Groundspeed does effect airspeed

[Tom Harper]

HarryC,

Not sure what there is to patent. Think you need to reread the posts. I am on the side of losses preventing the model from reaching the velocity of the moving air mass in which it flys. No perpetual motion there.

Being wrong is a matter of degree.

Tom

[Tom Harper]

Aerowhatt,

My models have most of their lateral area behind the CG. If there were a difference between the speed of the air mass and the model it would create pressure on the fin and rudder.

I don't think my first set of data will resolve the issue. But, it should provide something tangible for discussion (derision etc.).

I have been in engineering long enough to know that nothing works as planned. This is a low budget project with lofty (no pun) goals. The tough part is knowing the wind velocity. I will get an accurate measurement on the ground but still not know the wind at altitude.

This weekend I enlarged the camera bay of the model I call LD5 (ref LD4 build thread) and installed a small camera and handheld GPS unit. I removed the stops on the camera focus ring an set it to 3.5" distance so it records the GPS screen. The camera is 2mpx so the image is not bad. The fastest the camera will recycle is 2.5 seconds. I had previously modified it for electronic actuator. I added a free running 555 oscillator to snap a frame every 2.5 seconds. It all wieghs about the same as the camera it replaces.

This will give ground speed, altitude, heading, position and time. Cross wind passes will give model speed and upwind/downwind will show the impact of the air mass. It won't be a clean data set. There are a lot of noise sources.

I could add a leveler and GPS servo to remove pilotage. See what we get first.

Tom

[pmackenzie]

Tom,
Sounds good. I have a suggestion:
Go out on a calm night and trim the plane to fly large hands-off circles, so that to fly straight and level it requires a bit of stick force. Record the GPS info as a baseline.
Then take the model out on a day with some light wind and let it fly itself with the absolute minimum of control input.
If all goes well work up to days with more wind.
The ultimate solution for data gathering would be the Eagle Tree FDR with the GPS expander:
http://www.eagletreesystems.com/Plane/plane.html
Not cheap!
Pat MacKenzie

[Tom Harper]

Ah, a baseline - good plan!

[Tom Harper]

Pat,

The eagle tree equipment looks neat!

Price isn't too bad for commercial product. But, there's a lot of stuff there. Lot's of connectors and a down link. Simpler is often better. Hiigher MTBF.

You can do nice 3D plots with Excel. I wrote a simple program to plot Latitude and longitude paths. Surface plots with Excel could be used for well ordered passes.

[seaotter12398]

when i was taking flying lessons, one of the things they taught us was to navigate accounting for wind.

we had to calculate for wind angle relative to the course, and wind speed. we never had to account for drag, even for shorter flights where drag would be a more appreciable percentage of the total effect of the wind.

if you fly at 90 degrees to the direction of a 10 mph wind, at the end of one hour you will be 10 miles off course.

[seaotter12398]

ooooh! ooooh! oooh! i just remembered sumpin!

the effects of the airmass on altitude is easily provable. commercial airline pilots call it wind shear.

flying parallel to and in the same direction as the wind means that:

KIAS - WIND SPEED = effective airfoil speed

so for a given airspeed and aoa the wing will generate a certain amount of lift.

in a wind shear event, the airmass abruptly changes direction relative to the aircraft. if it changes by 180 degrees from the above example,

KIAS + (-WIND SPEED) = effective airfoil speed

when that happens, pilots have to immediately decrease speed and shove the stick forward cuz they are headed heavenward.

conversely, when the wind shifts from on the nose to on the tail, the wing behaves as if it lost the airspeed caused by the wind component and the pilots get to increase throttle and pull up cuz they are headed for an arrival.

the effect of airmass motion on airspeed (and ultimately on lift) is why we takeoff and land upwind.

i think i got all this right. i'm still on my first cuppa kawfeeee < yawn >

[Tom Harper]

SeaOtter,

More coffee! Thanks for the contribution.

Extreme wind shear is an acceleration issue. That would be similar to starting conditions - which could start another lively thread.

We are concerned with steady state conditions or as steady state as they can be. When you take wind speed into consideration for navigation where do you get your number? Do you think that number is good after traveling for an hour? It works because it is just one input and you make corrections as you progress.

The annemometer reading on top of the tower can be very accurate for that point on the tower. Theodolite tracking of a weather balloon is full of errors. Radar tracking removes only some of the errors.

Traveling in or near the ground/wind boundary layer adds problems. It's worth investigating.

[Tom Harper]

Did some Googling on the accuracy of wind velocity measurement.

1 percent in a laminar flow wind tunnel with a top of the line instrument.

2 meters per second for Navy navigational instruments - that's a 4mph error window

10% to 30% for shipboard annemometers

I assume weather balloons fall into the 25% error range.

http://oceanworld.tamu.edu/resources/ocng_textbook/chapter04/chapter04_04.htm

[biber]

With a swivelhead probe like the one on the pics an accuracy of about 0.2m/s is possible.
I've calibrated it, so that I know it's quite good. ;)

biber

[Sparky Paul]

We had a special nose cone made up for the Tristar, with 4 pressure sensors.. two in the vertical plane, two in the horizontal plane, to measure alpha and beta ahead of the influence of the airplane, for flying into turbulence..
Flew up and down a few ski slopes in Utah...
On one flight, the copilot asked Salt Lake Center if they had any information on turbulence...
An airliner on the frequency came back with..."Fly over my house. It was pretty turbulent when I left there this morning." :)

[Tom Harper]

Biber,

That's a spiffy gadget. A bit large for a model. What's the swivel for?

[biber]

It weathervanes into the oncoming airstream, so that the head of the probe does not see any varying angles of attack or slip angles.
That is an absolutely neccessary feature if signifficant errors are to be avoided.
The head is basically a Pitot tube.
The small hole at the center tip gives the total pressure (= static + dynamic) and is not prone to errors until quite some degrees of bias airflow.
The 4 static pressure holes are about 5 diameters of the probe behind the very tip.
They are much more sensitive to varying flow angles, so the swivel is needed for them.
The pressure difference between static and total pressure, the dynamic pressure, is proportional to (airspeed)².

biber

[Tom Harper]

Thanks - impressive!

[Sparky Paul]

You can see the difference between the plane's flight angle (pitch) and the angle of the inflowing air, which is the alpha, and deviations laterlly are beta.
Back in the olden days there were two small vanes on the nose-mounted probe for this purpose.

