Quote:
Originally Posted by eh?
Here's a bit from Eric Hsu from cosworth/XS engineering.
"This is an inertia type chassis dynamometer and does not have "load control", but is the absolute most accurate and sensitive dynamometer available. The method of horsepower measurement is fixed and is unaffected by PID closed loop controls from modern chassis dynamometers with eddy current power absorber units. In other words, there is no computer algorithm required to control the load and therefore possibly affect the horsepower measurement. The intertia type Dynojet is very simple and has a pair of 4,000lb rollers (a fixed mass) and the time needed to accelerate this fixed mass is the amount of power the car generates. It is so sensitive that it even measures bent wheels, cylinder misfires, rough engine running conditions, etc. that other chassis dynos cannot measure. If you can tune a car on an inertia type Dynojet with a perfectly smooth dyno graph at 0 smoothing on the graph, you can bet for damn sure that the car will run smoother than a baby's ass. Even to this day, the Dynojet remains my favorite chassis dyno and I've used them all.
While a dyno with load control is certainly convenient for mapping an engine from scratch, it isn't "necessary" for mapping AT ALL. If you want to map specific cells on your ignition or fuel tables, all you need to do is vary the throttle in different gears and you can very easily map entire fuel and ignition tables in no time. If you need more load at part throttle go up a gear. If you need more load at full throttle, go up a gear again. The Dynojet 248C has a 200mph speed limit. Don't let all those young fools or FFF tell you that a Mustang, Dyno Dynamics, or Dynapack is better because "it can control load". If you really think about it, when are you driving on the road and the car is held at a specific rpm or speed? The only time that happens is when you're doing 170mph in top gear on the way to Vegas and your STi is aero and/or gear limited. Otherwise your engine is ALWAYS accelerating and decelerating isn't it? This is also the case on an inertia type Dynojet. How's that for real world conditions?
Anyhow, this may sound like I'm giving a full on sales pitch here, but I'm not. A lot of what Koji and I learned when we were tuning regularly at XS was because of the sensitivity and accuracy of the inertia type Dynojet. While forums and younger tuners may have you thinking that "loading" dynos are better, the reality is that it's how good the tuner is. If you gave Tiger Woods some Wal Mart golf clubs and you had the latest and greatest Ping irons and Callaway drivers, you can bet your ass he'll still smoke your ass in 18 holes (probably even just 1 hole). "
Dynojet 248C Chassis Dynomometer |
Some of the points that Eric makes I agree with, however, steady state tuning becomes more important the lower the engine speed and load, due to lower port velocity. Real world, as Eric says, is a combination of various engine loads, speeds and acceleration rates. When you are cruising down the freeway at 75 , that would be considered a steady state load, as you are running at a specific rpm, and are at a specific load. The problem with sweeping through light load lower rpm map area's is that you get transient fueling which will greatly skew what the steady state fueling requirements are at a specific rpm/load site. You can see this clearly on a steady state dyno when you bring the engine up to a specific rpm/load site and hold it there it will take a couple of seconds before the mixture stabilizes. In the real world, I would agree with Eric, that tuning a car on a inertia dyno by someone who knows how to tune, is much better than having an incompetent person tune your car on a steady state dyno.
- Chris