12/1/1997
Building Horsepower Through Dyno Testing
By Dave Emanuel
Pick up a copy of any racing-oriented automotive publication and
as you thumb through its pages you'll notice a smorgasbord of
ads from engine builders. Most of them will have an assortment
of horsepower figures large enough to rival the national debt.
Of course, all the numbers are suspect because, in most cases,
they were obtained on the advertisers' own engine dyno. But that
doesn't alter the fact that if a shop is to successfully sell
high performance street or race engines, it has to have horsepower
data at hand. Prospective customers expect to have at least an
approximation of the power their dollars are purchasing. But there
are other reasons for getting friendly with a dyno. They have
to do with knowledge and long term success.
Although they try to convince their customers otherwise, engine
builders are human and as such are subject to make mistakes. Those
mistakes, even if relatively minor, can have a significant impact
on power output and engine longevity. If an engine leaves a shop
without ever having been run under load, virtually every aspect
of its performance profile is questionable. There's simply no
way to know if all the critical mating surfaces are sealed properly,
if the rings are seated, if the cam is broken in or if the engine
makes enough power to fall out of a tree. This type of knowledge
is not only technically enlightening, it's also extremely useful
in avoiding embarrassment.
For years, the majority of NASCAR Winston Cup engine builders
used their dynos solely as quality control devices (thereby minimizing
embarrassment). Engines were broken in and checked for power output
and if all parameters were within expectations, the engine was
deemed race ready. It wasn't until the mid-1980s that the Winston
Cup contingent began using dynos extensively for research and
development projects. That change of direction came about through
necessity; as the level of competition touched new heights, complacency
and luck were no longer sufficient to win races.
The situation in NASCAR racing has its parallel in the independent
high performance engine building business. So many shops now use
dynos, it has become almost impossible to compete without one.
Mike Osucha of MORE Performance, Charlotte, NC, has been using
a dyno for years to develop and refine both the high performance
late-model street engines and full-tilt race engines he builds.
"An engine dyno is an essential tool for a race or high performance
engine builder because it gives you the most efficient means of
developing optimum spark and fuel curves and establishing a baseline
before an engine is installed in a car. Even if you're building
high performance street engines, you have to know the amount of
power they make and how they need to be tuned. If you don't give
your customers some guidance, they'll wind up looking pretty foolish
and so will you."
Those have become important considerations because over the past
few years, the only way for an engine builder to be able to deliver
a competitive product is to continually refine each package according
to the demands of a particular market. It is therefore, no longer
possible to simply assemble a group of components, call it an
engine, and leave the tuning to someone else. Customers expect
a complete combination that works. But customer expectations aside,
once an engine builder develops that "ideal combination,"
it's back to the dyno to determine how to take it to the next
level.
With the advent
of electronic engine management systems, dyno testing has taken on a new dimension.
Special air/fuel and ignition calibrations are routinely developed for engines like this
highly modified L98 small block Chevy which is "on the pump" at Cottrell Racing Engines of
Chaska, MN.
Assuming you're not a member of the "Flatheads Forever Society,"
and have recently installed a dyno or are about to, the next task
at hand is to wrestle it into generating enough income to pay
for itself. That's partially accomplished as soon as a dyno is
up and running. Like all other machine shop operations, dyno testing
expenses are added to the cost of an engine. Most check-out/base
tuning sessions last between one-half and a full day and are billed
at $350 to $500. If a customer wants to do additional development
or tuning, a daily rate is applied.
Some shops also rent out their dyno facilities, charging a daily
rate for the dyno and operator, and adding the costs of any additional
services required. Dyno rental is a double-edged sword. On the
one hand, it puts the dyno to work generating income while it
would otherwise be standing idle. On the other hand, it ties up
shop personnel and can be disruptive to other activities. Most
times, successful rentals involve a shop's existing machine work
customers, e.g., other engine builders in the area who purchase
machining services and do their own assembly or installations.
Racers and street performance enthusiasts are rarely profitable
renters because they expect too much and want to pay too little.
Mike Osucha of MORE Performance checks the plugs
between dyno runs. Osucha has relied heavily on dyno testing in the development of both naturally aspirated and
supercharged LT1 engine combinations.
Although the return on investment may not be as direct or easy
to see, using a dyno to improve horsepower and quality ultimately
generates the most profitable results. Many shops make it a policy
to dyno test every complete high performance or race engine before
it goes out the door. This practice ensures proper break-in of
all parts, allows for optimization of fuel and spark calibrations
and establishes horsepower and torque profiles. Lessons learned
on the dyno can then be applied to subsequent engines to increase
power, improve durability or both.
With known power figures on file, determining the cause of poor
performance, once an engine is installed in a car, truck or boat,
is much easier, and is also instrumental in minimizing finger
pointing. If a vehicle doesn't meet performance expectations,
and the engine's power output is known, the engine builder is
pretty well off the hook - assuming rational minds prevail.
