From: marc_zeitlin@agilent.com Subject: COZY: CSA newsletter - Roll over protection and Rod end size Date: Fri, 27 Apr 2001 15:49:59 -0600 People; Many of you are CSA (Central States Association) Newsletter members (if you're not, you should be). You may have noticed in the last issue (CSA #62) that the January Issue (#103) of the "Canard Pusher", put out by RAF has a couple of mandatory changes for Long-Eze's, involving roll-over protection under the canopy and rod end size for the aileron pushrods. First, roll over protection. In the COZY plans, it is explicitly stated that the headrests are NOT roll-over protection, which of course leads one to ask "what is?" and the answer is "nothing, at the moment, unless you've invented it yourself". Now, we're obviously _not_ seeing people dying left and right from rolling these things on their backs, so it's not clear how big a deal this is, but RAF has seen fit (15 years _after_ ceasing plans sales) to mandate a roll over structure, and has designed and tested one, as well as making plans for it available. Should we expect to see an equivalent mandate from COZY Development corp. (in both senses of the word - is it necessary, and will it occur?). Second, rod end size. Again, RAF has mandated a change from -3 (#10-32 thread) rod end bearings in the CS-1 inserts in the aileron pushrods to -4 (1/4"-28 thread). There are at least two methods of modifying the existing rod ends presented in the CSA newsletter, and apparently Brock has been instructed to drill and tap new CS-1's for this size. Since the COZY uses the same inserts and rod ends, this brings up a few issues (in no particular order: a) will we have a mismatch between CS-1 and rod end bearings if the COZY sticks with the smaller threads? b) will the COZY plans be modified to also call out the larger bearings? c) will existing COZY's be instructed to modify their CS-1's and bearings to the larger size? d) will Brock make a different CS-1 for COZY's if the size change isn't mandated? Apparently a -3 bearing actually failed in one L.E., so this is not just a theoretical issue, although the cause of the failure (possible misuse?) isn't stated in the CSA newsletter. To join CSA, see: http://www.canard.com/csa/member.html -- Marc J. Zeitlin marc_zeitlin@alum.mit.edu | http://cozy.canard.com Non Impediti Ratione Cogitantonis (C&C) Date: Fri, 27 Apr 2001 20:46:55 -0500 From: David Domeier Subject: Re: COZY: CSA newsletter - Roll over protection and Rod end size Marc, re "d) will Brock make a different CS-1 for COZY's if the size change isn't mandated?" The fitting is already available. In fact we use it in the elevator push rods. CS1 and CS1A are the same except CS1A is drilled out to 1/4 x 28. dd Date: Sat, 28 Apr 2001 10:45:10 -0400 From: "Marc J. Zeitlin" Subject: COZY: Rod end size Dave Domeier wrote: >The fitting is already available. In fact we use it in the elevator >push rods. >CS1 and CS1A are the same except CS1A is drilled out to 1/4 x 28. Yes, that's true, but if the plans still call out CS-1, people will purchase the wrong item for the ailerons. Basically, it's a question of cooperative relabeling of common parts between L.E.'s and COZY's - someone's got to make some decisions and ensure that Brock and the builders know what the "right" answer is. -- Marc J. Zeitlin mailto:marc_zeitlin@alum.mit.edu http://www.ultranet.com/~marcz/ From: "Nat Puffer" Subject: Re: COZY: CSA newsletter - Roll over protection and Rod end size Date: Sat, 28 Apr 2001 08:43:03 -0500 Dear Builders, The post by Marc sounded a little unfriendly, don't you think? Marc correctly observed that RAF finally is providing roll over protection for the Long EZ. Please be aware that we have provided our builders with roll over protection since our 3-place model first came out in 1982. We not only have 2 head rests (the Long EZ only has one), but we have a rigid and strong turtleback (if built according to plans) which overlaps the canopy above the pilot and co-pilot's heads (providing additional strength), and also a very strong bulkhead, TB-1, right behind the headrests. This was even tested by Mr. Harris, in Memphis, who had an engine out and put a Cozy III down in a very muddy field, and which subsequently overturned. He was protected by our roll-over protection, wasn't injured, and the canopy didn't even break. We have an additional safety feature, as compared to the Long EZ, in that we have our fuel tank vent lines trapped (rather than coming out the top), so that in the event of a roll over, the fuel does not drain all over the ground while you are trying to extricate yourself from the cockpit. As for the subject of rod ends, we believe the most important control is pitch control. We have redundant pitch control, with two (one from each stick) pushrods going to the elevator torque tube and attached to it separately. In spite of this, we changed from AN-3 to AN-4 rod ends because the pushrods and rod ends are especially vulnerable to damage (bending) when removing and reinstalling the canard. The AN-3 is strong enough for the push-pull application, but if it is repeatedly bent and