Galloping Ghost crash Reno 2011

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

Joined
Sep 21, 2011
Messages
287
Location
Orlando, Florida
It could be said that the engines were too tightly cowled, so minimizing volume for low drag might have ultimately caused the engine fire. If that's the case, the aerodynamics could have been a problem.
In any case, I don't think they ever got over 400 mph, so the aerodynamics were unproven at 500 mph.
How the hell do you too tightly cowl water cooled engines?
 
Joined
Sep 21, 2011
Messages
287
Location
Orlando, Florida
Did one guy do the CFD, propulsion integration, stability, design? Bet they had a top aero guy, or guys, that understood when and where the CFD was lying to him/them, top propulsion guy/s, etc, all with experience in their field to pull it off.

The statement was, 'Man can progress'. Some design space is too complex for a man to succeed in his lifetime.
They had ONE guy -- Ron Ayers. When the CFD results agreed with rocket sled tests using a scale model, Ayers felt confident to proceed.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
How the hell do you too tightly cowl water cooled engines?
Not all cooling on a water cooled engine comes from the water jacket or oil - virtually all water cooled engines require a specific amount of air flowing through the engine compartment, especially around the case (or block), in order to get the full benefit of the "total" cooling system. A tightly cowled engine, especially one where the cooling design is optimized for high speed operation, can cook itself without that peripheral flow.
 

Toobuilder

Well-Known Member
Log Member
Joined
Jan 19, 2010
Messages
5,364
Location
Mojave, Ca

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
I think that Scaled Composites is the right organization to design an Unlimited and that they should go conservative on the engine, such as an R2800, and design and build a nice airframe to go around it. They should finance it in-house or partner with an intelligent person who won't rush it and has no incentive to (Red Bull, hey...). They have demonstrated that they can do high speed flight and if the developement program is not botched it could be a spactacular result.

I say this not because I don't think Orion can do it, but because of my general lack of "inside" knowlege and because Scaled Composites achievements are very visible even to someone like me. Orion's team's choice of the i0-720s is heartening because it is conservative and I hope that his team is wise enough and financially strong enough to give him the autonomy he needs to get the job done successfully and without cutting any corners (pun semi-intended)!
 

topspeed100

Banned
Joined
May 4, 2009
Messages
4,063
Location
Oulu/Finland
I think that Scaled Composites is the right organization to design an Unlimited and that they should go conservative on the engine, such as an R2800, and design and build a nice airframe to go around it.
I say this not because I don't think Orion can do it, but because of my general lack of "inside" knowlege and because Scaled Composites achievements are very visible even to someone like me. Orion's team's choice of the i0-720s is heartening because it is conservative and I hope that his team is wise enough and financially strong enough to give him the autonomy he needs to get the job done successfully and without cutting any corners (pun semi-intended)!
Clever move to use an aviation engine in an aeroplane ! There are no unknown factors involved !....eh now I see the problem..it doesn't fit in the Super Sport nor the Unlimited ?!
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,775
Location
Rotterdam, Netherlands
Designing something new for 500 mph today is riskier and the tools are the problem. The tools mean one person can get more done and fewer people are involved. It means someone with no prior experience designing Reno racers can design one in their living room, or small shop, and using compsoties, can see it fly in their lifetime. The tools also mean that the inexperienced person wuold rely too heavily on the tools, and tend to not know when not to believe them during application. This could attract someone with the capitol, who doesn't know any better, to make it happen if the guy acts confident enough, though it would be a false confidence since he's never done it before. Now this person would be excited about actually getting to build a Reno racer, and would spend most of his time on drawing, integrating systems, and building, to see it come true. He would be so busy getting the final product, that he would overlook most of the knowledge available to him. An example would be not understanding that the current breed has spent the last 1000hp to get that last 10-20 mph battling compressibility. But in his design, he was caught up in getting the pretty laminar flow and spent no time looking at compressibilty, which would be available knowledge that wasn't tapped, so he thinks with laminar flow he can clock 500 mph laps on 2000 hp. The end result wuold have him doing 480 mph, maybe, after ten years of development.

