What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Avoiding another bent front gear leg - help needed

MrNomad

MrNomad

Well Known Member
In an effort to help "A" model owners avoid a bent front gear leg, Chet and I built a test fixture on a boat trailer which we will tow behind my truck to simulate rolling down the runway. Our goal is to observe the load as we tow the fixture thru dips and over bumps, and to measure the gear leg flexing and bending that occurs.

The next phase of this build will be to add 250lbs atop the front wheel to simulate the load supplied by the airplane.

img2204l.jpg


While Chet is a retired EE, we could use some help with those devices we should attach to the fixture to get measurements and decide where to stiffen the front gear & with what materials. If you'd like to help us by loaning us measurement equipment, we promise to handle whatever you loan us for these tests very carefully and to return it to you when the tests are completed.

Chet and I both wrote to Vans and received a polite reply, but no solutions were offered beyond what's already known. If you're tired of reading about bent gear legs and you?d like to help us with test equipment, please contact Chet:

Oldak ?at? Comcast.net or 520-797-7161 8am to 6pm Tucson time.

Thanks.
 
Barry, this is a noble undertaking, but I can't help but wonder if you are really duplicating the RV's. Wouldn't it be better to have the leg attached to an engine mount? It seems to me that attatching it the way you have done, with what appears to be a weld at some distance down the leg, you have actually changed the harmonics of the gear leg itself. You might discover a problem, and engineer a fix for it, but it might not necessarily work on the aircraft due to the different configuration you have here.

Vic
 
Thanks Vic for your comments.

It is important that the tests have meaning and produce results that benefit our community so we appreciate any and all comments.

The portion of the gear leg that extends from our fixture is the same as the portion that extends beyond the engine mount on my 9A. We replicated the angle too. Yes, I guess the engine mount might flex too, but (hopefully) not enough to skew the results.

The first test run had me (200lb old man) riding on the trailer fixture but we know that's unsafe so we hope that our fellow builders will loan us devices to video and measure. Tomorrow we will weld some steel onto the fixture and bolt 5 Chev cylinder heads to the fixture to add weight. Chet knows about D/A converters but we don't have any such devices so we're hoping this will become a collaborative effort with results that will benefit all of us.

Basically, we're eager to see where this takes us and how receptive our community is to the idea.

Thanks again.
 
Last edited:
Barry,

When you get it set up, try it with different tire pressures.

It would also be interesting to see if the Matco axle makes any significant improvement over the stock Van's axle.
 
Test with different tire pressures? Sure!

We will definitely try it with different air pressures. I can readily see where a soft front tire will set up different harmonics than a tire with 35lbs.

We don't have a Matco axle but if someone wants us to mail us one for testing and comparison, we're happy to accommodate. Neither one of us are mechanical engineers so we're looking for ideas and direction from folks more experienced. Basically, we're two old guys who know how to weld.

After I bent my nose gear 200 hours ago, we eliminated the drag the stock seals add to the front wheel by inserting a steel spacer on the axle. Using an air conditioning spacer, we machined the length of the spacer so that the seals still "seal" but no longer add drag.
 
Gear leg

I think what you guys are doing is a great idea and have a good Idea for testing. And best of luck.
And don't let anyone discourge you from experimenting to try to solve a problem that DOES exist.
 
That's exactly what that Matco axle that Bill mentioned does.. so if you made your own, you wouldn't need the Matco part..


MrNomad said:
After I bent my nose gear 200 hours ago, we eliminated the drag the stock seals add to the front wheel by inserting a steel spacer on the axle. Using an air conditioning spacer, we machined the length of the spacer so that the seals still "seal" but no longer add drag.
Click to expand...
 
Response to Vic's Comments.

Vic. The gear leg is enclosed in a hardened steel tube similar to that used in Van's engine mounts. The tube is welded at the bottom with welded stiffner supports at the top. In essence the gear support tube is positioned and supported as in the standard engine mount. The rest of the structure is 1 3/4 14 gauge box tubing with added stiffening structure. We feel it will not add any unsual harmonics to the structure and is stiffer then a standard mount.

In our preliminary tests we noted deflections of about 1.5 inches back and forth on a smooth surface and low speeds. Additional testing is planned.
 
Test equipment

What kind of test equipment are you looking for? How about a high speed video with a black on white grid panel mounted close to the assembly to show scale? Are you planning on testing to failure?
A few points, placing the trailer wheels at the same location as the aircraft mains, and loading the entire trailer as an actual aircrafts may provide the benifit of duplicating a bounce situation, if you want to go that far.
I myself am going with conventional gear, but enjoy the process of experimenting, looks like fun!
 