restraightened, it could fatigue and break. But even so, our design provides redundancy. Having spent 35 years in an engineering profession which involved continual problem-solving, I subscribe to the philosophy to first make sure you understand the problem before coming up with solutions, and not to come up with solutions to problems which do not exist. In other words, if it ain't broke, don't fix it. We have a redundant aileron control system going back to the firewall. The AN-3 rod ends have more than adequate strength for the push-pull duty in the aileron system. We use wide area washers to make sure that if a rod end bearing does loosen, it cannot escape. The rod ends are not subject to bending, as they might be with the elevator application, and we certainly advocate inspecting them as part of the annual inspection. Our design has similarities to, but is not identical to the Long EZ. We have no evidence that there is any design flaw which would require us to make a mandatory design change. If you wish to convert all of your AN-3 rod ends to AN-4, or if you are making a new installation and wish to do so, we have no objection. But unless we subsequently find a good reason, we will keep ours as shown in the plans, and we will not require all of our builders to make expensive changes. It goes without saying that if you ever bend a rod end, whether it is an AN-3 or an AN-4, you should discard it and replace it with a new one. I am sure I could make even an AN-4 rod end fail, if I used it in a way that wasn't intended. If a part fails, it is important to determine the reason. In the case of the 3 inch wide spar cap tape, we wasted a lot of time and money because we tried to come up with a solution before we understood the problem. It happens all the time. Best regards, Nat yourself from the cockpit. ---------- > From: marc_zeitlin@agilent.com > To: cozy_builders@canard.com > Subject: COZY: CSA newsletter - Roll over protection and Rod end size > Date: Friday, April 27, 2001 4:49 PM > > People; > > Many of you are CSA (Central States Association) Newsletter members (if > you're not, you should be). You may have noticed in the last issue (CSA > #62) that the January Issue (#103) of the "Canard Pusher", put out by RAF > has a couple of mandatory changes for Long-Eze's, involving roll-over > protection under the canopy and rod end size for the aileron pushrods. > > First, roll over protection. In the COZY plans, it is explicitly stated > that the headrests are NOT roll-over protection, which of course leads one > to ask "what is?" and the answer is "nothing, at the moment, unless you've > invented it yourself". Now, we're obviously _not_ seeing people dying left > and right from rolling these things on their backs, so it's not clear how > big a deal this is, but RAF has seen fit (15 years _after_ ceasing plans > sales) to mandate a roll over structure, and has designed and tested one, as > well as making plans for it available. Should we expect to see an > equivalent mandate from COZY Development corp. (in both senses of the word - > is it necessary, and will it occur?). > > > Second, rod end size. Again, RAF has mandated a change from -3 (#10-32 > thread) rod end bearings in the CS-1 inserts in the aileron pushrods to -4 > (1/4"-28 thread). There are at least two methods of modifying the existing > rod ends presented in the CSA newsletter, and apparently Brock has been > instructed to drill and tap new CS-1's for this size. Since the COZY uses > the same inserts and rod ends, this brings up a few issues (in no particular > order: > > a) will we have a mismatch between CS-1 and rod end bearings if the COZY > sticks > with the smaller threads? > > b) will the COZY plans be modified to also call out the larger bearings? > > c) will existing COZY's be instructed to modify their CS-1's and bearings > to the > larger size? > > d) will Brock make a different CS-1 for COZY's if the size change isn't > mandated? > > Apparently a -3 bearing actually failed in one L.E., so this is not just a > theoretical issue, although the cause of the failure (possible misuse?) > isn't stated in the CSA newsletter. > > To join CSA, see: > > http://www.canard.com/csa/member.html > > -- > Marc J. Zeitlin marc_zeitlin@alum.mit.edu | > http://cozy.canard.com > > Non Impediti Ratione Cogitantonis (C&C) From: "Nat Puffer" Subject: Re: COZY: Rod end size Date: Sat, 28 Apr 2001 09:02:08 -0500 Dear Builders, The usual proceedure, if a design is changed but the old one continues, is to assign a new part number to the new design. I will check with Ken Brock to determine whether he is assigning a new part number for the inserts for AN-4 rod ends and let you know. Actually, as David Domeier pointed out, the new inserts aren't new at all, because they have been used for years for the elevator push rods. So maybe no notification is needed. Best regards, Nat ---------- > From: Marc J. Zeitlin > To: Cozy Builders Mailing List > Subject: COZY: Rod end size > Date: Saturday, April 28, 2001 9:45 AM > > Dave Domeier wrote: > > >The fitting is already available. In fact we use it in the elevator > >push