The dangerous part is that the person would have no experience, or time, to become an expert in control, flutter, Mach buffet, etc, and would rely on tools. Then the owner would be anxious to get going at Reno to show his baby off, so the flight testing would be limited. And the absolute worse thing that could happen is if the owner decides he knows more than the engineer, since he only got 480 mph, maybe, and not 520 mph as advertised, and starts doing mods himself, without direction.

That's the quick answer.
I wouldn't exactly call people like Orion and many others who're currently developing unlimited designs "inexperienced".
But let's compare apples to apples. Any clue whether Rare Bear or any of the other top contenders were put through a full FEA/flutter analysis? Are they ground-tested (shaker)? Do full CFD models exist to check for transonic effects? Now these are things (except for transonic) I will do even for my comparably tiny design/project and I'm sure that people who go after a much more challenging design will certainly do so.

Sure, properly and safely designing a winning unlimited ship is a monumental task that's far from easy, even for the most talented professionals. But saying that it's inherently less safe than flying hot-rodded 70 year old designs? The Lancair Propjet cruises at the same mach number as the Unlimited's winner, so far for those "too complex" transonic effects...
 

Monty

Well-Known Member
Joined
Jul 15, 2010
Messages
1,298
Location
Fayetteville, AR / USA
Ground vibration testing will only tell you the resonant frequencies of the airframe. It will not tell you what the excitation is in flight. To do that you have to be able to model turbulence. Not something CFD is particularly good at. There are codes that can do it, but only for very specific cases where the codes have been validated by experiment. In a general case, flying around pylons, encountering wakes of unknown magnitude and direction, I'm not sure how useful CFD will be in finding problems. You have to be able to model the damping from the air as well. Then there is the problem of having the required computational horsepower to do useful work. I spent some time working on a CFD model of an airplane last summer, only to find that the number of cells needed to get the results I wanted was far beyond the capability of my license and my computing power.:mad2:.

To model the structure and the fluid flow together would be a task that I don't really think the state of the art is up to yet. Perhaps Boeing and Airbus with a team of PhD's and custom codes working on it may be able to approximate this, but to quote one of my mentors "you don't know $*&t until you test it"

Any program like this would have to use an existing racer. Instrument the thing extensively. Race it. Collect the data so you can know what your load cases are. Put this in the model. Use the model to make predictions. Test the predictions. Only when all of this matches up are you ready to use the tool for design. Even then you must validate your design before you fly.

I can see a small well funded team being able to design a clean sheet unlimited. The tools available now certainly make it feasible. The ever present impediment is "well funded".

I see the power plant as being the main problem, not the airframe. Propulsion drives airframe design. I think that sport gold class is a more realistic place to play.
 

Toobuilder

Well-Known Member
Log Member
Joined
Jan 19, 2010
Messages
5,364
Location
Mojave, Ca
If I recall the program correctly, yes you're right. When I saw it the plane was undergoing some prep maintenance - while the work looked good, talking to the folks there at the time I was not convinced that they did their homework on the structure and on the incorporation of the tail surfaces. Nothing specific but the answers seemed just a bit too vague to convince me that proper engineering analysis was in place prior to airframe assembly. The bottom line of their reasoning was "Well it works on the high speed Lear jet...." Too bad.
Thanks for that insight, Orion. I never saw the airplane in person, but I often wondered how well the structural integration of the Lear parts went into the "sort of" Mustang fuselage. As a scratch built fuselage, did they account for the different attach points/loads right from the start, or did they build to the North American prints and adapt after the fact... (not asking, just wondering). As for the comment that "it was good enough for the Lear..." The comment seems to overlook the significant difference between a relatively stable slipstream of a jet vs. flying through the wake of a 16 foot diameter, counter rotating propeller.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Couple of interesting points that relate to CC's post - often time in dealing with customers I'm put into a position where I have to educate said customer in order for them to understand the level of work or complexity that's involved in any new development. This can be dangerous since often times you have to tell said customer that his assumptions are incorrect and that the design project has to go in a direction that he may not foresee. The race plane is a good example - it takes a bit of talking to make folks understand that a plane designed for the course needs to be optimized more for the turn rather than for the straight. No, you don't ignore the straight but it is the turn that can make or sink a good round course race plane design.

