Veeduber[_2_]
August 4th 09, 11:51 PM
VW Heads - Part 1
One of the most common errors seen in VW engines converted for flight
is the failure to maintain proper valve train geometry. Valve Train
Geometry is based on a series of dimensions which use the center-line
of the crankshaft as their foundation. In basic terms, the valve
train takes the PUSH produced by the lobes of the cam shaft, transfers
it through 180 degrees and uses it to PUSH on the valve. Given that
the valve must be 'pushed' at exactly the right moment and the fact
that the power needed to actuate the valve must be transmitted through
arcs -- one on the Input side of the rocker arm, the other on the
Output -- an error of only a few degrees in the alignment of the
components may reduce the output of the engine by as much as twenty-
five percent.
And that's a bunch.
The reason the losses are so great is because they cause the engine to
work against itself. In effect, any error of alignment or volumetric
imbalance must be made up before any useful power can appear in the
crankshaft.
A nice example of this -- and an error frequently seen in VW engines
-- occurs when the sealing surface of the combustion chamber is re-
newed. This surface is a given distance from the center-line of the
rocker arm shaft, and this distance must be perfectly equal in both
heads.
To measure this distance you must first establish a Standard Surface
on the cylinder-side of the head. On older heads and some newer after-
market heads, the spigot bores are surrounded by a flat surface that
is used as our Standard Surface. On newer (stock) heads each spigot
bore is provided with three small lugs that are machined true when the
head is fabricated. These lugs serve as our Standard Surface.
Rather than measure to the center-line of the rocker-arm shaft, we use
the gasket rail. The height of the rocker-arm shaft is then
calculated rather than measured directly. But that would be an
exceptional case. In most cases all we need to know is the DEPTH of
the spigot bore relative to our Standard Surface, and the HEIGHT of
the gasket rail above our Standard Surface. This dimension should be
virtually identical on both heads, with a tolerance of about 0.0015,
plus or minus. This dimension -- the distance between the combustion
chamber sealing surface and the gasket rail -- is the foundation for
sitting-up our valve train geometry.
Google 'valve train geometry' and virtually all of your references
will be for the OUTPUT side of the rocker arm, with various
explanations as to why the mid-point of the rocker-arms arc MUST
coincide with the mid-point of the valve's travel. All of which is
true enough. But what goes unmentioned is that the same rules apply
to the INPUT side of the rocker arm. That is, the mid-point of the
push-rod's travel MUST coincide with the mid-point of the arc
prescribed by the rocker arm.
The input is measured in the same way as the output, and while you can
build a simple jig for setting the rocker-arm shaft height, you can
achieve the same result using the actual head.
WHY we are doing all this has to do with the losses that always occur
when linear motion is converted to rotary motion, as on the input
side, and rotary motion is converted back to linear motion on the
output side. Not only do we want to keep those losses as small as
possible, we want the motion to occur at exactly the right moment.
Unfortunately, when the spigot bores are opened up to accept larger
cylinder barrels, the depths of the four spigot bores are rarely in
agreement. As mentioned above, an error of plus or minus one and a
half thousandths is acceptable but this degree of precision is
virtually impossible to achieve unless the work is done on a milling
machine operated by an experienced machinist. Having the heads opened
up by the local VW guru using a drill press virtually guarantees an
inaccurate, imprecise job, making it almost impossible to dial-in your
valve train geometry. The solution is to have the work done by an
automotive machinist having the proper tools.
-R.S.Hoover
One of the most common errors seen in VW engines converted for flight
is the failure to maintain proper valve train geometry. Valve Train
Geometry is based on a series of dimensions which use the center-line
of the crankshaft as their foundation. In basic terms, the valve
train takes the PUSH produced by the lobes of the cam shaft, transfers
it through 180 degrees and uses it to PUSH on the valve. Given that
the valve must be 'pushed' at exactly the right moment and the fact
that the power needed to actuate the valve must be transmitted through
arcs -- one on the Input side of the rocker arm, the other on the
Output -- an error of only a few degrees in the alignment of the
components may reduce the output of the engine by as much as twenty-
five percent.
And that's a bunch.
The reason the losses are so great is because they cause the engine to
work against itself. In effect, any error of alignment or volumetric
imbalance must be made up before any useful power can appear in the
crankshaft.
A nice example of this -- and an error frequently seen in VW engines
-- occurs when the sealing surface of the combustion chamber is re-
newed. This surface is a given distance from the center-line of the
rocker arm shaft, and this distance must be perfectly equal in both
heads.
To measure this distance you must first establish a Standard Surface
on the cylinder-side of the head. On older heads and some newer after-
market heads, the spigot bores are surrounded by a flat surface that
is used as our Standard Surface. On newer (stock) heads each spigot
bore is provided with three small lugs that are machined true when the
head is fabricated. These lugs serve as our Standard Surface.
Rather than measure to the center-line of the rocker-arm shaft, we use
the gasket rail. The height of the rocker-arm shaft is then
calculated rather than measured directly. But that would be an
exceptional case. In most cases all we need to know is the DEPTH of
the spigot bore relative to our Standard Surface, and the HEIGHT of
the gasket rail above our Standard Surface. This dimension should be
virtually identical on both heads, with a tolerance of about 0.0015,
plus or minus. This dimension -- the distance between the combustion
chamber sealing surface and the gasket rail -- is the foundation for
sitting-up our valve train geometry.
Google 'valve train geometry' and virtually all of your references
will be for the OUTPUT side of the rocker arm, with various
explanations as to why the mid-point of the rocker-arms arc MUST
coincide with the mid-point of the valve's travel. All of which is
true enough. But what goes unmentioned is that the same rules apply
to the INPUT side of the rocker arm. That is, the mid-point of the
push-rod's travel MUST coincide with the mid-point of the arc
prescribed by the rocker arm.
The input is measured in the same way as the output, and while you can
build a simple jig for setting the rocker-arm shaft height, you can
achieve the same result using the actual head.
WHY we are doing all this has to do with the losses that always occur
when linear motion is converted to rotary motion, as on the input
side, and rotary motion is converted back to linear motion on the
output side. Not only do we want to keep those losses as small as
possible, we want the motion to occur at exactly the right moment.
Unfortunately, when the spigot bores are opened up to accept larger
cylinder barrels, the depths of the four spigot bores are rarely in
agreement. As mentioned above, an error of plus or minus one and a
half thousandths is acceptable but this degree of precision is
virtually impossible to achieve unless the work is done on a milling
machine operated by an experienced machinist. Having the heads opened
up by the local VW guru using a drill press virtually guarantees an
inaccurate, imprecise job, making it almost impossible to dial-in your
valve train geometry. The solution is to have the work done by an
automotive machinist having the proper tools.
-R.S.Hoover