The manufacturer claims 150 horsepower for the brand-new roadburner parked in your garage. The magazine tests say 130. After careful break-in you take your baby to the shop for its first tune-up, where the mechanic puts your bike on the dyno. "115 horsepower? What the...? I've been robbed!" The obvious explanation is different dynamometers measure power differently, and give varying values. But what's more important is that the peak horsepower number is a fraction of the information available to you from a dyno run. We spent a day at Factory Pro Tuning (manufacturers of both inertia and eddy current dynamometers) with owner Marc Salvisberg and got the lowdown on dynos--how they work, how to use them and what the numbers mean.
Torque and Horsepower
To get a handle on what a dynamometer actually measures, it is important to know what torque and horsepower are. Torque is the moment of a force, acting at a distance, which tends to cause rotation. The obvious analogy is a torque wrench: 250 pounds of force acting on a two-foot wrench is 250 x 2 = 500 foot-pounds of torque. Torque is what gives a bike that instantaneous lunge when you twist the throttle, for things such as wheelies and launches off the line.
Because we are dealing with spinning wheels and rotating parts, work is defined as torque exerted through an angular distance. Imagine a shaft that you turn using a torque wrench, and it requires a constant 500 foot-pounds to turn. Move the shaft a half turn, and you've done work. Power is work done over a period of time, or essentially, a measure of how fast you can spin the shaft while exerting that 500 foot-pounds of torque.
power = work/time = (torque x angular rotation) / time
Sorting out units and conversion factors, we find that:
horsepower = (torque x rpm) / 5252
In other words, horsepower is a direct function of torque and rpm, and for a constant torque, horsepower will increase as rpm goes up. Power gives a bike its roll-on times, quarter-mile and top speeds--anything that takes time to accomplish. Figure 4 shows a dyno chart for two bikes (each with identical peak horsepower numbers), but one bike makes its horsepower at a higher rpm than the other. The reason? Less torque with more rpm can make the same horsepower as more torque at less rpm. Other things being equal, which bike would you rather ride?
But what about the dynos?
Measuring crankshaft or countershaft power requires bolting the engine to a platform and connecting the dyno's input shaft directly to the engine--a time-consuming process that often requires custom-made fittings. The dynamometers available to most sportbike riders measure power at the rear wheel, which doesn't require disassembly because the motorcycle's rear tire can spin the dyno's input shaft. This number is lower than numbers generated on an engine dyno due to friction losses in the drivetrain--the spinning transmission gears, the chain flapping around, and so on. Manufacturers always cite the crankshaft number, because it is the highest.
An inertia dyno works on the theory of, um...inertia, which is a property of rotating objects determined by their shape and mass. A large-diameter object has more inertia than a smaller object of similar weight, and that extra inertia requires more torque to accelerate. On an inertia dyno, your bike's rear wheel is spinning a heavy drum that has a known inertia value, and the engine is used to turn and accelerate the drum. Accurately measuring the acceleration over a small amount of time gives a value for torque (torque = inertia x acceleration), and horsepower is determined from the rpm of the engine and torque value. A typical run starts with the motorcycle revving just over idle, in a high gear and turning the drum. The throttle is opened, the engine revs up to redline, accelerating the drum. Acceleration and rpm are monitored, from which torque and horsepower at any rpm can be calculated.
A brake dyno incorporates some form of varying load, or brake, that the engine is forced to turn. The load is controlled to turn at a constant speed, and torque is determined by measuring the force on the brake. While water and hydraulics are a common form of brake utilized, we are mostly concerned with the eddy current brake dynamometers popular in the motorcycle industry. The drum of the dyno--spun by a motorcycle's rear wheel--is connected to the rotor of an eddy current brake; the actual brake looks something like a huge car alternator. Coils in the stator are controlled by a varying signal, and the resulting magnetic interaction (the eddy currents) places a rotating force on the stator. This force is measured with a strain gauge, and is representative of a motorcycle's torque. A typical step test run on a brake dyno begins with the motorcycle at idle and in a tall gear, turning the drum. The throttle is opened, and the engine revs to a set rpm, say, 2500. The brake is applied by the computer so that the rpm is held constant at 2500 with the throttle wide open. When the load and rpm reach equilibrium, strain on the brake is measured (and, hence, torque and horsepower), and the throttle and brake are released. This is repeated for each rpm increment--often 1000--up to redline. A step test can be conducted using any throttle position.