One of the nice things about our customer is that he is not the gung-ho type and understands that with the relatively conservative choice of the TIO-720's the plane's competitiveness will be somewhat compromised. Even considering that we will eventually boost the engines to somewhere around 900 hp each, the plane is unlikely to be too competitive in the Gold Cup. I'm pretty sure we can win Bronze and be competitive in the Silver race but our projected top speed is in the 460 mph range (give or take a bit) which is well below the top speed performance of the top class winners.

The customer has "survived" about ten years on the air-show circuit so he understands that pushing things in aviation is only a recipe for problems.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Thanks for that insight, Orion. I never saw the airplane in person, but I often wondered how well the structural integration of the Lear parts went into the "sort of" Mustang fuselage. As a scratch built fuselage, did they account for the different attach points/loads right from the start, or did they build to the North American prints and adapt after the fact... (not asking, just wondering). As for the comment that "it was good enough for the Lear..." The comment seems to overlook the significant difference between a relatively stable slipstream of a jet vs. flying through the wake of a 16 foot diameter, counter rotating propeller.
Exactly - different airframes, different environment. But please let me point out that that was a perception I got from only two visits to their hangar. I was not the proverbial fly on the wall during the design/integration phase so do not have any deeper insight than that.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,775
Location
Rotterdam, Netherlands
Any program like this would have to use an existing racer. Instrument the thing extensively. Race it. Collect the data so you can know what your load cases are. Put this in the model. Use the model to make predictions. Test the predictions. Only when all of this matches up are you ready to use the tool for design. Even then you must validate your design before you fly.
There is software available that does flutter analysis. Not exactly Microsoft paint in simplicity, but there are companies that provide analysis in an "affordable" (6-digit) price range. I know several of the smaller GA companies outsource to them and if I'm not mistaken Hollmann does it too. Several people have promoted (PhD) on the flutter analysis of a certain design, using the research facilities of a major AE university.

As for turbulence and it driving flutter, that's virtually impossible to predict, given the non-linearity of it. Weren't the existing cases (during Reno) almost all "clean air" events?
I see the power plant as being the main problem, not the airframe. Propulsion drives airframe design. I think that sport gold class is a more realistic place to play.
There I fully agree. And don't worry about me, my design would be very competitive at Reno, somewhere at the bottom of bronze sport with a seriously boosted engine ;)
 

Radicaldude1234

Well-Known Member
Joined
May 30, 2009
Messages
402
Location
Front Range, Colorado
Well, I think the underlying issue at hand is bad engineering, something that is nothing new. I think the problem is with the mentality that new tools such as CFD and FEA are being used; that is, as a shortcut rather than another layer of insurance against physical failure. From what little I've seen of the engineering industry (as an engineering student), this is a mentality that is being more and more pervasive. I mean, simulation tools can be used to create an unbelievably optimized design if used correctly, but I think what is happening, at least with my fellow students, is that the underlying concept is approximated in the design and minute details are tweaked in an attempt to make it work. I mean, its understandable to try to obtain more results from less work, but if the basic concept is flawed...

I think what CC is worried about is the ability of someone using these new tools to disguise the shortcomings of new designs well enough to seem passable, but not being safe enough for the actual race. I fully agree with that in that simulation should be used an extra step between design and testing in the development loop; as a way to validate and test a sound design digitally before needing to do it physically.

