PATTAKON

GREECE

 

 

 

Variable Valve Lift

 

Honda VTEC: a valve is activated by one of two available cam lobes. Two distinct lifts instead of one in typical engines.

Toyota Celica VVTLi: one cam lobe activates a valve below 6000 rpm, and another cam lobe above 6000 rpm. Only two distinct lifts.

Both of them are impressive steps, not the solution of the problem, combining relatively good behavior at "partial loads and low revs" with high specific output at "high revs".

 

BMW 316ti: Continuously Variable Valve Lift from zero to a maximum.

This gasoline engine operates without throttle valve. Instead, the intake valve lift determines the amount of mixture left to enter the cylinder. The small valve lift at idling, light loads and low revs acts like a mincemeat mill (or "air-gasoline mill") for:

·        homogeneous mixture,

·        entry speed, into the cylinder, close to sound velocity,

·        turb.

All of them are, if not sufficient, at least necessary conditions for good combustion (efficient, smooth, clean).

The mechanism used in BMW 316ti is based on the oscillation of an "eccentric member". The angular amplitude, A, of this oscillation is about constant. What is controlled is the starting angle, F, of the oscillation. The "eccentric member" has an initial angular interval B1 to B2 offering zero valve lift and then smoothly follows another angular interval B2 to B3 offering a continuous progressive increase of valve lift from zero to a maximum Lmax. If the starting angle of the oscillation of the "eccentric member" is F (an angle in the B1 to B2 angular interval), then there is an initial angle B2-F of "eccentric member" rotation along which the valve stays closed, then follows an angle A-(B2-F) along which the valve is opening progressively to some maximum L(F), then follows the reverse motion of the "eccentric member" from A-(B2-F) to B2 along which the valve progressively closes and then follows the B2-F angle along which the valve keeps closed. Depending upon the selection of the starting angle F, the final lift L(F) of the valve is determined, as well as the angular interval 2*(A-(B2-F)) along which the valve is opened.


Depending on the starting angle of the oscillation of the "eccentric member" the valve: remains steadily closed OR remains initially closed and then starts a progressive opening, followed by a progressive closing and then stays closed till the end of the oscillation OR starts a progressive opening and continues with a progressive closing.

 

Together with the valve lift it is also changed the time or the angular interval along which the valve stays opened. As the desirable valve lifts gets smaller, smaller becomes also the time the valve is opened.

 

 

Looking at the diagrams taken from two relevant patents, one can spot on the fact that the valve at idling, low loads and low revs can stay open for very few degrees of crank rotation.

 

It is difficult to control the exhaust valves with such a system. It seems countereffective for an exhaust valve to stay opened for only 60 degrees of crankshaft rotation.

 

In fact it is difficult to control even the intake process with these systems, as the condition "high load at low revs" is far from being effectively manageable:

At low revs and light loads it is an easy job: all the intake valves need to stay opened for just a few degrees near TDC and the valve lift be small.

However, as the load becomes heavy at low revs, things are getting difficult, as it is necessary to enter and stay into the cylinder the maximum possible quantity of working gas.

At low revs and "heavy load operation", all it takes is a small valve lift, let say 1.5 mm, with a duration of about 180 degrees to completely fill the cylinder with fresh gas. But for a 1.5 mm valve lift the only available duration is 80 degrees and for a duration of 180 degrees the only available valve lift is 8 mm. In other words, the desired valve lift with the desired duration cannot be met simultaneously. This means that it is necessary either to use valve lift of let say 6 mm just to keep the valves opened for the necessary duration, or to open the intake valves only near BDC. In the first case, i.e. using much higher valve lift than what is really necessary, the benefits of good turbulence, homogeneous mixture and increased entry speed vanish. In the second case, which implies ability for extensive intake camshaft "advancing", the valves can open only as the piston gets close to BDC resulting in energy and oil losses as piston moves downwards with the cylinder actually empty.