Because of the way torque is measured, eddy current and inertia dynos can give different readings. One measures at a constant engine speed, the other with a rapidly varying engine speed. What goes on inside your engine is subtly different in each case, and each manufacturer's software accounts for these and other factors in different ways. Our TL1000R test unit (June 2000) measured 102.1 horsepower on a Factory Pro eddy current dyno, and 118.9 horsepower on a Dynojet inertia dyno. According to Salvisberg, Factory Pro eddy current dynos read 14 to 15 percent lower than say, a Dynojet inertia dynamometer in the 90 to 100 horsepower range, with even more difference at higher power levels.
An engine's output depends on the quality of air it breathes, and it is therefore essential to take into account variances in air pressure, temperature and humidity when measuring horsepower. Raw numbers are generally normalized to sea level conditions within a dynamometer's software using a standard correction factor. However, this does not mean you will get identical readings from two different dynos, or for that matter, the same dyno on two different days. While the dynamometer corrects the horsepower it reads to standard atmospheric conditions, it cannot account for jetting changes you should have made to account for the weather. For example, you could run your bike at the local dyno and see 100 corrected horsepower on a cold day and return--without changes--on a hot day for another run and get 98 corrected horsepower. Where's the two horsepower? To get back to 100 horsepower, you'd have to lean your bike out for the hotter weather.
Tuning using the dyno
With correction factors, weather conditions and not only the type, but also the model of dyno playing a role in reading accurate horsepower, you've probably figured out the most important rule of dyno tuning: Always use the same dyno, under conditions as close as possible to the original run. This means returning when the weather conditions are the same, running your bike at the same engine temperature every time and changing only those parameters you want to test, one at a time.
Because the eddy current and inertia dynamometers operate on vastly differing principles, using each dyno to tune a motorcycle requires an accordingly different approach. Figure 7 (on page 41) is a typical graph from an inertia dyno. Since horsepower and torque are graphed for virtually every rpm, a skilled dyno operator can tell by the smoothness or roughness of a curve if there are rich or lean spots in the jetting.
Acceleration runs to get horsepower readings on an inertia dyno take seconds, so there is ample time to experiment.
Trial and error can play a big role in tuning. For example, making a run, then covering part of the engine's air intake for a second run and comparing the charts will tell if an engine is running rich or lean.
An eddy current dynamometer, because it can provide a loading condition, can force the engine to hold a certain rpm during a test for several seconds at a time, which--while extending the time required for measurement--allows tuners to analyze the exhaust gases at that rpm. These gases take time to reach equilibrium, making analysis somewhat impractical for inertia dyno use, as there isn't time for the levels to settle. Carbon monoxide (CO) is the most common indicator used to determine a rich or lean condition, and a skilled operator can tell how rich or lean from the CO values.
Residual oxygen is also measured, and this is used to determine if a stagger in jetting for different cylinders is required. Additionally, an eddy current dyno can be programmed to act like an inertia dyno for quick runs, or to perform a "sweep test" for horsepower numbers throughout the rpm range.
Well, which is better then?
Because they don't have a brake or sophisticated feedback system, inertia dynos are simple and relatively inexpensive to manufacture--an individual run can be a fairly cheap proposition. A test takes seconds, and gives horsepower numbers for the entire rpm range. With no brake (although some inertia dynos do have a brake, it is only used to slow the drum after a run--it cannot be controlled as on a true brake dyno) it is virtually impossible to take reliable exhaust gas readings, and this makes carb or fuel injection tuning on an inertia dyno a somewhat hit-and-miss affair. One thing to watch for on an inertia dyno is the fact that it takes into account the inertia of the entire drivetrain (it all has to be accelerated), and adding or removing weight to the rotating parts will decrease or increase horsepower respectively. For instance, putting a lighter rear tire on your bike can give you a higher horsepower number on an inertia dyno. (In fact, your bike will accelerate faster as if it has more horsepower.) Eddy current dynos are more expensive to build due to the requirement for a brake and controller, and although each incremental run takes a few seconds, it adds up to several minutes to ascertain horsepower over a wide rpm range--expect each run to cost slightly more than using an inertia dyno. With exhaust gas data available, tuning is a more methodical process than on an inertia dyno, and fewer runs will be necessary to dial-in your jetting for a performance part. Additionally, eddy current brake dynos can perform a step test at partial throttle positions, which is essential for fuel injection programming. Because making any changes to your bike--a pipe, ignition advancer, cams and so on--invariably results in different jetting requirements, the exhaust gas analysis used most effectively on eddy current dynos can be an invaluable tool for carburetion or fuel injection setup.
This story originally appeared in the February 2001 issue of Sport Rider.