[Guest]

I dreamed this up at work the other day, and I don't know much about the subject so this is probably wrong.

Anyways I was mulling over the Honda VTEC system (what little I know of it). If you don't know how it works: 2 sets of "lumps" on the camshaft, 2 for low rpm, one for high rpm, with the high-rpm located between the low rpm ones. The high rpm lump has a greater radius from the center of the camshaftEach triplet corresponds to one valve, and each lump has a lifter that it strikes. During idle and low rpm, the low rpm lumps strike the low rpm lifters, which are attached to the valve. When the revs reach a certain point, say 4000 rpm, hydraulic pressure caused by the increased revs slides a bolt through the single high rpm lifter, which attaches it to the valve. That means the high-rpm lump comes into play, and you get increased airflow because of the increased duration and uhhh, i don't know the other one. Anyways, that's why VTEC engines are such screamers at high rpm, because they have switched over to more aggressive cams. Some of this is probably wrong, but you get the general idea.

So anyways, I was thinking this was pretty complicated, and there should be a simpler way of doing this. So, instead of messing round with hydraulic bolts and different lumps, why not move the camshaft itself closer to the lifters? At idle and low rpms, the camshaft would be at it's furthest point, and therefore the lifts would be at the least duration and extent. As revs increase, the camshaft would be moved progressively closer to the lifters buy some hydraulic pressure or something. This would increase the duration of the lifts as well as the extent of the lifts because the lumps would be striking the lifters sooner, for a longer period, and move them a further distance outward. It would function much the same way as the bigger lumps on the VTEC camshaft. An advantage of this over VTEC is that it's a lot simpler plus it's continuously and infinitely variable.

Now, I figure someone has either already thought of this, or it's completely unfeasable for some reason or another. Hopefully Mike93PGT can give me some answers as the resident tech-guru. But hey, if I thought of putting Air in the front of shoes before Nike did, then I can think of this right?

 

[Mike 94PGT]

will,

Not to burst your bubble, but this won't work.

The reason is that by moving the camshaft, you are not changing the distance that matters: the distance from the base circle to the tip of the lobe (the "lump"). The lifters, be they solid tappets, HLAs or even manually adjusted, use the base-circle as the "closed" reference. Theoretically, if you moved the camshaft closer to the valves, the lifters (HLA) would adjust the movement out, resulting in the same net lift or, in the case of a solid valvetrain, the valves would be held open on the base circle by the amount you moved the cam minus the lash that existed before the cam moved. Also, because one cam serves 3 cylinders (in the V6, for example), you'd be affecting valves on cylinders that are on different cycles.

One variation on this idea is something I think BMW played with and that involves moving the fulcrum for "rocker-arms" (followers, really) to effect move or less overall lift. You might have heard of aftermarket rockers for pushrod Chevy valvetrains that are "1.7:1". The stock rockers were 1.6:1, so the extra tenth resulted in the valves opening a bit further; a free lobe upgrade. The BMW idea took this one step further and put the lift in the hands in the PCM; by moving the fulcrum hydraulically, the valve would lift more or less.

 

[Stage36]

not a bad idea though... i think that he means that the cam would actually sit above (not in contact) to the lifters and the lobe would be ground so that it contacts the lifter "oh so gently". this could work on certain engines but is definately NOT infinitely variable. i read this thread thinking i might have a solution for my horrible idle... i hated leaving my foot on the gas at stoplights to bring the revs up to 1.25K.

maybe in order to keep the lifter in contact with the cam, a low tension spring is needed between the retainer and lifter. if im thinking correctly, the closed radius of each valve would have to be larger and not very steep.

what do you think mike?

 

[mxblitz]

Now I know very little about engines but here's an idea... why don't they just use computers to open and close the valves rather than relying on a camshaft? Wouldn't you pretty much get max performace at every rpm then? Just an idea.

 

[asso]

the valves are still mechanical... they'd have to redesign the entire engine..

 

[Mike 94PGT]

Now I know very little about engines but here's an idea... why don't they just use computers to open and close the valves rather than relying on a camshaft? Wouldn't you pretty much get max performace at every rpm then? Just an idea.

It's not such a wacky idea, and "they" are working on it. The most promising solution is to use powerful solenoids to act on the valve stem, opening the closing the valve using PCM commands. It would, in principle, give much more flexibility in valve control than traditional VVT schemes.

The problem is the actuators. Right now, the magnetic flux densities aren't high enough to make the system workable without huge, power-consuming solenoids. Also, the speed of the actuation is limited, putting a cap on engine RPM.

See http://www.engineering.ucsb.edu/~ywang/research.htm for some notes on the subject.