In general the management of such systems is difficult, making absolutely necessary the use of powerful controllers and intelligent software for the coordination of the subsystems involved. The "drive by wire" is not an improvement, just a necessity. Perhaps the main argument supporting HONDA 's claims for its one step system of its new VFR 800 versus BMW's Valvetronic.

Finally to overcome the strong resistance of the additional restoring springs opposing -through the lever- the Control Shaft, a driving electric motor is always necessary; it is an absolute necessity, nothing to do with steps ahead.

 

 

A reasonable further step would appear the creation of more suitable mechanisms, allowing stricter control for both intake and exhaust valves. 

 

 

The ideal Variable Valve Actuation system offers constant conditions to the mixture (intake manifold pressure, swirl, turbulence, scavenging of the cylinder during overlap, velocity around valve seats, mixture homogeny etc) for entering the cylinder in order to get burnt as effectively as it gets and then leave, at all revs and every load.

 

For an engine tuned for optimum performance at 6000 rpm where a 10 mm valve lift seems reasonable, it's as well reasonable that this valve lift is quite injurious for half load at 6000 rpm where the optimum lift is not more than the half (i.e. 5 mm), and equally important, also half the actual overlap. The entry speed of the mixture, the swirl and turbulence, the residual gas, the scavenging during overlap, the burning and the exhaust processes can keep similarity only by means of a really similitude in valve lift curves.

If the engine tuning at 6000 rpm proves to require a 10 mm valve lift, the same engine at 3000 rpm and full load needs only about 5 mm valve lift and at 1500 rpm and full load needs only about 2.5 mm valve lift. With half load, at 3000 rpm it calls for about 2.5 mm valve lift, and at 1500 rpm for 1.25 mm.

 

The rule seems general and simple:  "The necessary valve lift is about analog to both revs and load".

 

Applying this rule, the mixture's entry speed into the cylinder, the turbulence and swirl, the scavenging of the combustion chamber, the burning and the emissions, if not identical, are almost the same in all revs and every load. For the mixture what counts is its own kinematics, not the speed of the crankshaft or the load.

 

On the other hand, consider the case of a conventional engine tuned at 6000 rpm and full load. When this engine operates at 1500 rpm and at one third of the load, the entry speed into the cylinder is not 2 or 3, but more than 10 times lower. The overlap from blessing at 6000 rpm becomes curse at 1500 rpm. Needless to say more.

 

The advantages from the use of CVVL mechanism in an engine appear nowadays beyond any doubt. It takes only to spot on the "announcements" for the fist use of such a mechanism in a production car.

On the other hand the suitable CVVL may lie a goodly way ahead. Among its characteristics prevail:

·        the weight, height and cost,

·        the ability to work correctly at high revs, as now the engine can have good volumetric efficiency and effective combustion process from very low to very high revs,

·        the compactness, allowing space for spark plugs, intake and exhaust manifolds, cooling,

·        the effectiveness to control both intake and exhaust valve lifts,

·        the lack of weak points of excessive stress and wear,

·        the simplicity and the reliability.

 

 

The above diagram is from a proposed mechanism, which among others achieves Continuously Variable Valve Lift. In this case the angular interval of crank rotation, along which the valve stays opened, is fixed. Variable is the valve lift alone.

With such a system it is easy to control both, intake and exhaust valves, and get the optimum results according the existing operating conditions.

The exhaust appears having also considerable influence on the engine behavior. Instead the exhaust valves have to stay widely opened for some 250 degrees of crank rotation at very high revs, it seems wrong to keep the same timing/lift scheme at low revs/loads. "Ex-up" in 4-cycles and "Kaaden" in 2-cycles show the influence of exhaust control.

A more accurate and precise control over the valve lift is evident. This way it can be controlled the exhaust process too, the noise from the exhaust system, the exhaust gas recirculation and perhaps the overall performance of the engine.

