What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Errors in January 2006 Kitplanes "You Know Yo' Pitot" article

Kevin Horton

Well Known Member
There were several major technical errors in the "You Know Yo' Pitot" article in the January 2006 Kitplanes magazine. I have sent a letter to the editor, but I also want to cover the issue here, to try to give the errors some visibility.

Here is what I sent to Kitplanes:

============================

I was quite dismayed to read several major technical errors in the "You Know Yo' Pitot" article in the January 2006 Kitplanes.

1 - TAS is not equal to average of four ground speeds

I have spent enough energy dispelling mathematical myths about TAS vs GPS without having Kitplanes give more life to the misconceptions. Specifically, the author stated that the average of GPS ground speeds taken on four runs on the cardinal headings will be equal to the TAS. This seems intuitively obvious to those who don't understand wind triangles, but it is patently false.

For example, imagine an aircraft with a TAS of 150 kt. The wind is 270 degrees at 30 kt. This is not an unusual wind speed at altitude, where you often need to fly these tests in order to get good air conditions. The ground speed (GS) will be 120 kt on the W run, and 180 kt on the E run. These average to 150 kt - so far so good. But, on the N run, the wind is exactly off the wing tip, and the wind triangle is such that the GS is equal to 153 kt. The GS heading south is also 153 kt. The average of the four GS is 151.5 kt, which is a 1% error. The error is approximately the same no matter what the wind direction. If the wind speed is higher, the error is higher. A 60 kt wind (not unknown - I had 70 kt at 5,000 ft one day last month) with a 150 kt TAS gives an average GS of 156 kt, or a 4% error.

People come up with many creative ways to get TAS from GPS readings, but most of them are wrong. The best way that I know of is to use a mathematical method devised by Doug Gray, from Australia. It uses GPS GS and Track data from three runs on any widely different headings.

Method:
http://www.kilohotel.com/rv8/rvlinks/doug_gray/TAS_FNL4.pdf

Spreadsheet:
http://www.kilohotel.com/rv8/rvlinks/doug_gray/TASCALC.XLS

Both of the above in a zip archive:
http://www.kilohotel.com/rv8/rvlinks/doug_gray/TASCALC.ZIP

The National Test Pilot School came up with a useful twist on Doug Gray's method - they take data on four runs in a box pattern. Doug's method only uses data from three legs, so you do the calculation four times, discarding the data from a different leg on each calculation. If the four calculations yield about the same result, you know you have good data. If you get very different results from the four calculations, you know there is a problem - maybe you didn't fly accurately, or you copied down a wrong number, or the wind changed. But you know the data is bad so you throw it all away. If you only record data from three legs you have no way to detect a bad set of data.

2 - Airspeed errors don't only come from the gauge

It is stated that the TAS determined from the GPS data, when converted to indicated airspeed, can be used to check the accuracy of the gauge. This seems to imply that any error will be due to ASI instrument error. While it is true that ASI instrument error is an important source of error in the indicated airspeed, it is not the only source of error. The static source is almost certainly not reading correct pressure at all conditions, and this static source position error also contributes to the error in the indicated airspeed. It is useful to calibrate the ASI on the ground using a water manometer. Then, once you have completed the in-flight tests of airspeed system accuracy, you can determine how much of that error is due to ASI instrument error, and how much is due to static source position error.

3 - Flush static ports are not always best

The author suggests we should try to make the static port as flush as possible with the fuselage skin. While this may be good advice for many aircraft designs, it is not a universal truth. If the designer has provided specific recommendations, they should be followed. For example, Van recommends using a domed pop rivet as the static port on many (all?) RV models. This protruding static port works well in the recommended location. I have seen reports from several builders who second-guessed Van by putting in flush mounted static ports, then found that they had significant airspeed errors. The errors were reduced by modifying the static ports to obtain a contour closer to that recommended by Vans.

A static port would ideally be located in a location that was always at the free stream static pressure. But, the airflow accelerates and decelerates as it makes its way over the aircraft's skin. Bernoulli's law tells us that the pressure will vary as the speed of the air varies. So it is not possible to find a location that will give an accurate static pressure under all conditions. The designer may find that he has to modify the local pressure by having a protruding static port, or a fence ahead of, or behind, the static port. Follow his advice, at least until you have flight test data to tell you how accurate your static system is.

It is important that most builders ensure that their aircraft has a reasonably accurate static source. If the aircraft is flown IFR, errors in the static source lead to errors in the indicated altitude. This may reduce the altitude separation with other aircraft, or cause the aircraft to be lower than expected when at approach minimums. If the aircraft, flown IFR or VFR, has a transponder with altitude reporting, errors in the static system lead to errors in the altitudes reported by the transponder. The altitude reported by the transponder is used by the Traffic Alert & Collision Avoidance System (TCAS) in larger aircraft. The TCAS provides commands for the other aircraft to climb or descend to avoid a collision with your aircraft. If your static system is providing inaccurate altitude indications, this increases the risk that the TCAS commands will not be adequate to prevent a collision.

I have put together a guide to pitot-static system calibration, including information on how to check for ASI instrument error and instructions on how to determine the static system position error:
http://www.kilohotel.com/rv8/rvlinks/ssec.html

Please have any technical articles reviewed by someone with relevant technical knowledge. It is very disconcerting to find glaring technical errors. If the articles in my area of expertise have obvious errors, why should I have confidence that articles in areas where I have no competence are technically sound?
 
Back
Top