From: N11TE@aol.com Date: Wed, 21 Jul 1999 01:26:22 EDT Subject: COZY: Structural Question To all 3 and 4-place canard builders: I recognize that this message may elicit some controversy. I simply ask all to read and study carefully what is being presented before any flame wars are started. The whole purpose is to make a better and safer plane for all of us, including me. In January 1995, there was a Cozy MKIV crash in Florida. This was clearly the fault of the pilot who flew too low and hooked a power line and impacted the ground almost directly on the nose at probably near approach speed. The two in the front died but the two young men passengers in the rear seats survived with relatively minimal injuries. My first reaction, after my sorrow that this had happened, was that I was impressed that this structure was strong enough to save these two. I initially became interested in investigating this accident when I was told that the front seat belts did not hold. Some members of this group will remember the controversy that report caused on this forum. The response that I received encouraged several of us builders to go ahead and investigate much more thoroughly. Since there was the potential of a law suit being filed on this accident, we agreed to proceed quietly so as to not become involved. That's why I've waited until now to bring it to everyone's attention. As background, I received a copy of the pictures taken of the wreckage and, based upon questions that these pictures raised, several very experienced builders made a personal trip to inspect it. It was found that yes, the copilot's shoulder harness bolts did pull out. Plus, the whole center keel separated from the bottom and allowed both center seat belt hard points to pull loose. The accident investigators told us that the two men in the front seats were thrown thru the canopy and instrument panel and impacted the ground. They were found outside the fuselage. One investigator even told us he thought they would have had a good chance to survive if the seat belts had held. This, in my opinion, is pure speculation. However, I think you can understand why we started by looking at the seat belt attachments. However, the seat belt question may be only a symptom of what may be a more important concern. When the whole wreckage was studied, it was found that most of the structure between F22 and the forward edge of the strakes simply disintegrated with large delamination areas in the fuselage, itself. In other words, the bigger question was did the front seat belt attach system fail or did the forward structure fail around it? Surprisingly, the nose from F0 to F22 simply split down the middle and half was found with the nose gear strut and gear still attached. This side was still in almost one piece. It apparently did not absorb any of the forces and allowed the area between F22 and the front of the strakes to become the "crumple" zone. Unfortunately, this is where the pilot and copilot are located. Apparently the forces ended up as end compressive forces on the fiberglass foam sandwich fuselage ... which is not designed for strength along this axis. In comparison, during this investigation I talked to a number of canard airplane builders (including a structural engineer) who told me of specific accidents where they had personal knowledge of a VEZ or LEZ impacting on the nose at similar speeds and the nose forward of F22 absorbed the impact (became the crumple zone) leaving the pilot compartments intact and saving the lives of the pilots. Basically - IN MY OPINION - this indicates that the modifications made in the structure when it was enlarged from the LongEZ may have caused loss of structural strength in the area between F22 and the forward junction of the strakes (pilot seating area). It very well may be only due to the change to a removeable instrument cover. Mike Melville pointed out to me that Burt Rutan was very insistent that this cover not be made removable and stated it in the old Canard Pushers several times. In other words, for full strength the top of the box must be closed. Of course, the solution may involve other changes, too. Before everyone gets carried away, let me point out that while this airplane design has never had any structural engineering review, it was built and tested thoroughly. I am quite comfortable with the flying qualities. But, flying stresses may be very different from crash stresses. Even automobile manufacturers who have very sophisticated structural computer design software still go thru crash tests to determine what will actually happen in the event of a crash. This accident was one where the airframe impacted the ground head-on at somewhat near approach speed. Should we say that this was a unique accident and will never happen again? Or, should we look at it as an example of a crash that could be symptomatic of what might happen if one lost power