PATTAKON

GREECE

The proposed VVA as an intelligent Variable Valve Timing system.

The VVA can of course be combined to any VVT system, but it seems it manages alone the effective timing of the engine.

Let's see the operation of an engine without VVT system, of an engine with VVT system and of an engine with only the VVA system.

An engine without VVT system.

It is held that the overlap is constant.

But what is really the overlap? How can it be defined?

In each cylinder there are some intake valves and some exhaust valves. The camshafts act on the valves and open them, while valve springs restore them to their close position. The opening and the closing of the valves is a function of the crankshaft angle.

The lift of each valve multiplied by the periphery of the valve gives the opening through which the gas flows entering or leaving the cylinder. This is especially correct at the partial lifts near Top Dead Center (TDC).

The overlap starts the moment the intake valves open finding the exhaust valves still not closed, and lasts until the moment the exhaust valves close.

During overlap, the cylinder communicates with both intake and exhaust manifolds.

Overlap is commonly defined as the crankshaft angle, around TDC, during which the cylinder communicates with both intake and exhaust manifolds.

But does it give a good indication (description) of how easily communicate intake and exhaust manifolds through combustion chamber?

It is only the working gas that cares about the overlap, nobody else. If we had to describe to the molecules of the working gas the overlap, the degrees of crankshaft seem meaningless. The intake and exhaust valve lift during time is what counts, and the "time / valve area" plot around TDC is the tool.

A definition of the actual overlap could be :

å [ min { Pi*Li(t) , Pe*Le(t) } ] * dt,

where : Pi : intake valves periphery

Li(t) : instant intake valve lift

Pe : exhaust valves periphery

Le(t): instant exhaust valve lift

min : the minimum of the two

å : sum of

dt : infinitesimal time step

As shown from the above diagram, where the "time / valve area" is plotted for operation at 7500 rpm (the two curves near vertical axis) and for operation at 1000 rpm (the two "fat" curves), the working gas sees 7.5 times more actual ("time / valve area") overlap at 1000 rpm than what it saws at 7500 rpm!

No matter whether the engine operates at full load or at light load, the "time / valve area" at 1000 rpm is constant. Light load means that the intake manifold pressure is very low, suctioning back gas from the exhaust manifold. Not an ideal situation, at all.

So if at 7500 rpm the engine is tuned to work at its best, then don't ask too much at 1000 rpm.

The same engine with a typical VVT system.

At low revs and partial loads the camshafts are appropriately "angularly" shifted in order to reduce the overlap, as shown in the following plot.

The typical VVT system solves the problem of overlap around TDC but can't help spoiling, at the same time, the operation around Bottom Dead Center (BDC), because at low revs it keeps open the intake valves more than necessary and opens the exhaust valves sooner than necessary.

For instance, if at 7500 rpm the intake valves close 50 degrees after BDC, and at 1000 rpm the VVT system shifts the intake camshaft for 40 degrees to reduce overlap at TDC, then at 1000 rpm the intake valves close 90 degrees after BDC.

Similarly if the exhaust valves at 7500 rpm open 50 degrees before BDC and at 1000 rpm the VVT system shifts the exhaust camshaft for 40 degrees to reduce overlap, then at 1000 rpm the exhaust valves open 90 degrees before BDC. The ideal at 1000 rpm is to open exhaust valves at BDC and to close intake valves also at BDC. The typical VVT does exactly the opposite, as the reduction of the overlap at TDC prevails.

Rover's VVC, though complicated and expensive, is an intelligent VVT system improving the operation at TDC without spoiling it at BDC.

If the total intake "time /valve area" (represented by the total surface below the relevant curve) at 7500 rpm is the appropriate one for the gas to enter and fill the cylinder, then at 1000 rpm and full load it is offered 7.5 times more "time / valve area" to the same quantity of gas to enter and fill the cylinder. So the mean entry speed of the gas through the valves becomes 7.5 times lower at 1000 rpm than at 7500 rpm! Things get even worse as the load at 1000 rpm is partial.

The same engine equipped only with the VVA system.

