I just wanted to let you know how disappointed I am with the Tech Notes  article in the February 2005 edition of RRW&MT. With a tag line like,
“With Horsepower, Technique Matters,” and three images of dynamometers,  one could reasonably conclude that the focus of the article would be on  dynamometer operating and testing procedures. Indeed, I was looking  forward to reading what appeared to be four pages of technical notes on  this topic. Unfortunately, the technical portion of these notes never  materialized.
 
Apparently, the author feels as though operating procedures are  significant in the context of dynamometer accuracy. However, the  magnitude of this influence is never quantified in the text of this  article. We, as dynamometer service consumers, are encouraged to get  intrusively involved with the dyno run to make sure it’s done right,  but what’s the payoff? One percent tolerance? Twenty percent tolerance?

Perhaps, if the Test Methodology section comprised more than 20% of the  total word count, this critical detail would not have been overlooked.
 
 If errors of omission were not enough, there were also glaring errors in  the descriptions of the various dynamometer types. According to the  author, the two most common dynamometers are both of the inertial type,  but one uses an eddy-current brake, “to create a variable load which can be precisely matched to the mass, inertia and horsepower  characteristics,” of a test vehicle, provided one is willing to,  “accept the risk of error from incorrect calibration…” If, as one is  led to believe by the next paragraph, that the two dynos to which the  author refers are Dynojet and Factory Pro Tuning products, then this  description is completely wrong.
 
Dynojet’s test procedures are based on sweep tests, where horsepower and  oxygen levels in the exhaust gas are evaluated during a steady-throttle  run-up (sweep) through the engine’s operating range. In this procedure,  horsepower is derived from the angular acceleration of a ~900lb.  rotating drum that’s driven by the rear wheel.
 
This differs significantly from Factory Pro’s procedures where  horsepower and four exhaust gases are evaluated in quasi-static,
 steady-throttle, fixed-RPM step tests that increment through the  engine’s operating range. In this procedure, each fixed-RPM run is a  separate test and horsepower is derived from the steady-state force that the rear wheel exerts on a lightweight rotating drum.
 
Clearly, because the Factory Pro system calculates horsepower under  steady-state (zero acceleration) conditions, it is not an inertial type  system. Since it is not an inertial device, it need not be calibrated to the inertial characteristics of the test vehicle, thus the risk of error introduction is non-existent.
 
Note that Dynojet also makes dynos that include an eddy-current brake,  such as the model 250, but that addition brings with it both advantages  and disadvantages. The advantages are that the oxygen level readings  can be better synchronized with the power readings by controlling the  rate of acceleration, and by actively controlling the rate of  deceleration, wear on the test vehicles' mechanical systems can be reduced. The disadvantages include the fact that the horsepower readings are obfuscated even further. With the addition of an eddy- current brake to the system, the calibration software must take into  consideration both the inertial forces from the drum and the power  absorption of the eddy-current brake.
 
 What is probably the most important aspect of chassis dyno tuning was completely missing from this article. That aspect is how the various dynamometer systems are used to optimize power output. Power output  optimization is, after all, the ultimate goal of dyno tuning.
 
With the Dynojet system, horsepower is treated like a byproduct of proper air/fuel ratio selection. This is evident when a Power Commander map is developed, as this process is based on setting air/fuel ratios to  a pre-determined, arbitrary number across the RPM range. While this procedure can be accomplished very quickly using sweep tests and the integrated software suite that Dynojet has developed, the optimization criteria are based exclusively on circumstantial data. If the optimization criteria are changed to include maximum power output and smoothness, expect an exponential increase in testing time.
 
In contrast, the Factory Pro EC997 system uses horsepower figures directly. Each RPM/throttle position point is tested and optimized for peak power regardless of what the O2 numbers are. The exhaust gas chemistry is used only to guide the operator in his iterative quest for  maximum power. Compared with Dynojet's normal approach, this can be a very time-consuming process, but it always results in optimized power output. But, because the operator is presented with 200% more empirical data (power and oxygen plus three other exhaust gases and a "flux factor"), converging on optimal power and smoothness takes
 relatively less time.
 
Which system is better depends on whether you place greater emphasis on quickly achieving the same O2 levels in your exhaust gas as your competitors or getting the most power and drivability out of your
 motor.
 
Finally, I fail to see the certainty in spending a “lottery-sized” sum of money on a water-brake test cell and a staff of engineers to maintain and calibrate the system. (Use of hyperbole of these dimensions is typically discouraged within technical writings.) Clearly, if the author is willing to accept a five percent error band, a considerably less expensive and substantially more practical solution can be found.
 
 
 Sincerely,
 Tim Chin
 X Dot Racing
 Whittier, CA

 

Mr. Chin is a mechanical engineer, graduated USC, 1988.
He works for Boeing on their B-1 Program.
He owns  Dot Racing and it campaigning a gsxr1000.