In Part 1 I presented an argument for the need behind good quality flight testing, so now I'll offer a few specific techniques to gather the data needed to analyze your aircraft’s performance. Testing will fall into 3 main areas:
Aircraft behavior in normal, expected situations, like level and turning stalls, incipient spins and recovery.
Establishing the critical performance airspeeds, including stall speed, best angle of climb, best rate of climb and best glide speed.
Determining performance data for the aircraft’s Pilot Operating Handbook (POH), including rate of climb performance, takeoff and landing distances, airspeed calibration error, fuel burn and cruise speed tables. Additionally, several interesting things can be extracted from the data. These items might be of interest to some pilots, like glide ratio, glide polar, and max predicted service ceiling.
Simplified Flight Testing Spreadsheet
Over the years I put together a spreadsheet (download now) of various techniques to gather and analyze data for these requirements. Some of these techniques are my own, and others are modified versions of other people’s work that I’ve collected over the years. In any case, I’ve developed the spreadsheet to suite my needs, and to hopefully make them as simple and user-friendly as possible.
The spreadsheet is arranged into various tabs, each devoted to a particular task. Similar logic is used across all the tabs - the user is responsible to enter certain data, and then equations built into the spreadsheet calculate the results based on those user entries. User input cells are shaded yellow to call attention to them, but a general caution is offered that many non-shaded cells have equations embedded in them that are totally unprotected. Care must be used to not damage the equations by overwriting the contents of the cell, as well as shifting, deleting or rearranging cells that may break the linked nature of the equations. Simply put, be careful editing anything on the spreadsheet other than the yellow-shaded cells!
The remainder of this article details how to use the spreadsheet and accompanying data sheets. There’s quite a lot of text, but don’t let that prevent you from using the spreadsheet. The reality is that much of the best data can be gathered in only a couple flights, using just a few data sheets. The spreadsheet works its magic and does all the hard work for you.
Part 1: Aircraft Behavior and Stall Speeds
The first group of data that I recommend gathering is the various critical speeds of your aircraft. These speeds include the stall speed (Vs), best angle of climb (Vx), best rate of climb (Vy) and best glide speed (Vbg).
Stall speed can be determined in flight fairly easily and the observed stall speed simply recorded on a notepad in flight. A data sheet is included in the spreadsheet to simplify this process (Stalls Data Sheet).
Figure 1 – Example Stall Speed Data Sheet
The important thing to emphasize is to fly as smoothly as possible, decelerate into the stall slowly, and run the test a few times to average out errors in any one particular test. Second, don’t stop at simply testing the straight and level stall speed. Test the stall behavior in turns to the left and the right. This may seem uncomfortable, and you might want to delay the turning stall portion of your flight testing until after some time has been flown, but it needs to be done. The objective is twofold: determine that the indicated stall speed in a turn matches what you expect the stall speed to be (30 deg bank increases stall speed by 7%, 45 deg bank by 19%, and 60 deg bank by 41%), and become familiar with the feel and response of the airplane while approaching the stall in a turn. The familiarity gained during these tests is absolutely invaluable, and will make you a far safer pilot! For a great discussion of why stall speed increases in a turn, see the Bold Method article.
In addition to stall testing, this is the place to investigate the slow flight handling of your aircraft. Experiment at speeds 10-15 mph above stall speed, getting progressively slower. See how the plane feels, explore the onset of airframe buffet as the stall approaches, and get a good feel for control pressures and feedback in slow flight. Make sure you perform these tests with and without flaps – you’ll want to be well accustomed to the feeling of the airplane.
Part 2: Critical Performance Speeds
The next data set to gather is required for the various performance speeds of your aircraft. These speeds include the stall speed (Vs), best angle of climb (Vx), best rate of climb (Vy) and best glide speed (Vbg). Stall speed will be determined experimentally in Part 1, so we’ll skip over that discussion.
Best angle of climb and best rate of climb
Best angle (Vx) and best rate (Vy) are calculated from flight test data. A great way to gather the data required is to fly a “Sawtooth Climb-Descent Profile”. The aircraft is flown at a specified airspeed from a starting altitude through an altitude block (typically 500 to 1000 ft), and the time to climb is recorded. After the climb is complete, power is reduced to idle and the plane glides back down through the altitude block and the time is once again recorded.