MrNomad said:
It is important that the tests have meaning and produce results that benefit our community so we appreciate any and all comments.
Click to expand...

And it is in that context that this is offered.

Your effort is noble, but I don't think it is likely to provide any useful information. In fact it could actually cause a greater misunderstanding.

The reason I believe this, is the design of your test rig.

A tri-gear RV can bob up and down in pitch as the gear legs flex. If you tow the trailer by hitching it to a vehicle, you have removed any allowable pitch movement from the equation. The test rig would need to be designed so that you are simulating all three gear with the springiness (highly technical term) of the three gear legs (logical way would be use three actual gear legs and wheels/tires).

Then you would need to have the equivalent weight effected on the three gear legs at a typical vertical and longitudinal C.G. position so that a similar moment of inertia is simulated when the test rig bobs about the pitch attitude. And, it needs to be towed in a manor that would not induce any resistance to the pitching motion induced by the legs flexing.
 
Make me an offer I can't refuse......

Andrew M said:
How about a high speed video with a black on white grid panel mounted close to the assembly to show scale? Are you planning on testing to failure?
Click to expand...

If you are offering to loan us "a high speed video with a black on white grid panel", we accept.

Ideally, we won't take the tests to failure given that the front gear we are using is planned for Chet's RV6A which is rapidly approaching completion. I expect to be painting the 6A in the next week or so.
 
Long on talk

short on action;). I don't have the camera to lend. Like many folks here it is easy to armchair engineer. I will offer continued encouragement. I think you are on the right track for what you want to find out. I can be really easy to out smart one self or fault find.
How about using your rig to test allready failed gear legs, with additional "skis" attached that "would have rolled" down as the gear leg bent?
 
Promote thoughts and ideas? That's already happening.

rvbuilder2002 said:
A tri-gear RV can bob up and down in pitch as the gear legs flex. If you tow the trailer by hitching it to a vehicle, you have removed any allowable pitch movement from the equation. The test rig would need to be designed so that you are simulating all three gear with the springiness (highly technical term) of the three gear legs (logical way would be use three actual gear legs and wheels/tires)."

Thank you and we receive your comments in that same positive light. The fixture is not welded solid to the trailer. At the aft end it flexes so the jig is able to move up and down on its own based on the undulations of the road and those bumps we may place in the road to simulate potholes.

I have no doubt that a better fixture in a tightly controlled environment would yield even better results. Our sincere desire is to see others avoid the expense and downtime I suffered three years ago. If our activity stimulates thoughts and ideas which others pursue (including engineers who are more qualified than two old guys like us) and bring this matter to conclusion, our efforts were worthwhile.

With respect to the axle and the spacer, here is what we did to allow the front wheel to spin freely.

nosewheelspacer.jpg


Thanks to all for your helpful ideas.
Click to expand...
 
rvbuilder2002 said:
And it is in that context that this is offered.

Your effort is noble, but I don't think it is likely to provide any useful information. In fact it could actually cause a greater misunderstanding.

The reason I believe this, is the design of your test rig.

A tri-gear RV can bob up and down in pitch as the gear legs flex. If you tow the trailer by hitching it to a vehicle, you have removed any allowable pitch movement from the equation. The test rig would need to be designed so that you are simulating all three gear with the springiness (highly technical term) of the three gear legs (logical way would be use three actual gear legs and wheels/tires).

Then you would need to have the equivalent weight effected on the three gear legs at a typical vertical and longitudinal C.G. position so that a similar moment of inertia is simulated when the test rig bobs about the pitch attitude. And, it needs to be towed in a manor that would not induce any resistance to the pitching motion induced by the legs flexing.
Click to expand...

It looks like the jack holding it up could be removed and then weight added so there is 250 lbs on the front wheel. this would allow it to bob up and down much like it would on an airplane.
 
MrNomad said:
Thank you and we receive your comments in that same positive light. The fixture is not welded solid to the trailer. At the aft end it flexes so the jig is able to move up and down on its own based on the undulations of the road and those bumps we may place in the road to simulate potholes.
Click to expand...

My opinion is that unless you fully simulate the interaction that exists with the entire airplane (including the main gear) then you really don't have any way to correlate what ever your results are, to what would actually happen on an RV.
Example - The trailer wheels will also be going over bumps and undulations just like the nose wheel is, but with the trailer tongue attached to a vehicle you will not get the same reaction induced into the nose gear (the pitch angle can't change, instead the rear of the trailer will move up and down).