[Aerowhatt]

I've got a simple idea for an experiment which should convince most of the doubters.

Put a long lightweight boom out the nose of a pusher aircraft. Use a weather vane such as lightweight yarn (or something more sophisticated and steady) out at the tip and elevated so it doesn't physically interact with the boom. Point the camera at it. If Tom is right and the aircraft isn't moving in lockstep with the airmass you would see it on the weather vane.

Steady state in crosswind flight would show the weather vane pointing somewhere other than dead straight down the centerline of the fuselage. Turn around 180 degrees and you would get the equal but opposite reading from the indicator.

This is much simpler and easier to do with no built in error to color the evaluation of data.

What you will find is that in steady state flight in non turbulent steady winds (or in calm conditions too) The weather vane will point straight back except when manuvering or before reaching steady state.

Seaotter: Thanks for reminding me about navigation training. Ground school was a long time ago!

He is right you figure the cross wind vector and find out how far you will be off destination if you fly the actual destinations compass heading. Then you make that many miles to the other side of your actual destination your compass heading destination. Fly that adjusted compass heading and you end up where you wanted to be in the first place.

Aerowhatt

[Tom Harper]

Aerowhatt,

The yarn idea is a bit tricky to set up. Sparky's L1011 was a sophisticated version. The vector is small. A few degrees would be hard to detect. I'll stick with GPS for the moment.

The problem still is knowing the actual wind velocity. Biber's magic boom only shows velocity through the air.

The list of errors for weather balloons was revealing. They refer to the vertical component as a major source of error. Also the degree of turbulence. It looks like an air mass in the 100 to 200 ft height above terrain is not a very clean environment.

Tested my rig on the bench and it all works. I tried hi intensity LEDs for a reliable light source. Plenty of light but the lens causes saturated spots on the image. Diffusers don't work. Gave up and used a diffused window on top of the model. I'll check it our in full sun today.

[Tom Harper]

Aerowhatt and SeaOtter,

Of course you use a cross wind vector in navigation, but it is an approximation. Ever read about the Dole Pineapple Derby?

[seaotter12398]

tom: actually no, i'd never heard of it till you mentioned it. from what i found on the web it sounded like a bad idea.

i dunno if anyone else is having this problem, but after reading all the pages of this string i'm having a hard time figuring out what the exact topic of this conversation is.....

instantaneous effect of airmass on things in the air?
cumulative effects of airmass on things in the air?
instantaneous and/or cumulative effects of airmass on lift and/or altitude?


btw, lots of military airplanes have lengths of yarn or string attached to the fuselage just forward of the cockpit to give the pilot a rough idea of beta.
you can see this in the movie Top Gun and in other aviation films.

a slightly dazed and caffiene deficient otter :3~~~

[Tom Harper]

seaotter,

The basic issue is whether a model flying in an air mass matches the velocity of that mass. If it does not, then what are the effects on the model.

I say it cannot - simple matter of physics.

Others say it must - simple matter of physics.

[Aerowhatt]

The bottom line is that there is nothing to prevent the object from matching velocity of the air mass. Every force on the object from the air flowing over it accelerates it towards the velocity of the air mass. Those forces decrease as the velocities approach equality. Once they do, then the forces reach zero and you have equilibium with the object moving at exactly the velocity of the flow.

The only thing inhibiting this matching of velocities is the momentum of the object. The mass of the aircraft can only delay the reaching of equilibrium not prevent it. I read over and over about some mystery "drag" keeping the aircraft from reaching the velocity of the air mass. Drag from what?

All of the drag vectors act to accelerate the aircraft downwind. Until the velocities match that is. Then they cease to exist.

Your setup has too much error built in to it to be useful to detect any mystery, yet undiscoverd (read as non existant) force acting on the aircraft against the crosswind.

I would like to hear an explanation of the proposed force preventing the aircraft from matching velocity as something other than "drag" in the opposite direction of a flow. Drag from what, the attraction of the moons gravity?

Aerowhatt

[biber]

Back in the olden days there were two small vanes on the nose-mounted probe for this purpose.biber

[Tom Harper]

Aerowhatt,

My rig has numerous sources of error. Fewer, however, than tracking a weather balloon. I don't anticipate shocking revelations. This equipment should indicate an altitude loss, if any, on the downwind leg and altitude gain, if any, on the upwind leg. It will also give accurate ground speed measurements. It will be interesting to see what the cross wind readings are.

You have pointed out that there is a vertical component to the moving air mass. The vertical component is also listed as an error source for weather balloon tracking. Turbulence is a source of error in wind velocity measurement. So what wind velocity are you going to match?

Let's see what kind of data we get. If the aircraft does not behave as it should then there is a topic for discussion.

[Aerowhatt]

[QUOTE=Tom Harper]
So what wind velocity are you going to match?
QUOTE]

In the real world, it would have to be the composite, average wind velocity at the point of interest . . . of course.

Aerowhatt

[Tom Harper]

Aerowhatt,

Yes - If the flow is not laminar that value will constantly change.

Buttoned up the model and tested the system in full day light. The illumination, with the diffuser, is adequate. Get a little contrast loss at some sun angles.

Need to raise the mount a little. Also flat black the focus ring to get rid of the reflection.

[Neil Walker]

The basic issue is whether a model flying in an air mass matches the velocity of that mass. If it does not, then what are the effects on the model.


What a terrific thread, I've really enjoyed reading it and am also quite impressed with how everyone has maintained a nice air of civility with what I would expect to become a heated discussion. Well done.

But I think (at least for me) you need to clarify the above into a concrete question, because I'm not really sure what it is you're asking. You say "whether a model flying in an air mass matches the velocity of that air mass"... what do you mean by "flying in an air mass"?

I think my confusion stems from the fact that some of the hypothetical scenarios in this thread seem to be talking about objects being accelerated by the wind in the same direction (dandelion seed, hot air baloon), and the ensuing argument that they could never reach the same velocity as the wind pushing them. But this is not flying. You are simply talking about one object (the baloon, or whatever) receiving an impulse from a bunch of different objects (air molecules). When I fly my gliders around I can trim them into various steady states where they fly (groundspeed or airspeed, take your pick) much, much faster than the air in which they are travelling. But then their speed (ground or air, it doesn't matter) has little to do with the motion of the air mass through which they are flying w.r.t. the ground. Whether they're going upwind, downing or crosswind, I can make them fly faster than the air through which they are moving. Now I don't quite think this is what you are talking about, but it's not clear to me what you are asking when you bring the word "flying" into it.