 

[Guest]

Another problem with the solenoids is they're either open or closed, theres no middle ground. They might be able to work around that, but currently that makes any sort of VVT impossible with those engines.

the distance from the base circle to the tip of the lobe (the "lump"). The lifters, be they solid tappets, HLAs or even manually adjusted, use the base-circle as the "closed" reference. Theoretically, if you moved the camshaft closer to the valves, the lifters (HLA) would adjust the movement out, resulting in the same net lift or, in the case of a solid valvetrain, the valves would be held open on the base circle by the amount you moved the cam minus the lash that existed before the cam moved.

That could be worked around by having the base-circle not come into contact with the lifters at all at the furthest setting of the cams (low rpms), and when the cams are closest the base circle can come into contact with the lifter (high rpms), but not extend it past closed.

Of course having the lifters actually striking the camshaft instead of rolling on it would build up some friction which would be a problem, and might cause some wear and tear. I guess they could be made from titanium or something. Does oil circulate around the cam system? I'd say, if it doesn't then an extension of the oil system could be put in the heads to reduce that friction.

Also, I'm not sure how it would matter if a certain cylinder was on a different cycle. Do you mean firing sequence? I guess I'm not sure what a "cycle" is. Cylinders don't generally have different air and fuel intakes and ratios between each other do they?

And Mike, you didn't burst my bubble because I didn't have one to begin with.

 

[Mike 94PGT]

Another problem with the solenoids is they're either open or closed, theres no middle ground. They might be able to work around that, but currently that makes any sort of VVT impossible with those engines.

Well, VVT would be this system's strong point. Since the actuation of the valves wouldn't be tied to the crankshaft position (like a cam belt/chain gives), the PCM or valve controller has the ability to open the valves at any point in the cycle. Early, late or anywhere in between. All it would take is a VVT is what this system lives for.

In addition, by using small springs that the magnetic force would have to work against while opening the valve and by modulating the strength of the field, the valve opening distance could also be variable.

That could be worked around by having the base-circle not come into contact with the lifters at all at the furthest setting of the cams (low rpms), and when the cams are closest the base circle can come into contact with the lifter (high rpms), but not extend it past closed.

Of course having the lifters actually striking the camshaft instead of rolling on it would build up some friction which would be a problem, and might cause some wear and tear.

I don't think friction would be the problem. I think the simple "impacting" force would ruin the surfaces in short order. Even on "regular" engines...if the valve clearance is too loose, the pounding can break through the hardened surfaces and cause metallurigcal damage. The damage would be much like that seen on chisels: mushrooming on the valve stems.

In old engines, often you need to use a file to remove this from the tips of the valve stems before you can slide the stem through the guide to remove it.

I guess they could be made from titanium or something. Does oil circulate around the cam system? I'd say, if it doesn't then an extension of the oil system could be put in the heads to reduce that friction.

There are few materials hard enough to withstand that kind of punishment in the long term.

Think about rods bearings: when one gets loose (and I'm talking about just a few thousandths-of-an-inch loose), the fact that there is pressurized lubrication there doesn't stop the knocking and damage that occurs.

Also, I'm not sure how it would matter if a certain cylinder was on a different cycle. Do you mean firing sequence? I guess I'm not sure what a "cycle" is. Cylinders don't generally have different air and fuel intakes and ratios between each other do they?

I was thinking about the camshaft moving "en masse" to suit the needs of, say, cylinder #2 while on the intake stroke. The other cylinders on that bank are at different points in the cycle (i.e. intake/compression/power/exhaust)and so moving the cam, without special attention paid to how lash is taken up, will result in those intakes opening.

I think I see that I misinterpreted your original idea (I think...) so this may not actually apply.

 

[mxblitz]

ok here's another idea then.... what if instead of using a variable valve timing system, they created an infinitly variable transmission? I know this probably sounds pretty stupid but if that could be done, then the engine would only have to run an one speed and the valve timing could be designed to give max performance at that speed.

 

[Mike 94PGT]

ok here's another idea then.... what if instead of using a variable valve timing system, they created an infinitly variable transmission? I know this probably sounds pretty stupid but if that could be done, then the engine would only have to run an one speed and the valve timing could be designed to give max performance at that speed.

Honda had one with the Civic HX and Subaru had/has a CVT for the Justy. Don't know if they still make the car or import it to North America though.

Or, you could think about hybrids that use a small engine, running at a fixed and very efficient speed, to power a generator that charges batteries that power electric motors.



[Edited by Mike 94PGT on 05-14-2001 at 08:29 PM]

 

[mxblitz]

how come everything I think of has already been invented?