So long as a ContinuouslyVariableValveLift mechanism is:

·        reliable, simple, light and cheap,

·        capable for high revs,

·        capable for zero lift, to wipe out the throttle valve,

·        short and compact to leave space for other vital systems,

·        able to control both, intake and exhaust valves,

it seems to be the solution.

It could be used in gasoline and in Diesel engines, in two cycle poppet valves engines, in motorcycle engines, singles or multicylinders.

An engine qualified for specific uses due to its high power output, could turn into an all purpose engine with the system proposed, without loosing anything in performance.

A 600 cc straight four motorcycle engine produces easily more than 100 PS nowadays. If such an engine could offer its torque from very low revs and if it could be good at partial loads and idling, then it could be used as the prime mover for cars, offering excellent economy in town traffic and keeping the 100 PS ready for the moment they will be asked.

 

According CrankshaftAngle-ValveLift diagrams, as the valve lift gets continuously smaller, the actual overlap is accordingly reduced, even without variable timing mechanism. The typical overlap - as long as the valve lift is constant - is changed only by changing the angular phase difference between intake and exhaust camshafts. When the valve lift is dramatically changed, even without a variable valve timing system (VVT) the actual overlap is changed. It is another thing to have at Top Dead Center both intake and exhaust valves opened at 4mm and completely different thing to have at Top Dead Center both intake and exhaust valves opened at 0.25mm. The first case suits to high revs and loads, the second one favors low revs and partial loads. The actual overlap is what counts. On the other hand, the typical VVT system is not a panacea, as the change of the overlap at TDC spoils relatively the status around BDC.

 

The CVVL is a simple mechanism for better control on the working medium flow (air or mixture) in internal combustion engines. Looked at from a today perspective, it seems strange the absence of successful alternatives to the standard valve train system, for more than a century.

 

 

The assembly and the basic parts of a Continuously Variable Valve Lift (CVVL) mechanism.

 

 

Basic parts from different Points of View of a CVVL mechanism

 

A CVVL at three "ControlShaft" angles, in two camshaft angles each.

 

 

 

A CVVL having roller as cam follower.

 

 

A CVVL with rocker arm and plain cam follower.

 

A CVVL at three angles of the "ControlShaft", shown right, for a straight four, 16 valve engine.

 

 

A CVVL at three angles of the "ControlShaft", shown right, for a straight four, 16 valve engine.

 

 

A CVVL at three angles of the "ControlShaft" (one per row), for five angles of the camshaft (one per column).

 

As comparing to the existing ones, the proposed type of CVVL mechanisms appears more suitable in terms of:

1.     simplicity (it has one and a "half" parts per valve or per pair of valves, plus a control shaft per row of valves),

2.     compactness (it leaves space for other vital components),

3.     height (it is too short, about a valve diameter higher than standard),

4.     weight (it has fewer and of lower weight constituent parts),

5.     reliability (there are no wear concentration points),

6.     cost (construction plus assembly plus maintenance cost),

7.     controllability (for both intake and exhaust valve lift),

8.     completeness (the variable timing becomes less important),

9.     applicability (easily applied in engines of every cost and use).

 

As long as the "reduction in consumption and pollution combined with improved performance" is (?) the first goal for all engine manufactures, the effort for effective CVVL seems worthwhile.

 

The ordinary people ask questions like:

·        Has really the mechanism so impressive influence?

·        Is it possible to use the same engine for smooth polite motion in town traffic and the next moment to use it, as is, for racing?

·        What are the implications, the negatives?

·        Does it really combine lower consumption with higher performance and easier use? Is the difference remarkable?

·        What is the overall cost, or profit, at long term use?

·        If everything is positive, then why it is not widely offered yet?

·        How much it costs?

 

Thank you for your time.

 

PATTAKON

mailto:vva@pattakon.com

Fax and Tel No:    +30 210 4934402

Post address :        Lampraki 356, 18452 GR

Nikea Piraeus

GREECE