Second Example - Watch an RV (or video of one) taxiing over uneven ground. The tail is constantly moving up and down. This induces a moment of inertia on the nose gear, constantly changes the load carried by each gear leg, and effects a number of other possible factors. Your test rig design totally disables this interaction to your test nose gear leg.

As noble as your effort is, I think any information that think you gain from your tests, can not be directly compared to what actually happens to the nose gear of an RV taxing through / over the exact same surface.

My opinion anyway....
 
What's the scissor jack used for?

rubber314chicken said:
It looks like the jack holding it up could be removed and then weight added so there is 250 lbs on the front wheel. this would allow it to bob up and down much like it would on an airplane.
Click to expand...

The only purpose for the scissor jack is to raise the jig and lift the wheel off the ground while we drive to/from the test area. When we arrive at the test area, the jack is relaxed allowing the wheel & tire to follow the road.

I'm too old to try to lift 250-300lbs once we add the weight to the jig (probably tomorrow).

With respect to those who question the knowledge our efforts will contribute, I agree. A much better jig is indicated to fully simulate the front gear issues, but Chet and I don't have access to such equipment. Our hope is that someone who does have the appropriate resources will join/take over our efforts and solve a problem that has caused too much damage, injury, and heartache. While we wait for that day, we will add our findings to those who have conducted other experiments for these wonderful experimental airplanes.
 
I have to agree with Scott here, the trailer is not going to replicate the complex monkey motion of an actual airplane LG.

That said, your setup will get some data-----just how relevant it is, well probably better than what we have now??

How about towing a plane with the truck, with a couple cameras mounted on it----looking at LG from side, and longitudinally.

At least this will be addressing the issues Scott mentioned.
 
Good progress

Barry/Chet,

Looking good. Keep in mind what I told you in the past. It is important to actually fail the gear and document it (video etc). Then when you test the fix in the same way under the failure loading and terrain/speed you will have a good idea if the fix improves the design.
 
After a day of welding.....

Thanks Paul. A little levity is in order. We'd be happy to tow your recently completed 7A behind my truck until failure....... nah...... just kidding Paul.

Given how close you live to us, please plan a visit. You are precisely the right kind of engineer we need to improve the results. You were helpful when I bent the 9A, I'm certain you could be helpful now.

Bottom line? If I never saw another incident like Bob's friend, that would be great.

sfriend.jpg
 
MrNomad said:
img2204l.jpg
Click to expand...
Looking at this photo, here's my 5-minute Mechanical Engineer's thoughts...

If the silver portion is hinged at the back of the trailer, then it can move up and down independent of the rest of the trailer, right? So if the scissor jack is removed, and that portion is allowed to "bob" up and down on its own, I think that would be a reasonable simulation of the freedom of movement of the nosewheel with respect to the aircraft.

Add your 200 lb weight to the silver portion (simulating noseweheel weight) and tow away. As the main trailer wheels bounce over potholes, the nosewheel should be able to bounce semi-independently like it would on the airplane.

As others point out, it's not an ideal recreation, but I think it's not too far off. If you could get a camera videotaping it at a high framerate and cause it to fail somehow, even if it's not an exact recreation of the airplane's configuration, I think that video would answer a lot of questions... and may go a long way to explaining a lot of failures.

I suspect that you will probably need to cause a failure before you'll learn anything... So plan on buying a new nosewheel assembly and gearleg at some point. That might be prudent in any case... If you're going to do a lot of testing on this *not* to failure, do you really want to go and put that part on your RV, knowing how it's been stressed? I don't think I would.
 
moment of inertia and weight both important

As Scott pointed out, the moment of inertia is also important.

Since the fixture with the nose gear is pivoted on the trailer at the back, you will have a reasonable approximation of the motion, especially if the distance to the rear pivot is about the same as the distance to the main gear struts on the airplane.

It is still an issue that the trailer wheels crossing bumps will add an unrealistic motion and load, but you can get around that by towing over very smooth surface, with simulated bums that only the nose wheel goes over, not the trailer wheels.

In addition to simulating the static weight on the nose wheel, you can simulate the moment of inertia of the airplane by attaching a beam with weights at both ends, or just a long beam sticking out the front with the weight well forward of the nose wheel. This would have the added advantage of requiring less total weight. For example, if the beam extends twice as far forward as the distance from the pivot to the nose wheel, then only half as much weight would be required to simulate the static load. The trick is to calculate the right combination of weight and beam length that would simulate both the static weight and the moment of inertia.

