PART 5 - Making Sense of the Data
Story and Photos by John Copeland
Over the past few months we've been looking at different kinds of dynos and how you might find a place for them in your karting program. Finally, last month we talked about what kind of information the dyno can give you and how to collect it. Now it's time to figure out what that information means. If the object of all this effort is to make you faster on the racetrack, then you have to translate the data from the dyno into usable tuning information. Let's begin by looking at what we have to work with.
The most basic information you'll have to have is a listing of Torque values and the RPMs at which they were measured. And, of course, you'll want to be able to identify the engine and it's basic configuration, as well as the general conditions under which the data was collected (temperature, humidity, etc). Even though we know that torque is a more valuable consideration than Horsepower, you ought to go ahead and calculate Horsepower anyway, using the formula HP = Torque X RPM \ 5252.1 to get the result. Your finished table should look like this:
Now you could go ahead and graph this to take a look at it, but you'll remember last month we talked about applying the proper correction factors to the raw data so that you can compare runs made under different conditions and still have meaningful results. In fact, if you don't adjust the data for the atmospheric conditions, then, unless you happen to race under exactly the same conditions that you dynoed (very unlikely), your dyno results don't mean anything! Fortunately, the Society of Automotive Engineers (SAE) have developed a set of conversion tables to help you what they call "normalize" your data. The SAE has defined the "normal" atmospheric conditions to be 60 degrees Fahrenheit, 0% Relative Humidity, and 29.92 inches of Mercury Barometric Pressure. If you run your dyno and take your data under these exact conditions, that's great. But of course the conditions when you take your data will not be exactly like this. But by using the SAE correction factors, you can normalize your data to these conditions. What that means is, once your readings are normalized, you can accurately compare data collected on any day, under any conditions, regardless of the temperature, etc., and make meaningful analysis. Without this data correction, comparing data collected on a hot summer day to that collected in mid-winter would be about as helpful as using your kart set-up records from a 1/4 mile dirt oval to set up for the big track at Daytona! Most folks using dynos today know they have to correct their data, and most use the SAE tables, compare their conditions, and then multiply it all out. That works fine, but the tables are too lengthy to reproduce here. If you need them, give me a call at 317-742-0935 and I'll help you get them. A much slicker way to handle this conversion is by using this formula: F = 29.92 / (BP-Vp) x the square root of (460 + T) / 520. Using this formula, you'll only need the table given below.
PD Table (Figure 1)
It gives you the saturation pressure of water for your known temperature, a value we shall call Pd. Having looked that up, the rest of the factors are: Vp = Pd (from the table below) x RH, T = the observed air temperature, BP = the observed barometric pressure, and RH = the observed relative humidity. F = the correction factor that you must multiply times your torque readings to get corrected data. And, as you'll remember, since horsepower is derived from torque by the formula HP = T x RPM / 5252.1, using corrected torque will give you corrected horsepower too. Of course, if you are using automatic data collection, and if you are inputting the temperature, humidity, and barometric pressure when you make your dyno run, its a snap to program the software to use the formula and table above and make all the corrections automatically. That way, every run you make will already be normalized and your comparative data will be much more useful.
Here's the way the corrected data looks:
So now you've finally got data that is (hopefully) accurate, with all the supporting information you'll need to make sense of it later when you refer back to it, normalized (corrected) so that you can compare this run to any other run, regardless of the conditions when you made it. This, my friends, is what you came for. Armed with this kind of information you can really test RLV's newest pipe, or Crane's latest cam, or anything else, compare it to previously collected data from other pipes, cams, etc. and make an educated assessment about whether it will help you on the track. At this point you may find it helpful to get a piece of graph paper, plot the data points, and draw the torque and horsepower curves for the run we've used here. There is value in looking at both the raw data as it's shown in the table above, and the plotted curves. It's easier to compare absolute numbers using the data table, but the curve will give you a better feel for the rate of change in torque and horsepower throughout the RPM range. Next month we'll look at these tables and plots and discuss what they're trying to tell you. We'll also try comparing data from different set-ups and see how we can use the dyno data as a diagnostic tool to start tuning before you get to the racetrack. See you then.