| Author | Message | ||
     Gary Markwardt |
I have the shifted forward main gear & my nose wheel weight is 466 lbs. empty. Since I was out of the factory just ahead of Phil, I presume it represents the latest design. I second Jim Butler's comments re: Phil's analysis. Thanks Phil for a thorough look at the causes. I feel a little better about my nose gear today!!! Sorry for your loss, however. Gary | ||
| Ali Moghaddas (Ali) |
Phil I also like to thank you for being candid and informative about the accident. I am glad you are OK. But I feel for your loss, since I got a chance to meet you, your wife, and your beautiful airplane at Oshkosh. Since I am an engineer working for Boeing Co (18 years experience), and used to do stress/ structural analysis on Rocket engines, and have been involved in many failure investigations, I thought I throw my two cents worth of thought into this. This opinion is strictly based on very limited knowledge of what I know about your accident & the hardware that was used in your airplane. I also am a part time flight instructor (for the past 13 years), and used to own and teach in a Gruman Yankee which is known to have a "marginal" nose gear. First let's talk about the failure: Based on my knowledge of operation of the landing, this was caused by either a Low Cycle Fatigue, LCF( for sake of argument: 10,000 cycles or less) or an overload. It would not be a High Cycle fatigue situation, since it did not involve a high cycle loading or harmonics(such as an engine type component). Looking at the fractured surface(possibly with some magnification), a metallurgist should be able to tell you whether it was LCF or overload. Again, based on very limited knowledge of what I read, I would guess(long distance diagnostics) it was caused by an overload type of loading. What this means, is that the load exceeded the allowable limit for your hardware. This can be cause by one or combination of several factors: A- Hard landing caused shock load (in g's) to exceed design allowable load. For example, if your nose weight is 500 lb. and you experienced 3 g's during landing, it would cause a 1500 LB loading on the landing gear. If the landing gear has an allowable load of 1000 LB, the applied load has exceeded the ultimate strength of the part (Ftu), and the part would fail. B- Having a manufacturing defect in the hardware: - Geometrical discrepancy: The hardware has local wall thinning, diameter undersized, improper angle, length, weld length, weld width/ throat, etc. - Weld contamination, Flaws larger than CIFS (Critical initial Flaw sizes),etc. C- Metallurgical issues: Incorrect welding rod material, basic material flawed (microfissures, improper grain size, incorrect material), incorrect heat treat/ aging resulting in low strength and/ or ductility. D- Marginal basic design. - No analysis done to adequately design for 95 percentile pilot - Not designed for heavier engines - Not designed to account for minimum material properties, minimum geometry (wall thickness, etc) Let's talk about NDT(non-destructive) testing and see what than can tell us: - Weld inspections: There are generally 3 classes of welds (I, II, II, some call them A, B, C) - Class I, gives you the best weld quality. This is done by inspecting both surfaces through Mag particle or Pent inspection and inspection for flaws with the wall through X-ray(single wall or double wall depending on what the fracture mechanics/ CIFS analysis indicates or eddy current inspection(UT). In this case the designer or analyst would use 100% of property. - Class II, provides you the 2nd best quality. This is same as above, except without the X-ray. In this case the designer or analyst would use 60% of material property. - Class III: you just do visual inspection, where the designer or analyst would use 40% of material property. - Basic material inspection: - Almost all aerospace quality material that is purchase with proper Cert papers has gone through inspection of some sort (generally UT or X-ray to verify lack of flaws). This might have been on a sample from the same batch going through the same process . Generally, you pay a premium for the CERT. One thing to bear in mind is that if you do a single wall X-ray, you can only detect flaws larger than ½ wall thickness with 95% reliability. What this means is that one can not detect small embedded flaws using X-ray with a high degree of confidence. Bear in mind that these figures are good for a Certified inspection house using Certified technicians. X- ray type inspections are rather subjective. The interpretations has to do with the inspectors experience having the proper reference X-rays for that particular material with similar thickness to compare. Now, if one goes to Joe Smith's inspection house in their neighborhood, his diagnosis would not be that reliable. What can we do to prevent this from happening: - Information to determine the cause: - Metallurgical and geometrical inspection of failed part/s: - Determine material type, composition, heat treat (using tensile test or Rockwell hardness test), sign of corrosion. - UT and or mag particle inspection - Wall thickness, and