I think you need to state it as a question, and if you're going to use the same terminology ("a model flying in an air mass") define exactly what you mean by flying. And exactly what you mean by "match the velocity". Are we talking about the "being pushed by the wind" thing or something else?

I also do sort of agree with pmackenzie that we need some physics and balanced equations and force diagrams - at least enough detail to clearly define the question. Right now you seem to be constructing some sort of experimental rig in a model with GPS receviers and digital cameras in an attempt to prove your point, but if you're going to do that you should first state the question clearly and allow your "adversaries" to agree on whether or not they think your apparatus is acceptable - or if your question is even unambiguous.

[Tom Harper]

Neil,

Good points!

The discussions stem from interpretation of observations. The downwind turn, altitude loss when running downwind, altitude gain on the upwind leg, weather vaning....

I am not trying to prove anything with the data gathering rig. I would like to get some data that is measured rather than simply observed. We will know the velocity, altitude and heading of the model. We will also know position and time. I can measure wind velocity and direction at ground level.

If the model shows no gain or loss of altitude when flying upwind and downwind it would seem that there is no argument in which to engage. If that is the case then I can drag out the data everytime the issue raises it's ugly head. If there is a gain and loss of altitude then we can rationalize the cause.

In this experiment, 'Model' refers to a draggy, engine powered RC model airplane. 'Wind', in this discussion, is a mass of air moving relative to the surface of the earth.

[Neil Walker]

Neil,

Good points!

The discussions stem from interpretation of observations. The downwind turn, altitude loss when running downwind, altitude gain on the upwind leg, weather vaning....

I am not trying to prove anything with the data gathering rig. I would like to get some data that is measured rather than simply observed. We will know the velocity, altitude and heading of the model. We will also know position and time. I can measure wind velocity and direction at ground level.

If the model shows no gain or loss of altitude when flying upwind and downwind it would seem that there is no argument in which to engage. If that is the case then I can drag out the data everytime the issue raises it's ugly head. If there is a gain and loss of altitude then we can rationalize the cause.

In this experiment, 'Model' refers to a draggy, engine powered RC model airplane. 'Wind', in this discussion, is a mass of air moving relative to the surface of the earth.

I still don't know *why* you are flying this model and gathering the data that you are gathering. Specifically what question are you trying to answer? Can you understand what I'm getting at? It's clear to me that you have the question in your head (even if you haven't directly asked it of yourself), but I do not think that you have adequately expressed that question to the rest of us. And if you don't, any data you come back with is meaningless.

Your post says "in this experiment..." and you also talk about the things you are going to measure. So you're conducting an experiment and making measurements. Measurements (or observations - same thing) are one part of an experiment. The others are a hypothesis, a conclusion, a method and an apparatus. You are sort of defining your apparatus here, and I have a vague idea of the method, but what I'm after is the hypothesis. I get the feeling you have already made your conclusion and are working backwards to the hypothesis!

[Tom Harper]

Neil,

Model = a powered model airplane that has been trimmed for hands off, level flight in dead air.

Question 1 = will the model gain altitude when flying upwind?

Question 2 = will the model lose altitude when flying downwind?

Hypothesis:

If RC models respond to wind in the manner described by some pilots, then the gain and loss of altitude can be measured.

How's that?

When I was in school, an experiment was defined as a question. A hypothesis may be required to explain the result.

All measurements are observations. All observations are not measurements. Not the same thing.

[Sparky Paul]

Back in the olden days, we flew a laser airspeed, alpha and beta instrument in the #1 Tristar. It used a green laser with the beam split into 3, and aimed thru a special window with lenses designed to focus the beam about 3 feet out from the fuselage.
Where these beams met, a sensor "looked" at the intersection of the beam, and measured the particles of whatever in the air.
The device was enormous, and used a 500 gallon water tank for cooling the laser.
Although the system did work, it had a sideeffect.. :)
We flew at night over the San Joaquin valley... and the phones at the local PDs were deluged with calls about the airplane with the green fire coming out the side!
And when we flew thru a cloud, the whole cloud lit up green!
Apparently the intent was a probe-less surface mounted airspeed-alpha-beta system for a Low Observable... but reason led to the idea that emitting anything from such a craft wasn't a good idea.
From the ground it looked like this.. and from the air..

[Tom Harper]

Sparky,

Spectacular - did they pay you to do that stuff? Looks like fun!

Saw a couple of similar systems on the web. They were getting 3 to 5% accuracy on low velocity flows.

[Neil Walker]

Neil,

Model = a powered model airplane that has been trimmed for hands off, level flight in dead air.

Question 1 = will the model gain altitude when flying upwind?

Question 2 = will the model lose altitude when flying downwind?

Hypothesis:

If RC models respond to wind in the manner described by some pilots, then the gain and loss of altitude can be measured.

How's that?

When I was in school, an experiment was defined as a question. A hypothesis may be required to explain the result.

All measurements are observations. All observations are not measurements. Not the same thing.

Now we're getting somewhere, although I don't think that's really what you are trying to prove, is it? I think it's what you think one of the ramifications of your idea will be. Back to this "The basic issue is whether a model flying in an air mass matches the velocity of that mass", and going back to your original post now it's starting to make a little more sense - it's all tied into this idea you have that airspeed *is* somehow linked to groundspeed because induced drag is somehow tied to gravity. Basically you are trying to chop down the "the observer on the ground is just one arbitrary frame of reference" claim... is that it?

I also think a glider is a much simpler system to consider. Making sure the engine/prop doesn't have any unwanted influence on the system is going to be difficult. A glider doesn't have to worry about always having exactly the same amount of thrust driving it forward...

[vintage1]

Model = a powered model airplane that has been trimmed for hands off, level flight in dead air.

Question 1 = will the model gain altitude when flying upwind?

Question 2 = will the model lose altitude when flying downwind?

The definitive answer to both questions is, if the air is smooth, and turbulence, ground effects, and incompetent pilots making tirns that upset the model, and so on are ignored, is 'no' to both questions.

