Broken Tail wheel bolt

[skywagon8a]

Like Tim, I too am wondering how we can both have a bolt tensioned enough so cyclic loads never are reversing, as Gilbert correctly states is the key to long fatigue life, and yet have the two springs slide during flexing as they should (otherwise I suppose they attempt to go "S" shaped). I grease between the leafs to promote sliding. Bill at Bushwheels said he currently puts springs together dry but is considering using a graphite paste in the future. Perhaps the relative spring movement creates a moment on the bolt and is the force that leads to nuts loosening. All speculation, but I see the need for high torque on the bolt to prevent bolt fatigue and the movement of springs against each other as an overall fault in the original design.

The slotted hole is sandwiched between the round hole and the head of the tailwheel with a bolt that is torqued to 70 ft lbs. That is a very tight clamping on the upper spring leaf. The highest landing impact will move the upper leaf to the maximum travel available. Now, with this high clamping, what force is going to move the leaf back to or near it's original position? The highest reverse load will only be the rebound of the tailwheel assembly. Won't the clamping action with the high torque on the bolt hold the leaf and prevent further motion? This means, to me, that the leaves of the spring will only flex after the initial large impact has set the position. Thus there will be no further motion and no need of any lubricant to aid in the leaf sliding which it will not be doing.

What am I missing? Will the rebound load be as high as the impact? Will the leaf still move back and forth with the 70 ft lbs torque? Should the bolt be re-torqued without loosening after the high impact landing? If it is loosened when re-torquing then it will have slid back to the relaxed position.

[Steve Pierce]

Do the two main springs actually slide or do they bend in the middle. I have seen the evidence of the third spring sliding but not sure on the two mains.

[sierra bravo]

The top leaf would have a larger radius in the arc so as it flattens out it would have to slip to allow the arc to straighten out, right? If there was no slot in the upper leaf the differential movement of the springs at the bolt hole would cleave the bolt or elongate the hole. If it slips it would wear the aluminum tail wheel casting unless there was some provision made for clearance. So how does it work?

I need coffee.

[Steve Pierce]

Good point Stewart. The spring should not slide on th head. I think the slot is a manufacturing design to insure proper hole alignment after they are formed in the furnace. Neat video from Univair showing them make these.

[sierra bravo]

The top leaf would have a larger radius in the arc so as it flattens out it would have to slip to allow the arc to straighten out, right? If there was no slot in the upper leaf the differential movement of the springs at the bolt hole would cleave the bolt or elongate the hole. If it slips it would wear the aluminum tail wheel casting unless there was some provision made for clearance. So how does it work?

So for you guys who've seen failed conventional bolts in the tailwheel, did they break at the head like the eye bolts? If the system fixes the bottom leaf to the tail wheel head with close tolerance bolt holes and allows movement of the intermediate member, the middle leaf spring, and then binds the assembly with tension of the bolt, isn't it completely plausible that the bolt is being sheared, even tilted in the stack? It it tilts the place we'd expect it to fatigue and fail would be at the bolt head, right? Does anyone install the bolt from the bottom with the nut on top, as would be the case in Spinners ring nut? Any failures? Where? Bottom (bolt head) or top (under the nut)?

[cruiser]

I looked at Tim's broken bolt this morning. At the break on the eye piece it appears that the bolt portion has a bit of a taper on the edge like it is threaded into the eye. I am ready to be corrected on that but we agreed that is what it looks like with what little is showing. A magnifying glass would show more which we did not have. If it is threaded there perhaps that is a reason for the break. Jim

[Steve Pierce]

Suppose to be a forging and then the shank is machined.

[Ursa Major]

I've never broken a TW bolt, even though anybody who has seen me land will attest that the bolts on my cub receive lots of stress. When I lived in Alaska, the cub was tied outside and I would have liked to have an eyebolt for tiedown purposes. ABW had not yet marketed the current bolts, so I used an alternative. I took a ski tab and bent it into a 90 degree L and used one end as a washer under the existing bolt head.

I would clip a snaplink or carabiner into the upright portion and leave the link attached to the tiedown rope or strap at the tiedown. I would clip the link to the ski tab when tieing down and snug the rope or a cam lock strap. Sure saved a lot of hassle with stiff frozen fingers trying to fight frozen rope at the end of the day. The key was to make sure you removed the caribiner or snap link from the ski tab before flight. The ski tab would stay attached as a washer/spacer.