diameter inspected at various locations - Have someone analyze the part to determine allowable load/ structural margin - Redesign/ modify the part if the part design is determined marginal - Modify piloting technique: - Having been an instructor, including aerobatics instructor, I would highly recommend against moving weight/ CG aft, specially for an airplane like Expess with a C tail (which is already marginal in spin recovery). - One way to minimize your rate of decent during landing is to use max flaps, speed brake, etc, and use a bit of power on short final. I thought this technique to several students in a Gruman Yankee, and none had any problems landing that airplane. This opinion/ input is getting long. This was a very short version of several technical subjects. I'll be happy to elaborate/ explain on any of these subjects. Please contact me via e-mail. Thanks, and Good luck | ||
| Phil & Margie Hodge (N410mp) |
Ali - Thanks for your long and detailed response. I hope you saw a different Express at OSH, as my maiden flight was last August, and she's been in Olympia until last month. You will see my beautiful plane at OSH, I just don't know when. The divot in the runway, plus the bend in the nose gear makes it pretty clear it failed due to overstress. The break is clean, consistant and straight, precluding a local material defect. All this will become clearer to you all when I get pictures back, but is pretty clear to me now. Like a few before me, the plane pitched nose down a little too much at the wrong time, and the design of the nose gear caused the yoke to dig in. Avemco has offered to pay for whatever analysis I recommend, but I think the cause is so obvious that it would be a waste of money. If the nose gear had not bent and broken, either the firewall would have pushed in to my feet, or the airplane would have cartwheeled. I'll give up a nose gear and propellor, thanks. By the way, I will post a summary of my experience with Avemco when this is all settled, but as a previow let me say that they have been a pleasure to deal with, and so far have earned a lifetime customer. Larry has already redesigned the nose gear to reduce the chance of it hitting pavement, but the new gear is taller, requiring taller mains also. This he accomplishes with a different bend on the new aluminum mains. I have EDI steel mains, so was forced to use the previous nose gear. During the rebuild I am definitely going with the newest nose gear, and will either re-bend my steel legs or replace them with new aluminum. I wouldn't feel comfortable if I just rebuilt her with the same problems waiting to bite me again. Your comments on cg location and pilotting technique are both correct. In order to get into my short grass strip I need to be able to cross the threshold safely and reliably at 65 kts (1.2*Vs). With my monster IO550 she jumps immediately to 80 kts the instant I add a little power, so that option is out. What I think I have learned is that there are two cg ranges to be concerned with - flying and landing. They do overlap, but not throughout their range. So the best way I've come up with is to have a weight that can clip into place in the baggage compartment, back seat, or front seat. This will allow me to move the cg depending on passenger/baggage load to keep it within allowable limits for both phases of flight. Again, pictures will be available when I'm done. Continental says that any prop strike requires an engine tear-down, no exceptions if I want to keep my warranty, so the engine comes off monday. There is a similar requirement on the prop govenor, I don't yet know about the alternator, mags, vacuum pump or prop hub. The fun continues, hopefully she'll fly in a month or so. Phil | ||
| Ali Moghaddas (Ali) |
Phil: Thanks, for the response. I apologize about getting you mixed up with the folks from Dakotas whom I met ar Oshkosh. I hope to meet you some day. One addition/ correction to what I wrote last night is that UT is ultra-sonic inspection, and is different than eddy current inspection. UT is mainly used for embedded flaw detections, and wall thickness measurements. Your point about having a failed nose gear vs firewall is correct. It is preffered that the gear "structurally yield"(bent,or in worst case scenario shear off) rather than having the failure at another point such as the firewall, therefore, you don't want to have your nose gear built like a tank (similar to a skyscraper in a high wind situation. ie, the building should be able to sway in the wind, otherwise if it is too stiff, the high load at the base/ foundation could cause failure). One thing you might want to look into, is adding VGs to the tail and wings to tame some of the low speed handling characteristics. Based on what I have read, VGs do wonders even for an airplane ike Cessna 411 (single engine MCA problem). Also, it might be a good idea to install a LRI(lift reserve indicator) or an angle of attack indicator. and optimize each landing using one of these instruments. Thanks again, | ||
| Terry Sack (Terrair) |