Ther model flies relative to the air mass, what the ground is doing is completely irrelevant unless it actually hits the model.

[Ollie]

"Downwind turn problem" is just about pilot's point of view. I bring out my story about pilot's point of view:


TK was dragged out to a power field on a Sunday morning to see an old
friends immaculate 1/3 scale Piper Cub that he had just finished over the
winter fly for the first time. Well, after they got the motor and
everything straightened out on the ground, he asks TK to take the first
flight, "just in case". He wasn't expecting to do the test flying as it
was a big power club and there were lots of "experts" in attendance, but he
didn't want to see his friend break his model either so he agreed to take
the first flight. So here is a senior guy, a brand new 1/3 scale Piper
Cub, a glider guy to do the test flying, and the club "experts" have now
broken away from running gas thru their engines on the ground to checking
this out like a bunch of vultures.

TK taxis the Cub out, takes off, trims it out, starts backing off the
throttle a bit and the engine sputters and quits. Next thing he knows guys
are screaming up and down the flight line, "DEAD STICK - - DEAD
STICK." He's like thinking to himself, no big deal, got flying speed, got
altitude and over the field, why are these guys screaming... so he goes to
set up an approach and the wing gives a bobble. He sets the Cub in a
thermal turn and spends the next 10 minutes specking it out. Brings it
back down, sets up a nice approach on their paved strip, and rolls it out
to his feet... and yells DEAD STICK.

See:
http://www.silentflight.org/officers/TK_Bio.htm

[Tom Harper]

Neil,

You are partially correct. I have always defended the position that the model moves in a bubble. However, after a flying session in high wind I had some doubts. The model definitely gained altitude on the upwind leg. As Sparky points out this could be due to terrain. So I started this thread with conjecture and wanderings as to what might be the mechanism.

The discussion here took numerous paths so I gave thought to how I could get firm data points. Something other than judgement. And, since some statements seem to be challenging, I reduced the issue to a basic question - does the model gain altitude on the upwind leg.

With an investment of ~8 hours and zero dollars I have a system that will record velocity (speed and heading), altitude, time and position. Gotta be more fun that taking an RTF out of the box to bore holes in the sky.

Once we have data points we can haggle over what they mean. I will record the points in Excel files and can email them to anyone willing to participate.

We are not dealing with a bouyant object in a lab tank full of glycerin. We are observing a self powered device operating in turbulent (a matter of degree) air near the boundary layer. Not a lab situation.

So, my plan is to keep it simple until I have data.

[Tom Harper]

Vintage,

How about this:

You are working on a project for some fantastic airborne device. You have a brilliant revelation. You assume that the path of the device over the ground always matches the moving air mass. You thus eliminate a bunch of electronics and cut the cost of the system in half.

Would you bet your job on it?

[Tom Harper]

Ollie,

Ted Nelson told me a related story.

He used to fly a welding rig into job sites in the Sierra Nevada mtns in an Aeronca C2. To get out he would slope soar the C2 up the face of the cliffs then just head downhill back to the Merced airport. He claimed he could get back faster than the mine owner who flew his stagger wing Beech down the canyons.

[Tom Harper]

Neil,

Why would a glider be a better vehicle for this test than a power model?

[Sparky Paul]

Sparky,

Spectacular - did they pay you to do that stuff? Looks like fun!

Saw a couple of similar systems on the web. They were getting 3 to 5% accuracy on low velocity flows.
.
Paid?... I'd fly in experimental airplanes for fun... but they pay, and handsomely!
What accuracy do you expect to achieve, seeing as the effect you're looking for is way below 1%?

[Sparky Paul]

Vintage,

How about this:

You are working on a project for some fantastic airborne device. You have a brilliant revelation. You assume that the path of the device over the ground always matches the moving air mass. You thus eliminate a bunch of electronics and cut the cost of the system in half.

Would you bet your job on it?
.
How do you measure the velocity of the air mass?

[Tom Harper]

Sparky,

It's a hypothetical. Assume some magic, military thing does it.

[Tom Harper]

Sparky,

Getting paid for having fun! That's what I did when I worked at the IBM R&D lab.

I think we'll get a data set that is accurate. The GPS velocity and altitude readings are good at model airplane speeds. Altitude change, if any, is integrated over several seconds so, even if the effect is small the change should be significant.

I'll borrow my neighbors hand held wind guage and take samples at some interval like 5 seconds.

I think the data will be good. Figuring out what it means may be another issue.

[Roj]

Hi Tom.

Excuse me for jumping in late, I want to add my 2cents to your original proposition:

Let's throw a balloon at 10mph into dead calm.
It deccelerates - its kinetic energy is traded off for friction against air particles so it slows...
It has to eventually stop, to propose that it would keep moving forever relative to the air implies it has infinite kinetic energy.
I (hope!) we all agree on this - it is elementary physics.

Now - similarly:
Imagine there is now a 10mph wind. You are standing still (relative to the ground), and you release a ballon without throwing it.

Guess what? It will accelerate at exactly the same rate as the example above.
Furthermore - it has to reach the same velocity as the wind eventually - anything more or less than 10mph would require it to have infinite kinetic energy.

[Tom Harper]

Roj,

Welcome - Thanks for jumping in.

I just went out in the back yard and made some air speed measurements with a Brunton Sherpa. The sample rate was as close to 3 seconds as I could manage. Here's what I got:

1.3, 1.1, 1.3, 1.5, 1.6, 1.6, 1.6, 1.8, 1.1, 1.8, 1.1

The wind is varying +- 20%. The 3 second period is long enough that the airplane will respond to these variations. The wind is constantly changing the speed of the model relative to the ground. Since the airplane has mass and drag, any speed change involves a loss. Since these losses result from a change in wind speed, energy must be extracted from the wind to balance them. Therefore, the model cannot match the average speed of the wind. There has to be a difference.

I agree that a model floating in a lab tank full of glycerine will match the flow rate of the fluid. But that is not the environment in which an RC model operates.

[Roj]

Hi Tom.

I'm not exactly sure what the test above shows other than in real life, wind is rarely at constant speed or direction.

What is beyond debate is that:
1. In a constant wind (no turbulance or gusts), any flying object - helicopter, bird, RC plane, blimp, ballon, or whatever - at equilibrium (not accelerating) will observe the same air speed regardless of whether it is moving upwind or downwind or crosswind.