What I don't agree with, though, is that the current racers are and would be any safer than any new design. The fact is that the aircraft as a system is not designed to go that fast. I know it's been often quoted that those WWII fighters are designed to go ~450mph, but that's at higher altitudes of 20,000ft and above. The fact is that those old warbirds usually didn't exceed 350 mph down on the deck. So the teams might focus their modification on certain critical areas, like the wing and empennage, but it's easy to overlook little details that have huge ramifications when you are essentially fixing problems as you see them with on-the-spot solutions...

There's nothing wrong with that with many engineering solutions, and that's what inspires innovation. The problem with current air racing, though, is that the validation of said innovation occurs at low altitudes and in front of thousands of spectators. I'm sure there's limited flight testing done privately by the race teams, but you should not be able to test in the race environment if you are following the law.

But this thread is about solutions, not a critique of the current air racing and engineering mentality, so what I propose (and I'm just writing off the top my head here) is that they (the race officials) require standardized flight tests be conducted some time before the races. The tests would prove the airworthiness of the aircraft in a race environment whilst endangering only the pilot. In other words, a flight test program would identify the flight envelop at which the aircraft could be safely operated. Only the race council would be privy to the results before the actual race, and any deviation (within a certain percentage) from the flight envelop would be disqualifying.

So the reasoning is that race aircraft should be tested in a safe environment, IE without spectators, in order to prove they can function properly in the actual race. If their performance in the race is below the demonstrated flight envelop, who cares. But if it is obvious they were sandbagging it in the flight tests, then the race team has willfully withheld information pertaining to safety. In essence, they are have designated themselves test pilots and the audience as crash test dummies during the race.

On top of that, certain performance requirements can be mandated and enforced, because everyone has to flight test. I think a good idea would be to mandate a minimum speed, as that would would automatically dictate reasonable wing loading. I think I read somewhere (I think in the EAA magazine) that the Galloping Ghost had approach speeds close to that of a MiG-21. Another would be some control requirements, such as roll, pitch, and yaw rates to indicate sufficient maneuverability, along with favorable control forces for the pilot.

Would it be more work? Of course. More red tape to climb through? Hell yes. But I see it as one of the few ways the air racing sport can avoid government regulation and remain self-policing...
 
Joined
Sep 21, 2011
Messages
287
Location
Orlando, Florida
Slow the Unlimiteds down simply by going back to the Thompson trophy race distance of 300 miles. People aren't going to push WWII airplanes to 500 mph if they have to make them last for a race that takes 45 minutes.
 

jlknolla

Well-Known Member
Joined
Jul 9, 2009
Messages
311
Location
San Diego CA, USA
Too many solutions looking for a problem. This was a freak accident, nothing more.

Making rules to feel better will do nothing for real safety or innovation.

I finally got a debrief from a friend who was on-site, he reported that GG appeared to develop an issue at the end of its run in the Valley of Speed, GG basically had passed Rare Bear but looked to be developing a low cycle pitch oscillation, then went wide into 7 and 8 where Rare Bear passed GG, then GG pitched up violently and the rest is history.

NTSB Preliminary offered nothing (as expected), will be interesting to see what information comes out.
 

topspeed100

Banned
Joined
May 4, 2009
Messages
4,063
Location
Oulu/Finland
It's also fun to work with doers, if they've designed and built past projects. If anybody's claiming they can do 520 mph at Reno on 2,000 hp, then they've picked the wrong aero guy for the program. The inexperienced and over confidence comments would fall on the aero guy in that case.
You could also say he is a genious. Bugatti Model 100 Racer Specifications

Also He 100D plane if given change to fly at same altitude ( higher ) as Me 109 R would have reached 480 mph with well less than 2000 hp with a conventional looking fighter design ( and boil off cooler ) already in the thirdies. H-Model Mustang went past 480 mph at 2250 hps and that is pretty much 30% bigger than He 100D was. Neither of these ever raced in Reno.

I agree with the doers etc.
 
Top