I’m trying to devise a clever way of relaying the speed of an RC model aircraft while flying. The only ways I can think to do this are as follows:
GPS Receiver to record the plane’s position and relay the position data back to a base station. However, anything GPS-related seems prohibitively expensive.
Radio Triangulation. Have a simple transmitter on the aircraft and 3 receiving points. Problem with this is…I have no idea how this works.
Have a small propeller (no, not the one that makes the thing fly) mounted somewhere on the plane. This prop will spin when the plane is flying. The spin rate could be encoded and transmitted easily enough, no triangulation necessary. The problem with this arrangement is that some elaborate form of callibration would be necessary and its likley to be highly inaccurate regardless.
So, any ideas on how I could (cheaply) determine the speed of this model aircraft? Am I missing something obvious?
(PS. There likely are geometric/observational-based methods of doing this. However, such a technique would not really be useful for our purposes. The ideal method is some sort of wireless telemetry. Also note I have electronics knowledge and could build whatever circuits may be necessary.)
The easy way, if what you are interested in is speed in level flight - calibrate the airplane itself. Set the throttle at various positions, and fly the airplane through two points and calculate the velocity. Record the speed at various throttle settings. Measure the wind speed and direction, and add or subtract the wind velocity vector to come up with the airspeed. Now you’ve got a good idea of how fast the airplane goes at various throttle settings. Then when you need to know the speed of the aircraft in the future, you can figure it out just from your throttle settings and the wind velocity.
If you need a more instantaneous value than that, at different angles of attack, your job is much tougher. The GPS method is the obvious solution - I imagine there are some pretty small GPS encoders out there now. That will give you groundspeed, and not airspeed, so it’s not much use for determining things like best rate of climb speed or whether you’re approaching a stall.
For measuring airspeed, you really need a wind speed guage on the airplane. You don’t need a full pitot-static system unless your RC plane will fly to very high altitudes, but you will need a barometer on the ground to correct your readings.
The real problem with a small speed measuring device is wind blast from the propeller. Ideally, you want the windspeed sensor out on the wing somewhere, but on a small airplane that might be tough.
Something I just thought of… I don’t know what your application is exactly, but would a radar gun work?
If you want to measure the airflow, I don’t think I’d use a propeller. I’d be thinking about a small flat plate connected to a strain guage maybe. Put a little plate on the end of a stiff wire, and a strain guage wire making a triangle between them. I’ll bet you could make a nice little speed sensor that way. You’d just have to calibrate it.
In a 1:1 scale plane, typically what is doen is a static/pitot system.
The static port is a small hole on a plate on the side of teh airframe.
The pitot port faces straight forward typically on the wing of the airplane
(the L shaped tube)
By comparing the pressure difference, you can determine airspeed.
As Sam Stone points out, indicated airspeed depends on pressure (and temerature). However thats what I use for landin etc because the whole plane is affected by pressure/temp by about the same amount.
I wouldn’t surprised if there arent RC airspeedometers avaialable.
I once read about a simple method for measuring air speed. An electrically heated wire was used. The resistance of the wire varied with its temperature. The current through the wire was measured to determine the air speed. Calibrating the sensor might be a bit tricky.
There used to be, and still may be, a device we used on R/C sailplanes called a “Thermal Sniffer”. It used two thermistors and a calibrated leak to determine rate of climb or decent. The info was sent back to a ground reciever as a varying tone. Rising tone = rising aircraft.
Anyway one group of flyers made a small propeller mounted on the A/C that generated a digital signal. They then used this signal and the existing telemetry electronics to send the signal to the ground unit. The ground unit had been modified to display airspeed on an analog meter.
Don’t know how accurate it was as this was back in the late 80’s. Just wanted to get across that there was a method of sending signals back to the ground from an A/C rather cheaply.
If you can get your hands on some small accelerometers that are practical to fix to the plane, then that might do it. I have no idea what the practicality of that idea is, though.
I think you guys are over-thinking this, though the ideas are fun to read. I did this once with a friend and we got some pretty consistant numbers using two guys on the ground with watches a known distance apart.
You have to make a few runs to average out reaction time errors, but if you get the two guys far enough apart, those errors become negligible.
I think the OP wants instataneous results and thats why just timing over a distance won’t work well. Could be wrong.
One airspeed meter I’ve seen in a ultralight or glider (I forget which)
an L shaped tube with both ends open and one end facing forward.
In the vertical section was a ball and etched speed marks.
I’m wondering if you could vae a horzontal tube with a spring plunger attached to the back. A potentiometer is also hooked up.
the faster you go, the more the plunger is pushed back.
Calibrate by mounting on a car.
If you don’t need to know the speed at any given instant, I would just take the plane to a football field and a few friends to time how long it takes to go 100 yards. You should do several runs in each direction to help eliminate any wind.
If you have enough people, you could stagger them about 20 yards each, and get a decents approximation of the acceleration as well.
Time over measured distance. One place I used to fly, there happened to be a football field adjacent to the landing strip (though this method works ok without). We measured the length of a football field. Two of my friends had stopwatches (one in a digital watch, one in a mobile phone). They started the watches together, then walked to the goalposts and used them as sighting lines. I flew the plane the length of the pitch, and they stopped their watches as the plane crossed their respective sighting lines. The difference in the recorded elapsed times on the two watches allowed the speed to be calculated. We repeated several times and in opposite directions.
GPS: I have a handheld GPS with a ‘maximum speed since last reset’ function. I fitted it inside one of my 1/4 scale planes. Although this method sounds ideal (assuming the thing will fit in) the results I got over several flights showed an unreasonable amount of variability, so I don’t trust it. Maybe a different GPS would work better.
Radar gun: This is the method favoured by the dynamic soaring guys. You have to fly straight at the guy holding the gun - a bit dangerous really.
Audio Doppler: This works really well on calm days. Make some full throttle passes (engine speed must remain constant) and record the sound (I used a video camera). On a PC, use an audio program (I used Sound Forge) to measure the change in frequency of the fundamental component of the exhaust note. A few sums and you have your answer.
If you have a video camera, it can be used as a timer. NTSC video runs at 29.97 frames per second. Simply watch the video frame by frame as the plane passes between points of a known distance apart. Much easier than the audio doppler interpretation.
I tried this, and it doesn’t work. First you have to fly really low, so as to have some background objects. Secondly, you don’t know how far apart trees etc. are. Third, and this is the killer, depending on how far away the plane is, you can cover the ‘distance’ between landmarks in different times. If the plane is 10% further away on a pass, it will appear, by this method, to be going 10% slower.
You are measuring how fast a plane subtends an ANGLE in the camera’s field of view by this method - not how fast it covers a DISTANCE.
The radar gun method is impractical simply because we don’t have access to one N9IWP’s R/C Flight recorder device looks idea, but is beyond our budget.
I think we’ll just have to stick to the ol’ v = distance / time method.
Don’t give up on audio doppler. I was thinking of programming a laptop to give a live speed reading on the screen as a plane passes. I just need a good mic that picks up engine sound well even when the wind is blowing a bit. The rest is just some fast fourier transforms and some trivial math.
This only works though if the plane is making enough sound for the microphone to pick up and the engine revs remain constant. This method is no use for gliders.
