Valve train forces..

[JustGav]

Out of interest,

Does anybody know how much force the cam pushes the valve down with?

I've seen figures of around 70lb per valve (for the springs I'm using), obviously I've not specified the engine as it is still just a thought, but I might try it on a spare a-series engine I have...

This is not on a 2JZ engine.

[Kranz]

Rough calculation... 50g valve & tappet mass, 7000 rpm then peak desirable force (for good cam wear) = approx 400N or 40Kg

[JustGav]

Rough calculation... 50g valve & tappet mass, 7000 rpm then peak desirable force (for good cam wear) = approx 400N or 40Kg

Hmmm, thanks for that, it is a good start for the idea..

[Kranz]

Hydraulic or pneamatic actuators????? :)

[Digsy]

I assume you mean 50g each for the valve and tappet, not 50g for them combined? :)

400N is pretty conservative, too. I'd say you could go to 500N without issues. I've done a valve spring recently that was over 600N at its fully open height, but that had a relatively high installed (valve closed) force, so the working stress range was within normal limits. if you try to force the spring through too large a stress range when opening and closing you can get fatigue problems.

Its also down to what contact stresses on the tappet face you can live.

If you have an idea of cam lift and duration, maximum engine speed (overrun), and cam base circle, and some basic valvetrain info (solid or hydraulic tappets, etc, etc) I can have a bash at working it out.

Just noticed you mention the A-series. That'll be pushrods then, so things will probably start to load up at high RPM because of the high valvetrain masses.

[JustGav]

Hydraulic or pneamatic actuators????? :)

Bingo.....

[JustGav]

Just noticed you mention the A-series. That'll be pushrods then, so things will probably start to load up at high RPM because of the high valvetrain masses.

Theory is most of the valve train will disappear, just the valves and springs will remain.

[Digsy]

Cool :)

Well, if you are going to some kind of actuator, the force will depend almost entirely on how quickly you want to open the valves.

[JustGav]

Cool :)

Well, if you are going to some kind of actuator, the force will depend almost entirely on how quickly you want to open the valves.

A couple of more critical problems such as finding 8 actuators that will be small enough/cheap enough to work and then additionally they need to take the heat and duty cycle required by something like an engine...

[Digsy]

A couple of more critical problems such as finding 8 actuators that will be small enough/cheap enough to work and then additionally they need to take the heat and duty cycle required by something like an engine...

I think you'll have to add the mass of the moving parts of your actuator onto the valvetrain mass, too.

Sounds like an interesting project, though :)

[JustGav]

I think you'll have to add the mass of the moving parts of your actuator onto the valvetrain mass, too.

Sounds like an interesting project, though :)

It is all theory and something I've been bouncing around for a few months in my head... the obvious advantages of something like this is that it is a infinitely variable camshaft emulator...

Downsides are
1. Controlling it (Not overly difficult firing 8 actuators tho)
2. Costs

The main question of course being, why haven't engine manu's used it, they must have thought of the idea in the past.

[Digsy]

The main question of course being, why haven't engine manu's used it, they must have thought of the idea in the past.

Its currently an area of massive research - as far as getting a system into volume production on an in-vehicle engine is concerned.

We already regularly make and sell single cylinder engines with electro-hydraulic valvetrains for reserach purposes.

Advances in "conventional" variable valvetrain like VVTL-i (dual cam phasing and dual cam profile switching) and Porsche's Variocam systems have offset the necessity for a mass production fully variable valvetrain. Also, BMW's Valvetronic system is already pretty close to what a camless valvetrain can do anyway, at a fraction of the complexity.

[Kranz]

My tutor/head of department at Uni was working on this back in 1990!!

Maybe he's got it working now??? Professor Sheldon, Coventry University. There may be some SAE papers written by him & others that would give you more info?