Ducati was rumored to be working on such a thing (i would be surprised to learn that Honda wasn't). Computer controlled cam timing would be quite nifty - it would make the mechanically based variable duration/timing system on the new Porsche turbo seem mickey mouse (Road and Track had a write up on the Porsche system). One problem - with solenoids/whatever controlling the valves, Ducati could no longer be distinguished by their "desmo" trademark. C'est la vie.
Electromechanical valves do indeed use solenoids to control the valves. The main advantage is that it allows valve timing to be varied almost indefinitely. It has huge implications - think of Honda's VTEC system taken far, far further. Valve timing can be immediately changed to the best timing for any operating condition ... for maximum power at full load, or for best efficiency and emission control at part load.
There are two problems with all this. One of them has been the power consumption of the solenoids. To solve this, some systems use hydraulics (basically engine oil pressure) to do the actual work of moving the valve, with small solenoids controlling what the hydraulics are doing. Anyone out there with a Ford Powerstroke diesel? It's rather similar to the way the HEUI (hydraulically powered, electronically controlled, unit injector) system works for the fuel injection on those. (In fact, Navistar is working on a "camless diesel" which applies this principle to the valvetrains on their next generation diesel engines. No more dirty diesels - it's capable of reducing emissions on those engines to practically nil.)
The other problem has been getting the system to work fast enough. Making it work on a 4000rpm diesel is one thing - making it work on a 14,000 rpm gasoline engine is quite another. It may come eventually, but it looks like the first production application will be Navistar's diesel, and it is still a few years off.
It may happen in Formula 1 sooner than that, for the simple reason that those teams have a LOT of money available. It's a different story when trying to make something that is cost-effective.
Problem #1: High electric/hydraulic loads: thereby requiring really heavy-duty alternator or oil pump, or air compressor if an electro-pneumatic system is used instead of the proposed electro-hydraulic system. How much energy is used to turn a camshaft? Helical gears, chains or belts turning a shaft should be more efficient than massive alternators or oil pumps, shouldn't they?
Problem #2: Electrically- or electrohydraulically activated valves will work best on low-speed, high torque engines where valve timing is slower and can be more easily controlled. Meaning that this technology is best suited to Harley-Davidson, the least likely manufacturer to use it.
If I'm to understand correctly, Benetton has already ran and tested their electro-magnetic valve system with lap times that are right along side those produced with their Supertec Playlife engine. The Supertec engine already had a year of racing development. So needless to say, for a brand new engine technology to be immediately competitive is very promising.
The electrical load problem is one of the reasons that BMW (among others ) is looking to change to a 42 volt system. I believe that they have announced a 4 cyl. engine that uses the intake valves instead of throttles to control the engine speed - and this uses continuously variable timing.
When the Honda V5 GP motor was announced, I was curious if it might incorporate tricks in the valve train, such as the pneumatic springs used in F1 (AFAIK, F1 uses conventional camshafts with pneumatic valve springs, not pneumatic actuation). After close inspection of the photo, I'd guess not. In fact, it appears rather ordinary (except, of course, for the extra spark plug in the front cylinder head ;-).
Using the oil filter for scale, it appears to be about 12 inches wide, roughly the same as the VFR/RC30/RC45, with front bore center spacing of 85-90mm, precluding massively oversquare (2:1 bore/stroke ratio) F1 dimensions. A 990cc five with a more typical 1.6:1 ratio, about same as current Superbikes, would have a 74mm bore, which would fit comfortably within the apparent bore spacing.
The resulting 46mm stroke--same as the GSXR750--would allow revs up to 16,500 assuming 5000 ft/min mean piston speed. That could be supported by conventional valve springs rather than requiring F1
With ample power on hand now and no practical limit on displacement for street bikes, I don't see high revs and the attendant valve train gizmos as an economical performance development strategy. However, variable valve timing--and perhaps even electo-magnetic actuation--could be the key to maintaining performance under increasingly strict emission regulations.
As I understand it, intake/exhaust valve overlap--a period between exhaust and intake strokes when both valves are open simultaneously--confers considerable power gain at peak RPMs but dirties the exhaust with unburned hydrocarbons at lower revs. Variable valve timing, whether mechanical like Honda's VTEC or something more exotic, might permit the relentless pursuit of horsepower to continue at environmentally friendly emission levels.
I saw these things on a Speedvision show a while back, some mad scientist sort of stuff they were making fun of. None the less, it worked, it was loud as hell!, and according to the designer/engineer it could "theoretically" rev as high as he wanted it to. His were completely electric, and they were very, very loud. If I remeber correctly, he had it on a small Ford motor, a 2.3 pinto motor or something, and it had no oil on top of the motor. Kind of weird, but it appeared to work.
Diesels run on compression ignition so there is no spark to advance. Engine speed is thus limited to the speed at which the fuel can burn, generally around 4500 rpm for a car sized engine. As stated above, the main purpose for the camless engine is emission control.
Direct injection 2-strokes are anything BUT simple. Why do you think we haven't seen them yet?
Complexity in one area sometimes exists to create simplicity in another. Look at an old fashioned ignition system: distributor shaft, rotor, cap, condenser, points. All of these things are failure prone (except the shaft). Look at an advanced electionic system: crank trigger and a "black box." Seems simpler, but in fact, the "black box" has dozens of components and thousands of lines of software code. Still, it is far more effective and reliable.
Good analysis, Dan. An important thing to remember is that F1 engine lifespan makes ducati's WSB engines seem ultra reliable. An F1 team will go through an engine in qualifying alone. One team's F1 engine budget is larger than the whole GP industry. When Honda got out of F1 ten years ago, they spent $100 million per year. Costs have increased dramatically since then.
Hence, your observation that we have plenty of power capability as is is significant. On my way to work this morning, I was thinking about brakes vs engines. Any sportbike can stop much faster than it can accelerate. Much of this discrepancy is due to an inability to apply the power the engine has available.
This is the most significant area of improvement we will see in the next generation of sportbikes.
As for your thoughts on overlap, you are right, but there is more. High overlap at low RPM is kindof like a big valve, high lift and huge carb at low rpm. The engine makes less power and runs rougher. Variable valve timing (like my Pathfinder has) allows high torque down low and high hp up high and smooth running everywhere.
Your last point about the valve overlap is very interesting. I believe that Toyota's newest engines (used on the ECHO and Prius) have zero degrees of valve overlap at idle, but the VVT-i system increases that valve overlap at higher RPMs depending on the engine speed, throttle position, and a host of other variables to deliver more power. That makes them clean, and powerful.
Actually the Honda VTEC system is quite a bit more exotic than the Toyota VVT-i, making use of two very different cam profiles changed at a set engine speed. That's above and beyond what would be required to simply change the valve overlap.
1 - 20 of 37 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.