and had to put down and then ran head-on into a truck, wall, rock, house, etc.? I know, I know, the perfect solution is to never crash! I would prefer to make the structure as safe as I can make it for me and my family. To cover the event that I hope never will happen. While we have several potential changes in mind (none that seem too difficult or heavy), we are not aeronautical structural engineers. Is there anyone on this list that has these qualifications and would be willing to do a structural review? Or, do you know of any professional aero structural engineer who would already have some experience in canard aircraft construction and could be hired at a reasonable rate? Tom Ellis N11TE@aol.com From: David Joelsen Subject: RE: COZY: Structural Question Date: Wed, 21 Jul 1999 09:02:38 +0200 Tom, Very interesting story about the crash of a Cozy. I have always been curious about what type of caculations went into the design of the Cozy and other canards. However, no one doubts that the design has plenty of safety margin, but this margin might be lost at any slight change to the design (like altering the instrument panel or LG bulkhead). For your information, all homebuilts in Denmark are all subject to have the seatbelt attachment tested to 9G. Here I know that if I make a change it will have to "survive" this test, and suddently the alteration might not be that attractive. Med venlig Hilsen / Best regards David Joelsen Sales Support, Purup-Eskofot Phone: +45 44 73 66 73 Fax: +45 44 73 67 67 From: DougSheph@aol.com Date: Wed, 21 Jul 1999 12:59:36 EDT Subject: Re: COZY: Structural Question Tom, It's an interesting case you make, and our planes aren't going to improve if we never ask questions like that, but you didn't convince me of your point. Look at it this way: If you don't make the instrument cover removable, you're guaranteed to have a struggle on your hands every time you need to install, modify, test or repair any of your avionics. A serious inconvenience on a fairly regular basis. On the other hand, you THINK that a permanent cover MIGHT stiffen the fuse enough that it could save the occupants in a head-on collision with an immovable object at approach speed. If you had good reason to believe that that was true, you might have a case, but anecdotal incidents and idle speculation don't make a compelling point. Design limits are what they are, and I don't think ANY plane is designed to preserve occupant safety in that kind of incident. I don't think that's a reasonable or necessary design goal, considering the weight and cost increases that would be necessary. My final point: Suppose you put off upgrading or repairing your avionics because it's such a nightmare getting behind that panel without the removable cover. You could wind up pancaked on a missed approach sometime. The odds of that happening are probably in the same ballpark as the accident you cite. The point is that the price you pay (in terms of more difficult maintenance) is too high, considering the payoff, in my opinion. Doug Shepherd From: N11TE@aol.com Date: Wed, 21 Jul 1999 14:15:09 EDT Subject: Re: COZY: Structural Question Hi, Doug! First of all, I want to thank you for your response. You make some valid points. However, it also indicates that I did a very bad job of communicating with you. To a certain extent I expected that this would happen as it was impossible to cover all the information in one short Email. Let me address each area separately: In a message dated 7/21/99 12:01:02 PM Central Daylight Time, DougSheph@aol.com writes: > Tom, > > It's an interesting case you make, and our planes aren't going to improve if > we never ask questions like that, but you didn't convince me of your point. No question but that I was only able to give a rough outline of the information available. Remember that we have been studying this situation for several years. Every builder to whom I have been able to present the whole picture in detail (including a structural engineer) has agreed that it needs to be investigated and changes made. > Look at it this way: If you don't make the instrument cover removable, > you're guaranteed to have a struggle on your hands every time you need to > install, modify, test or repair any of your avionics. A serious > inconvenience on a fairly regular basis. On the other hand, you THINK that > a permanent cover MIGHT stiffen the fuse enough that it could save the > occupants in a head-on collision with an immovable object at approach speed. I did list the removeable cover as one major possibility for loss of structural strength. I did not mean to imply that it was the only one. We see a number of other changes that have been made which could be equally responsible. I mentioned it because it was the first thing that RAF focused upon when I discussed it with them. You are also assuming that there is no way to keep the benefits of the