Design by the Numbers?

Technical discussions on this Forum sometimes make me uneasy.

Once someone posts some numbers, or predictions about a given design based on numbers, the discussion soon drifts into the "nihilist zone." That is, the zone where theory means nothing. Where the simple, universal laws of nature mean nothing. Where only one's own direct experience matters. Where eggheads get beat up on the playground.

Some good ol' boy pilots go so far as to refer to design theory as the realm of "paper a--holes."

Folks, we'd never have advanced beyond our caves if had not been able to work out the laws of nature. The laws of nature ALWAYS work. You don't have to keep dropping the rock over and over to see if it will fall this time. It will. As Ayn Rand once said, the universe is not a haunted house.

A force X, pushing on a lever Y feet long produces a torque of X x Y foot pounds. Always. Even when the "bestest" good ol' boy (G.O.B.) is at the stick. The G.O.B. may be able to stay alive in an aircraft that is trying desperately to turn upside down, but that doesn't make it a good design.

We all recall the joke about the guy who jumped off the Empire State Building. He commented, as he went past the 50th floor, "So far, so good." Only in "theory" was he in any danger at that moment; he wasn't dead yet. Was he safe?

Aircraft that are "proven" safe by cut-and-try alone should not be deemed adequate, IMHO. They may be right against the edge, like the Empire State jumper. The next clueless newbie may just find the coffin corner that was there all along.

A craft that "proves" itself by testing should also "prove out" on paper. That's where the marginal nature of a particular approach may show its ugly head -- or OTOH where we might learn that the craft has very generous margins and isn't close to the edge at all.

I'm not aiming this rant against any particular aircraft. I'm targeting a particular form of toxic thinking that can impede progress in aircraft design.

And, by the way... the advocates of numerical design (the "paper a--holes") tend to be high-time gyro pilots with 40 or so years in this activity and/or a few thousand hours under their belts.

Designing aircraft by numbers (which today are calculated by computers, not on paper...;-) is state of the art in almost every branch of the industry. The performance and handling qualities of an aircraft or helicopter are calculated to within less then one or two percent routinely and this is the reason why airlines can order aircraft for certain mission profiles before even the first prototype has taken to the air. The highly complex interdependencies of the physics of flying machines can only be expressed and explored using mathematical models. The old naca reports are a vivid testimony to engineers strife of casting the behaviour of aircraft into numbers. An example of this is one of the earliest stabiliy critira for helicopters namely that the curve of normal accelaration become concave downward a maximum of two seconds after a stick deflection. Even test pilots will ultimately try to quantify theire findings by using flight data recorder to log such things as linear and angular accelerations over time. Unfortunately the early attempts to do the same for autogyros have not been followed up and today gyros seem to be designed much more by guess and by golly than any other type of aircraft at least for them, nothing like the collection of some hundred stability and performance coefficints that exists for e.g. a Beach 58 has ever been published by any gyro designer. If the flight characteristics of gyros would be discussed on the basis of hard and fast numbers we'd be spared quite some of the hulla-balloo discussions we've recently seen in e.g. the Magni thread.
Unfortunately interest in this research is very limited. I have started to write a math simulation program for rotary wing aircraft dynamics and it would be great if members of the forum could try to find Universities and people who are intersted in the subject and who would support the effort

PS: the math model for a UH1-H upon which my program is based dates from 1977 and even this old model is close to within 10% of some of the accelleration time histories recorded for the forward flight case. Wouldn't it be great if we could predict the combination of HTL and H stab volume that makes a gyro prone to PPO to within 10%?

PPS: one of the reasons why the Wrights succeded in designing an aircraft was that they calculated the thrust of theire propellers with an error of less than 5% as far as I remember.

kolibri282 said:

...If the flight characteristics of gyros would be discussed on the basis of hard and fast numbers...

Click to expand...

Bingo. The only people who have numbers are the ones who have them on paper only. There's been no regimented testing of gyros reported to demonstrate the correlation to design numbers, other than hypotheses based loosely on accident stories.

Many of the people buying Magnis and MTs are relative newbies drawn to showrooms by the aesthetics. If they can be turned loose with a little training and go on to live long, healthy lives, there's something wrong with an assumption that anything more than a 2" HTL makes for a deathtrap. Yet armchair critics worldwide jump to say as much, with no data, whenever a new HTL machine appears.

I would love to have confidence in a set of design parameters which would accurately predict the behavior of real-world gyros. So far, nobody has demonstrated more than pieces of it in isolation. And the dynamics of gyroplane stability appear more involved than those of simple fixed-wing airplanes, so comparisons to the Boeing 787 aren't convincing.

Testing actual gyroplanes could only refine and improve the theory. So, why don't any of the designers do it? If someone can debunk the voodoo, please, feel free to begin.

Well...

Data about airfoils can come from NACA's published figures.