Unfortunately, that's not always the case.
Whether doing development work or qualifying a completed engine
prior to shipment, it's important to remember that absolute power
numbers are less important than comparative data generated by
in-house testing. Owing to the equipment itself, the facilities
in which it's installed and the integrity of the operator, variations
in power readings from one dyno to another are commonplace. Rather
than chasing a ghost, it's far more productive to compare results
from a single dyno. The focus should be on verifying that an engine
makes the power it should, not on whether it makes as much as
those of a competitor.
Dynos are not engine specific,
but some of the equipment needed for testing is. That includes headers, starter and bellhousing. The owner of this dyno tests
a wide variety of engines so all the necessary equipment was on hand before this small block Ford came through the door.
Correction factors are supposed to bring all test results to equality,
irrespective of atmospheric conditions or test site location.
A "standard correction" factor is commonly used and
equates output to the results that would have been achieved had
the test been run at sea level, with an inlet air temperature
of 60F and no humidity, and a barometer reading of 29.92
in/Hg. But the correction factors computed based on existing ambient
conditions aren't always entirely accurate and if conditions are
measured incorrectly, the correction factors will become an incorrection
factor. Even with fully computerized dynos, errors, be they unintentional
or otherwise, can be made.
That's one reason to monitor all conditions with precision when
doing development work. Garry Grimes of Grimes Automotive Machine,
Alpharetta, GA, notes, "Every temperature and pressure that
an engine is subjected to is important, and the smaller the gains
you're looking for, the more important it becomes. In many instances,
engine builders are now trying to find three or four horsepower,
but you can gain or lose more than that if certain variables vary
significantly from one run to another.
"As an example, oil temperature is critical. If it varies
more than a few degrees, it will have a noticeable impact on horsepower.
Most dyno systems record just about all the data you need, but
many experienced people overlook a lot of that information. They
look at horsepower and torque and if the numbers go up, they think
they're on the right track."
Keeping accurate records of test conditions and results is also
important because it allows comparisons to be made between an
engine's output when it was freshly built and when it's ready
for reconditioning. Many engine builders have learned a lot about
building reliability by testing race engines when they come in
for a rebuild. In so doing, it's possible to determine the parts
or preparation techniques that minimize power loss as mileage
accumulates. Lessons learned from that exercise can be put to
good use in improving future engines.
In the electronic age, dyno beget computers. Three are shown in this photo, taken at Grimes Automotive Machine. One computer
runs the dyno data acquisition system, the second monitors engine operating conditions
and the third reprograms the PROMs that hold air/fuel and ignition calibrations.
Note that in addition to the electronic sensors used by the dyno's data acquisition system, a full complement
of AutoMeter gauges are mounted in the control console.
One facet of dyno-less engine building that compromises a shop's
ability to produce competitive horsepower is that there's no means
to evaluate the effectiveness of components. Without dyno testing,
it's extremely difficult to determine whether one camshaft produces
more power than another and the rpm range in which power is concentrated.
Obviously, in-car testing at a race track is an alternative, but
it's very time consuming and results are often influenced by the
car itself.
Having built a strong case for dyno testing, I would be remiss
if I didn't include a quote from noted engine builder Bill Jenkins.
A fixture in Stock, Super Stock and Pro Stock drag racing from
the 1960s through the early '80s, Jenkins has said on more than
one occasion, "Once an engine goes in a car, forget everything
you learned on the dyno and start all over again." Jenkins
wasn't stating that dyno testing was of no use, he was noting
that what works on the dyno may not work on the race track.
This appears to be the case more with drag racing than with oval
track or road racing because of the extremes in acceleration rates
and starting from a dead stop. In fact, since the advent of dynos
with programmable controllers, the gap between race track and
dyno cell has narrowed considerably. And intelligent interpretation
of data narrows that gap even further.
Far too many people, racers and engine builders alike, look only
at the biggest number in the horsepower column. But if an increase
in peak horsepower is achieved at the considerable expense of
power in the heart of the curve, it's a good bet that on-track
performance will be worse instead of better.
Mike Hedgecock of Eagle Racing Engines, Knoxville, TN, demonstrated
that point a number of years ago. He built a 358 cubic inch small
block Chevy and dyno tested it with a number of camshafts. One
camshaft increased peak horsepower (at 7000 rpm) from 536 to 555.
But between 5000 and 6000 rpm, the "big cam" was down
at least 11 hp at every test point. This particular engine had
been built for short track racing and it spent most of its life
between 5000 and 6500 rpm. So the driver could never take advantage
of the higher peak horsepower of the longer duration camshaft
and, in fact, lap times were better with the cam that made less
horsepower.
Like any other piece of shop equipment, an engine dyno is a tool
that must be used correctly for best results. When it is, the
outcome can be as enlightening as it is profitable.