removeable cover but still strengthen the fuselage. If you study the Cosy Classic, with their front-hinged canopy, you would notice that they may just have accomplished this goal in their design. > If you had good reason to believe that that was true, you might have a case, > but anecdotal incidents and idle speculation don't make a compelling point. > Design limits are what they are, and I don't think ANY plane is designed to > preserve occupant safety in that kind of incident. I don't think that's a > reasonable or necessary design goal, considering the weight and cost > increases that would be necessary. I do believe that there is a major problem that needs to be addressed. I would not have posted this message if I did not have this confidence and the confirming response from a number of other builders and engineers. > My final point: Suppose you put off upgrading or repairing your avionics > because it's such a nightmare getting behind that panel without the > removable cover. You could wind up pancaked on a missed approach sometime. The odds > of that happening are probably in the same ballpark as the accident you cite. > > The point is that the price you pay (in terms of more difficult maintenance) > is too high, considering the payoff, in my opinion. > > Doug Shepherd I consider the payoff to be added family safety for which cost becomes an interesting viewpoint. Again, I believe all objectives (including no more difficult maintenance) can be accomplished with minimal weight gain or construction problems. We have a number of simple changes in mind which should solve the concerns.... but we are not aeronautical structural engineers. Therefore, my original post to try and find such an individual. To make sure the revisions are going to be done correctly. All suggested revisions will be shared with all builders. As you are your own manufacturer, you have the right to use or not use them. I would be very happy to explain the problem in much greater detail with you or any other builder if you want to give me a telephone call. Tom Ellis N11TE@aol.com 812-867-2275 Date: Wed, 21 Jul 1999 13:50:47 -0500 (CDT) From: Tom Brusehaver Subject: Re: COZY: Structural Question >> a permanent cover MIGHT stiffen the fuse enough that it could save the >> occupants in a head-on collision with an immovable object at approach >speed. > >I did list the removeable cover as one major possibility for loss of >structural strength. I did not mean to imply that it was the only one. We >see a number of other changes that have been made which could be equally >responsible. I mentioned it because it was the first thing that RAF focused >upon when I discussed it with them. I think you cannot focus too much on the instrument cover either. That would possibly strengthen things between the instrument panel and F-22 (or F-28). The initial post also mentioned the "structure between F22 and the forward edge of the strakes simply disintegrated", indicating that between the instrument panel in the front of the strakes (couple inches) would need strength also. Having a crush zone is a good thing. It will take lots-of-G's(TM) and convert it into not-so-many at key places. That is the key, that the key places have to be where the people are, and the crush zones have to be where fixed parts of people aren't. I would focus more on the seat belt anchors. Let the airplane crush, and have my feet push back (or break), but keep my body strapped to the seat away from the crush. Indy and F-1 cars crash at speeds over 150mph all the time, and drivers mostly walk away. The crush zones allow this. From: DougSheph@aol.com Date: Wed, 21 Jul 1999 16:08:34 EDT Subject: Re: COZY: Structural Question Was said: >Indy and F-1 cars crash at speeds over 150mph all the time, and >drivers mostly walk away. The crush zones allow this. Sure -- and I'm not saying that crumple zones aren't a good thing. But remember a few other things: 1) Those cars also have welded steel cages to protect the driver. The cage and crumple zones work together for maximum effectiveness. 2) They push their machinery to the absolute limit every time they race...and sometimes well beyond! 3) They're expected to crash with some frequency -- something I hope we're striving to prevent. 