rods. > >CS1 and CS1A are the same except CS1A is drilled out to 1/4 x 28. > > Yes, that's true, but if the plans still call out CS-1, people will > purchase the wrong item for the ailerons. Basically, it's a question of > cooperative relabeling of common parts between L.E.'s and COZY's - > someone's got to make some decisions and ensure that Brock and the builders > know what the "right" answer is. > > -- > Marc J. Zeitlin mailto:marc_zeitlin@alum.mit.edu > http://www.ultranet.com/~marcz/ From: alwick@juno.com Date: Sat, 28 Apr 2001 12:56:44 -0700 Subject: Re: COZY: CSA newsletter - Roll over protection and Rod end size On Sat, 28 Apr 2001 08:43:03 -0500 "Nat Puffer" writes: > Dear Builders, > The post by Marc sounded a little unfriendly, don't you think? No, I personally don't. It's thought provoking and opens subjects that may be significant. > We have redundant pitch control, excellent! Certainly reduces risk. > In other words, if it ain't broke, don't fix it. Since your retirement Nat, the above adage has been identified as a common thread to inhibiting product improvement. You'll never hear that quoted any more in modern businesses. The adage assumes that failure risk is black and white, rather than a "normal probability curve". Moving risk farther along the curve greatly reduces risk.... ie eventual failures. > We have a redundant aileron control system going back to the firewall. True, but does not apply as the weak link (an-3's) are not redundant. They likely see the same forces as the VE and LE. Basically guys this is a no-brainer. If you are building your plane, substitute an-4's for the an-3's. FWIW, I found and replaced a bent an-3 (aileron) while doing my final wing assembly. Don't have a clue how it happened, I was very very surprised. An assistant thought they had done it. Once again, thanks Marc for risking criticism. Sure miss those days when we used to DISCUSS items such as this. Happy building! -al wick Artificial intelligence in cockpit, Cozy IV powered by Subaru 2.5 Done building! Ground testing. ________________________________________________________________ GET INTERNET ACCESS FROM JUNO! Juno offers FREE or PREMIUM Internet access for less! Join Juno today! For your FREE software, visit: http://dl.www.juno.com/get/tagj. From: "Larry & Jenny Schuler" Subject: Re: COZY: CSA newsletter - Roll over protection and Rod end size Date: Sat, 28 Apr 2001 20:01:25 -0500 > The post by Marc sounded a little unfriendly, don't you think? No I do not think so; however, the first line of your response most definitely was. Great discussion items Marc. Reminds me of the old days when the list was young and no one felt bullied by Co-Z Development for offering our honest opinions and trying to help us all make our own minds up concerning our own experiment. Nat would do well to join the discussions rather than fight them or try to lead them... There are days I wish I could reverse my vote... Nat, I'd suggest you re-think your response concerning roll-over protection provided by the headrests. Your plans, as written, disagree with your statement here as was so kindly pointed out by Marc. You have staunchly defended the plans as written for the sake of your legal toosh. Does this mean you are now changing the plans and we will see this in the next newsletter? Personally, I wouldn't rely on those 1-ply headrests as protection from much of anything. Rutan has also stated many times that the headrest in his other designs is not intended as roll-over protection either. As for the rod-ends I intend to go with the Rutan change as well since the Cozy is his basic design anyway; for which, we all know he receives royalties. I think (FWW) we would all benefit if the Cozy plans were adjusted (roll-over and rod-ends) as Rutan has obviously intended for his other designs. Larry MK-IV plans #500 Date: Sat, 28 Apr 2001 23:55:06 -0400 From: "Marc J. Zeitlin" Subject: COZY: CSA newsletter - Roll over protection and Rod end size Nat Puffer wrote: >The post by Marc sounded a little unfriendly, don't you think? It was not my intent to sound unfriendly, and if I sounded so, I apologize for my tone. >...... We not only have >2 head rests (the Long EZ only has one), but we have a rigid and strong >turtleback (if built according to plans) which overlaps the canopy above >the pilot and co-pilot's heads (providing additional strength), and also a >very strong bulkhead, TB-1, right behind the headrests. Having built all these parts over the past 6 years, I certainly understand that there's more structure in the COZY than there is in the L.E., and that's why I posed the issue in the form of a question, rather than a request for a mandatory change. I'll quote from Canard Pusher #44: BEGIN QUOTE ------------ VARIEZE/LONG-EZ ROLLOVER/HEADREST We have received a letter from Andrew Detroi of the FAA concerning the forced landing/crash of a Long-EZ that he investigated. This crash involved a Long-EZ that lost power after takeoff. The pilot made a successful 180 degree turn, landed long and left the runway. The nose gear collapsed, the nose dug in and the aircraft flipped inverted with enough