Several tabs in the spreadsheet are related to this task, including “Sawtooth Climb Test” and “Sawtooth Data Sheet”. Tests iterations are repeated for speeds a bit above stall speed up through a slow cruise-climb airspeed, such as 55 mph to 110 mph for example.
The key to gathering good data is getting the plane established in the climb or descent and ensuring airspeed has stabilized before starting through the beginning altitude. A way to accomplish this is to start the plane climbing 500 ft before your altitude block, let things stabilize until you hit the starting altitude, then start the timer. Continue climbing as stable as possible through the top altitude, and record the elapsed time when passing the top, but keep climbing for an additional 500 ft or so. The extra climb gives you time to make the power reduction and establish the glide speed before coming back through the top altitude of the descent block. Once again descend lower than the bottom by 500 ft to get ready for the next series.
To illustrate this, consider the following example. Start at 2000 ft at 60 mph. Apply full power and start climbing, holding speed at 60. As you pass 2500 ft (bottom altitude of the test block) start the timer. Keep climbing at 60 mph, but as you pass 3000 ft (the top of your 500 ft altitude block) stop your timer and record elapsed time. Continue climbing another 500 ft to 3500 ft, then level-off while reducing power to idle. Establish a glide at 60 mph and stabilize. As you descend back down to 3000 ft start the timer, and record elapsed time when passing 2500 ft. Continue descending until back to 2000 ft, then level off. This series in now done, and you are ready to start the next airspeed in the series, say 65 mph. Apply full power, establish a climb at 65 mph, and the process starts all over again.
Figure 2 – Example Sawtooth Climb Data Sheet
Fly each of these airspeeds going up and down gathering data each direction. This is where the sawtooth name comes from, as the profile seen from the side looks like teeth on a saw blade. You’ll know you gathered quality data when you keep your airspeed constant during the middle block of the test, and the elapsed time each way (going up and down) is recorded in the data sheet. Once back at your computer the data can be entered into the spreadsheet (tab “Sawtooth Climb Test”) and the analysis is conducted automatically.
Figure 3 – Sawtooth Climb Data Entry Tab
A few last-minute notes about gathering the data. All altitudes are “Pressure Altitude”, which is simply the altitude read off your gauge when the baro is set to 29.92". Ensure you record outside air temp (OAT) for the altitude block you selected. It’s not likely to change over the course of 20-30 minutes while you’re gathering data, but if it does you should at least test a few times just to be sure. This can be as simple as a cheap thermometer stuffed out the window, or even approximated by estimating the temp drop at your altitude (3.5 deg F per thousand feet) subtracted from the ground temp at the airport.
The last part of working the data into the spreadsheet is establishing the tangent lines. Refer back to the “Sawtooth Climb Test” tab and scroll to the bottom of the sheet, around row 56. You’ll notice the data has populated the charts automatically, but the user must look at the chart above and “guess” at the tangent line. Trial and error is how this is done, so play with it until the magenta line just barely touches the charted curve (e.g. is tangent to the curve). Read the chart and pick out the tangent point airspeed (Vx) and the peak of the curve (Vy) and enter them on the sheet in the yellow shaded cells. That’s it!
Figure 4 – Guessing at the Vx “Tangent Point”
Best glide speed
Best glide uses the same data gathered in the sawtooth climb, but analyzes it for glide performance. The only task here is to again “guess” at the tangent point of the descent curve, and once you have it to read the tangent point (best glide airspeed Vbg) and peak (slowest descent or “min sink” airspeed) and record them in the shaded cells. Best glide should be easy to find, but min sink is likely to be the slowest airspeed you tested. This is due to the way we gather this data – we don’t spend a lot of time climbing or descending at speeds just above the stall speed!
Part 3: Performance Data
Rate of Climb Table
A useful addition to the POH is a simple table depicting rate of climb (in ft per minute) versus density altitude (in ft). This can be collected pretty easily once the best rate of climb has been determined. The chart won’t be as fancy as those produced by a GA manufacturer, but they are very useful nonetheless.
