Variable Valve Actuation

[PetrolTed]

Got sent this over the weekend. Might be of interest to someone...

www.pattakon.com/

[deltaf]

Thats bloody good!!! I couldnt see how it worked at first, but the animationsa make it clear.
Im sold on it! I particularly like the fact that the lift is in fact continually variable infinitly.
What a smart guy. Surely some manufacturer will bite his hand off for the manufacturing rights???

[dinkel]

So, Ted, how much did ur TVR gain with this gadget?

[funkihamsta]

I'm not sure.
The forces acting on that spindle thing that doesn't seem quite attached at either end looks a bit frail. And what exactly makes the spindle holder return to its original location? Springs? Where?

[Mad Dawg]

Thats clever. Hats off to the guy. Simplicity is the hallmark of genius, or so I am lead to believe.
springs? no prob, you could just have your usual spring retainer and collet at the top of the valve stem, along with a cup for the, er, connecting balljoint type thing. As for its strength thats more a development issue than a design issue - get the right steel with the right heat treatment and you're sorted!
Wonder if it can be adapted to pushrod V8's....

[vrooom]

Hmm

its similar no valves or anything
just sphere thingy that spin.

www.coatesengine.com

pic of it

www.coatesengine.com/eGallery/pages/RotaryValveSystem.htm

[funkihamsta]

I like that a lot. If you have seperate sphere sets for exhaust and intake you could have some sort of gearing to alter durations.
However looks like it would knacker any power gains from air inertia. And if the sphere holds the air charge and rotates to present it to the piston chamber, why should the air come out of that sphere? What about swirl turbulence? I can't see how you'd scavenge the chamber properly, so you'd always leave burnt gases around. Doesn't look like a performance item to me.

[Mr2Mike]

Rotary valved engines have been around since before the war. The big problem is getting the valves to seal reliably. A moving seal which is exposed to combustion temperatures/pressures is a huge challenge

[Manolis]

funkihamsta said: I'm not sure.
The forces acting on that spindle thing that doesn't seem quite attached at either end looks a bit frail. And what exactly makes the spindle holder return to its original location? Springs? Where?

Yes, initially the return springs seem necessary to restore the ‘control lever’, just like the additional strong springs in the valvetronic system of BMW. As you can see in the animations and images, the ‘control lever’ is ready for locating and securing restoring spring (it is formed like bird beak to secure the one end of restoring springs).
Return spring for restoring the control shaft to its rest position seems also necessary.

In practice the normal valve springs proved more than adequate for the complete mechanism. The engine works having its original eight springs, one per valve, and nothing else.

As you can see from the photos of the engine, the only thing that acts on the two control shafts is a string, previously rotating the butterfly of the carburetor, coming from accelerator pedal. No restore springs at all. You simply release (partially or completely) the acceleration pedal, and the control shafts return immediately.

For the oscillation of each control lever, a cam lobe of the camshaft is acting at one direction, while the valve spring acts at the opposite direction to restore control and valve levers. The system works perfectly from below 500 rpm to 7000 rpm. Analyzing carefully the forces acting on the control lever, one can explain why.

As regards the strength of the constituents components:
The valve lever – that small pushrod with ‘sphere like’ ends – is about 25 mm long and has 4 mm diameter. The contact surface at one end to the relevant control lever and at the other end to the ‘cylinder like’ valve adjuster, is many many times larger than the contact surface between upper valve stem and rocker arm ‘valve pin’ of the normal systems, like Toyota Celica VVTLi. You can laugh, but when the parts were given for surface hardening, one of the eight valve levers was lost. Even if they were sorry for that, the engine could not work with only seven valve levers, so another one were made and is currently working inside the engine without any hardening!
The control lever works just interposed between cam lobe and valve lever. The success is that the loads pass immediately from the cam lobe to the valve lever (and then to the valve). There is no significant stress to the control lever, so it can be light. And less weight of the control lever means lower inertia loads and higher revs limit (as regards valve train system).
There is no need to mention anything for control shaft strength, I hope.