In this case, the plot (at 7500 rpm and 1000 rpm, all at full load) is like this

The "time / valve area" during overlap, at 1000 rpm and full load, is exactly the same to the "time / valve area" at 7500 rpm. If this "time / valve area" at 1000 rpm and full load is compared to the "time / valve area" of the conventional engine without VVT working at 1000 rpm, it is 7.5 times lower.

So the system does change the timing too, reducing the actual overlap around TDC. And looking more carefully, it does not spoil but on the contrary improves the operation around BDC, too, so it behaves as an intelligent VVT.

The curves of "time / valve area" could be thought (envisioned) as openings, through which the same quantity of gas has to pass in one second.

With one quarter of the full load at 1000 rpm, shown in the following plot, the "time / valve area" during overlap is a quarter of the "time / valve area" during overlap and full load at 7500 rpm.

With one quarter of the full load at 1000 rpm the VVA system achieves  thirty times lower "time / valve area" during overlap compared to conventional engine without VVT, operating at a quarter of the full load at 1000 rpm, as shown in the plot below.

The following curves are for full load at 7500, 3700, 1800, 900 and 500 rpm.

and below is the same plot with the VVA removed or locked at its maximum lift position.

The effective restriction of the actual overlap at low revs and partial loads combined to the constant pressure at intake manifold change dramatically the behavior of the VVA engine.

In the following two plots, the curves are for operation at 4000 rpm (above) and 1500 rpm (below), for 5/5, 4/5, 3/5, 2/5 and 1/5 of the full load.

From the preceding analysis of the VVA system, it seems that the mean entry speed of the gas trough valve openings is substantially constant at all revs and every load. So the turbulence and swirl into the cylinder as well as the homogeny and atomizing of the mixture become essentially independent upon load and revs.

The effect of VVA on Diesel operation can be explained by the previous plots, as well. It seems that the torque of the conventional Diesel engine falls quickly at low revs (compared to the revs of maximum torque) due to weak turbulence and swirl into the combustion chamber, caused by the extreme "valve time area" offered to the air for endering and for leaving the cylinder. Enough turbulance and swirl and good volumetric efficiency from very low revs are exactly what the VVA provides ungrudgingly.

Another feature of the new mechanism seems to be its capability to implement "internally" the EGR (exhaust gas recirculation) : keeping short the exhaust valve lift, compared to the intake valve lift, a good portion of "burned" (residual) gas can stay into the cylinder for the next cycle. Honda's "CRM 250 AR" project (advantages from "activated radical" combustion) opens the way, and the VVA fits to this technology.

The internal combustion engine operation does get more complicated with the additional parameters of intake and exhaust valve lifts, but the research and development of the new - if any - capabilities looks promising.

Since the new VVA manages also the effective timing of the engine, eliminating the need for additional VVT system, the result is less complicated, less expensive and more reliable. Taking into account that all present VVA systems (with either infinite or two step lift) need extra VVT system . . .

In the "Actual Overlap versus Load and Revs" plot above, they are compared the Actual Overlap (i.e. the "time valve area" during overlap) in Conventional engines, in engines having multiple cam lobes (like Honda's S2000) and in case of the VVA.

The S2000, with its two step operation, achieves weaker actual overlap than Conventional at medium to low revs, and stronger actual overlap than Conventional at high revs. This explains, partly at least, the ability of the S2000 to combine extreme power concentration at high revs, with driver friendly operation at medium to low revs. The step of the actual overlap of S2000 at 5.000 rpm is not the ideal way of operation, however it is better than nothing.

But only in the case of the VVA the actual overlap is reduced (about linearly) with the load.

Compared to S2000, the VVA combines even stronger actual overlap at high revs, with significantly weaker actual overlap at medium to low revs.

Besides, as depicted above, the transition of the VVA from high to low actual overlaps is smooth and stepless.

The actual overlap of the VVA becomes many times lower (trivial) compared to S2000 and to Conventional engines, at extreme low revs and light loads.

That is why the VVA can combine idling at less than 300 rpm, with maximum power well above 10.000 rpm.

It seems that the VVA is more intelligent.  

PATTAKON

GREECE