Power Calculation Question

[jbatch]

If drag force exerted on a model that is moving through still air
at constant speed and direction can be measured, then it can be assumed that the thrust exerted on the model equals the drag.
If the speed also can be measured, then the actual power of the
propulsion system can be calculated. Finally, if the model is driven
electrically, power consumed by the motor can be measured and
a realistic picture of efficiency can be drawn.

Question: Is there any practical way for an RC'er to do the
necessary measurements without expensive equipment?

[bartje]

Hey Jbatch


there's no easy way to do what you're describing unless you have a very accurate windtunnel or something. :D

Have a look at
http://www.image-maker.demon.co.uk/pages/spd_400.html
At the bottom page there are some good rules of thumb you can start with to make a good guestimate about what your plane will need. Hope it helps!

bartje

[Andy W]

Most of us find the estimates provided by the popular calculation programs are close enough to get the model airborne. Then we tune the power system to achieve the performance we're looking for.
..a

[jbatch]

Bartje-
I very much appreciate your reference to URL giving actual S400
and generalized guidlines for mechanical power exerted by thrust
systems. But I still wonder if it is possible to measure thrust even
without a calibrated wind tunnel. For example, drag may be as-
certained by pulling the model through still air, and measuring
drag force by means of an accurate spring scale. The path of the
model could be linear or circular. This method would provide drag
measurements over a range of speeds. If compared with power
consumed by the electric motor, a curve could be plotted and it
would illustrate the efficiency of power usage of the model over
the full range of its speeds. What do you think?

[Dick Huang]

jbatch,
See the Eagle Tree web site www.eagletreesystems.com
They claim $149.99 + SH. It logs altitude,airspeed,+abunch of rc stuff and is expandable to measure RPM, temp and battery current and voltage. It weighs about 3 oz. All data can be down loaded to a lap top and analyzed later. Sounds like this will answer your question.
Dick Huang:)

[mark_q]

There are at least half a dozen threads on ezone showing different ways to accurately measure static thrust using simple tools and scales.

Really though, I can't see any practical reason why you would need the level of detail measurements you seem to be looking for. I'm with Andy as we all usually only need enough info, as supplied by ecalc and motocalc, to get us flying and then of course we spend a bunch more money and over power everything anyway.

Really though, what is the reason for your need of the nearly unmeasurable? Just curious or are you into computational fluid dynamics research?

Mark

[jbatch]

Mark-
Thanks for your helpful advice referring to easily measured static
thrust, atual measurement of which gives a critically important
factor in determining a model's performance.
The focus of my interest apparently was not well communicated.
What I wanted was suggestions on how it might be possible to
measure dynamic forces in situ, using inexpensive and readily
available tools.

Now to your difficult question on the nature of my interest. Briefly,
I am well into my third childhood, playing with toy airplanes, and
getting all I can out of the hobby. My knowledge of fluid dynamics
is adequate to understand the mechanics of flying. My knowledge
of physciatry, however, is inadequate to explain this obsession
with models. Sorry!

[vintage1]

Jbatch:

I did some very similar 'thought experiements' and ended up deriving teh 50W/lb rule...I won;t actually repeat teh calculations but here are some pointers so that you can excercise the little grey cells and come to the same conclusions.

Lift to drag ratio on a well trimmed airplane is the inverse tangent of the glide angle.

Drag times speed plus weight times rate of climb is output power of prop.

These two together will tell you how much output you need from a prop to get a given model to fly level at near stall speed, and climb at a given rate.

You then have to guesstimate on prop, gearbox motor and drie train efficiencies - motocalc is pretty good here.

Darg at higher speeds is probably calculable from stall speed drag, but I forget the formulae. Try Ollie in the modelling science forum. He seems to be the best aerodynamicist around.

[bartje]

Hey Jbatch

We've done some of those calculations in school concerning aircraft performance, bank performance, etc. But that was only possible because we had all the necessary engine power curves etc.

As far as I know the only direct ways to measure the drag of an antire airplane is to put everything in a wind tunnel and place a "wake fork" behind it to measure the pressures differences behind the plane (pretty complex stuff) or you could attach it to the force and moment measuring scales that a decent wind tunnel is supposed to have and derive the drag from those results.

Oh wait, I've got an idea... :D But you'll need a veeeeery high shed or an abandoned water tower or something to do this. Stay with me on this one.