[Manolis]

funkihamsta said: I like that a lot. If you have seperate sphere sets for exhaust and intake you could have some sort of gearing to alter durations.
However looks like it would knacker any power gains from air inertia. And if the sphere holds the air charge and rotates to present it to the piston chamber, why should the air come out of that sphere? What about swirl turbulence? I can't see how you'd scavenge the chamber properly, so you'd always leave burnt gases around. Doesn't look like a performance item to me.

When you select wild camshafts and extensive overlap to increase the power output of an engine, you just improve the breathing at high revs and heavy loads, degrading at the same time the operation of the engine at lower revs and light to medium loads.
At the pick power the inertia of the exhaust gas cleans the chamber and helps (with the suction) the initial acceleration of the fresh charge to the cylinder.
Looking at the plot ValveLift to CrankshaftAngle at www.pattakon.com ‘Theory, images, photos’ page, you can see that the top curves, red for exhaust and blue for intake, can be as wild as you like and have as much overlap as you decide, in order to achieve the scavenging of the chamber and tune the operation at high revs and heavy loads. So the system is able to scavenge the chamber and to exploit the inertia of the gases at high revs and loads. So you can get the power you like.
At the conventional systems at lower revs and light loads the extensive overlap and the wild camshafts spoil the behavior of the engine: it cannot work clearly, it cannot idle constantly, in the intake manifold a lot of burned gases return, the emissions are out of limits etc, as the standard system have only one way of operation, e.g. 11 mm of valve lift, 280 degrees of camshaft duration and 60 degrees overlap. The entry of the mixture into the cylinder is very slow, the turbulence weak and the mixture cannot be homogenous.
The new system corrects exactly this weakness. The entry speed of the fresh gas into the cylinder is about constant at all revs and loads! The engine can rev at e.g.7500 rpm and full load with valve lift of 11 mm and the same engine can work at 1500 rpm and one third of the max load with less than 1 mm valve lift and very small actual overlap. In both cases, full load at top revs versus light load at low revs, the entry speed of the gas into the cylinder is about the same, and the ability of scavenging the chamber by the exhaust gas action, similar. The same apply also for all intermediate stages in engines operation. Who needs 11 mm valve lift at 2000 rpm when 3 mm lift is more than adequate? If with 3 mm the cylinder is completely filled at 2000 rpm and you offer not 3 mm but 11 mm valve lift, the only result is an entry speed about four times slower.

[Manolis]

Mr2Mike said: Rotary valved engines have been around since before the war. The big problem is getting the valves to seal reliably. A moving seal which is exposed to combustion temperatures/pressures is a huge challenge

There is no relation of the Variable Valve Actuation mechanism with rotating valves and pure sealing. The proposed VVA system improves the operation of the normal valve train system introducing the ability to control the lift of the valves and the actual overlap, in order to have more power at high revs and at the same time: plenty of torque from extremely low revs, better economy, correct manipulation of partial loads, improved drivability and lower pollution. It seems like advertising spot, but yes, all of them.
If the two control shafts of the proposed system are locked in any available position, the system is in nothing different than the existing valve train systems.
A short definition of the system could be that it is like having infinite number of camshafts in the engine and select each moment the best couple of them – for intake and exhaust – to work.

[v8guinness]

How does the system account for wear, most valve trians allow for shims or use hydraulic tappets, there is a lot of sliding contact so there will be some wear, even with hardened surfaces...

And with regard to the pulling piston engine also detailed on the site, I would suggest that you don't have sinusoidal motion of the piston at all, and the motion is as a conventional reciprocating piston engine... the text implies the motion is sinusoidal.

Interesting concepts.

[funkihamsta]

Well i'd take the VV widgets over those orbs of delight anyday.

Given you have just appeared and are from Greece, l take it you are intimately involved with this product. Hello!

What electronics does your demo car use to control the lift?

[andytk]

The ball valves look like a great idea. Funkihampster, I think you're missing the way these valves work. The ball valves open and close like normal valves but they give greater areas for airflow thus reducing resistance and increasing flow. If you search around it says somewhere that the site the bloke who did all the leg work has worked for loadsa big companies like Rolls Royce, BMW etc so you would think he knows his way around a cylinder head.

