Aircraft performance/flight summary formulas?

[groundfx]

I would like to write a computer program that searches for an optimal power system (electric or glow) and aircraft size for a given set of tasks (E.g. a specified minimum rate of climb, maximum rate of descent in glide, maximum stall speed, minimum range/duration, maximum cost)

I can't really use the various *Calc programs because they don't consider things like overall cost of goods in their optimization calculations.

I understand the formulas for predicing motor (and battery systems) performance and I have power curves for the glow motors I'm interested in. However, I know next to nothing about predicting aircraft performance.

I think the formulas I need to know are:

1. How many watts to turn prop at X rpm?
2. What is the thrust from prop at X rpm at an airspeed of Y?
3. What is the lift of an aircraft at speed Z?
3. What is the drag of an aircraft at speed Z?
4. What is the rate of sink during a power off glide?
5. What is the rate of climb at full power?
6. How much power is needed for level flight?

Can anyone help me with these formulas or point me in the right direction?

Thanks

[Ollie]

The formulas you are looking for will analyze a design but they won't synthesize a design for you much less optimize a design. Seeking an optimum design is an art rather than a science because it involves subjective tade off judgements between conflicting objectives.

It has been truely quipped that," An aircraft is a set of compromizes flying in formation." It has also been said, of aircrat design, with only slight exaguration, that,"Everything affects everything else."

In other words, there isn't a deterministic solution to your objective. IMO, that's why you can't find a computer program that does everything you want.

Even on the analysis side there are parameters that are best handled emperically. Different design propellers of the same diameter and pitch rating will load a motor differently because of blade width and airfoil differences. They will result in different RPM and current from the same motor and battery. Collecting this data will be a formidable task in itself.

[Dick Huang]

groundfx,
All that Ollie said is true but if you are determined to proceed anyway I have enclosed some files that should give you a start.
Dick Huang:)

[Dick Huang]

groundfx,
Here is the second file.
Dick HUang:)

[groundfx]

Ollie:

Thanks for your insight. Yes, I'm naive... and probably trying to do something that I'm not supposed to be able to do. :D But maybe I made it sound like I was wanting perfect forumlas that give perfect answers... I don't... I just want some formulas that are good for comparing designs and will put me in the ballpark.

Dick:

Thanks! Your help has definitely pointed me in the right direction. I've been able to figure out power consumed by a prop, static thrust, stall velocity and sink velocity.

(Although... I think, your static thrust formula has a typo. I think the constant in the formula should be 1.0e-10 not 10e-10.)

I'm still stuck on a few things:

In flight thrust. The formula you give doesn't seem to make sense. It implies that thrust is infinite if the airplane is not moving! :eek:

Thrust and Drag. What's q and SW?

Rate of climb. I'm not really sure what I'm supposed to plug in for T, D, V, W.

[greenshamrock]

How about several ezoners teaming up and writing elements of a program so as to make one big open source program that we all could use

I havent much experience with programming but am busy trying to learn JAVA just for the hell of it


My particalar reasons that I find most programs defficient is I have no access to what formula was used and often no access to the vairables that are pluged into the formula

Often I have to take the output look up the formula reverse enginneer absurd equasions so as to figure out what vairable was used and then hand calculate that vairable

over aperiod of maybe a year a useable clunky open source program might evolve which with open source others who are more profient at programming could tidy up

Mark

[Dick Huang]

Originally posted by groundfx
Ollie:


Dick:



I'm still stuck on a few things:

In flight thrust. The formula you give doesn't seem to make sense. It implies that thrust is infinite if the airplane is not moving! :eek:

Thrust and Drag. What's q and SW?

Rate of climb. I'm not really sure what I'm supposed to plug in for T, D, V, W.

In flight thrust: at zero velocity the prop eff "eta" is also zero. this formula is not tobe used to calculate static thrust.

q is the dynamic pressure or 1/2 rho v^2 where rho is the air density. sw is the wing area.
if you will make a run in M-cal for your model and use the T = thrust,D=drag ,W= weight and V = velocity at a velocity printout; the T-D/W units in oz cancels out and when multipled by velocity in mph yuu have climb speed in mph which can be converted to ft/min (88 fpm=60mph). This is the numbers printed on the last input line; also m-cal provides sink speed on the last input line see if you can make the formulas give you the same answer.
Because MS *.doc files must be converted to txt files in order to put it in e-zone all the greek letters get screwed up.

Any way I think you could use M-cal to determine a lot of numbers.
For examlpe if you plot thrust and drag as a function of velocity you will see v max and max rate of climb.
Dick


:)

[Ollie]

There is some hope for writing programs that simulate the artistic aspects of the design process. These programs do it the old fashoned way, by evolution. The programs work by taking a seed design, analyzing it and comparing the results to a set of objectives. Then the program makes random changes in the seed design and reanalyzes. If the new design comes closer to the objectives it is kept but if not it is killed. The process is iterated until the results come arbitrarily close to the objectives or until successive iterations make no improvement. If you could write a program that made insightful rather than random changes, then you would have arguably created a form of artificial intelligence that rivaled human intelligence.

The human designer often is more efficient than the program's emulation of natural selection because the human designer's changes are driven by insight rather than made at random. On the other hand, random changes sometimes lead down an evolutionary path to success that would not happen by human insight.

[Mike James]

I was involved in writing "Aircraft Design" software for RC, back in the mid 80s, and this is how we did it.

First, every scrap of data we could find for each type of model (Pattern, Sport, Glider, etc.) was collected. It was separated broadly into "sizes". (.40-size, .60 size, etc., or if unpowered, the span was used)

The "program" simply took a given parameter, such as wingspan, combined with the "type", (i.e. a Pattern plane) and then spit out a list of averaged specs, which would get you in the ballpark, as far as basic measurements, weights, etc.. There's still PLENTY of room for variation, due to the things Ollie mentioned, as well as cosmetic changes.

You can do the same thing today with a spreadsheet, and average the specs from your favorite types of models. to come up with a general idea. But, as Ollie said, it's a lot of work for results which really only allow you to start copying what others have done.

The real proof of any design is how YOU feel about it. If you're happy, then it's "good". If you're not, then it's "bad". You could probably design and build 3 airplanes in the time it would take you to really test existing "design" programs, or develop your own.

[Ollie]

Design is where art and science meet. Each designer will bring a different level of artistic and scientific atainment to the design process. If you are dissatisfied with either side of your design abilities then you can improve the side that seems weakest to you. In the end though, as Mike says, good and bad are very personal judgements. Design, IMO, is at the top of modeling activities and can be enjoyed at all levels of sophistication.

[greenshamrock]

Here in Spain I have the opertunity to try and practice solor flight

In this area most all the programs are not able to deal with the minor vairables

The best aerodynamic program on the market I hava found is visioaero but its motor section is primitive

the other programs motor calc and e calc are more primitive foraerodynamic but have better motor prop cell predicion

using various foprmulas and tried and tested specification gathered from other modellers it is possible to fine tune variables with calculator for calculating solor projects before investing in gear

hope to purchace first sets of panels in 2004

There is a need for good programs open source but its far too big a job for one person so I stick to mini programs with calculator and spread sheets

Mark