Someone tell me the distance between the nose gear and main gear on an RV-(A) and tell me the distance from your pivot to the nose wheel, and I can take a crack at calculating that for you. I have a good enough idea of the airframe component weights to make a reasonable guestimate.
 
scsmith said:
Someone tell me the distance between the nose gear and main gear on an RV-(A) and tell me the distance from your pivot to the nose wheel, and I can take a crack at calculating that for you. I have a good enough idea of the airframe component weights to make a reasonable guestimate.
Click to expand...

For the 9A I found numbers that varied from 56.5" to 57.5" The nose wheel is 34.5" aft of the datum, and the cg range is from 77.95" to 84.84" aft.
 
More Armchair comments

First - I think this is a great idea.
Second - I'd NEVER put the unit, post test, on a flying airplane

and now for the armchair comments
- Raiz has posted some calcs and comments in the last few months. He defined some "failure points". I would think if you could test a series growing closer and closer to replicate the step function failure or other failure with tire pressure differences he discussed, it would fill in the gaps between wonder, calcs, and real world.

- someone else mentioned the black and white blocks for scale/displacement. You could make up a plate very simply with a piece of metal or wood and black and white paint. This test won't be exact,.. so as long as you make the blocks about right,... video should be good. You could mount so you could see the tire/gear moving up and down, and possibly back and forth.. (need to figure out if the view would be "one to one" or due to angle the actual displacement is different than "shown" against the black and white checkerboard.)

Another means to figure out the movement is to make a "record" of the movement. Can be done in several ways,... for instance could use gorilla glue and put small cross piece on the gear by the wheel. Then mount something next to it that the cross piece could push aside (think clay) to show the maximum displacement. what you are trying to do is "isolate" the actual movement,.. and not get into "inertia" of something being displaced and total movement being more than that of the nose gear or object being studied.

now,.. if you could really get some of that neat electronic gear,... you might not need any of this
 
Gear leg to donate

I have my old nose gear leg and fork sitting in the hangar. It's the original 6A gear that I changed out after the service bulletin. You can test it to destruction if you'd like. Actually I'd like to watch too! I'm kinda close, up at Chandler.
 
gear leg

Now you guys are getting with it.I would like to see a gear leg video to destruction also.
I think you will see these gear legs tuck under instead of flexing up.
When I built my 7A I had a lot of comments on the spindly gear leg.
Your doing more than anybody else has done to test the gear.
I do think that when the nose dips you have the moment of the tail adding to the weight on the nose.
 
Hopefully, our efforts will lead to an improved design

billnaz said:
I have my old nose gear leg and fork sitting in the hangar. It's the original 6A gear that I changed out after the service bulletin. You can test it to destruction if you'd like. Actually I'd like to watch too! I'm kinda close, up at Chandler.
Click to expand...

Bill: At the risk of sounding like Tony Soprano, this sounds like an offer we can't refuse. YES, CHD is a stone's throw from 57AZ and YES, we'd like to meet and discuss so please send an email/phone number to

oldak at comcast.net or mrnomad57 at gmail.com.

In less than 12 hours since the initial post, we've been offered a gear leg, fork, and video equipment from a friend currently in S. Africa. While we wait for this loaner equipment to arrive, the initial non destructive tests will continue.

Later today Chet and I will discuss the engineering comments we received. Not being mechanical engineers, we appreciate the ideas and hope that the engineers in our RV community will play an increased role.

Just like building an RV, we firmly believe in teamwork and collaborating with others with the hope that our combined efforts will lead to demonstrable improvements in an area that's vexed too many pilots.
 
You can get close to eliminating the flipover at ZERO COST easily. Starting today.

Land on well-maintained paved surfaces only.
 
Nose Gear

I applaud your efforts as will many 'A' drivers. But it is apparent from all the obsevations that recreating a failure many not lend much 'real world' information. There is likely gyroscopic loads involved in the equation that might be hard to reproduce in a static test stand under point loading.
Most of the rubber side up photos of the 'pole vaulting' RVs show the plane laying in grass. We know the gear doesn't do well off the payment. I hope your activities lead to a new design.
Your trailer experiment may only reinforce what we already know "Keep off the grass."
 
MrNomad said:
Bill: At the risk of sounding like Tony Soprano, this sounds like an offer we can't refuse. YES, CHD is a stone's throw from 57AZ and YES, we'd like to meet and discuss so please send an email/phone number to
Click to expand...

One thing to keep in mind, regarding older RV6A nose gear legs.........is how old is it? Around 1999 Van's issued an SB for older gear legs, which failed early. My 1996 kit/leg was replaced at that time. When testing with a high speed wheel & cam........the early legs failed quickly. The new models stood up to the abuse.