TW tiedown.jpgTW tiedown 2.jpg

[8GCBC]

This is what I do for a tiedown in moderate conditions....
tailwheel.jpg

[Tim]

Mike, how tight do you tighten the -7 bolt

Tim

[Darrel Starr]

Bill at Bushwheels told me that the eyebolts are a forging and all come from one lot.
I am about to torque an AN7-21 bolt to replace the eye bolt, still thinking about the AN8 bolt but haven't ordered any yet. AC 43.13-1B Table 7-1 shows a torque limit of 450-500 lb-inches torque (38-42 lb-ft) for a 7/16-20 (fine thread) bolt. This is for a full depth tension nut and a cad plated bolt. I don't think bolts are Cad plated anymore so that is an issue. Also, the elastic stop nut adds another unknown as to how much torque it takes just to yield the plastic. Will probably just check to see how much torque it takes to turn the nut before the assembly goes tight then add that to the 42 lb-ft. The recommended torque only stresses the bolt to 40,000 psi according to the same chart. Yet the bolt yield strength is 124,000 psi as I recall. The book allows overtorquing to get to the next hole in a castellated nut so on second thought I think I will just use the castellated nut/cotter pin with a bit of overtorque.
I can see why Gilbert is recommending a higher torque, according to the following reference ( http://www.turula.com/bolt.html ) 70 lb-ft would result in a bolt stress of approximately 107,000 psi which includes 35,700 psi shear stress due to torque added to the 79,400 psi tension. I have not researched this methodology but just offer it as one easy to access approximation. And I think Gilbert's point is to torque the bolt just under the yield point to maximize fatigue life.

[Gilbert Pierce]

Minimum yield strength of AN bolts is 125,000 pounds per square inch. A 7/16" AN7 bolt has a area of .6 square inches. 125,000 pounds per square inch times 0.6 square inches gives a minimum yield strength of 75,000 lbs. My calculations show torquing a 7/16" bolt to 70 foot-pounds would yield a clamping force of 9600 lbs. That assumes minimum loss to friction and a castalletd nut. CORRECTION-I used diameter when I should have used radius of the 7/16" bolt. The correct area is 0.150 inches squared for a yeild strength of 18,800 lbs. See next post.

This comes from a simplified equation Clamping Force F = Torque in inch pounds divided by .2 (diameter of the bolt) Clamping Force = 840/.2(.4375)
This does not apply to gasketed joints or joints that contain material of less compressive strength then the bolt. Per Bill Duncan I use a hardened grade 8s washer under the tail wheel bolts for the above reason.

I have seen cylinder base bolts fail from under torqueing. Wheel lugs, my own tailwheel spring at the airframe, machine base bolts, and few others I can't remember fail from fatique due to under torqueing or just using a wrench and calling it "good enough". I started going to the higher torque values in Table 1 after finding my tailwheel spring to fuselage bolt stretched 3 months after changing it. I do that at annual time. Since I went to the higher torque and grade 8 washer I have not experienced stretching. The Clipper uses an AN6 bolt at this location.

Just an poorly educated guess as to the Eye bolt failures; it looks like to me from the pictures of the bolts to be fatigue failure caused by under tightening or hydrogen embrittlement weaking the bolt caused by the Cad plating or a combination of both. When strutural steel parts are cad plated they must be baked to remove the hydrogen gas that is trapped between the steel molecules. Heat drives that gas out.

[12 Geezer]

Looks like a finger-slip on the calculator keys - -

The cross sectional area of a 7/16 diameter bolt is 0.150 sq in. Thus a 125,000 psi 7/16 dia bolt would have a yield strength of 18,800 lb.

[cruiser]

There are many applications on an airplane where a particular size bolt is not torqued to the limits on that chart. I do not see the benefit or need of 70 ft lbs on this bolt. We all understand the need to torque connecting rod nuts but a tail wheel bolt? To 70 lbs? I don't see it. Jim

[8GCBC]

If the TW did not shimmy (from incorrect installation angle and loose tolerence), a clevis pin and a cotter key would work. There is a plethora of conflicting stresss on this design and too much "play" in the dimensional tolerences of the TW "shoe" for a single bolt.

My MLG axles are held with a 4 bolt pattern. The TW having a single bolt pinning a loosely fitting "shoe design" is asking for failure. I am sceptical of the product and make extra effort to inspect, tighten or change the bolt.

[Gilbert Pierce]

Looks like a finger-slip on the calculator keys - -

The cross sectional area of a 7/16 diameter bolt is 0.150 sq in. Thus a 125,000 psi 7/16 dia bolt would have a yield strength of 18,800 lb.

You are being kind. It was not a finger slip. I used Diameter instead of Radius in the Area equation that should have used Radius. The clamping force is correct and you are right, the yield strength is 18,800 lbs.

