The Dyno and What it Measures
The dyno I use is not my own. It would be very nice to have one, but for me it would merely be a very expensive toy. Dynobike, located in Moorabin, in Melbourne’s south east, is where I go to test all my bikes. The dyno used by Dynobike is a Dynojet Acceleration Dynomometer. Most people will know it as a rear wheel style dyno.
This type of dyno measures the acceleration of a big steel drum that the rear wheel of the bike runs on. The acceleration is recorded by the data acquisition component of the dyno’s software. Given the inertia of the drum is a known constant, it is relatively simple for the program to calculate the Power delivered to the drum at any time during the run.
To calculate the Torque from this we need a corresponding engine RPM value. This is because Torque and Power are related by a simple mathematical equation. It is, however, a relationship that many people do not understand.
To put it simply, Torque is the amount of work done and Power is the rate at which the work is done (or the amount of work done during a given time). The equations for these calculations are as follows:
For the imperialists: Power = Torque X RPM / 5252, where power is horsepower, torque is foot-pounds.
This means, for the sharp eyed among you, that any dyno curves displaying both torque and horsepower for the same measured run, in imperial terms, will cross at 5,252 rpm.
For the metric people: Power = Torque X angular velocity, where power is Watts, torque is Newton-metres and angular velocity is radians per second.
At this point many of you may be lost, but this is the system I learnt at university and makes far more sense to me than the imperial system (which really does talk about a single horse pulling a load out of a mine shaft). Expanding on this, there is 2 X PI ( the number you use when calculating the dimensions of a circle, or 3.142 roughly) radians in 360 degrees. Therefore, the conversion from RPM to radians per second = RPM divided by 60, multiplied by 2 X PI. This works out at about 10 to 1. We then divide the result by 1000 to achieve an answer in kilowatts.
The purpose of explaining all this is to show that you cannot modify an engine for torque as opposed to power. Many people confuse the two terms in this way. An engine that makes lots of low rpm torque will usually be called a “torquey” engine. One that makes lots of high rpm torque will be called a “powerful” engine.
Digressing a little, it is also not true that a v-twin will produce more torque at the bottom of the rev range than a 4 cylinder engine due only to the configuration. Most 4 cylinder engines are tuned for lots of high RPM torque, at the expense of low end. Because the twins don’t rev so hard (as a general rule) they are tuned for max torque at lower revs. The length and shape of torque curves is getting much better for most engine types as development progresses, though, regardless of configuration. If you look at a dyno curve for an R1 Yamaha, it will out torque any production litre bike you put it up against at any engine speed.
The truth is, an engine of a given capacity will make a certain maximum amount of torque at some point in its rev range. Where this maximum occurs depends on the engine style and tune. The best example of this is a comparison between 2 valve and 4 valve per cylinder Ducati 900 cc engines (900SS and 916). These two engines make about the same maximum amount of torque, roughly 65 ft-lb. This figure, incidentally, is approximately 90 Nm, or about 100Nm per litre, which is pretty much a given standard no matter what kind of engine it is. You just get to choose the RPM at which you want it.
The 900SS will produce a rounded curve with a peak at around 6,000 RPM (70hp) in fuel injected form, whereas the 916 will produce an almost flat, gently climbing curve that extends all the way to 9,000 RPM (110 hp). At 9,000 RPM, the 900SS is making about 45 ft-lb (80 hp), and is just about to hit its rev limiter. A carburetted 900SS has long since packed up and gone home, due to its long inlet manifolds. At this speed it makes about 30 ft-lb. Something like a ZX9-R will probably keep its 65 ft-lb curve going strong to 11,000 RPM, which is the reason Kawasaki can extract around 130 rear wheel hp from them.
It is interesting to note at this point that an engine like the modern Honda CB750, with its twin cams and 4 valves per cylinder, has an almost identical torque (and consequentially power ) curve to a Ducati 750 SS, which is a single cam, 2 valve per cylinder. The reason being they are both designed to work up to around 7,000 RPM. A GSXR 750, designed to work to 13,000 RPM, will make nearly twice as much power.
So what does all of this mean? To make more power, either you work at increasing the amount of torque your engine produces in its given rev range, or you increase the torque at the top end of the rev range.
There are two main ways to do this. Improve the engine’s volumetric efficiency to trap more air inside the cylinder, or increase the amount the trapped air is squeezed. There is actually a third way, explored by restricted race class engineers the world over. This involves minimizing power losses throughout the engine, and is usually a very involved and time consuming process.
I try to improve the engine’s volumetric efficiency with simple changes, so my usual modifications don’t involve engine disassembly. The reason for this is that many customers are scared by the thought of their bike being pulled apart, and that it is time consuming to do correctly (which is money consuming)
I am quite happy to pull engines apart should you wish. I just don’t do it if it’s not necessary.
Another of the things I have learnt from the dyno is that you need to understand what it is telling you. A major point is that the highest power will come from an engine that is on the lean side mixture wise.
When I took my Guzzi Sport 1100 down to Duane Mitchell’s ( Duane is Ultimap, formerly Fuel Injected Motorcycles, or FIM) to remap the fuel injection to suit the engine, inlet and exhaust mods I had made, I finished the day with a bike that ran so nicely it was just wonderful. The sort of nice that puts a real smile on your face weeks later, be it cruising around or pulling hard past 7,000 between corners. The smile vanished though when I went to the dyno and found I had lost 5 hp. We found the hp again, by leaning the full throttle mixture out by about 7%. On the road, all was fine until I spent a day in the hills working the engine hard in 4th and 5th gear. It was very obviously flat at the top end. A flatness not obvious belting through the gears around town, but a very real flatness when working the engine hard. Relaying this info to Duane, I realised I was telling him something he already knew.
On the basis of this info, my method of dyno tuning has changed somewhat over the years. I now find the most horsepower I can using different fuelling and spark advance as required, make a nice graph of this power using a spreadsheet and then richen the mixture slightly to get the desired air/fuel ratio.
At this point, it is most important to thank Steve (now deceased) and Dave at Dynobike for all their help. Fourteen years ago (in 1994) when I first met them they were helpful enough to encourage me to learn more and tolerant of my many questions. The freedom they have allowed me in the use of their dyno is something you don’t get very often with such an expensive piece of equipment. Very much appreciated.