L.Adamson -- RV6A
 
Van's spec

for maximum nose wheel weight on the nose is 375#. As you progress with your tests, you might want to consider higher weights.

Would $5 or $10 from a bunch of us help to fund your experiment. If you set up a paypal donation account, Im in. Wish I was closer. Still have my old engineering text books :rolleyes:.
 
On second thought,,,

My RVr friend has a very capable machine shop in his hangar (Mill, Lathe, Plasma cutter, etc.). If there is anything you need that I might be able to fab up on his machine tools, please don't hesitate to ask. KHND to KTUS is only 2 hours:D.
 
Glad to see this.

I thought about building a rig like yours, but am still building the fuselage, so don't really have the time at this point in my project. However, please let me offer a few thoughts in defining the tests themselves.

Having studied most of the taxiing videos on YouTube, it looks like there is a lot of fore/aft motion of the nosewheel on fairly smooth pavement. I think replicating this motion should be your first test, that is find out whether it is low tire pressure or loose bearings, or a lack of a spacer between the bearings. All tests should be repeated at three or four speeds up to the stall speed of the aircraft, say 50 mph, or let's say at 12, 24, 36 and 48 mph.

Once the cause of this behaviour has been established and a remedy found, then progress on to some bumps. As another post mentioned, this should be a bump that only the nose wheel sees, not the trailer wheels. And it should be a well defined "calibrated" bump. For example, a 2x10 or 2x12 plank with a 1:4 taper ramp on one end. You might start with a single "bump" and then progress to a series of them. A "hole" of a similar size would be nice, but I suspect not a practical idea. Something to watch for is whether the nose gear reacts to the bump by a vertical movement or a horizontal movement (probably some of each). I think the horizontal movement is the most dangerous reaction as it is in the direction of folding the nose gear under. Raiz' analysis seems to indicate that the reaction is mainly vertical and the failure mode is tensile stress failure at the nose gear rod socket (if I interpret his results correctly). I suspect the "pole vault" tuck under failure is a combination negative bending moment (caused by aft movement of the nose tire) coupled with a column buckling tendency.

If the nose gear has not yet failed, you can consider testing on some "rough ground", but the problem is defining just how "rough" it should be.

I spoke with Van's about this issue, and in addition to the eccentric test rig (RVator Fifth Issue 1998), they did instrument an aircraft and landed/taxiied it on a grass strip. But I'm not sure how rough it was or the results of the instrumetation.

Well, you're getting lots of advice and encouragement, I look forward to some results.
 
Last edited:
Calibrated bump

Barry, my hat is off to you for undertaking this project!

I am in agreement with the "calibrated bump" suggestion. In order to have useful test information, it is important to make the test simple enough and the test protocol rigorous enough that the test is repeatable (the basis for the scientific method, i.e., some one else can independently replicate your test and get the same result).

This generally means that the test will be "oversimplified" to a lay observer. The best example I can think of is automobile barrier crash testing. Does the test represent a real world collision scenario? Of course not. However, because the test is so simple to conduct, results are repeatable. Over years of testing, much has been learned and the state of automotive crashworthiness has been greatly improved. If you want proof, go to the IIHS website and watch the video of an offset front-end collision between and 1957 Chevrolet and a 2007 Chevrolet.

Back to the calibrated bump. By varying the shape and height of the bump and comparing the response, a tremendous amount of information will be learned. I personally like the idea of the grid and video camera because it will capture the complex wheel motion. In automobiles, the natural frequency of the suspension is about 1 Hz. Standard video cameras run at 30 frames/second. Perhaps it would not take a really high speed video to produce good results. The vertical deflection is cantilever bending, so the moment arm is the distance from the engine mount anchor to the axle. The "tuck under" is the local moment from the rearward directed force just above the ground about the attachment to the end of the gear leg superimposed on the static moment (moment arm = trailing arm length; F= nose wheel weight).

It will be exceptionally valuable to vary tire size in addition to tire pressure. Speaking from personal career experience, auto manufacturers spend tons of money trying to address "impact harshness" issues, defined as the ability of the suspension not to isolate the "impact" of striking a step in the pavement from the vehicle's occupants.

This might (probably is) be more information than most forum members want. If you would like some additional input regarding testing and analyzing data, I would sure be willing to help. Please contact me by PM because the it will probably take something more interactive, like telephone conferences, to explore the possibilities.

LarryT
 
Impact Harshness

I am clearly linguistically challenged (I am not a cunning linguist). What I was trying to say was:

Good impact harshness = good isolation from occupants

Bad impact harshness = bumps are transmitted to the occupant compartment.