4) They're incredibly expensive. A few thousand bucks spent on cages and crumple zones means nothing to a sponsored racing team. A more realistic comparison might be Volvo's use of the crumple zones in their production cars. It's not a BAD feature, I just wonder if the cost is really justified in terms of lives saved. In response to Tom Ellis's further comments: If you can make my plane safer, at minimal cost, with no significant weight or performance penalty, go for it. I'm still building my plane, and I'll be the first one to incorporate your changes if they're legitimate. But if it costs me $20 in additional fiberglass/epoxy, or adds one pound to my plane, or takes me more than five additional hours of construction time, I'm not going to do it. You're solving a nearly nonexistent problem. The certain daily rewards of being one pound lighter are greater than the benefits of a possible solution to a 0.01% problem. Remember, the one incident you can point to was caused solely by pilot error, leading me to believe I'd be much better served to take my $20 and five hours spare time and get 15 minutes more training time with my CFI. That will directly address the LEADING cause of airplane fatality (pilot error) and the benefits of that are undeniable. I'm not a naysayer, just a skeptic. And I think you're chasing a mouse with a red Ferrari. Doug Shepherd Date: Wed, 21 Jul 1999 15:40:52 -0500 (CDT) From: Tom Brusehaver Subject: Re: COZY: Structural Question >additional hours of construction time, I'm not going to do it. You're solving >a nearly nonexistent problem. The certain daily rewards of being one pound >lighter are greater than the benefits of a possible solution to a 0.01% >problem. Remember, the one incident you can point to was caused solely by >pilot error, leading me to believe I'd be much better served to take my $20 >and five hours spare time and get 15 minutes more training time with my CFI. >That will directly address the LEADING cause of airplane fatality (pilot >error) and the benefits of that are undeniable. I mostly agree with these sentiments. This is where engineers get all torn up. There needs to be a balance, it isn't a perfect world, how much change can make things better. If the change makes the device perform poorly in some important respect, but fixes a defect, is it worth it. The airplane might be made safer by attatching a parachute system to it, Cirrus is trying this (and other features). Will the extra cost every be tested, or prove better? The Jury is still out, it has worked for smaller lighter airplanes, evidence says it will help. Airbags could have saved lives in cars before they were put in. Once they were put in, they actually had another defect hurting and killing short people (kids mostly). It seemed like the right thing to do, but until someone tested it, the full effects of the change were not realized. Date: Wed, 21 Jul 1999 14:25:35 -0700 From: Stefan G Siegel Subject: Re: COZY: Structural Question While I understand the desire to improve the design towards more safety, I would like to point out that it is not as easy as just hiring "someone who knows about those things" for a couple of hours, as suggested: At 1:26 AM -0400 7/21/99, N11TE@AOL.COM wrote: > >Is there anyone on this list that has these qualifications and would be >willing to do a structural review? Or, do you know of any professional aero >structural engineer who would already have some experience in canard aircraft >construction and could be hired at a reasonable rate? Being the targeted aerospace engineer and having done structural composite design, here are my two cents worth: 1. If you strengthen the structure, you do *not* necessarily improve the crash safety. In the described accident for example , a stronger fuselage might have killed all occupants because it would have decelerated the airplane faster, thus increasing the g loading on all occupants. Your body can only withstand a certain amount of acceleration, if this is exceeded, you are being seriously injured. 2. To figure those things out, the car industry undertakes countless crash tests on top of computer simulations. During the course of these tests, the structure is often deliberately _weakened_ in certain spots to _improve_ the crash behavior. Also, you always have to make compromises during this optimization. You gain in certain type crashes by a certain modification, but in other type crashes this modification might make things worse. In order to make decisions on design improvements, you need not only know what failed during a crash (by looking at the wreckage), it is also crucial in which order things failed, and most importantly, what the accelerations and movements of the persons (dummies) involved were like during the crash. Having said all this, I do not think that one can make sound engineering conclusions as to how to improve the design from the observations described in the original post, with one exception. In all crashes it seems to be beneficial to keep the occupants inside the vehicle at all cost. Therefore, improving the seat belt attach points should be a sure means to improve the crash worthiness. As far as strengthening the fuselage goes, chances are high that this will cause more additional problems than it will solve. None of us will be able to afford the testing necessary to make sound structural improvements towards more crash safety to the design. While designing a structure to take the loads encountered while the structure is intact is enigineering wise relatively simple, it is very