I have a project involving a small single engine aircraft to measure the distance between a specific spot under an aircraft fuselage and the ground (less than 24 inches) when the aircraft lands. I need to know how close the fuselage comes to the ground should the aircraft incur a hard landing.
I need advice on the best sensors to use to measure the distance (10 inches to 36 inches) when the landing gear touches the runway. I would like to package the sensor with a power source and data recording device by taping the package below the fuselage and record the data during several landings. If necessary, I can run cables (power & computer interface) from the measuring sensor to the cockpit.
I am new to this technology so any advice is appreciated.
The speed of the aircraft will be between 50 to 100 mph. Yes the runway will be either concrete or asphalt. The aircraft has fixed landing gears and each gear strut is a leafspring
which deflects upwards as the load increases.
We have considered using an accelerometer to give us the g load timeline to determine the point of highest deflection. We are not expecting g loads to be 1.5 to 2.0 at most. Will the laser range finder handle a 2 g load?
But the accelerometer won’t provide distance measurements which we need to know.
It looks like a sensor like the PBS-03JN may be worth considering. It has an RS-232 interface and can be operated by battery. What type of controller and data storage method would work with the PBS-03JN?
Have you thought about doing this the old fashioned way.
At the locations you are most interested in measuring, attach a block of wood with many sticks (wire or wood) pointing down of progressively increasing length.
After landing measure the longest stick that has not been deformed.
Of course these ‘sticks’ would have to be able to endure the forces during flight.
You could probably make a dozen such measuring devices in a couple of hours for a few dollars. I don’t know how you would attach them, but you would have to do that for the other ‘device’ as well.
In our technology centric world I see many Rube Goldberg machines that are equivalent to a ‘stick’.
Hi cetech, interesting project. This is going to be tricky.
The first thing that comes to mind in terms of sensor technology is ultrasonic rangefinders. Specifically the Maxbotix XL-MaxSonar-WR1 Weather Resistant Ultrasonic Range Finder. There is however a problem with ultrasonic rangefinders and fast moving wind across the sensor and this may require experimentation. The XL series seems to be better with this but results may vary. If this doesn’t work, maybe a laser rangefinder may be a better option (not sure if they will handle rough landings). It is obvious more research as to the appropriate sensor type will be required.
The next step will be to capture the data this sensor provides, we suggest a microcontroller for this. We have a large selection of microcontrollers, perhaps you can select one that suits your fancy. For beginners we suggest Basic Stamp or Arduino based microcontrollers.
Then you may need a way to store this data to analyse it later. Maybe you will want a time stamp on the range readings. You can get really creative using a variety of sensors such as real time clocks, temp/humidity and so on to gather all sorts of useful data. Also you can record this to microSD cards to easily read it on a PC. We so have a variety of external data storage solutions for use with microcontrollers. You may even opt to save the data inside the microcontroller, but this may fill up fast.
You can really play around with this technology and get exactly what you envision.
What about quantifying the impact by measuring the G-Force of impact from time of impact to the point where the dampeners (shocks/springs) start to push the plane upwards? This data should be used to get a decent approximation of distance. This would also not be affected by environmental conditions, or forward aircraft speed.
You can get a measure of distance simply by knowing the linear acceleration.
x2 = x1 + v1t +1/2a*t^2
You would assume x1 and v1 are both zero at the first impact when the wheels touch the runway (you would see a small spike in acceleration, or manually/visually determine when the wheels have made contact). The easy way would be to take an average value of the acceleration over the time frame. However, as you mentioned, you would not get the distance measurements directly. If you are interested in the PBS-03JN, we can contact the manufacturer on your behalf to see if they have any insight into using it for your application.
What is the speed of the aircraft as it lands? I assume the runway will be ashphalt or concrete as opposed to dirt? The choice of technology (infrared seems out of the question, and ultrasonic may not be fast enough) will be critical - possibly laser. Did you consider adding sensors to test how far the landing gear retracted instead?
Scanning laser rangefinders are used on autonomous vehicles which tend to travel at higher speeds than normal robots. Yours is one of only a few applications where the target surface is moving at almost 100mph. The scan takes ~100msec (less depending on the model). The rangefinder should certainly be ableto handle a 2g load (read the specifications sheet before selecting, for example the URG-04LX can handle 19.6m/s2, but is “not intended for outdoor use”). The accelerometer option may be far less expensive and reasonably accurate.
Hmm… very interesting. I may have another approach. What about using no sensors at all ? Generally, I think the land of an aircraft is a constant line going down since the beginning of the landing track. Or, it may be near en average line I think. What about putting some line with duck tapes on the landing track and measure the distance it takes since the beginning of the landing track to the landing of the aircraft.
With some measures on the landing track and a little crono, you could have a nice idea of a good average of what you want to learn. In the end, and some basics formulas, you’ll have the time it takes to land, the distance. With the two parameters, you can have the speed. And… you can calculcate the degrees. With that, it will be possible to extrapolate for the averages measures nice enough for a good landing in each hight, wind speed, snow, etc.
But, it is not perfect. It’s really a cheeeeeap way of doing it but not perfect.