Examples:

MacPherson struts have poor impact harshness inherent in the design.

SLA suspensions (upper and lower control arms) have the capability for good isolation

LarryT
 
Gear Leg

This may sound far out , but why not contact Myth Busters out in California for some ideas on what they might use for test monitoring, measuring stresses, etc/ They seem to exhibit a vast knowledge of physics.
Just a thought
Jerry:cool:
 
Larry, I couldn't find the video

LarryT said:
Barry, my hat is off to you for undertaking this project!

The best example I can think of is automobile barrier crash testing. Does the test represent a real world collision scenario? Of course not. However, because the test is so simple to conduct, results are repeatable. Over years of testing, much has been learned and the state of automotive crashworthiness has been greatly improved. If you want proof, go to the IIHS website and watch the video of an offset front-end collision between and 1957 Chevrolet and a 2007 Chevrolet.

LarryT
Click to expand...

Maybe this is the one you were looking at:

http://www.youtube.com/watch?v=joMK1WZjP7g

kent
 
Last edited:
Serious reservations about test method

One of the really overlooked factors in the great RV nosewheel failure debate is the effect of torsional moments on the nose gear.

When a two seat RV nose gear yields and fails it results from the combined effect of all applied loads...typically compression force, bending moment, and torsional moment.

Torsional moment (twisting of the gear leg) occurs when there are lateral (sideways) forces applied to the nose of the aircraft. For example, push sideways on the engine of an RV7A and the nose of the aircraft will readily rock from side to side (like a great big jelly in fact). That sideways movement induces twisting in the gear leg....lots of it.... due to the fact that 1) the gear leg is a very long spindly rod 2) the nose wheel is not concentric with the axis of the gear leg.

Unless the test rig can simulate all major loads on the gear leg then it cannot hope to simulate the real world.

I cannot see how the test rig as it stands will enable the simulation of torsional moment in the gear leg because there is no possible lateral (sideways) force being applied to the upper gear mount as there would be in an RV aircraft. On the test rig the nose gear is effectively located centrally between four road wheels ( 2 on the trailer and 2 on the pulling vehicle) and therefore is is not subjected to any lateral forces.

That alone suggest to me that any results from the proposed tests are unlikely to be valid unless the test rig is seriously modified.
 
What I don't understand yet is what this information is going to be used for.

Are you going to find a set of controlled conditions that cause this leg to fail and then use that as an acceptable criteria for a redesign? Because if that's the plan, how will you know that the criteria is adequate? All you'll know is that the old leg failed with it and the eventual new one didn't. You won't know how it relates to flying an airplane.

Information is primarily used to make decisions. What decisions will be made using this information?

Not trying to ruffle feathers here, just trying to find out the purpose of the exercise.

Dave
 
Torsional Moment

Sorry to disagree with you Bob, I do not think the the torsional moment contributes to a primary failure mode. You are right about pushing sideways on the nose of an aircraft, but in this condition, the machine is at rest. So the weight on the nose tire contact patch prevents (unless you push really hard) the tire from rotating around the castor axis which is about a foot in front of the contact patch. If you put a ball bearing under the nose wheel and tried again, then all you would have to overcome is the breakout force on the castor axis preload springs.

Watching the taxiing videos, when the aircraft is moving there appears to be no noticeable torsional moment in the gear leg, that is no deflection, as the tire castors in response to the turning of the aircraft.

Again from watching the taxiing videos, it appears that the primary failure mode of "pole vault/tuck under" failures is longitudinal load through the gear leg plus a negative bending load from either a bump resisting the forward motion, a harmonic vibration, or from the gear leg nut digging into the surface. The combination of longitudinal (compressive) load plus negative bending moment can produce a column buckling failure consistent with the "pole vault" pictures we have seen.

In any case, "a test is worth a thousand opinions". Even if this test fails to identify what the culprit is, it will probably identify what the culprit is not. And this will probably lead to another test program. You always learn something from a test program. Sometimes it is not what you expected to learn.

So as before, I await the test results with great interest.
 
Donations???

I certainly am not any kind of expert pilot or even a judge of good/bad landings and only speak for myself here. I have set down beside the landing zone of several runways at fly ins and watched the RV-As as they landed. Many RV pilots appeared to land hot and immediately put the nose wheel on the runway - (in order to begin hard breaking?) - thus causing a fore and aft oscillation of the nose gear that was clearly visible. Thus, Van's admonition not to land too fast and not to put the front wheel down until there is insufficient elevator authority to keep it off of the runway.

I watched Van himself land his RV-10 as he came to work at the factory last year and his landing roll out I guess was about 600'.