difficult to predict how it will behave when parts of it start to fail. It would require another type of NASA program funding this undertaking, in the same way as they brought the AGATE program into life. Stefan --------------------------------------------------------------------------- Stefan Siegel email: stefan@u.arizona.edu 2813 E Elm Street http://venus.ame.arizona.edu/~fury/ Tucson, AZ 85716, USA Phone: (520) 321-9217 Long Ez N234BW, hangared at KRYN --------------------------------------------------------------------------- From: cdenk@ix.netcom.com Date: Wed, 21 Jul 1999 21:12:52 -0500 (CDT) Subject: Re: COZY: Structural Question Stefan said I concur with his remarks. But I wish more pilots would pay attention to the details of safety of flight. 1: Equipment - All complaints get fixed promptly. 2: Attitude - Listen to others, seek advice. Landed at Pueblo Co. once. A local asked where we were headed, I indicated going diect Page, Az. He suggested a slightly different route, which would keep us closer to civilization. Thats what we did the next day. 3: Always get a standard weather briefing, unless very local flight. If using DUATS and weather is not super, talk to a briefer before departing, he may have some trends, insight and words that weren't obvious with the dumb computer. If "VFR flight not recommended", think twice, consider alternatives. Consider what risks are added with that comment. i.e. I will under some conditions go if its say less than 20 miles of ground fog, with clear above. The risk I accept is if a forced landing is required, I won't be able to see to select a good landing site. But I will carry additional altitude, possibly enough to glide to a clear area. 4: File a flight plan. This has 4 advantages a: Forces some planning and thought of whats ahead b: Someone will try to give you a weather briefing, which includes pireps and notams c: Of course notification of search and rescue. d: Make position reports, I usually do 1 per hour, more frequently if flight conditions are not great. Briefer will update weather. THis worked great once going in to Jackson,Wy., where we learned of a recently issued thunderstorm watch, and with a discussion with the briefer, found a neat place for the night. 5: I don't fly at night over unfamiliar terrain, that is more than about 100 miles from home, don't accept the added risk of electrical failure (never had one), or forced landing, within known terrain, I carry extra altitude and probably could make it to an airport or road that I know well. 6: No night IFR (or near), do not accept risk of electrical failure, and trying to navigate, fly plane, while holding flashlight in mouth. 7: No continous flight in any icing, may with much thought do an easy approach through a thin layer or 2, with enough fuel to get to clear conditions. 8: No flight with possible thunderstorms anywhere near, unless able to get high in the clear with excellent visibility and see possible storms. And then stay in the clear! This is only a few of the major items that go into my Go/No Go decision. THe idea is minimize the risk, when several items happen together, its time to land. Way home from OSH, it was 3 strikes, and land: sun setting, tired, couldn't read the identifier for Grand Rapids on the chart fold. We had planned on Grand Rapids for the night, landed at Muskegon, Mich. The bonus the next morning was tower said they had never see a plane like ours, could we offset toward the tower after lift off, we did a 45 degree bank at about a 200' radius around the tower at their elevation (buzzed the tower) and got thanked! Has anyone heard that well know fatality recently even did something free, and takes only a few minutes, like a standard briefing? Date: Wed, 21 Jul 1999 22:57:34 -0600 From: James Russell Subject: Re: COZY: Structural Question >>> a permanent cover MIGHT stiffen the fuse enough that it could save the >>> occupants in a head-on collision with an immovable object at approach >>speed. >> >>I did list the removeable cover as one major possibility for loss of >>structural strength. I did not mean to imply that it was the only one. We >>see a number of other changes that have been made which could be equally >>responsible. I mentioned it because it was the first thing that RAF focused >>upon when I discussed it with them. > >Having a crush zone is a good thing. It will take lots-of-G's(TM) and >convert it into not-so-many at key places. That is the key, that the >key places have to be where the people are, and the crush zones have >to be where fixed parts of people aren't. > >I would focus more on the seat belt anchors. Let the airplane crush, >and have my feet push back (or break), but keep my body strapped to >the seat away from the crush. > >Indy and F-1 cars crash at speeds over 150mph all the time, and >drivers mostly walk away. The crush zones allow this. Hi all: This is certainly a good