h a l f b a k e r yWhere life irritates science.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Variable valve timing is so baked, its burnt, but I have a slightly different way. At least this isn't the way Toyota or BMW does it.
The system is based around a pulley with a differential inside of it. One side of the differential is attached to the cam. The other side to a gear. The gear
is turned by a worm gear attached to an electric motor (servo).
The pulley rotates, driven by the motor's crank via a timing belt (like a normal cam pulley or sproket). The differential will try to split the rotations between the cam and the gear. Since the gear is locked in place (because its attached via a worm gear) the differential will transfer the rotations to the cam (the cam will rotate 2x as fast as the pulley). This is normal operation when the timing is being held constant. Not nuetral just no further advance or retard from its current state.
When the timing is to be retarded or advanced, the motor just needs to spin the worm gear one way or the other. To visualize this consider the pulley stopped. When the gear rotates the differential housing (pulley) is locked in place by the timing belt. This will cause the cam to rotate an equal and opposite amount as the gear. This same action occurs when the pulley is turning, it just happens relative to the normal rotations of the system.
For packaging the setup could be rearranged so that the pulley is attached to one side of the differential and the housing is turned by the worm gear. Either way works, its just a matter of packaging.
Differential gearing
http://en.wikipedia...i/Differential_gear
[BJS, Feb 21 2007]
Epicyclic gearing or Planetary gearing
http://en.wikipedia...wiki/Planetary_gear
[BJS, Feb 21 2007]
http://en.wikipedia.org/wiki/Hub_gear
http://en.wikipedia.org/wiki/Hub_gear
[BJS, Feb 21 2007]
[link]
|
| |
I'm having trouble visualizing exactly what you describe.
As I see it, in terms of a car differential, you have one axle being driven by the cam pulley, the other axle drives the cam and the position of the drive shaft (the worm gear) is adjusted by a servo. If this is so the cam does not turn at 2x the pulley speed. It turns at the same speed but in the opposite direction. Seems like it would work if the gears could be designed to hold up at engine speeds, not absorb too much energy and not make too much noise. |
|
| |
This is quite a good idea, I think. If I may:
As I see it, the servo motor is connected, via a reducing gearbox, to where the driveshaft would connect to a differential gearbox.
The 'left wheel end' would be connected to the cambelt pulley, and the 'right wheel end' would be connected to the camshaft.
Turning the servo to a new position would make an angular difference between the cambelt pulley and the cam belt - and hey presto: valve timing change.
The camshaft would be going backwards, but I think this problem could be sorted out.
[SublimeGTP], a pity no one took you up on this.+ |
|
| |
I'll try to clarify the mechanics a bit.
If you've ever seen or driven a car without a locking or limited-slip differential on ice or snow, you know that when you apply too much gas, one wheel does nothing while the other spins. This is done through the "spider" gear setup within the differential. When one output is stopped, the other output will spin at twice the normal speed. |
|
| |
What this setup does is essentially mount the whole assembly by one of the outputs, and drive the spider with a belt or chain. The unrestrained output will turn at twice the angular velocity of the driven spider gear. |
|
| |
Normally, the output will remain properly phase-aligned. By rotating the "fixed" side of the spider, in this case by servo gear, the mechanism will shift the phase of the output, changing the timing. |
|
| |
This is a very clever way to address variable valve timing, but it's been done before. I've seen it used on large sterling engines to tune the phase angle to optimize torque or velocity. |
|
| |
The main problem is that on today's smaller, high-revving engines, you'd need to keep this assembly well lubricated, and that means more complexity. A failure here would also mean severe engine damage. |
|
| |
[+] for the idea, but I don't know if I'd want it in my car, considering that there are other fail-safe methods of doing the same thing. |
|
| |
Extra: if a twin cam engine was used, then the inlet and outlet valves could be timed independantly. |
|
| |
I think I understand your setup now and you're right. it is 2x the pulley speed. Still, the idea of adjusting the timing by rotating the differential case might have advantages in that it would be easier to lubricate the gears inside a fixed (except for adjustment) differential case vs trying to get oil to those gears in a rapidly rotating case. |
|
| |