Some pilots have suffered the nose wheel tuck under while taxing at low speeds = bad luck? I gotta line up with those who say that that particular failure should not happen. So how is it possible that a well engineered nose landing gear could suffer a collapse at a low taxi speed? Yeah - I know that RVs are not Cessnas or Pipers but I'll bet that most of us 'graduated' from Cessnas or Pipers and none or few of us flipped one of them over.

I am glad that you guys are trying to do something about what is a problem (in my opinion). Keep up the good work and I hope that you will find an answer that will reduce the accident rate.

I am happy to send you guys some $$ in order to help defray expenses. Just tell me how to send my $20 donation to you. Hope that will help.
 
I will give you a solution that will result in an immediate reduction in these events of perhaps over 90%:

Operate off of well-maintained paved surfaces.

Throw in converting to the newer gear leg/fork that provides 1" more clearance of the bottom of the gear leg, perhaps reinforcing the lower part of the front lower wheel pant to aid as a skid and recommended landing techniques and the reduction in these events could be over 95%.
 
Ron Lee said:
I will give you a solution that will result in an immediate reduction in these events of perhaps over 90%:

Operate off of well-maintained paved surfaces.

Throw in converting to the newer gear leg/fork that provides 1" more clearance of the bottom of the gear leg, perhaps reinforcing the lower part of the front lower wheel pant to aid as a skid and recommended landing techniques and the reduction in these events could be over 95%.
Click to expand...

I can offer a solution that will result in a 100% reduction:

DON'T FLY AT ALL

Seriously though, neither "solution" is going to happen. We like flying and many folks don't have the nice runway option. Not logical to operate an aircraft that is "safe, as long as you restrict normal operations". It behooves us all to support nose leg design improvements.
 
Paypal Fund

HI Barry & Chet,
Outstanding!!! Disregard all criticism and plow ahead ,Any effort will help the cause and the rig looks great,its a start as time gos by you can make changes and include more accurate simulations but you have to start some where. I suggest a I beem with two posts to hold bar bell weights side by side of 45 pounds each till you reach desired weights for testing.A papal account is in order,a request to Vans for the donation of gear legs is in order,Fun n Sun is right around the corner and they are our captive audience if 100 customers request their help,it will put this effort on the map.Best of luck and here is my $20 and Thanks.
Bob
 
This rig has so many options...

...that allow tests to destruction...exactly the physical proof-of-concept beginnings.

Leland Snow at Air Tractor did landing gear drop tests on the new Air Tractor 1002...a thousand gallon airplane (1,000 gallons in the hopper!) and dropped the whole landing gear assembly until the spring gear buckled inward and the wheels leaned in at more than 45 deg at impact. Everything survived and was analized (sp) for structural damage.

Seemingly Rube Goldberg devices are often very valuable!

Best,
 
pierre smith said:
... Everything survived and was analized (sp) for structural damage...
Click to expand...

I think that "analyzed" is the word (a verb) you were looking for Pierre, but you may have just invented an adjective that best describes some of the discussion that goes on here...;)
 
PROGRESS REPORT ON FRONT GEAR TEST

Thanks to (almost) all for your comments. Given that my nose gear folded on a “well maintained” paved runway in San Diego, avoidance of grass fields is not an adequate solution.

PROGRESS REPORT: This week we welded additional steel onto the jig, mounted 5 Chevy cylinder heads adding 210 lbs, and towed it for test #2.

In total, using my wife’s bathroom scale (she was none too happy), the nose gear weight including the front wheel is 252 lbs. However, as one of the engineers in this thread pointed out, the see-saw inertia caused by an airplane as the mains bound over bumps, changes the amplitude of the load on the nose gear. I am not a mechanical engineer so I apologize if my terms are not exactly perfect but we think we need to change the configuration of the jig so that more of the static weight is in front of the attachment point which will (hopefully) induce the see-saw effect and increase the bending of the gear leg. Thus far, all we are observing is the tire with 38lbs of pressure compressing as it hits small bumps. We have not created any simulated bumps yet. Yes, we will vary tire pressure once we accurately simulate the action of an A model.

How can people help? Given that we are type-A old men and there are only 27 hours in every day, here’s what we need. First, we could use some engineering to tell us how to induce the see-saw and side loads. My catastrophe occurred when my front wheel followed an intersecting runway and bent over sideways. If you have the engineering skills we need, send us some design considerations otherwise we will simply waste time and steel with experiments.

