty pissed off devices.
The best thing we could do would be to move to some kind of rotary valve system - any kind, really. Because reciprocating valves have problems. They are what limits RPMs, which is why rotary engines have been known to reach over 10,000 RPM, and why a [very very built] small block tchevy :) with a Coates rotary valve system [coatesengine.com] has reached over 12,000 RPM (can't find cite for that, but their page says "The comparative efficiencies of the spherical rotary valve combustion engine have enabled engine speeds of 14,850 RPMs."
We could make smaller, even more efficient engines by increasing RPM, but we don't do that because it causes valvetrain death. In order to get high RPMs, you need to be able to open and close the valves faster. Cams only open valves; springs shut them. This has two effects; one, there is a hammering process that goes on between the valve and the seat. Two, if the springs are not strong enough, they do not push the valve closed fast enough, and you get a phenomenon called "valve float". Solenoids can provide infinitely variable valve timing and duration, and through a shifting system (where the whole actuation system moves) you can provide variable lift. But as you increase RPMs, you need to increase the spring rate, and therefore you need stronger and stronger solenoids.
A solenoid valvetrain has been used in racing (I forget by who) but no one has managed to make a system suitable for the street yet. That's really too bad, because you could eliminate most of the valvetrain that way. But there are definitely serious implementation issues. Rotary valves are here now. There are competing designs, but none with pictures as pretty.
Re:Nothing new (Score:5, Informative)
This is not news. BMW has been playing with this for years. So has Mercedes -- they call it EVT, for Electronic Valve Train. And next year it will ship in the 2008 C-Class sedan.
Re:Hope they patented it. (Score:3, Informative)
http://www.uspto.gov/web/offices/pac/mpep/documen
A person shall be entitled to a patent unless -
(a) the invention was known or used by others in this country, or patented or described in a printed publication in this or a foreign country, before the invention thereof by the applicant for patent, or
(b) the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States...
emphasis added
Re:Variable valve timing (Score:3, Informative)
The trouble with direct valve actuation (Score:5, Informative)
The trouble with direct valve actuation is making an actuator that's fast enough, powerful enough, small enough, heat-tolerant enough, and reliable enough to do the job. Cheaply. This is not easy. Prototypes have been built [fev.com], but it's still not something that's easy to do. BMW did quite a bit of work in this direction, but backed off to their "Valvetronic" scheme, which still has a camshaft with other components to give some adjustment potential.
Most of the existing schemes for tweaking valve timing still involve camshafts, but there's an additional mechanical linkage which allows adjustment of phase angle, valve travel, or both. That's an idea which goes back to steam engine design. Most of the gear on the side of a steam locomotive is there to adjust valve timing. Steam engines are controlled by valve duty cycle, not throttling. This was the original pulse-width-modulation system. On steam engines, valve phase can be adjusted far enough to reverse the engine, which is how locomotives back up. Some newer marine diesels have that feature, too. Eliminates the need for a reverse gear.
So this isn't a new idea. It's an old idea that's hard to make work cost-effectively. Somebody may crack this thing; it's a tough mechanical engineering problem, but not an impossible one.
Ah, the mandatory conspiracy theory (Score:4, Informative)
Ah, yes, the mandatory conspiracy theory. Get this, this is just variable valve timing [wikipedia.org] which by now a _lot_ of car manufacturers _already_ use, with various degrees of sophistication. This one may be slightly more efficient, but the important thing is that steps in that exact direction have been made, and there is already a healthy competition in that domain.
If you'll kindly read that Wikipedia page, you'll notice that both Ford and GM, since you name-and-shame them, _already_ offer engines with variable valve timing. GM has worked on theirs since 1975, and built automobiles equipped with, say, their Northstar System [wikipedia.org] since at least the 90's.
So, you know, even as conspiracy theories go, this one... shall we say, fails to be entertaining at least. It is lacking in the suspension-of-disbelief quality. It's akin to asking me to believe that Boeing is trying to kill the jet engine... never mind that they're already using them.
Re:Killed in "development"? (Score:3, Informative)
2006 Malibu sales = approximately 175,000.
2006 Silverado sales = approximately 675,000 (with a shortened model year.)
On top of that, over half of the Malibu sales were to fleets.
I think we know which side Detroit's bread is buttered on.
Re:Killed in "development"? (Score:2, Informative)
Don't be so naive. Both American Car companies AND oil companies have a vested interest in keeping things status quo. They are greedy bastards and want to wrench every penny from you. Don't believe me? Go watch this movie [apple.com]
"no other explanation" (Score:3, Informative)
engine displacement (Score:5, Informative)
Well, no actually. A rotary such as the current Mazda 1.3 litre simply spins faster than the equivalent piston engine. The volume passed per unit of time is the relevant comparison, not the static displacement.