Data about what load a H-stab actually experiences in given conditions can be gathered in flight with strain guages. It ought not to be very challenging to modify off-the-shelf, battery-powered digital scales for this purpose.

When it comes to pitch or roll stability that is to be provided by rotor damping, testing gets more complicated. We need a few data (or really sets of data) about the rotor involved. We need to know:

1. The ratio of the rotor's rate of precession to various rates of cyclic pitch input. It'll always be less than 1 if the rotor has any mass at all, but how much is the question. This rate can be estimated using lift curves, RRPM and rotor moment of inertia, but there are enough variables (such as over-and under-balance of blades and their chordwise rigidity) that it would be better to test.

2. The rotor disk's instantaneous lift curve, or at least the slope of it. By "instantaneous" I mean over a time interval too short for the RRPM to change significantly.

3. The rate at which #2 changes as RRPM adjusts to the new AOA.

I'm not sure we can get far without this information. I believe that obtaining it would require, at minimum, an instrumented rotor-testing rig built on a heavy truck. It would need to be able alter the rotor spindle angle precisely, and at various rates. It would need to be able to sample RRPM, rotor thrust and rotor disk angle perhaps a few dozen times per second.

The test truck itself would have to be built with guards to protect the occupants from disintegrating rotors, because it will happen.

The technology isn't very complicated or novel. Cost is the issue. One would need several sets of blades of any given model, not to mention the truck. Both McCutchen and Carter built versions of such a contraption, but I don't know what information they were set up to gather.

I built a very light-duty version, suitable only for testing H-stabs. See photo. A beefed-up, scaled-up version of the same thing would work.

I think the real issue is that gyros have been such a small industry until recently that there wasn't much engineering brought to bear on the subject. Now that there is a growing market for moderately expensive 2 place gyros, there is a vacuum of knowledgeable experts on the subject being filled by amateurs. I am not saying that someone has to be an aeronautical engineer to be an expert on the subject, or that someone with a passion for gyros couldn't become an expert through their own study, but there are very few true aeronautical engineers in the field now, and many people who's knowledge doesn't extend past what they read on the Rotaryforum or the sales brochure of a fancy 2 place.

Another issue is the people some consider experts in the field having conflicting motivations. GG has been a tireless supporter, but what chance was there of the Magni's failing the stability tests when the guy who sells them is so involved with writing the stability tests? I am in no way saying Magni's are unsafe, their record speaks for itself. I frankly believe the fact of their safety record trumps the theory of their unsafeness.

There is also the concern of certain innovations that promote safety may also be proprietary. I guess it could go back to the prop guard/horizontal stab of the Bensen. The offering of a piece of the Magni rotor to Chuck Beaty to show it was balanced, but deliberately withholding other sections of the rotor that were overbalanced. The problem with that is, you have to fight for the acceptance of HTL gyros even though PPO has arguably been the number one killer of Gyro pilots, and you have to do it knowing other manufacturers of HTL gyros might not have the very things that makes the large thrustline offset tolerable in the gyro you sell, the overbalanced rotor and large downloaded stab. At the same time, promoting your relatively safe gyro while promoting the allowance of the sale of less safe gyros. Once you've gone down that slippery slope, it isn't much more to float absurd theories like the ' the drag of the long legs of LTL gyros make them less safe than HTL gyros'.

Thanks guys . . .this will be a precious thread!
Heron

John:

To be very clear, stability theory does NOT say that Magnis are "unsafe." If the moments are always sufficient to snub PPO very early in its development, it doesn't necessarily matter whether they pop up as the PPO is trying to develop, or whether they are present all the time. It's just much harder to measure the former circumstances.

Without certain facts that we don't have in hand, we cannot say that the Magni is either safe or unsafe from our "paper a--hole" viewpoint. We need numbers that describe how exactly the rotor behaves when you start trying to unload it. Unfortunately, that means measuring things that happen in fractions of a second. That takes some electronics and a sort of armored test rig.

What I've tried to describe is within the reach of many people on this Forum. It would not take a PhD engineer with Fortune 500 money backing him. A rattly-bang dump truck, a few hunks of used structural steel and some electronics similar to Ralph Taggart's Digipod (for the Gyrobee) would take care of most of it.

Of course, you'd also need some rotors, spindles and bearing assemblies that might get martyred.

A rattly-bang dump truck, a few hunks of used structural steel and some electronics similar to Ralph Taggart's Digipod (for the Gyrobee) would take care of most of it.

Click to expand...

What you want to have ultimately is rotor forces and moments. Unfortunately on a truck you would not be able to measure these since your values would be obscured by inertia loads generated by your rattly bang truck, so let's rather consider a static test rig at some windy corner of America (Kill Devil Hills...;-)

Good thread subject Doug. You should be commended.

Another issue is the people some consider experts in the field having conflicting motivations. GG has been a tireless supporter, but what chance was there of the Magni's failing the stability tests when the guy who sells them is so involved with writing the stability tests?
John, just for starters, the [ apparent] hypocrisy in his preachings iv been tryn to get him to explain for one.