Also the engine they had running was turing out 470hp at 5500rpm. Not bad for a 5 litre ford clunker.
Even for a completely modern 4 valve per cylinder V8 thats really good going for the revs. Any N/A engine that kicks out 100bhp per litre normally only does so above 6000rpm.
OK so you would lose the air inerta effects but if the head is designed properly I don't think that scavenging would be a problem. Certainly doesn't seem to be on their test engines, re. emissions etc.

I love simple but brilliant ideas and this is deffinently one of them.
I think that a conversion kit on a Chevvy block in an ultima would be great.
They even said that they'd gotten the (ford V8) engine up to 14000rpm.
Didn't mention the power though but still thats pretty amazing. I'd like to see standard roadgoing valve train do that. (motorbikes and F1 engines excepted)

Andy

PS and this other thing looks cool too but its complicated. Complicated things break a lot. KISS.

[deltaf]

Very elegant concept Manolis. I like the fact that it would seem possible to transplant it onto most engines. Have you approached any manufacturers yet? Surely theyd bite your hand off to be a part of this idea?
Once again, well done mate! I like it!

[Mr2Mike]

Manolis said:There is no relation of the Variable Valve Actuation mechanism with rotating valves and pure sealing.

Yes, I understand the principals behind this system. My comment was directed to vroom who put the link up about the coates engine.

[ultimasimon]

Wow! This new technology is definately radical. If they are available for a SBC then I am seriously interested.

The only thing I didn't like about the chevy engine is its inability to rev without iminent destruction being a likely contender. This changes things.

Thanks for the interesting facts Manolis.

[Manolis]

v8guinness said: How does the system account ....Interesting concepts.

The small cylinder between valve lever and upper valve stem is the valve adjuster. For the moment it is mechanical, but its form and location is proper for hydraulic too. To achieve the appropriate valve clearance, you have just to slightly shorten the height of this cylindrical adjuster at its lower side. It acts like the shims in conventional engines.

As regards sliding contact and wear:
The valve adjuster pushes the valve stem only along the valve guide direction, so the valve guide can long live.
The valve adjuster slides along a ‘hole/slider’. At high lifts the thrust force is negligible as the valve needle reciprocates nearly parallel to the ‘hole/slider’ axis. At low lifts the thrust is small partly because the inertia loads are small, partly thanks to the only-partially compressed valve spring, and partly due to the small angle between valve needle and ‘hole/slider’ axis.
As typical engines operate most of their life at partial loads and low to medium revs (i.e. calling only for low to medium valve lifts) the mechanical friction falls and the expected life of the valve train system becomes much longer than conventional, where the strength of the valve springs –necessary to return the valves to their seats at maximum revs- loads excessively everything (cam lobes, cam followers, valves, valve guides, valve seats, timing belt) at all revs and loads, even at idling.
It would be helpful to compare the expected sliding contact and wear of the VVA system to that of the valvetronic of BMW, which is currently a commercial success.

As regards pulling piston engine, you are right in that piston motion is not sinusoidal. The pure sinusoidal motion is not related to the pulling piston engine but to the harmonic engine, the ‘father’ of pulling piston engine.
The harmonic engine is described in the last page of the web site. This engine has absolutely harmonic reciprocation of its piston and, without using any external balancing shaft is compared as regards inertia vibrations only to the Wankel Rotary engine. Even twelve cylinder engines cannot beat this single cylinder in inertia vibrations! We bet on it. We still have the prototype of this single cylinder engine. It is 350 cc, with 80 mm stroke. It revs from 1000 to 9000 rpm (24 m/sec mean piston speed), free standing on the floor, without any tend to leave its place. You have just to download and run the www.pattakon.com/educ/harmonic.exe Quick Basic program and see the way it works. Besides its perfect smoothness, it also revealed a significantly improved behavior in the combustion process, too. The additional time offered to the mixture – compared to the conventional engines where the piston stays for less time near top dead center – for combustion is the main reason.
In the pulling piston engine the time the piston stays near top dead center (where the combustion takes place) is longer than harmonic engine and much longer than conventional engine. Actually the motion of the pulling piston engine around its top dead center is identical to the motion of the piston of the conventional engine around its bottom dead center, supposing same stroke to connecting rod ratio. So the mixture has more time and better conditions to be prepared and burned effectively. Start thinking of a Diesel engine able to burn efficiently at 6000 rpm (now the limit is around 4000). Think of a two cycle diesel, self-scavenged by the action of its piston, without addition pump. Think of a racing spark ignition engine, able to increase its revs limit. The compactness and lightweight of the star arrangements… If you are familiar to the Fourier Analysis (mathematics) you can use the Balance program at last page of the web site to see all above mathematically.