2. If airspeed of a flying object at equilibium is not zero, energy is required because the object is doing work (by definition - in the physics sense)!

[Neil Walker]

Neil,

Why would a glider be a better vehicle for this test than a power model?


Because then people can't claim that the engine had anything to do with your results - i.e. variations in rpm for <insert whatever reason here> causing more/less thrust, causing altitude gain/loss etc.

With a glider there are no such problems. One less source of error to worry about.

[Tom Harper]

Roj,

You are correct. We are not in a physics lab. We are discussing the dynamic case.

The steady state case described by you and others does not exist in air movement near the earth's surface. Due to the constant variations the model is never (seldom) at equibilibrium. As you point out the object is doing work. To extract that energy from the wind there must be a net difference in average velocity between the model and the air mass it flys in.

[Tom Harper]

Neil,

No argument.

But, the only models lying around here are the paper and foam slimers I use for aerial photography.

[seaotter12398]

Neil,

Model = a powered model airplane that has been trimmed for hands off, level flight in dead air.

Question 1 = will the model gain altitude when flying upwind?

Question 2 = will the model lose altitude when flying downwind?

Hypothesis:

If RC models respond to wind in the manner described by some pilots, then the gain and loss of altitude can be measured.

How's that?

When I was in school, an experiment was defined as a question. A hypothesis may be required to explain the result.

All measurements are observations. All observations are not measurements. Not the same thing.

experiments are not questions and hypothesis explain nothing because they aren't facts.

observations may lead to questions

hypothesis are possible but unproven answers to the question based on theory, observation and guesswork

experiments are repeatable procedures intended to attempt to prove the hypothesis

facts are the repeatable results of the experiment

when i was in school they called this the scientific method. :)


picky otter :3

[Sparky Paul]

And a properly set up experiment has ALL the variables controlled except the one whose influence is being tested .
Wind speed isn't controllable.
The variable is the airspeed of the plane.
Which is affected by the power used, and the trim.
Power used and trim can be eliminated by use of a glider.
The manuver might possibly be wide 360s. instead of upwind-downwind, with the time for the turns measured.
But it still depends on knowing the windspeed the plane is flying in.
A remote reading anenometer attached to a tethered balloon at the height the plane will be flying..
It will be fun.. :)

[Tom Harper]

Sparky,

I've thought about using a leveler and then making 360 rudder turns. The radius is about a quarter mile.

Another possibility would be to use a GPS servo and program to fly a square. Then it flys hands off with the throttle in a fixed position.

For the first flight my plan is:

Wait until the wind is up to 15 to 20 mph. It needs to be a significant percent of the model's flying speed.

Make upwind and downwind passes. Reduce the throttle setting for a low cruise speed. Trim for level flight.

Make at least four upwind and downwind passes without changing throttle setting and using only R/A control.

Look at the data and see what we get.

[Tom Harper]

Otter,

This is the learning phase. Data first - facts later.

[Sparky Paul]

yeah, first you have to explore what you can do, then how to get what want out of that.

[Tom Harper]

Especially considering the budget.

[Neil Walker]

Otter,

This is the learning phase. Data first - facts later.


To be ultimately followed by the Defining of the Great Question... now where have I heard that before? ;)



Sorry Tom, just being silly. I'm looking forward to seeing what comes of all this.

[Tom Harper]

Clearly, a bit of humor is not out of place in this operation.

[Tom Harper]

Closed the hatches and tested it with an 8:00 AM sun angle. Worked OK but I have to raise the mount more to get 'heading' in the frame. I cropped the frames shown. There's enough room for the display.

Picked up the model and walked around the yard a bit. Reached a speed of 2.3mph (at that speed the error is great). The two images below are #s 10 and 26. The sample rate is close enough to 3 seconds that the 16 samples show a time difference of 48 seconds. I set the model down at a different point than I picked it up so there is a difference in Longitude. I didn't move .001 degree but I crossed it's quantizing level.

Altitude readout shows a difference of seven feet (carried it over my head). The GPS unit's software smooths the altitude reading. A 7ft change in altitude will change the readout a foot every 6 seconds. It's almost linear rather than exponential. In flight it will damp out minor changes and lag the actual altitude.

[capncrunch]

Altitude readout shows a difference of seven feet (carried it over my head). The GPS unit's software smooths the altitude reading. A 7ft change in altitude will change the readout a foot every 6 seconds. It's almost linear rather than exponential. In flight it will damp out minor changes and lag the actual altitude.

I think that could be because they're typically designed for ground vehicles where you wouldnt expect big altitude changes.

do you know how many satellites it can track at once? I have an older unit that plugs into my palm pilot that is incredibly accurate, and I think its because it has a bulkier antenna than other models.

-b

[Tom Harper]

I think you are right. You're not going to change altitude very fast on a hike.

I wasn't complaining, just observing. I think it is integrating the altitude value for accuracy. Detecting a 5 ft change is pretty impressive. I think it will do fine.

[Aerowhatt]

"The wind is varying +- 20%. The 3 second period is long enough that the airplane will respond to these variations. The wind is constantly changing the speed of the model relative to the ground. Since the airplane has mass and drag, any speed change involves a loss. Since these losses result from a change in wind speed, energy must be extracted from the wind to balance them. Therefore, the model cannot match the average speed of the wind. There has to be a difference."

In your cross wind scenario. There are losses but their are also gains. The wind becomes a power source for downwind acceleration and upwind deceleration and Mass (momentum acts as drag to this action). What you will find is that the average speed of the model in the direction of the varable velocity wind is an exact match.

In a variable velocity wind (with quick variations) the model will never reach the maximum wind velocity or the minimum velocity of the wind because of the momentum from it's mass. However the average downwind velocity due to the wind will be an exact match.

Aerowhatt

[Tom Harper]

There may be a match between the model speed and the average wind speed. In that case there will be constant acceleration and deceleration forces acting on the model. Anytime the wind is different than the model there is an energy loss.

Where would there be an offsetting gain?

[vintage1]

There may be a match between the model speed and the average wind speed. In that case there will be constant acceleration and deceleration forces acting on the model. Anytime the wind is different than the model there is an energy loss.

Where would there be an offsetting gain?

Loss to what in respect of what?

Mechanical energy is 1/2 mv^2, and v is a relative quantity..

likewise potential energy is mh, where h is a relative entity (height).