discussion - even though I hope no one has to benefit from it during an impact... Racing cars are tanks when compared to aircraft - because they crash alot ( compared to A/C ) - I have a section of a March Indy car tub: .250" of uni- and bi-directional carbon. Compared to the old sheetmetal tubs, composite cars are much stronger overall ( esp. in torsion ) and lighter. But, composites fail by fracturing vs. bending for sheetmetal ( remember the flood-damaged Pipers that NASA? crash-tested and saw the huge temporary deflection of the fuselage top almost hitting the seats? ). I have seen carbon racing car noses fail completely @ the dash bulkhead leaving the driver's feet hanging out. Rule-making bodies have been moving the driver( and his feet ) back to provide the room to absorb the impact and most composite tubs use Al honeycomb ( the cells aligned w/ the cars' centerline ) to provide crush space in the nose. I plan to include core aligned up-to-down under the seats to help w/ compression fractures... If the harness mounts are failing before the surrounding structure, then I will share the engineer's cost to review them - please email me. Harness attach point failures are vanishingly rare in racing cars. You ( and the belts ) will strech an amazing amount under load - one of my drivers hit a wall and his helmet impacted a mirror mount 34" away! I certainly know it is easy to get fixiated on one concept and not view the entire package - I do it all the time! I also am very aware that any design is a necessary mass of compromises - the unbreakable 14 lb. steel snowshovel that you can't lift comes to mind. I am very glad that Nat "ported" the VZ/LZ design over to four places. I was tired of being crammed into racing car cockpits designed for 150 lb., 5' 6" drivers...So, having more panel space, getting to see your "copilot" (thanks, Shirley!), real luggage space, occasional 4 person trips, etc. are all the reasons I am going to build a Cozy. Regards, James From: "Nat Puffer" Subject: COZY: Stress calculations Date: Wed, 1 Dec 1999 12:10:43 -0600 Builders, Burt Rutan told me that he supplied stress calculations to all foreign countries requesting it for the Long EZ, and he probably also supplied the confirming static tests. He told me that for the 3-place Cozy (We are a licensee) to tell these countries to use the same information as he supplied for the Long EZ. In the Mark IV, because of its higher gross (even though Long EZs have been operated at a higher gross than the Mark IV), we increased the spar cap thickness by 20% and added other reinforcements in critical areas, and lowered the category to "normal" from "utility", mostly because we wished to discourage aerobatics in a family cross country airplane. Most accidents in Long EZs were caused (in our judgement) by reckless behavior. My information is that Burt designed the Long EZ wings to 12+gs, the centersection spar to 18gs, and the canard to 14+gs and these were confirmed by static testing. The "normal" category is limited to 3.8 gs, and I doubt if we have seen more than 2 gs in normal operation. Regards, Nat PS: There are no know cases where either a Long EZ or a Cozy, built per plans and operated within critical limits, has come apart in the air. Date: Wed, 01 Dec 1999 18:50:32 -0800 From: "J. D. Newman" Subject: COZY: Re: Stress calculations > Nat Puffer wrote: > My information is that Burt designed the Long EZ wings to 12+gs, the > centersection spar to 18gs, and the canard to 14+gs and these were confirmed > by static testing. Based on a 1325 lb. gross weight Long-EZ. Date: Tue, 14 Dec 1999 10:32:56 -0500 (EST) From: Richard William Crapse Subject: COZY: G-loading comparisons Ok guys, I tripped across the discussion of normal versus utility category because of the higher Gross weights. So here is what I did in Excell: Long EZ: Wings 12g * 1325 lbs. = 15900 Canard 14g * 1325 lbs. = 18550 Spar 18g * 1325 lbs. = 23850 (Assumption: 1325 lbs. is gross weight of Long EZ) Cozy: Wings 15900 / 2050 lbs. = 9 G Canard 18550 / 2050 lbs. = 7.75 G Spar 23850 / 2050 lbs. = 11.6 G Heavy Cozy: Wings 15900 / 2150 lbs. = 8.6 G Canard 18550 / 2150 lbs. = 7.4 G Spar 23850 / 2150 lbs. = 11.1 G Assumptions for Cozy and Heavy Cozy: 1. Does not take into account upgraded spars or wings. 2. Uses published numbers from Rutan for analysis. 3. Pilot could inflict such loads under normal flight or control inputs. 4. Same style of attachment for engine mounting. 5. Original user typed correctly and excel does correct math. Rick Cozy #706 (1 flood, six hurricanes, and more miles on the road than most GA aircraft fly in one year, .... the joys of homebuilding) 'All the power in the Universe, but an itty,bitty,little living space'---- Aladin(genie) 'If it 's got wings, I'll fly it'--Pilot 'And let me know when Elvis shows up...'----Beldar Conehead 'If a shoebox had wings, I'd make it fly!'---The Rocketeer