What you need:
* your airplane
* Callibration weight with no aerodynamic drag but exactly as heavy as your plane
* 2 pulleys
* a very long wire
* a dropweight heavier than your plane with virtually no aerodynamic drag


Now, attach the pulleys to the ceiling of your very tall building a bit seperated so the dropweight can't hit the plane. Pull the cable through it and make it so that when the weight is near the ceiling, the other end of the cable is on the ground. Attach the plane without the to the cable and you're set. Now if you release the plane, the dropweight will fall down and the plane will go up.
In the mean time you need to know the exact distance the weight has dropped, and how long it exactly took to fall down.

Of course you need to do a dry run with the callibration weight to figure out how much you loose in the friction of the pulleys. Now, if all this goes well, the weight should fall slower with the plane attached because the drag is braking it a bit. If you know how much longer it takes its a matter of a few calculations to figure out how much the drag really is. Although I wonder how high the dropheight has to be to make the drag measurable...

The next step is to vary the weight of the drop weight to get readings for different speeds.

Btw, the reason why I'm not advising to do this with a dolly on the ground is that you are likely to get plane/dolly and plane/ground interference that will disturb the results of your experiment.

bartje

[jbatch]

To all who have contributed advice-

The original question has been muddied by a flood of comedy.
The intent of my posting was to challenge the ingenuity of those
who have aerodynamics background and experience in the area.
Let me frame the question in terms of Newtonian Mechanics:
Using inexpensive and readily available tools, how can one
measure the work being done, per unit of time, on the model plane by virtue of its propulsion system? The answer must lie
in determining the force of thrust driving the model forward
through some measurable distance in some time.
Maybe someone out there has done it already?

Thanks to all for their thoughts.

[Andy W]

In that case, this really belongs in Modeling Science, where these matters are discussed in all seriousness! :)

I'll move it over there for you..
..a

[jbatch]

Andy W-

Thanks for the rescue!

[JMP_blackfoot]

If you can hand glide the model, or from some height, and the time it takes to landing. This, with the launch height gives the sinking speed. Sinking speed times the model weight is the power spent by the model drag (= power provided by gravity). Measuring the glide distance will give you the flight speed. Power spent by the model drag divided by the flight speed gives the drag force value. It is best to glide the model without prop, with an equivalent weight attached. This method is as acurate as your stopwatch and tape measure, and your ability to consistantly launch the model at its own glide angle and speed.
Hope it helps.

[Dick Huang]

[QUOTE]Originally posted by jbatch
[B]If drag force exerted on a model that is moving through still air
at constant speed and direction can be measured, then it can be assumed that the thrust exerted on the model equals the drag.
If the speed also can be measured, then the actual power of the
propulsion system can be calculated. Finally, if the model is driven
electrically, power consumed by the motor can be measured and
a realistic picture of efficiency can be drawn.

jbatch,
Your assumption above is only true at maximum possible speed in level flight at constant altitude. See Perkins and Hage, Airplane Performance, page 156. The point that I was making in my previous post is at Vmax, Thrust and drag are equal. Therefore, if you can measure one then you know the other. Also Drag=[(sigma*f/391)*V^2] + [124.5/sigma*e*(w/b)^2*(1/V^2)]. The first term is the parasite drag and the second term is induced drag. All you need to do is instrument your model as I suggested and measure the in-flight total drag or in flight thrust at max velocity. If $149 is too much then I suggest that you use either M-Cal or E-cal to get your answer.
Dick Huang:rolleyes:

[Plantoflap]

Hello jbatch,

I understand the value of what you want to do. I have thought about using ground testing to get good data many times.

I think the Wright brothers flew their gliders as kites while measuring the angle of vertical lines that held the glider down.
The lift would be trying to make the line straight up while the drag would pull the line horizontally. The resulting angle from vertical is the glide angle.

If you look at their wind tunnel balance, they used parallelograms
to hold the test models at a constant AOA while the leverage against a drag element varied. The L/D was found using trig.

I plan to construct an open air version that would use a vertical parallelogram with the model mounted on top and balance weights on the bottom. all 4 corners would have ball bearings as well as having ball bearings at the center of the vertical rods to attach the setup to a support. The support would also be on bearings and have a vertical stab on a boom to keep it pointed into the wind. A plate on an arm submerged in water may dampen out oscillations.
Connections to a computer could give 24/7 data.

I will try it myself when I have time.

George