If you are flying into wind, encounter a gust and the model thereby climbs 100 feet, it's gained energy relative to the ground. It has in effect greater (self) destructive potential... if it hits the ground....on the other hand its airspeed relative to the ground will diminish, so it has less kinetic energy, with respect to the ground...

With respect to an F-18 doing a low pass and mach 0.7, it has MUCH more kinetic energy of course. Scads of it.

Your confusion lies in misunderstanding what energy actually means.

It is the work needed to be done to bring the model to some state with respect to some other reference point. Discussions about what energy an airplane at the equator which is already doing 3000 mph or whatever the equator does as the earth spins (relative to its orbit), should be moved to the as yet unformed modelling pseudo-science thread.........along with discussions about what the kinetic energy of such a model is relative to the integral center of gravity of the known universe.. :rolleyes:

[biber]

Well said, vint, except it is E_kin = mgh

;)

But apart from that, you still have a point.

biber

[Tom Harper]

Vint,

Thanks for the comments:

"It is the work needed to be done to bring the model to some state with respect to some other reference point. "

Yep.

The model is flying level in an air mass whose speed is varying +- 20%. The reference point for our discussion is the average velocity of the wind parallel to the ground (thus the flight path of the model). Since the period of the variations is long compared to the response time of the model, the model will constantly gain and lose speed. In each case it is due only to the force of the wind. Energy is required in both cases.

In both cases work is done by the air mass. Work requires a difference in potential - height, pressure, voltage, velocity etc. - hence the speed of the model must lag behind the average speed of the air mass.

My confusion lies in trying to understand how your model constantly changes speed without work being done.

[Tom Harper]

We are ready for a first flight this morning - and there is not a breath of wind.

It will be a base line day unless something changes.

[Tom Harper]

Turned out to be a 30 second shake down cruise.

I walked west to set it on the runway. Decided not to hand launch - just let it drag itself off of the ground. It accelerated south and was airborne by nine seconds at a speed of 14.2 mph. At 18 seconds it was nose high so Carl dropped it into a right turn. It accelerated to 25.5 mph and (we discovered later) the fuel line slipped off inside of the tank - aaaaarrrrgh. The nose dropped and it was in the creosote by 33 seconds. Did enough damage to stop flying for the day.

But, we learned quite a bit:

The thing works.

I am surprised at the low take off speed. The model flys slower that I thought.

Vibration is not a problem. A few numbers were elongated. Nothing serious.

Glare is a problem. As soon as the model rotated for take off the upper portion of the screen washed out. I can add layers of diffuser paper.

May try again tomorrow.


Time Altitude Speed Heading
0 4787 0.0 253
3 4787 7.1 160
6 4787 11.0 161
9 4798 14.2 162
12 glare 21.8 153
15 glare 22.9 166
18 glare 18.9 195
21 4856 24.7 280
24 glare 25.5 215
27 glare 22.0 250
30 4847 15.6 298
33 4845 0.0 298
36 4844 0.0 398

[vintage1]

Well said, vint, except it is E_kin = mgh

;)

But apart from that, you still have a point.

biber

I left the constants out...;)

[vintage1]

My confusion lies in trying to understand how your model constantly changes speed without work being done.

And wind gusts do not do work on the things they act upon?

I am sure that next time you see a tornado rip up a telephone pole and drive it through the side of a house, your understanding will improve ;)

[Tom Harper]

Vint,

So, you agree with me. The constant acceleration and deceleration constitutes work.

Now the next step. How will that work be accounted for in the performance of the model. I say the model will have a speed lower than the average speed of the air mass. If you do not agree than tell me where the energy comes from.

[Tom Harper]

Vint,

Can a sailboat on a dead run equal the speed of the wind that drives it?

[Aerowhatt]

Vint,

Can a sailboat on a dead run equal the speed of the wind that drives it?

It could if it were floating in the same wind.

Seriously, the sailboat is a terrible analogy. It has the drag of the water against the hull working against the wind.

The choice of analogy is telling here too I believe.

You asked earlier about where the work is being done in a variable wind. It's simple, when the wind strengthens it does work to accelerate the plane downwind. When the wind lulls the plane acts to accelerate the wind. Both processes have losses which are equal and opposite. Conservation of energy!

What is missing in your observation and analysis, is vector analysis. You must make the airplane the frame of reference and choose positive and negative directions. Some of the force (and or velocity) vectors have positive value and some have negative value.

Aerowhatt

[vintage1]

Vint,

Can a sailboat on a dead run equal the speed of the wind that drives it?

No. Sailboats are faster across the wind than running before..the wind in the latter case is the 'pitch speed' pf the combination.

[vintage1]

Vint,

So, you agree with me. The constant acceleration and deceleration constitutes work.

Now the next step. How will that work be accounted for in the performance of the model. I say the model will have a speed lower than the average speed of the air mass. If you do not agree than tell me where the energy comes from.

I don't understand the question. Speed of model relative to what?

Eeverything you post suggests a deep lack of appreciation of the relativity of motion and kinetic energy.

You cannot talk about the 'energy of the model' unless you state with respect to what frame of reference you are measuring it.

Kinteic energy is all potential until the model hits something...then what is released is a function of the speed difference between the model and what it hits..

In the case of a model glider circling in a steady wind, it has two speed vectors - its own motion, realtive to the wind, and that of the wind, WITH RESPECT TO THE GROUND.

It's crash energy INTO THE GROUND is the vector sum of these..

If you like, the wind speed relative to the ground represents an energy gain, or loss...by slowing the wind down with a windmill, that energy can be extracted..

[Tom Harper]

Aerowhatt and Vintage,

The model vector is tangent to mean sea level. Assume the wind vector is also tangent to mean sea level.

The length of the wind vector is constantly changing. The model is a mass bearing (gravity) on the interface between two surfaces (model and air), hence friction. The changing wind vector accelerates and decelerates the model. In each case energy is transferred from the wind to the model in the presence of friction. Work is done.

The problem:

In your scenarios the energy conversions or transfers are 100% efficient. You acknowledge that work is being done but do not account for it's cost. In the case of the windmill the wind is slowed because work is done. The same must be true for the model.

A 100% efficient energy transfer would allow a pendulum to swing forever. That is the stuff offered by Prof. Dimento in the psuedo-science lab.