Add my thanks to the others for sharing information on your mishap, Phil. I very much appreciate the advice and insight the other contributors have provided as well. An unfortunate but valuable use of this site!! (Thanks, Tom) I hope to see you at Sun n Fun, Phil. You sat next to my wife at last year's dinner. I also have the CT, steel legs, and screamer IOF 550 N. Whatever you learn will be of great benefit to me. Terry | ||
| Anonymous |
I'll start this out with the fact that I am not the most experienced pilot around, so take everything I say with a "grain of salt". We now have 222.4 hours on our Express. I have found that the vertical sink rate can be quite high when you slow the airplane down. The airplane hasn't stalled, it just has a big sink rate when flown slow with a low power setting. Due to this fact, I don't try for full stall landings, but fly the airplane to the runway, similar to flying a Cessna 210. This point was driven home (no pun intended) one day when I was landing with a full load with high density altitude. I was making my normal landing, when all the sudden it felt like the bottom fell out. It felt like I dropped two feet, but I'm sure it was only a few inches. I am still flying the fiberglass main gear, which held up nicely. Now, I plan on flying down final at 100 knots, probably doing about 90 knots over the numbers, flare and fly parallel with the runway, allowing the speed to bleed off. I use the same power setting it took to maintain the glideslope or VASI until touchdown. You can make "squeaker" landings with this technique. I am quite concerned about trying to land at 65 knots over the numbers. I believe if I tried this, I would have to make a perfect landing everytime, including allowing for density altitude changes. Jim Butler | ||
| Phil & Margie Hodge (N410mp) |
Finally got pictures back. WARNING - these pictures are not for the weak of heart. Follow this link: http://www.snapfish.com/share/p=34191014179866620/l=7174590/t_=7777 to see them. You will have to register with Snapfish, but in over a year of using them they have never shared my address with any of their advertisers. The day after the incident my wife and I both knew I needed to get back in the saddle, so I climbed back in my trusty Archer. When I pushed the throttle forward on take-off, it just sort of sat there. Slooooowly it started to creep forward. Sure enough, long before the end of the runway it was airborne, but after just flying 30 hours in the Express that seemed the longest, slowest, take-off ever. I want my Express back! The engine shipped out today, due back in about 3 weeks. If the prop and all the EAC parts are timely she'll be back in the air soon. Phil | ||
| Sven-Ake Eriksson (Sven) |
Phil From your pictures it is easy to see that there has been a severe overload on the nose gear mainly due to the spindle impact to the pavement. Questions: Did you hit the runway with a steep nose down attitude or did the spindle hit the runway due to the impact compressing the tire? I am interested because I have heard earlier that the distance between the spindle and runway is marginal. What would then happen if someone made a slightly hard landing with no air in the wheel! Another topic: There have been people trying vortex generators on the horizontal stab to get better performance from the tail. I do not remember who it was but they have published the result on the net. Mr Dick Lind added some area to the horiz stab by building up the leading edge. My first suggestion would be to use turbulator tape(zigzag type) on the underside of the stab and also to add some gap seal to increase elevator efficiency. Turbulator tape acts very similar to VG:s. This technique is used by the glider people to get the best performance. Probably you encountered a stabilizer stall due to fwd CG and low speed maybe also induced by the known "tail blanking" phenomena common to all CT expresses. I am myself building an original wheeler CT and I am very interested to get these problems straightened out before I will fly my own example. Good luck with the rebuild. Sven Eriksson | ||
| Phil & Margie Hodge (N410mp) |
Sven - I had bounced slightly on landing, and when the tail stalled that would have put me in a nose down attitude, but neither I nor my pilot spouse thought is was particularily steep. The nose tire was fully inflated, and still is. And the height couldn't have been very far off the runway, nor the impact very large, as the ELT didn't activate (yes, it does work). The distance between the bottom of the standard spindle and the runway is marginal at best. Some builders have cut off as much of the spindle as possible, and used a thin nut to help. Larry has a new yoke that is angled up, for that reason. I certainly recommend doing something about it on yours. I plan on replacing mine with the new angled version. If anyone knows who or where the info on vortex generators is, I'd much appreciate knowing. I don't want to increase the tail size as that will lower cruise speed, but properly placed vg's or turbulator tape shouldn't hurt. Phil | ||
| Terry Sack (Terrair) |
Phil, go to the test flight topic area of this list and you will find a link to the testing Ed Bernard had done. Terry |