[12 Geezer]

There are many applications on an airplane where a particular size bolt is not torqued to the limits on that chart. I do not see the benefit or need of 70 ft lbs on this bolt. We all understand the need to torque connecting rod nuts but a tail wheel bolt? To 70 lbs? I don't see it. Jim

This Wikipedia article gives a good explanation. Torquing to a spec has to do with reducing the cyclic (fatigue) loading on the bolt. http://en.wikipedia.org/wiki/Bolted_joint

[Gilbert Pierce]

There are many applications on an airplane where a particular size bolt is not torqued to the limits on that chart. I do not see the benefit or need of 70 ft lbs on this bolt. We all understand the need to torque connecting rod nuts but a tail wheel bolt? To 70 lbs? I don't see it. Jim

Any bolt loaded only in shear does not and in most cases should not be torqued to any value over finger tight. The wing attach bolts, strut bolts and any bolt holding a pulley are shear loaded only. Especially the strut to fuselage attach fitting should not be torqued. However any bolt loaded in tension should be torqued, especially if is cyclic loaded.

[12 Geezer]

I used Diameter instead of Radius

Heheh - - - I give my students extra credit for catching me in mistakes - they love it when I do something like that!! Thanks for your input on this, Gilbert - good food for thought.

[Darrel Starr]

Removed the eye bolt and discovered the torque was only 20 lb-ft -- less than half the torque I used 105 hrs ago.
The eye bolt seat and shank is rough with tool marks in contrast to an AN7 bolt that is smooth -- this is not "AN Quality". I don't know if the roughness has contributed to the low fatique life but it can not help. Poor quality control!

P1170208a.jpg
P1170215a.jpg

[cruiser]

After looking at Tim's broken bolt under a magnifying glass it is not threaded on the ring end. Jim

[Bill Ingerson]

I jacked my tail wheel of the floor and sure enough the whole assembly was loose again. It moves about 1/4" from side to side. be sure to check it if your running baby bush wheels about every 10 or 15 hours. This is the third time in about 6 months that I have had to tighten it and I am really getting a tight wrench on it about 60# would be my guess. I removed the eye bolt tonight and washed it in solvent and put a good light on it. The bolts are cast steel then machined down with threads all one part. Then they are cad plated, I checked for any grooves or problems and found none. The cad plating had worn through in two small spots, I think that was from being loose. I was curious, should a thin washer be put under the Eye ring there is none now. I put a dab of grease on the bolt and put it back in and tightened down again. I can always tell if its coming loose if someone is in the back seat and I set the tail wheel down early on landing, it feels like the tail is shaking really bad, I move the stick forward to get the tail weight off and it usually goes away. When I first put the eye bolt on I did put it on, and it was upside down with the eye pointed down. On the way back from Johnson Creek I noticed Bill and Wup looking at it, Probably thought This guy has to be from Seattle. Didn't seem to matter to the bolt, just made it harder to tie a rope to the eye.

[spinner2]

I like the washer/ring idea and made a sketch of what would work. This would use all AN hardware.

The ring is .250" dia., the 'washer' plate is .125" think and the two would be welded together. The weight calculates at .10 pounds.


tail_bolt_eye_r.jpg

[Darrel Starr]

Looks really good -- suggest you check for box end wrench clearance inside the ring. I am assuming the plan is to have an AN7 nut inside the assembly with the bolt inserted from the bottom.

Here are zoomed pictures that I posted before that concerned me about the Bushwheels eye bolt. Random lathe tool marks in the critical transitional radius between the shank and top surface of the bolt are evident in the eye bolt but not in the typical AN7 bolt. This could be a random contributor to lowering the eye bolt fatigue life compared to the AN7 bolt.
P1170208c.jpg
P1170215b.jpg

[N5126H]

We Who are not worthy salute you

I really like the eye bolt and tie down more often and quicker because of it but this has me wondering if there is a better mouse trap... Maybe a stainless steel tang that a standard bolt passed through so you can regurally replace the bolt and still have a loop to tie to.

Like maybe nothing attached to the plane, but rather attached to the tie down rope, like say an oversized carabiner.
I use one on the end of my winch, mounted on the back wall of the hanger to pull the 185 in during the winter. I would not recommend this for serious tiedown applications, but for the more laid back applications it is quick and easy.

[mike mcs repair]

..... When I first put the eye bolt on I did put it on, and it was upside down with the eye pointed down. On the way back from Johnson Creek I noticed Bill and Wup looking at it, Probably thought This guy has to be from Seattle. Didn't seem to matter to the bolt, just made it harder to tie a rope to the eye.

thats the way I put them, bellow..... or else its gonna eat you rudder for lunch when heavily loaded/rough....(the tail wheel housing can already munch a rudder WITHOUT that big eye on top!)

[mike mcs repair]

So for you guys who've seen failed conventional bolts in the tailwheel, did they break at the head like the eye bolts? ...

no, middle of bolt, it would shear between the spring leafs(front/fuselage bolt)... we used to change them every other year or so....