[Roj]

Tom,

There are many wise people on this forum. You seem like an earnest, genuine guy. They are in unison, trying politely to tell you that you're simply wrong.

Tom, what is being debated here is not some recent controversial concept in astro-science. It is elemetary 14th century physics. It is beyond debate.

No amount of hiding behind loose terminology can change that.

Here is the cold hard reality:

Your floating object, in equilibrium, can ONLY have ZERO velocity relative to the wind.

To us (on the ground) looking at a dandelion floating past in a wind, it might look as if work is being done to move it, but NO work is being done.
It needs NO energy to move along at constant wind speed.
It would actually REQUIRE energy to move slower than the wind.

I suggest you forget turbulance and even gravity for now - just get this basic concept of objects in equilibrium mastered first.

This thread is going nowhere - i'd advise all except those of you who still believe the Earth is flat to bail out.

[Tom Harper]

Roj,

You and others have been kind.

"Your floating object, in equilibrium, can ONLY have ZERO velocity relative to the wind."

If one eliminates turbulence and gravity your argument is correct. You have the ideal theoretical case. All objects reach equilibrium. I give you (for the moment) that the dandelion is falling through an air mass that may incidently be moving. It is in equilibrium. The system is 100% efficient.

Therefore we have agreed and set aside the case where the system is in equilibrium.

Here on earth, we do have gravity and we do have turbulence.

We are discussing the case where the model airplane is in an environment of costantly changing wind. The system is seldom, if ever, in equilibrium. Work is required to change the velocity of the model. How do you account for that work? The system is not in equilibrium.

Your 14th century physics may explain your stand. But by the 17th century Newton had figured it out. In the proportion F=Ma, the 'a' term has two components 'a1' and 'a2'. The 'a2' term is a vector opposing 'a1'. The 'a2' term accounts for friction which is a defined as a mass bearing on the interface of two surfaces. Forgive me if my terminology is loose. Perhaps you can tighten it up a bit.

Since the model is accelerating and decelerating it is not in equilibrium. The 'a2' term becomes significant. Work is being done. For work to be done there must be a difference in potential. That difference will be seen as between the average velocities of the wind and the model.

Please explain how work can be done without a difference in potential.

[Aerowhatt]

In your variable wind velocity scenario you keep argueing that everyone else is representing a 100% efficient process. That's simply not true. What we are argueing is that all energy is conserved. Some is "lost" to heating and redirecting air molecules. Some is "lost" to heating the skin of the model.

Efficiency and conservation of energy are seperate issues entirely. Energy has substatively been shown to always be conserved (accounted for). Efficiency is just a measure of how much of the energy used ends up accomplishing what we wanted it too.

Gravity has no bearing on the wind vectors what-so-ever. It is perpedicular to the wind vector and therefore has a zero component in the plane tangent to the earths surface.

You are correct that the system is not in equilibrium when the wind velocity is variable. However, the average of those variations would produce an equilibrium state.

Basically it sounds to me like your perception is that work is being done on the model by the wind as the wind accelerates it, or decelerates it with it's changing velocity. This is where your thinking is off.

Let's look at kinetic energy. The mass of air in motion in the area of the model has kinetic energy. The model being blown by the mass of air around it also has kinetic energy.

If the air is "gusting" faster than the model then fricton transfers some kinetic energy to the model (some is also converted to heat).

If the wind "lulls" to a lower velocity than the model then the model imparts some of it's kinetic energy back to the air through friction (some is also converted to heat).

Which ever actor (the air or the model) with higher velocity represents what you are calling potential to do the work.

Now I know you are thinking if there are "losses" to drag then they can't reach an equal velocity. This is incorrect though because the "losses" to drag are equal from both "potential" sources. As the model is accelerated by the air, or the air is accelerated by the model. Frictional "losses" reduce to zero as the velocities converge.

In addition, I think it is clear that the transfer of energy from one body to the other cannot end until their "potential" in that exact direction is equal. As long as the transfer hasn't ceased (yet) the velocities are forced to converge by that transfer.

Aerowhatt

[Neil Walker]

Tom, awhile ago in this thread pmachenzie said:


I will say again, the time to make verbal arguments is over. They are of little value.
Do the physics and show the math that supports your claim. Be sure to show that it works from all reference systems.
From the point of view of the pilot on the ground. From the planes point of view. From an observer at the north pole who can see the earth's rotation.
From an observer in space who can see the motion of the earth around the sun.


.. to which you replied "Sorry, I'm not a physicist." That doesn't quite wash. You obviously know quite a bit about physics. Your first post was full of equations, you're quoting Newton, talking about work, energy, mass, velocity, acceleration and arguing strongly about physics but using loosely-phrased english pesudo-physics to do so. The argument can go on and on forever because there is so much room for ambiguity and misinterpretation in everything you say. There's enough physics and a few equations scattered throughout your text to sound all serious and formal, but your point always evenutally boils down to some english-language hand-waving arguments which could go on and on forever.

So I think I have to repeat what pmackenzie said. Let's see the force diagrams and the math. Make a clear, concise, unambigous claim and show the proof.

[Sparky Paul]

I...
Gravity has no bearing on the wind vectors what-so-ever. It is perpedicular to the wind vector and therefore has a zero component in the plane tangent to the earths surface.
...
Aerowhatt
.
The wind is made up of particles which -are- affected by gravity.
These have velocity and mass.
Vector-wise there may be no effect, but physically there is.

[Tom Harper]

Neil,

True - maybe we can get the model in the air toward the end of the week. It is time for data.

I am not a physisist. I am an electronics applications engineer.

[Tom Harper]

Aerowhatt,

"Gravity has no bearing on the wind vectors what-so-ever. It is perpedicular to the wind vector and therefore has a zero component in the plane tangent to the earths surface."

That issue is going to enter this disscussion somewhere down the road. But for now:

For level flight, T=D and L=W

In that basic equation L is the force that opposes gravity - not the lift vector of the wing. The equation must balance even in knife edge flight.

Drag (D) is the force that opposes thrust. It is a vector whose direction is tangent to the surface of the earth. A significant component of the term 'D' is Induced Drag.

For a given airplane the variable in the equation for Induced Drag is Cl (coefficient of Lift). In order to maintain level flight, the model must vary it's Cl until the lift generated equals the force of gravity. As it does so, the Induced Drag vector will vary in direct proportion.