Second, if someone wants to set up a PayPal account and collect $$ to be used towards the steel and other equipment we need, please set it up and hang onto the proceeds to be used for that equipment beyond our social security funded budget. We think we need D/A equipment so that we can get precise measurements but haven’t a clue where to get it. If you have other ideas or measurement jigs we can attach, please send photos and/or the actual devices.

Third, remote cameras are not my forte so if you have access to photo equipment we can use to record the effects of engineered designed bumps, please loan it to us. If we destroy it during the tests, we will ask the PayPal teammate to send the $$ and offset your contribution. Right now, a senior citizen is sitting on the trailer taking pictures while it’s towed down the road. Obviously, that’s not a good idea.

Thanks again.
 
Bravo

you guys - this is great. The only thing I would suggest is to formulate as many hypotheses as you can to test. For example, what is the effect of weight?, or what is the effect of pot hole depth on failure. It may sound silly, but in the end you will have a list of specific hypothesis that you tested, and the results. Maybe others here can suggest some. Other than that, duplicating as closely as possible the RV-XA landing gear as possible I think is critical. For example - here's a hypothesis... At or above gross, when the main gear hit a pot hole the increase in downward force on the nose wheel will cause a failure. To test that hypothesis you'll obviously need actual main gear and wheels. Also, I feel that some of the failures are due to the airplane "pole vaulting" over the nose wheel, placing the entire airplane weight on it with additional force (F=M*A, remember) and causing it to fold under. Your rig may not be able to duplicate that scenario.

You get the point... good luck!

Tom
 
Another thought has just made it way through....

I know, at my age it can be a shock, but it does happen from time to time.

I also may have missed it in the proceeding messages, but....

With increased speed the weight on the wheel will be lessened (IE we are holding the stick back). How much effect does that have on the nose wheel weight?

Will this be accounted for with math or will your remove some of the weight when you try higher speed runs with the 'rig'?

Kent
 
Run over a "calibrated bump" ...

MrNomad said:
Thanks to (almost) all for your comments. Given that my nose gear folded on a ?well maintained? paved runway in San Diego, avoidance of grass fields is not an adequate solution.

PROGRESS REPORT: This week we welded additional steel onto the jig, mounted 5 Chevy cylinder heads adding 210 lbs, and towed it for test #2.

In total, using my wife?s bathroom scale (she was none too happy), the nose gear weight including the front wheel is 252 lbs. However, as one of the engineers in this thread pointed out, the see-saw inertia caused by an airplane as the mains bound over bumps, changes the amplitude of the load on the nose gear. I am not a mechanical engineer so I apologize if my terms are not exactly perfect but we think we need to change the configuration of the jig so that more of the static weight is in front of the attachment point which will (hopefully) induce the see-saw effect and increase the bending of the gear leg. Thus far, all we are observing is the tire with 38lbs of pressure compressing as it hits small bumps. We have not created any simulated bumps yet. Yes, we will vary tire pressure once we accurately simulate the action of an A model.

How can people help? Given that we are type-A old men and there are only 27 hours in every day, here?s what we need. First, we could use some engineering to tell us how to induce the see-saw and side loads. My catastrophe occurred when my front wheel followed an intersecting runway and bent over sideways. If you have the engineering skills we need, send us some design considerations otherwise we will simply waste time and steel with experiments.

Second, if someone wants to set up a PayPal account and collect $$ to be used towards the steel and other equipment we need, please set it up and hang onto the proceeds to be used for that equipment beyond our social security funded budget. We think we need D/A equipment so that we can get precise measurements but haven?t a clue where to get it. If you have other ideas or measurement jigs we can attach, please send photos and/or the actual devices.

Third, remote cameras are not my forte so if you have access to photo equipment we can use to record the effects of engineered designed bumps, please loan it to us. If we destroy it during the tests, we will ask the PayPal teammate to send the $$ and offset your contribution. Right now, a senior citizen is sitting on the trailer taking pictures while it?s towed down the road. Obviously, that?s not a good idea.

Thanks again.
Click to expand...

... at an angle. This is would be very similar to your mishap in San Diego. Please, please avoid designing a test with too many variables (refer to my previous post re simple tests). The calibrated bump can be as simple as a strip of 1/2" plywood secured to the pavement that engages only the nose wheel. By varying weight, tire pressure, approach speed and approach angle w/ multiple runs at each configuration, you are going to have a lot of data to analyze. Once you accomplish all of the above, you can change the height of the bump to 1", etc. etc. That is a lot of testing. If possible, I would consider one camera at a fixed location on the ground ("global reference") and one camera on the trailer ("Local reference")

LarryT
 
You must log in or register to reply here.
Back
Top