[Manolis]

funkihamsta said: Well i'd take ...
What electronics does your demo car use to control the lift?

Hello from Greece.
What electronics we use at the prototype car: nothing.
The foot of the driver presses the accelerator pedal, the string from the accelerator pedal rotates the two control shafts and the engine provides more power to the car.
What is necessary for a system like this?
The spark advance has to be remapped as the operational conditions are different from the original engine. The spark must be retarded especially at low lifts and partial loads as the combustion is more effective and quick, otherwise it knocks.
The carburetor also has to be replaced with multipoint injection system near intake valves or even inside cylinder (always with lamda sensor at exhaust).
The intake manifold has to be replaced with a set of bell mouths (butterflies are useless now).
The relation of the rotation angles of the two control shafts has to be also controlled to take the best result according operational conditions.

[Manolis]

andytk said: The ball valves look like ...
Andy
PS and this other thing looks cool too but its complicated. Complicated things break a lot. KISS.

Looks complicated, agree.
But is it?
Compared to what?
And what it offers in exchange of its complication?

Remember: the system offers infinite valve lifts, from zero to a maximum for both, intake and exhaust valves, and as it controls the actual overlap too, it does not need variable valve timing system or other supporting systems. It works with pure mechanical control (the rotation of the control shafts). No hydraulics, no pumps, no pipes.

Compare it to the Toyota’s Celica VVTLi, which is a ‘two steps’ system.
VVTLi needs camshafts with double lobes.
It has rocker arms with both, slipper follower and roller follower.
It has sliding pins, springs and hydraulic system to control its condition.
It needs a Variable Valve Timing system to support it.
If we had to choose to built one of the two, which is the simpler?
The result also counts, as the now wild and unpredicted Celica could turn into an even faster but also civilized car, with enough torque from idling to the revs limit and without the need to work the engine between 6000 to 8000 rpm to take the best.
The proposed VVA is like having not two cam lobes per pair of valves, as Celica does, but hundreds, and activate each moment the best one.

Compare the VVA system to the Honda’s VTEC, which is a two or three step system.
VTEC needs camshafts with multiple lobes.
It has complicated rocker arms.
It has sliding pins, spring and hydraulic system to control its condition.
It uses, in the latest versions, Variable Valve Timing system to support it.
Again, if we had to choose to built this or the proposed system, which is the simpler? And which one could work better?

Now compare the system to the BMW valvetronic, which is the real competitor as it is a continuously Variable Valve Actuation system.
Valvetronic involves too many parts.
It has lots of interfering joints conveying the motion from cam lobe to valve stem.
It needs extreme accuracy in manufacturing of the constituent parts.
It can work only in close cooperation with an effective Contiuously Variable Valve Timing system (Vanos).
It needs additional strong springs.
It has wear and stress concentration points.
The valvetronic can only control the intake valves. It is not proper for exhaust valves.
The valvetronic changes together the lift and the duration, which makes it unable to effectively manage the condition of heavy load at low revs.
The valvetronic needs drive by wire and electromotors.
The valvetronic cannot reach really high revs.
The valvetronic cannot be applied on existing or on small cheap engines.
Even though, the valvetronic is absolutely successful. It is currently used in all cars of BMW with 4, 8 and 12 cylinders, doing what it promises: lower pollution, better economy, smooth idling, easy starting, it is driver friendly etc, etc.
We could never build a ‘valvetronic‘ working prototype. Only well-equipped industries could.
Finally, yes it is complicated but it does not break at all.