forward velocity to break the canard in half and rip one wing off at the end of the centersection spar. The rollover/headrest was broken off. The pilot and passenger received minor head cuts, scratches and bruises. This letter has been distributed to the various FAA offices and in some cases redistributed with some inaccuracies. This has caused some consternation among the local FAA and among groups and individual Long-EZ builders. We have spoken to the FAA in Chicago and they have agreed with us that obviously the pilot's head rest is not, nor was it ever intended to be strong enough to resist the forces imposed in an inverted crash with any appreciable forward speed. It is a roll over structure, and has proven that it will remain intact in the event that one of these aircraft should roll over with little or no forward speed. This was in fact the case, when Ken Swain flipped his EZ in a corn field near Oshkosh after an engine failure. His aircraft ended up resting on the rollover structure (canopy broken), the firewall and two broken winglets. He was not injured, but had to wait for others to lift the aircraft to get out. The rollover has provided this protection in at least two other cases, one example is in CP #14. -------------- END QUOTE As we can see, RAF explicitly states that they never designed the headrests to withstand a rollover with appreciable forward velocity, although they have withstood roll overs that were slower (as I believe the COZY Harris roll over was). So, after a number of years, maybe what's happening is that RAF is merely bringing the L.E. up to the rollover standards of the COZY. However, if the canopy does break, the first thing that'll hit the ground are the same headrests that are not strong enough in the L.E., and even though there are two of them, if the forward velocity is appreciable, it won't matter how many there are. Maybe this is a minor issue and we should ignore it - the _one_ data point we have suggests there's no problem. I asked the question to explore the issue, since Burt and RAF decided to make a change in a substantially similar aircraft. >...... In other words, if it ain't broke, don't fix it. On the other hand, things aren't necessarily always black or white ("broke" or "fixed"). Shades of grey exist, where something generally works, but could be better/more robust, etc. Al Wick stated the situation well. > We have a redundant aileron control system going back to the firewall. Yes, but the AN-3's are behind the firewall, and are not redundant. If any one fails, at least one aileron will not work and the whole system can jam. >The AN-3 rod ends have more than adequate strength for the push-pull duty >in the aileron system...... In tension and compression, yes. However, unforseen circumstances, or slight builder error might cause misalignments and bending forces. >....... Our design has similarities to, but is not identical to the Long EZ. It's dang close, except for their being more AN-3's in the COZY, I believe. >........ We have no evidence >that there is any design flaw which would require us to make a mandatory >design change. Was there a design flaw in the L.E.? The change was made apparently because an AN-3 bearing had failed in at least one aircraft, although the root cause of the failure isn't clear. Maybe builder error, maybe a stackup of tolerances, maybe an alien subterfuge. Burt Rutan and RAF apparently thought that it was important enough to mandate a change - if the COZY is significantly different enough from the L.E. to allow it to use less robust components in a similar use model, so be it - I'm just asking the question to explore the possibilities. >..... If you wish to convert all of your AN-3 rod ends to AN-4, or >if you are making a new installation and wish to do so, we have no >objection. Good - glad to hear it. >..... But unless we subsequently find a good reason, we will keep ours >as shown in the plans, and we will not require all of our builders to make >expensive changes. For new builders, it's hardly expensive - it would be a couple bucks more at the most per bearing. Even as a retrofit, it hardly falls into the category of expensive, compared to most of the stuff on these aircraft. I would think that at least a recommendation to make the modification at the next conditional inspection (or yearly equivalent in countries other than the USA) wouldn't be an unreasonable burden on builders. Al Wick wrote: >Once again, thanks Marc for risking criticism. There's no risk here :-). I get criticized at work continuously (heck, at home too :-) ) - a few critical emails won't make much difference - keeps things interesting :-). >..... Sure miss those days when we used to DISCUSS items such as this. There's no reason not to discuss anything that pops into anyone's mind, as long as it has to do with COZY's. A good Failure Modes Effects Analysis (especially AFTER a couple of failures of components) can't hurt. -- Marc J. Zeitlin mailto:marc_zeitlin@alum.mit.edu http://www.ultranet.com/~marcz/ From: "Nat Puffer" Subject: COZY: Pre-flight & annual inspections Date: Sun, 29 Apr 2001 13:04:58 -0500 