Automotosports Ford 2.3L Turbo Cam
This data was originally on the AMS webpage; it has since been removed.
Preserved here for posterity and for those who wonder about JR's cam.

I. Specifications:

Intake Valve Lift: 0.420"
Exhaust Valve Lift: 0.420"
Intake Valve duration @ 0.050": 215
Exhaust Valve duration @ 0.050": 235
Lobe Separation Angle: 116°

Cam Material: Special cast alloy designated for roller follower type cam use
and currently used by TRW, Federal Mogul, and other OEM roller cam makers.

Automotosports Original Writeup:

The decision to design a camshaft for the turbocharged Ford 2.3 engine was made
after we realized that the Ranger roller cam was one of the more effective
street camshafts available for this engine. The Ranger cam has less than 200
degrees of duration at.050 inches of lift and the maximum lift is less than
0.400 inches, by anybody's standards this is a mild camshaft. However practical
experience has shown that the Ranger cam has great low RPM drivability and
performs well in drag racing motors even those that have been modified. The
Ranger cam is effective not because it is optimized for the 2.3 turbo engine,
but rather because it doesn't do anything that is too inconsistent with making
power at low RPM in a turbocharged motor.

More than likely the Ranger cams good performance comes not from small lift and
duration but rather from valve opening and closing events that are reasonable
for a boosted engine running at low RPM. It seemed reasonable to us that the
Ranger cam probably wasn't optimized in terms of lift, duration and maybe the
particular timing of the opening and closing events. Who would argue that the
Ranger cam wouldn't work better with just increased lift? Additionally isn't it
reasonable to think that some increase in duration would strike an effective
compromise between drivability and performance? Possibly the events could even
be improved.

It might seem reasonable that we would have just contacted any number of
experienced cam grinders and asked them to design us a good camshaft. We tried
this but found that some grinders suggested we take what works on a naturally
aspirated engine and apply it to a turbo charged engine. Other grinders
suggested we pay large sums of money for an un disclosed design process but
didn't seem to be aware of the fundamental differences between boosted and
non-boosted engines.

Many of the people we spoke with didn't mention two considerations that we find
essential when designing a street turbo cam:

1. Cylinder pressure and fluid mass is greater not just during the filling and
compression phase of crank rotation but also during the expansion and emptying
phase of crank rotation. This puts an emphasis on cylinder emptying and imposes
certain limitations to valve timing that are different than those of a naturally
aspirated engine.

2. In the Ford 2.3 motor the exhaust manifold pressure is very high and can
exceed 2 times the intake manifold pressure at maximum RPM. This condition works
in opposition to emptying the cylinder and again imposes certain limits on
cylinder emptying.

The two observations above have important consequences. The first being that
in order to empty the cylinder the engine must work to push a larger mass of fluid
out of the cylinder in a given amount of time and it must push this fluid into a
highly pressurized manifold. Both of these conditions create a pumping loss and
a larger fraction of the engine torque is used to empty the overfilled cylinder
against the high exhaust manifold pressure. This condition is aggravated when
cylinder filling is improved or boost is increased.

The second consequence is that valve overlap (that is having exhaust and intake
valves open at the same time) is less effective and the enhanced filling that is
normally brought about by mass flow inertia in the intake and exhaust tracts is
diminished. In a worst-case scenario the gas flow will reverse as cylinder and
exhaust gas following its downward pressure gradient enters the intake tract.
One way to envision this is the following: If the cylinder has mostly emptied,
both valves are open, intake pressure is 15 PSI and exhaust pressure is 30 PSI
what is the direction of flow? The inertial momentum of the flowing intake air
can and does partially offset the reversed pressure gradient, however at street
engine speeds the effect of manifold pressures can predominate. This reversion
phenomenon not only reduces cylinder filling but also can increase the
likelihood of detonation because the cylinder retains hot exhaust gases with
reactive combustion end products.

We hypothesized that improved cylinder emptying was paramount to improving
performance of the Ford 2.3 turbo motor. One clue we had towards this was the
observation that performance increased very significantly when improvements were
made to the exhaust tracts of this motor. Another way of considering this is: If
you artificially fill the cylinder shouldn't you also help to empty the cylinder
in order to maintain the designed relationship between intake and exhaust flow?
Isn't boosting somewhat analogous to increasing the intake valve size and if you
did that wouldn't you also increase exhaust valve size or somehow compensate
exhaust flow? Its useful to remember that Ford, when designing this motor,
probably had two considerations that they held as higher prioritizes than we do.
One being non-boosted performance and the other being economy of not redesigning
the exhaust tract to accommodate 15-20 PSI of boost. Ideally we would offer a
new cylinder head, camshaft and exhaust manifold but a relatively inexpensive
camshaft is a cheaper way to solve some of the problem.

So we had our hypotheses. Now how did we go about testing them and developing
the camshaft?

Initially we collected specifications on as many 2.3 camshafts as possible, if
possible we also collected torque curves for the various camshafts. We compiled
this data and then tried to make sense of it using knowledge gained from an
extensive literature review, cam design seminars and our knowledge of hydraulic
and fluid flow theory. At that point we started to see some trends that we felt
supported a performance hierarchy of valve event parameters for the turbocharged
2.3 motor.

At this point we were ready to test some ideas. In order to reduce cost and
speed up the design process we used engine simulation software. We began by
entering the known camshaft data into the simulation program in order to verify
its use for our situation. The torque curves generated were quite accurate when
compared to the known torque curves. We then tested our ideas on what the
important design parameters were. After evaluating many parameters such as lift,
duration, intake opening and closing, exhaust opening and closing and overlap as
well as how these parameters interacted we came up with a couple designs that we
thought would work well. We ran these designs in the computer to verify them and
then set off to have them ground.

There are two ways to go about having a camshaft ground: The first is to have
the grinder try and match your design parameters with a cam master that is already
available. Often times this master is not for your particular valve train and the ramp
designs don't work very well. Additionally you'll probably compromise to some extent
on the exact events. This method seemed unacceptable to us in the long run not
only because you take what you can get but also because it produces a situation
in which valve train stresses are unknown and durability becomes a question.

Alternatively, you can design the lobe so that it has appropriate ramps for the
particular valve train and then have a master ground to your specifications.
This is ideal and is also expensive but once you know what you need it is the
most direct and accurate way of producing the desired cam.

We chose to test our proposed cam by grinding prototypes with currently
available masters and then once we had verified the computer simulation data and
the effectiveness of the cam we paid to have custom masters ground.

In the end we are very satisfied with the result. The cam is ground on a new
cast alloy that is specific for roller followers. The ramp design is correct for
the 2.3 valve train and promises good long-term reliability. The performance is
a great improvement from the Ranger cam. It produces great low RPM performance
that is equal to the Ranger cams off idle performance. The valve events that
determine dynamic low RPM compression are actually very similar to the Ranger
cam. By 2500 RPM the torque is better than the Ranger cams and this torque
advantage gets greater as the RPM increases. The torque peak is about 1000 RPM
above that of the Ranger and the peak torque, total horsepower and area under
the power curve are all greater than that of the Ranger cam. The best part for
many is that our turbo cam is a bolt-in on stock cylinder heads and requires
no additional machining or parts. However, we suggest that you always verify
clearances when installing any camshaft; machine work or large cylinder head
manufacturing tolerances could cause differences that could lead to inadequate
clearance. If you maintain stock maximum engine speeds the stock valve springs
are even sufficient. Running a softer spring has the advantage that it reduces
parasitic power losses from the increased work necessary to open stiffer
springs, however an improved quality spring with small increases in stiffness
over stock is not a bad idea.