Since the RX8 competes with similar HP sports cars by guzzling at SUV rates, it indicates Mazda's best effort so far is still inferior in power conversion of the gasoline. (Though the smoothness is great fun.)
As for turbines, same deal really. The aircraft turbine has yet to match piston engines on efficiency for short flights. You have to run long-haul at cruise altitude before the overall fuel consumption is lower.
The idea of a completely spinning engine is very seductive, but the actual results of forty years of careful research has not delivered a spinning engine that's better than the 'tossing potatos'. This is counter intuitive, and it's entire worth your while to dig into the studies to find out why that is.
Re:Nah (Score:1, Informative)
I have to squeeze into most urinals stalls at an angle slightly sideways. That doesn't work so well in vehicles, especially when driving.
No other explanation? You must be an arrogant, all-knowing eco-snob. Can I just call you shit-for-brains, instead?
Desmodromic.... (Score:2, Informative)
Re:Nah (Score:2, Informative)
Re:What will they do with this efficiency, though? (Score:4, Informative)
"Reprocessing plants around the world have exhibited poor records of occupational safety, pollution control, waste containment, and security. For example, at the Hanford military plutonium reprocessing plant in Washington State, over a million gallons of high-level liquid waste has escaped from steel-and-concrete tanks into the soil. One gallon of this waste is enough to ruin an entire city's water supply. Hanford workers have also shown a significant increase in the incidence of cancer. In Russia, an explosion involving high-level liquid waste contaminated hundreds of square miles and hospitalized thousands of people. In the UK, a small explosion in 1973 occurred at the Windscale reprocessing plant [now known as Sellafield], and radioactive effluents have been substantial. In the US, large quantities of plutonium are missing and "unaccounted for" -- enough to make several hundred atomic bombs." (http://www.ccnr.org/AECL_plute.html)
The UK has been reprocessing recently at their THORP plant, and it suffered a significant leak in 2005, so it's not just ancient history.
Granted, much of the current concern about reprocessing is about making weapons-grade Plutonium, but the concern is generally that a country like Iran might use U.S. reprocessing to justify their own, but tweak their own plant to create Pu-239.
Re:Killed in "development"? (Score:4, Informative)
Don't be so naive. Both American Car companies AND oil companies have a vested interest in keeping things status quo. They are greedy bastards and want to wrench every penny from you. Don't believe me? Go watch this movie [apple.com]
You want to know the deal with the EV-1? California bureaucrats thought they could wave a magic wand (i.e. use laws) to make technology advance. GM whipped up a quick electric car out of off the shelf parts, but at the same time sued to have the mandatory production rules reversed. The problem with fielding a vehicle like the EV-1 is that GM is then required to support that car for ten years after the date of manufacture. The owner of the vehicle has to pay for it, but GM would be required to maintain a full supply of parts and a staff of mechanics capable of servicing those vehicles. This would not be a trivial expense. The more EV-1's sold, the more expense. The arbitrarily short timetable mandated by the CARB made ramping up such a service system doubly expensive. The temporary rental scheme was something of a short term loophole that put them technically in compliance and bought them enough time to litigate the CARB into submission without incurring a long term maintenance liability. The thorough destruction of said vehicles is to ensure that no one will ever be able to hold them to their mandated parts and service obligation should they ever somehow get their hands on one.
Re:repairng your own vehicle (Score:3, Informative)
BTW, the compression tester, vacuum gauge, dwell meter, oil pump primer, timing gear puller, and piston ring compressor from rebuilding the '78 aren't useful for much of anything outside of auto repair... Nor can you transfer the skill required to install that stupid short hose connecting the water pump to the intake manifold on a Chevy big block.
Re:What will they do with this efficiency, though? (Score:2, Informative)
Basically it isn't reprocessing in itself that is dirty, it is the PUREX scheme that needs to be phased out. Reprocessing facilities like THORP may be modern, but they still use PUREX, with all its disadvantages.
