Re: VANOS and Pneumatic Valves

[MICHAEL_KOHLBRENNER@xxxxxxxxxxxxxxxxxxxxxxxx]

From: "Frank Natoli" <frankn@domain.elided>

> Can anyone provide a modest summary of the VANOS system? As
> RPMs increase, does it change (1) valve opening time, (2)
> valve closing time, (3) valve lift? I would think that
> higher RPMs need (1) and (2), but cannot imagine a
> mechanical system that can do that, short of a variable cam
> lobe configuration. Thanks.

VANOS is a system that varies the timing of the camshaft(s).
Single VANOS varies the intake only; Double VANOS does both.

Essentially, the cam profile(s) is(are) selected to provide good
high rpm performance, which typically includes a good amount of
overlap between the intake and exhaust lobes.  By varying the
timing of one or both camshafts, this overlap is changed (or
nearly eliminated) to get rid of the "lumpy idle" usually associated
with aggressive high rpm cams.  So, it really does both (1) and (2)
above at the same time.

It is a little bit more detailed than that, but the end result
is a significantly wider torque band -- good high rpm power without
the usual penalty of poor low rpm performance.

The implementation in BMW's case (many other car companies use
the same principle with varying implementations...) is through
the use of something akin to moveable timing chain tensioners.
For example, in the single VANOS case, the intake cam is driven
by a chain off the exhaust cam.  There is a movable device, similar
to a chain tensioner, that varies the length that the chain has
to travel between the two cams, thereby changing the timing of the
intake cam relative to the exhaust.  Porsche's implementation is
similar as is Audi's, if I recall correctly.  Other companies use
a device that resides within the cam drive gear "assembly" which
continuously varies the camshaft timing relative to the drive gear.
Toyota's VVTi does this.  I think it was Alfa, also, that used a
similar method in one of the very earliest implementations
somewhere around 15-25 years ago.

As for variable lift, this is another desirable trait in such
systems and they actually exist.  The most familiar example is
Honda's VTEC, but Mitsubishi has a similar system, MIVEC, and
there are probably others.  The main issue here, however, is that
these typically consist of two completely separate cam profiles
that are switched depending on rpm.  So, rather than a continuously
variable timing system, you have a two state "on/off" system.
Based upon the shape of the resulting torque curves, neither
approach seems to be significantly superior than the other.

Clearly, the ideal case would be a system that is fully variable
in timing, duration AND lift.  Supposedly, there are plenty of
efforts underway to accomplish this, but they all seem to involve
a move from the familiar cam actuated valve systems we are
familiar with to something like computer controlled solenoid
actuation.

In another message:

From: "Frank Natoli" <frankn@domain.elided>

> Are there BMW racing engines that do not use valve springs
> to close the valves?

I am willing to bet that the upcoming BMW F1 engine uses
pneumatic valve closure.  Can't say about any of the others.

> I understand they use a conventional
> camshaft to open the valves, but compressed air to close
> them. Is this true?

Renault, if I am not mistaken, was the first to use this
principle and it has been adopted by all of the F1 engine
manufacturers since it is the only way to eliminate the
inevitable valve float you would see at the ridiculously
high rpm ranges where these engines operate -- currently
well over 17,000 rpm.

The pneumatic valve "springs" significantly reduce the moving
mass in the valvetrain which is the mainly determinant of valve
float.  Also, pneumatic "springs" provide a rising rate spring
constant rather than the constant rate associated with linear
"mechanical" springs.  This is a "good thing" in a valve spring.

> What is the source of the compressed air? Turbo? Tank? Thanks.

As far as I know, they use compressed nitrogen which I seem to
recall is stored in a reservoir (tank) to keep the pneumatic
springs pumped up.  Perhaps someone else will confirm or refute
this part -- I'm a little sketchy on it.

Regards,
Mike Kohlbrenner

------------------------------