I am in no way saying Magni's are unsafe,
Neither am i.
their record speaks for itself.
The inherant nature of records is that they are broken.
I frankly believe the fact of their safety record trumps the theory of their unsafeness.
Natures laws of physics dont give a stuff bout records. So the record dont diminish the theory thats based on the facts of nature.

**
One thing you can be sure of is this;
Wen GG says [ and he has said] that wen he shoves the magni stick forward to unload the disc................ .
His definition of "shove" could mean anythn. Be i do know that it dont mean any thing resembling a high cyclic rate, which is wots needed to unload the disc. Coz you simply CANT shove the magni stick anywhere.
This is obviously a good thing, coz it means no pilot can deliberately or accidently unload the disc.
But nature certainly can, and will, eventualy.

Of course, you'd also need some rotors, spindles and bearing assemblies that might get martyred.
And of corse, an unbiased volenteer.

Juergen, I think you could filter out the "noise" as long as you made plenty of runs on a smooth runway. When we tested a H-stab on the rig I pictured, we used a mechanical bathroom scale to measure lift. We drove on a straight public road. There was plenty of "noise," but my friend and I took turns flying the wing and reading the scale (to reduce operator bias) and tried to record the numbers in the center of the dial's swings. The lift curves we got were surprisingly close to what the theory predicted -- including the losses associated with low (2:1) aspect ratio.

If you wanted to get fancy, I suppose you could rig a digital G-meter and record the difference between its output and the output of the one connected to the rotor.

Or just fill the dump truck with sand.

Kill Devil Hills does have a nice historic resonance, though.

Cost has certainly been an issue, along with the suitably qualified people who are willing to become involved.

It has been reassuring to know that in the case of one new recent entry to the field the designer is an aeronautical engineer, and his product would seem to be proving that in it's performance and demonstrated stability.

Another very recent entry has been doing rotor tests with a car, which looked quite reassuring

I have thought about this some and believe alot of the data needed could be gathered from a few strain gauges placed on a machine. Several companies make small stick-on gauges that are very accurate. There are even a few "coatings" that can be applied and then viewed with a "Polariscope" to determine stresses.

Vishay, bachmann, Pulstronic, HBM and MicroDAQ are a few widely known brands.

I started working on a test plan but got side tracked.

birdy said:

I frankly believe the fact of their safety record trumps the theory of their unsafeness.
Natures laws of physics dont give a stuff bout records. So the record dont diminish the theory thats based on the facts of nature...

Click to expand...

Birdy, if only it was that simple. Magnis appear to have had no stability-related accidents, and have passed Section T, while at least one "CLT" machine which has not been put through regimented stability tests looks to have entered an unrecoverable dive. This is the kind of thing we need to resolve.

The problem with old data is that it could be off by 30% in the math dept - even with computers :rant: If the old data is off and used as a foundation for further use the following math will also be off, and that is why you have flight tests to conferm the results, Lots of test aircraft made smokin holes after the numbers added up.:typing: The Wrights had found a lot of Iron rules of thought that had a lot of holes,:twitch: Too much thinkin gives me a headach. Back to the cryons.....

Juergen, I think you could filter out the "noise"...

Click to expand...

Dough,

the propositions in your post #10 are sound and definitely worth a try. To get more data one would probably use electronic data loggers. I think that measuring truck accelerations could greatly enhance the quality of the data obtained, IMUs (inertia measuerment units) have become so cheap and precise over the last few years that theire use is well within reach of anyone. I'd still like to advocate a two pronged approach: take as many measurements on the ground and in flight as you possibly can but in parallel build the most sophisticated math model you can muster. The two lines of attack on the problem will mutually push each other and in the end only a math model is capable of predicting the qualities of a new design ( rumor has it that each Messeschmitt plane killed at least one good test pilot, this beeing the 21st century we have the means to glean insight at a less hefty price)

Juergen, you’re going to put gyroplane “designers” on a diet of differential equations when most can’t grasp the concept that “free bodies rotate about their center of mass when acted upon by a moment of torque?”

I've heard that holograms, rare metals, and a firm opinion increases strength, balance and stamina, that's why I have incorporated them into my new gyro design. I know this to be true because my pet rock told me so.

.

I would imagine that Nicolas would possibly grasp that concept CB, though he probably wouldn't know a pet rock if it hit him on the head.

Magnis appear to have had no stability-related accidents
Duno how many times i gota say it, but ill say it agin, iv never said the magni is unstable.

while at least one "CLT" machine which has not been put through regimented stability tests looks to have entered an unrecoverable dive. This is the kind of thing we need to resolve.
Got any more info of this incident Paul?
I know of one ere in oz too, but it had nuthn to do with stability.

Why is so dificult to discuss stable x stabilised?
I don´think Magni owners will discard their machines OTOH prospect buyers will make a more educated decision.
Heron