[Patrol Guy]

here is the waitress that broke Bill Tracy's bolt.

waitress.jpg

[mike mcs repair]

Here are zoomed pictures that I posted before that concerned me about the Bushwheels eye bolt. Random lathe tool marks in the critical transitional radius between the shank and top surface of the bolt are evident in the eye bolt but not in the typical AN7 bolt. This could be a random contributor to lowering the eye bolt fatigue life compared to the AN7 bolt.....

I did also see some stretch, looseness of tail wheel head in first 50 hours on the one I installed.....

so how many have broke? how many have seen stretch/looseness?? and what type of service were they in?? normal or real rough off airport?? this one I installed(and got field approved) was not in very rough use......

[Bill Ingerson]

Mike, my eyebolt has come loose three times since I put it on. In a time span of less than a year. I remove it, look it over good, then put it back in and tighten it real good. Bushwheels told me to check them for coming loose every 10 - 15 hours of use. My plane is never loaded heavy, and almost all my landings are two wheel. Always taxi with the stick forward to keep the weight off the tail wheel as much as I can. You mentioned something I never thought of Maybe the bolt is Stretching some ?

[skywagon8a]

I am not a metallurgist. It seems to me that if these bolts are stretching enough to loosen in such a short length of time and as often as seems to be the case, then they are being made of the wrong alloy of steel. Are you using stop nuts that are rotating or castle nuts with cotter pins? Do the pinned nuts loosen also?

What about the idea of placing the eye bolt at the forward hole in the spring where it attaches to the bottom of the fuselage? No greasy rope, no damaged rudders and no possibility of the rope bending the tail wheel steering connections.

[mike mcs repair]

...
What about the idea of placing the eye bolt at the forward hole in the spring where it attaches to the bottom of the fuselage? No greasy rope, no damaged rudders and no possibility of the rope bending the tail wheel steering connections.

I like that, but is it the right length and diameter?

[mike mcs repair]

I am not a metallurgist. It seems to me that if these bolts are stretching enough to loosen in such a short length of time and as often as seems to be the case, then they are being made of the wrong alloy of steel. Are you using stop nuts that are rotating or castle nuts with cotter pins? Do the pinned nuts loosen also?...

all steel nuts, they aren't turning....

why I get nervious about this is, I had an extremely wrecked airplane job from that tail wheel head bolt breaking.. all 4 spars bent, fuselage bent in half....(was sloped beach, it pulled rudder one way, the wrong way..or such)

[skywagon8a]

I like that, but is it the right length and diameter?

Of course it would need to be of the correct size for the application.

[skywagon8a]

Another question. Are the bolts being installed in the correct FAA head up position or are they being installed with the nut up? Do you all think that there would be any less chance of the head breaking off if it was on the bottom against the spring instead of on top against the tail wheel housing?

It seems strange to me that the heads are breaking off. You would think that the bolt would shear at the spring instead. Perhaps the bolt should be of the high strength type, such as NAS624 which has a large radius at the head and requires a special MA20002C beveled washer?

[n40ff]

Just read the entire thread again. Now have a headache and blood is beginning to shoot out of my eyes. A tailwheel bolt should not seem to be exactly rocket science and I would tend to agree that while a torque wrench is needed on rod, prop and other parts there are many areas where "tight enough" "should be "good enough". If not, maybe a poor design ?

That said, I think specific design is more complicated than it seems. I've had issues with the tailwheel getting loose on my homebuilt which "should" be little different than a cub. I found it seemed to be soft washers in my case. Rather than trust another lot and have to wait I made one out of 4340 and hardened/tempered the darn thing. Has stayed tight for 15 hours now where before I found it loose in 5-10 hr.

My basic issue is too high tail/short gear, maybe dragging in tail first is placing more load(tension) than I would think on smooth grass.(I do intend to get a straight spring and different angle tailwheel to solve that)

FWIW

Jack

[sierra bravo]

The engineer types are all focused on the bolt. I still want somebody to explain how two arched leaf springs through-bolted to a tailwheel and pinched together with a torqued bolt can work well. In the Cub tail assembly the top spring would have to slip but the nature of the application is to impede any slippage/movement. Broken bolts are a symptom, not the problem. That the eye bolts are breaking may be another issue but the design flaw remains. There's a potential Cub mod for you experimental guys to explore.

[Steve Pierce]

This can be beat to death but if properly maintained, bolt kept tight, I have seen no real issues. Springs loose their arch and tail wheel mechanisms wear but it is a mechanical thing that gets the most abuse of anything on the airplane in my opinion.

[8GCBC]

Flew 20 hours of light landing and the TW loosen uped. I guess that is abuse. Scary stuff Brah.

[Steve Pierce]

What was the bolt torqued to?