Finally missing plutonium, while of course very serious, can't be used to make bombs unless it is weapons grade, and even then it isn't something you just don't do in your basement, you need machinery capable of cutting high power plastic explosives with the same accuracy as lenses for binoculars, not to talk about the difficulty in designing a functional device. Highly enriched U-235 would be a greater concern as a gun-triggered uranium device is comparably easy to design ( but by no means trivial ). Plutonium is a lot more difficult to use and if it is not weapons grade it is unlikely that anybody other than existing nuclear powers will be able to obtain the necessary data to make it work. You are talking about technology necessary to compress a metal sphere the size of a grapefruit into something the size of a golf ball. Have any idea what kind of force is needed to compress a metal ? Well, here's a hint, the entire weight of the Eiffel tower doesn't manage to significantly change the density of the steel it rests on. Now keep in mind that any deviation from a perfectly symmetric compression will screw over your bomb. If any one of the detonating charges is a fraction of a second late, or if it detonates with a few percent less force than the others, or if the machine that cut the shape of the explosives did not cut it in a perfectly parabolic shape, then all that will happen is that the metal will fly out where the pressure is lower. To actually get a detonation out of the damn thing you need to produce a perfectly spherically imploding shock wave. Keep in mind, this will involve knowing exactly how to shape the explosives ( you need at least two types, and you need to determine their parabolic shape based on the exact detonation velocities of each ).
Basically people like to think that once you got plutonium you are two minutes short of building a bomb. Its not quite that simple. If the plutonium isn't pure, or if it contains too much Pu-240, or if you failed to alloy it with just the right amount of gallium, or if you got the power of your explosives wrong by just a tiny bit, or if you failed to shape them perfectly right ( to the same accuracy as an optic lens ), if any of your neutronics calculations went wrong, or if your neutron initiator triggers at the wrong point, or if your tamper was slightly too heavy, or didn't reflect just the amount of neutrons you thought it did, the thing won't work. The probability that anyone other than an existing nuclear power would be able to use lost reactor grade plu
Re:Nah (Score:5, Informative)
I easily outmaneuver SUVs on "all season especially winter" tires in my little FWD car with maybe 5" of clearance tops. Because I have enough sense to put proper snow tires on in the fall.
4WD does nothing to help you stop, either.
Re:Nah (Score:3, Informative)
IIRC, 4-5 years ago Top Gear (or 5th Gear, forgot, one of the UK TV car shows) did a test comparing the following: Landrover Freelander (typical POS SUV), Skoda Octavia 4x4 (VW group estate with a 4x4 gearbox), Renault Scenic 4x4 (Medium family MPV with 4x4 gearbox). All in roughly the same weight and engine class. I recall some of the results:
Motorway Handling - dodging a cardboard box dropped from a lorry in front:
- Worst - Freelander. It nearly flipped over when doing the exercise. Screaming passengers, Flying luggage. Typical SUV style.
- Scenic - tolerable, considerably better than the FreeLander.
- Best - Skoda. Stayed glued to the road.
Towing a trailer on wet grass field :- Worst - Freelander. All show, no substance. SUV all the way.
- Skoda - close second to the Scenic overtaking the FreeLander by far.
- Best - Scenic.
And so on. Overall the "classic" SUV fared worst even in the dirt tasks they advertise it for, while the two 4x4 family cars more or less tied the first place (with Renault winning by a small margin).Re:I RTFA yesterday when I saw it on the Firehose (Score:2, Informative)
Hmm, my inline 600 ccm Yamaha R6 motorcycle engine tops out at 15.000 RPM, and it got nothing special to it. Just good old valves and springs.
And if you want to overcome the problem of floating valves at high RPMS you can do like ducati and have a forced closing of the valve (desmodromic valve) [wikipedia.org].
The limiting factor in engine RPMS is the piston mass, you can't have a piston that travles too fast and wieghs too much, or it will knock the crankshaft silly and break. That is why high performance, high reving engines has started to go "oversquare" (bigger bore, shorter stroke).
Re:I RTFA yesterday when I saw it on the Firehose (Score:1, Informative)
Has been done, is in production, called desmodromic.
The valve spring isn't there to keep the valve shut. It's needed to make sure the valves don't float. Actually it isn't the floating that is bad but the slamming of the valve into the valve seat at the end. A camshaft has a profile that sets the valve down gently to prevent wear. So the valve springs have to be strong enough to force the valves to follow the hump of the camshaft. And the higher the rpm the stronger the spring has to be. Come to think of it, that's not quite correct either. The limiting factor is the acceleration of the valve, so a combination of rpm and cam profile.
Re:I RTFA yesterday when I saw it on the Firehose (Score:3, Informative)
On another front, cams can close valves too. Ducati uses a desmodramic(sp?) system that doesn't use valve springs at all. It was developed at a time when the metallurgy to make springs meant lots of compromise. Desmodromic means no vavle float. Ducati uses it in their production and MotoGP motorcycles. Right now in MotoGP, Ducati has a big horsepower advantage over Yamaha & Honda (both with valve springs I think) and Suzuki (with pneumatic valves).
In the 80s, Maico had a 4 stroke that used premix and reed valves. The air/fuel mix came into the crankcase via reed valves. Another reed valve from there went to a tube that lead to the more conventional overhead valves. It gave it a supercharger effect.