This is an interesting idea, It could be combined with ball differential technology for slightly better efficiency. There are two ways this could be mounted. The cam can be mounted on the "left-wheel" side, the pulley on the "right wheel" side with the differential case being moved to advance/retard the timing. The method I think would work better, would be to mount the cam on the "left wheel side" the casing of the differential can be rotated by the cam belt and the timing can be adjusted by the "right wheel" side, this means the cam will rotate in the correct direction. |
|
| |
Doing away with a mechanical linkage to the valves would seem to be the ultimate goal, as an electronic valve could have its timing AND throw altered constantly for optimum performance. |
|
| |
Evinrude has done something similar to that (but with fuel injection) on its "2-stroke" E-TEC outboard. |
|
| |
I would think it would also be possible to vary valve timing by adding two idler pulleys to the timing belt, one on each side of the crankshaft pulley. By moving these pulleys, one would vary the amount of belt on each side between the crankshaft and camshaft(s). |
|
| |
i like this idea, it looks like you put a lot of thought into it [bun] you should maybe draw it out in 2D or 3D and post a link to a picture. that would probably help everyone (myself included) understand your desctiption a little better. |
|
| |
This idea would not work with the specifications [SublimeGTP] specified. |
|
| |
Engines work by spinning the cams at half the speed of the crank shaft, not twice the speed or the same speed (like [hangingchad] explained). |
|
| |
In order for this idea to work, the pulley would have to be connected to one side of the differential, and the cam would have to be connected to the differential casing, and the gear and worm gear would be connected to the other side of the differential. |
|
| |
If it was done exactly this way then you would have to figure out how to connect the motor to the gears or the wires from the motor to the electrical system while still allowing the differential to rotate around the gear and worm gear. |
|
| |
I suggest using a planetary gearing system instead of a differential. The annulus would be spun by the crank shaft, the planet carrier would be connected to the cam shaft, and the sun gear would be connected to the worm gear. Or, the sun gear could be spun by the crank shaft, and the planet carrier would spin the cam shaft, and the worm gear would spin the annulus. |
|
| |
I think this mechanism should also include a rotation sensor connected to the motor or sun gear. |
|
| |
It could still work as described. All that would have to be done is have a 4:1 gear reduction int he timing belt instead of the current 2:1. |
|
| |
The dictionary called. they want their word "simple" back. Apparently you have been abusing it. |
|
| |
I suppose you are right [Hunter79764], but I still suggest using a planetary gear system instead of a differential. |
|
| |
I've had this idea myself. The obvious way is the one first described: cam connected to one side gear, pulley connected to the other side gear, control mechanism connected to the planet carrier. |
|
| |
A conventional differential is a planetary gear train, though it is symmetrical in that the "sun" and "annulus" are interchangeable. Imagine a planetary train morphing: the sun gets bigger and the annulus gets smaller; both become conical so that the planets remain the same size. Eventually it becomes a differential. If the process continues it will become a planetary train again, but the roles of sun and annulus will have been reversed. |
|
| |
I've had [supercat]'s idea, too. Great minds and that. |
|
| |
"Simple" can be regained, and then some, by doing the adjustment by cable to a lever on the dash. It'll never be quite spot on, but it may be closer than you'd think if one goes to the trouble of learning a whole new driving skill. But some of us like that. |
|
| |
Now that cars have such sophisticated
electronics, wouldn't it be simpler just to
implement the whole engine in software? |
|
| |
What do you mean by that? |
|
| |
Hi, I voted against. I realise its easy to be critical of other's ideas..but here's the rub. The system you describe would actually rob the engine of power to make it work. A better but more complicated system uses the torsion in the camshaft to achieve the same aim and will be in the 2009 fords and jaguars before too long. I applaud you going with a direct drive system that has more accurate control but I fear this is occuring with a penalty in the form of lost power. |
|
| |
consider supercharging vs turbo charging. both use output from the engine to produce more to compensate, but the turbo actually uses the heat from the engine in the form of expanding gasses that would be lost unless harnessed in this way. The supercharger however robs peter to pay paul. In this way, the turbo is superior rather than the gear/direct drive supercharger, in much the same way as your system is flawed through being direct drive and taking power. The best inventions are the simplest, so consider what you lose in terms of power to achieve what you are aiming for. Ford's idea of using the torsion in the camshaft to alter timing on 2009 models apears a better idea.
I dont want to stomp your creative ideas...keep goin' |
|
| |