The procedure is simple. Establish a climb at best rate of climb airspeed, stabilize everything, then start your stopwatch when passing a known starting altitude and continue climbing. I recommend starting the climb at least 500 ft lower than your desired starting altitude to give time to stabilize. If you don’t do this, the plane will start the climb with excess airspeed that will artificially inflate the rate of climb early on during the test. Continue to climb until you see a reduction in climb rate down to only 100-200 fpm, until you get bored, or you run out of time. Go as high as you can, and get good data! Record the data using the example data collection sheet on the tab entitled “ROC Data Sheet”.
Figure 5 – Example ROC Data Sheet
Once complete, the test data can be transferred to the tab “ROC vs DA (950 lbs)” for analysis. The spreadsheet includes two tabs for ROC vs DA – one for a lightweight configuration, and one for gross weight. You can choose to collect data at both weights for completeness, or a single weight for simplicity. The spreadsheet will analyze the data and automatically graph the results. It’s normal to have data points that are somewhat scattered around, so the chart adds an average “trendline” to help show the average climb performance.
Figure 6 – ROC vs DA Data Entry Tab
You’ll need to examine the chart and determine the values on the trendline equation. Enter the slope and intercept from the equation into the yellow shaded cells, and the charts update automatically. There’s even a clean chart that you can copy and paste into your POH using the data you entered. It’s really that simple.
Figure 7 – Entering ROC Trendline Slope & Intercept
Takeoff and Landing Distances
Takeoff and landing distances are best determined by actual measurement for test runs. Flying 3-4 takeoff and landings and recording the distances will establish a baseline performance, and the spreadsheet will calculate additional performance estimates for you. This is especially useful when filling out a table of performance values at several different density altitudes, and estimating performance over a 50 ft obstacle.
Record the takeoff and landing distances using the data sheet in the “TO-LNDG Data Sheet” tab, then transfer the average performance numbers to the “TO & LNDG Distance” tab. The only numbers required are the approximate takeoff and landing distance at sea level for a light-weight condition (optional) and a gross-weight condition. The spreadsheet uses the analysis already conducted for the climb-descent tests to approximate the 50-ft obstacle performance, and fill those numbers in directly. These numbers are only estimates, so you should test them out in your particular airplane for accuracy.
Figure 8 – Example Take-off & Landing Data Sheet
Figure 9 –Take-off & Landing Data Entry Tab
Percent Power, Fuel Burn and Cruise Speed Tables
Power Tables are difficult to find, and require extensive experimentation to develop on your own. The spreadsheet includes a sheet to assist in generating your own power tables. These will be approximations only, but will help get you close as a starting point. To use the sheet, you’ll need to measure fuel flow, rpm and indicated airspeed. The spreadsheet will complete the tables once you provide some baseline starting data. Cells shaded yellow require user input, and the rest are calculated automatically. The “% BHP” and “Time, Dist, Fuel” tabs contain several useful products that may be included in your POH.
Figure 10 – Percent Power Table Data Entry Tab
Figure 11 – Example ROC Chart Suitable for POH
Airspeed Calibration Error
Lastly, the spreadsheet includes a method for checking the accuracy of your airspeed indicator. The procedure requires flying the plane at a constant altitude and airspeed along three different legs of a triangle. Each leg records the direction and GPS ground speed, then the spreadsheet resolves the geometry to determine the wind present at the time. Once wind is known, the spreadsheet compares the indicated airspeed to the calculated actual airspeed and determines the error present. The method works really well, and if very easy to gather the data.
Miscellaneous Items of Interest
The analysis conducted by the spreadsheet can shed insight to some interesting things. Sprinkled throughout the tabs are calculated items like glide ratio, glide polar, and max predicted service ceiling. Exploring the formulas and calculation might yield additional insights as well.
The Simplified Flight Testing Spreadsheet automates much of the data analysis normally required to translate flight test data to usable products that can be placed into the aircraft POH. The test profiles are simple to understand, and simple to fly. Upon completing the flight, the data is entered into the spreadsheet and analyzed, and then used repeatedly throughout for subsequent products. As with any analysis program, results are only as good as the data entered. If a particular run of data collection is suspected to be poor, the user can always try again and gather new data that is of a higher quality. I hope this spreadsheet allows pilots to approach flight testing in an efficient, methodical manner, and produces useful end products that enhance the pilot operating handbook. Give it a try – I think you’ll like it!
Download the Spreadsheet (Excel)
Back to Part 1