The drag vector is parallel to a line tangent to the earth surface and is proportional (among other things) to gravity.

[Tom Harper]

Aerowhatt,

I am close to agreeing with you. Yes, it is an issue of the efficiency of energy conversion. Yes, I will point out that the frictional losses cannot be recovered. Once converted to heat the energy is removed from the application.

If you could alternately store and release energy with 100% efficiency, a pendulum would run forever. It does not.

[Neil Walker]

I am not a physisist. I am an electronics applications engineer.

Nobody claimed you had to have a PhD in physics and be actively conducting research. I'm a computer scientist by trade... but I did get through 3 years of an EE degree before I realized I was far more interested in combinatorics than calculus! ;) They made us learn quite a lot of physics during those 3 years, so I know that if you're an EE you should have more than enough to formalize what you are trying to say.

[Neil Walker]

The drag vector is parallel to a line tangent to the earth surface and is proportional (among other things) to gravity.

This is the core of your misunderstanding. The drag vector is not neccessarily tangent to the surface of the earth. And it is not in any way proportional to gravity. In the simple case you have chosen it is, but only because you have set the conditions up for it to be so. I could just as easily say that the drag on my favorite glider at some particular airspeed and angle of attack is an even multiple of the volume covered by my aunt's tea cozy times the square of Tom Cruise's show size. It may be true, but it doesn't follow that my aunt or Tom Cruise have anything to do with how my glider flies.

Lift-induced drag has absolutely nothing to do with gravity.

[eflightray]

I've only just started reading through this thread, but it feels like i've been here before, many times.

The usual comment that someone comes up with is to do all the data recording using a free flight model. Not one that has any control input during the data acquisition.

That way there can be no calls of 'outside interference caused such and such to happen'.

[Tom Harper]

Neil,

For an orderly discussion it is necessary to set the conditions. The "Ohms law" of flight is the L=W and T=D illustration. That is specified for level flight. I think it is the best case for discussion.

Since you do not agree with my discussion, can you explain how the induced drag vector is other than created in opposition to gravity and felt in opposition to thrust. If you wish to use knife edge flight, fine. Just use the lift vector that opposes gravity. That is the only one that satisfies the L=W, T=D equations for level flight. I am eager to see your presentation.

I haven't found the TC (Tea Cozy) term yet.

[Tom Harper]

eflightray,

I am too old to chase free flights.

With an RC you can somewhat determine it's path relative to the wind.

[eflightray]

eflightray,

I am too old to chase free flights.

With an RC you can somewhat determine it's path relative to the wind.

It's also a great way to introduce other factors. At least fly your data acquisition flight in 'free flight' mode, then use controls to bring it back to land.

[Tom Harper]

eflightray,

First we'll get data. If we need to go further I can add a GPS servo and fly a fixed path.

[Neil Walker]

For an orderly discussion it is necessary to set the conditions. The "Ohms law" of flight is the L=W and T=D illustration. That is specified for level flight. I think it is the best case for discussion.


It may be the easiest case for discussion, but most importantly it is the one that allows you to do the magic hand-waving term substition that fits your logic. Bank your wings to make your downwind turn and it all falls to pieces.


Since you do not agree with my discussion, can you explain how the [...]


Sorry, I think I'm done here. This discussion was interesting due to all the hypothetical thought experiments in it, but after following it for a few days I have to agree with Roj that it's not going anywhere. Good luck with your data collection model though, I hope you learn something from it.

[biber]

Well, many cores of misunderstanding seem to have been detected already... :rolleyes:

I wouldn't say there is no connection between cwi and gravity vector.
It's only that it has nothing to do with the wind vector.

biber

[Tom Harper]

Neil & Roj,

Perhaps it's time for a break.

SYAL

[Sparky Paul]

The "free-flight" idea has some merit.... with the plane trimmed to circle, hands off, the resultant flight path when plotted for a few circles should show the wind drift.. comparing that to the measured wind speed.....

[Tom Harper]

That is true. I think the GPS servo would do the same thing - especially with altitude hold. Not sure all of that stuff will fit in this model. It would be fairly easy to connect a parallel servo to indicate the altitude hold corrections.

I could use a rig like that for AP.

[Roj]

Tom I'm back.

i thought of a simple thought experiment to clear up some misconceptions.

Let's start with a plane flying in calm air, at 100mph groundspeed..
Let T=D and L=W, like your example.

Now a 20mph tail wind develops.
The plane's ground speed increases - but you make the proposal that it can't accelerate to 120mph groundspeed, because of small but definite energy losses. So it ends up with an airspeed of just under 100mph.
Right? This is what you're proposing!

Your explaination is that the air mass isn't 100% efficient at speeding up the plane. You say some kinetic energy from the plane has to be lost, to heat up air particles by friction or whatever.

Ok, go back to the calm air, plane flying at 100mph again.
This time, a headwind develops at 20mph.
The planes groundspeed slows. But by how much?

This time, the air mass isn't 100% efficient at slowing the plane down.
This time, your theory predicts that the plane will have to gain from the deal. It ends up with increased airspeed for free!

I'm keen to hear your thoughts on this Tom!

[vintage1]

I'm keen to hear your thoughts on this Tom!

Somehow I think that is one of life's pleasures I will have to regretfully decline... :rolleyes:

[biber]

Good one, Vint! :D

Relaxing this thread a bit, hehe.

biber

[Sparky Paul]

Roj, I think the difference between wind speed and airplane speed Tom would expect in the 100 knot-20 knot example might be only a mph or so, not 20.
If there is such an effect, I'd expect it to be .001 mph-- or less. :)

[Aerowhatt]

.
The wind is made up of particles which -are- affected by gravity.
These have velocity and mass.
Vector-wise there may be no effect, but physically there is.

I didn't say they weren't effected by gravity I said the wind vector was perpendicular to the force vector of gravity. Gravity has no component in the direction of the wind or any other direction in the tangential plain (to the surface of the earth) that the wind vector resides.

The air molecules and the model are acted on equally by gravity in an irrelevant direction to the "problem" being discussed. Your comment is irrelevent and moot.

Puting mud in the water doesn't change the current or clear the view!

Aerowhatt

[Sparky Paul]

Winds around here go up and down... directionally wise speaking.