Builders, If you refer to the Owner's Manual Appendix 1, Initial Systems Check out, Control Systems, page 34, you will note it says, "Check all rodends. Reject any with evidence of bent tangs." Then on page 50, Annual/100 hour inspections, it says you should also check all items in Appendix 1, (starting on) page 33, which includes rodends. Congratulations to Al Wick, who found a bent tang on a preflight inspection and Carl Denk, who found one during his annual inspection. At least this means some builders are checking them. In the next newsletter, we will remind builders to make these checks, and report to us any discrepancies they find. We need to know if it is a wide spread problem, and if so, whether it is an installation problem or operational problem. Regards, Nat From: "david vollrath" Subject: Re: COZY: CSA newsletter - Roll over protection and Rod end size Date: Mon, 30 Apr 2001 12:25:45 >It is redundant only if there is someone flying the other control. If the >left side breaks, which is the most probable since most of the time >that is the side that is flying the plane, then the pilot would have to >have >the >presence of mind to reach over and grab the other stick ================================================= Let me add an embericing real life experience to this thread in the interest of safety. I had been working on the panel of my Cozy III, and had replaced the little rectangular Hobbs meter with another instrument. I left the old instrument laying on the forward armrest on the right side of the aircraft, and did not see it on the preflight inspection. You can probably imagine the the next item in this scenario as after takeoff the instrument slides back the armrest, and drops into the gap between the right control stick and the stick well cutout. With full forward stick against this stop, the aircraft would fly, and climb at minimum airspeed, and roll movements felt high friction as the right stick rubs against the blockage. After about twenty seconds of just "fly the aircraft" to a safe altitude, I was able to confirm the blockage in the right stick, and reach over to remove it as I pulled back on the left stick. Big, big lessons learned--- 1. Never fly without protective boots installed on the control sticks. 2. Always preflight for loose items in the cockpit and cabin that my jam against the other control stick, or between the stick and the sidewall. If this blockage had jammed itself into the roll side of the stick I may not be here writing this now. Be careful out there, and share your safety experiences. David Vollrath _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com From: alwick@juno.com Date: Mon, 9 Jul 2001 18:06:00 -0700 Subject: Re: COZY: Flight training, safety and other stuff... On Mon, 09 Jul 2001 09:32:25 -0500 David Domeier writes: > How come so many airplanes are being destroyed? This would be a good > material for a doctoral paper. > > It is easy to speculate and I'll do it. These type issues I dealt with often in my occupation. I've found it's best to ignore the human failure part of the problem. Why? Because you really can't do anything lasting to improve it. Frankly aircraft systems are the crudest, ineffectual you can devise. It's absurd to expect the pilot to notice 1 of 15 gages in time to respond. There are many easy improvements that would save lives each year, but fear of litigation rules. If you make a design change, it implies the previous product was defective. So you lose big bucks. Guaranteed. Over time you have an entire industry afraid of change. I can give you lots of examples. Flight from vfr into imc is a very common cause of deaths. Those hung up on the people part of the problem will say "what an idiot, I'll never do that". So no one improves the system. I have a friend who's modified a laser to broadcast a narrow red gyro horizon along the entire width of the cabin. Do you have any idea how intuitive level flight would be with that device? You would not need to seek an instrument, you would notice thru your peripheral vision that you have just initiated your "graveyard spiral". Guarantee John Kennedy would still be alive. Fuel exhaustion. What idiots we are to be running around with only passive gages. Hello? The gages know the fuel level. They see it getting lower and lower, near the bottom of tank, and they say nothing. We don't have to tolerate stupid gages. Smart gages that compensate for our distraction and oversights are available. But here we have: "Good pilots KNOW their fuel level.... I don't need a smart gage." Once again, we don't improve the system. So guys die year after year. I've found my best asset over the years is that I think I'm an idiot. (yeah, I know I'm not the only one). Sorry for preaching, but we are so backwards........... -al wick Artificial intelligence in cockpit, Cozy IV powered by stock Subaru 2.5 N9032U 80+ hours on engine/airframe ________________________________________________________________ GET INTERNET ACCESS FROM JUNO! Juno offers FREE or PREMIUM Internet access for less! Join Juno today! For your FREE software, visit: http://dl.www.juno.com/get/tagj.