Broken Tail wheel bolt

[8GCBC]

What was the bolt torqued to?

Good question, Steve. The IA who helped with the STC's installed the TW. It may have not been installed right. The IA lost his license the next year. He also screwed my brand new logbooks up! The whole reason I bought a new airplane was to get fresh clean logs! Oh, well. The grass is always greener...

[Patrol Guy]

Think that I heard some snickering over in the 172 section.

[8GCBC]

Think that I heard some snickering over in the 172 section.

I hope they all have Lord shimmy dampeners.

[cgoldy]

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.

I have been wondering the same thing for 18 months now. Concisely put SB

[8GCBC]

A complete re-work in a stronger lighter config is welcome here. Another thought...equal ABW-TWLG surface area weighs more than my 8GCBC-ALUM-MLG, not good.

Screwing around with a bolt does NOT help the design, just more cheap talk. Similiar to when Mike Busch says an engine with no rings develops the same HP? Glad he does not do diesels.

[Steve Pierce]

I don't throw anything away. Anyone who has been through my three hangers can attest to that. I went and grabbed about 10 tail springs after I ate my peanut butter and jelly sandwiches today. Noneof them showed signs of sliding where they bolt to the tail wheel. Actually they have paint transfer from one spring to the other from being clamped together. I can see where they where moving in the middle of the spring as in bending the spring but no sliding at the bolt holes, some actually have rust there where the slot is. Looking at them and the aluminum head and spring adapter I don't believe there is any movement there.

[sierra bravo]

Heck, they may as well weld the two leaves together then, right?

[Steve Pierce]

They bend in the middle. Two springs probably have better bending properties than one thicker spring I would assume.

[highroads]

Time to throw in my 2 cents worth. I believe the failures shown thus far indicate the peak stress is occuring due to a torsional load at the wheel assy attach point, causing a bending moment on the bolt. Contrary to first intuition, the tail wheel is not just an up and down motion but is also experiencing side load and torsional loads which probably induce the greater stress component on the bolt. Having said all that, it is intuitve to me that a loose bolt could cause increased stresses due to more allowable motion and related higher attach point torsional loads countered only by bolt bending. I am a little anxous on why a preliminary metalurgy report hasn't been produced as a simple hardness test and a 10X glass would suffice for most of the findings.

[highroads]

Just a side note, I don't believe the head to shank grain flow requirements of high strength bolts could be duplicated in this eye bolt design. This would result in a lower load capability in the eyebolt.

[cgoldy]

When I assembled my tail wheel for the first time I asked around if i should lube the spring leaves. I was told not to.

[nanook]

I believe the most common cause of tw attach bolt failure is hitting an object like a rock or an abrupt edge like a hole. The spring is not causing the failure in this type of shear force. The bolt can only take so many of these shots before it fails. The eyebolt that failed appears to have been cracked by a previous shear shot...bigger/wider and softer tw tires help alleviate these types of bolt failures by bouncing over and or absorbing these shots. Get rid of your small/narrow and hard tw tire and make sure you put a large area washer under the nut on the spring side....

[cubdriver2]

When I assembled my tail wheel for the first time I asked around if i should lube the spring leaves. I was told not to.

The hotrod guy's ( 32 Ford ) use a thin Teflon tape between the leafs to smooth out the ride.

Glenn

[mike mcs repair]

Kinda what I was thinking... How is normal AN bolt head made.. Forged also? Or???

Just a side note, I don't believe the head to shank grain flow requirements of high strength bolts could be duplicated in this eye bolt design. This would result in a lower load capability in the eyebolt.

[8GCBC]

cleat.jpgrodend.pngsamson.jpgswages.jpg
sch-64187.jpg
An eyebolt would never be used on my sailboat. Cleats, bulwarks, samson posts, chainplates, and linear rod ends/swages only.

P1010012.JPG (dropped my hat)

[sierra bravo]

Time to throw in my 2 cents worth. I believe the failures shown thus far indicate the peak stress is occuring due to a torsional load at the wheel assy attach point, causing a bending moment on the bolt. Contrary to first intuition, the tail wheel is not just an up and down motion but is also experiencing side load and torsional loads which probably induce the greater stress component on the bolt. Having said all that, it is intuitve to me that a loose bolt could cause increased stresses due to more allowable motion and related higher attach point torsional loads countered only by bolt bending. I am a little anxous on why a preliminary metalurgy report hasn't been produced as a simple hardness test and a 10X glass would suffice for most of the findings.

I'll float a different theory. Since we have two arched springs in a stack each component leaf has a different radius and therefore are different lengths. When the arc is flattened the length of the assembly is limited by the lower (shorter) spring. While it may flatten the upper spring can't. Since it's longer it must maintain an arch and that arch is not the same as what it was bent to since the two ends are bolted tight and those contact surfaces aren't available to bend. That's asking a lot of the upper spring when it's forced to flatten. At that moment the tensile load on the bolt must be incredible. And the load wouldn't be transmitted in a perpendicular plane to the upper spring since the resistance to the arch is greater at the bolted end than in the middle of the spring's length. If the eye bolts aren't equal in tensile strength to the typical AN7 bolts you'd likely see some stretching of the bolt, and coincidental to that stretch the torque on the nut would decrease. That exact thing has been described by multiple eye bolt users. The AN7 bolts have worked fairly well for a long time so it stands to reason an eye bolt of equal strength should work equally well. I'd predict the eye bolts aren't up to the AN standard. No matter, the design is not very good. Pinning both ends of a stacked, arched leaf spring is a poor design.

Hits to the tailwheel are transmitted to the spring, which in turn absorbs the shock by flexing, or flattening the arch. Don't put the cart before the horse. I'd submit the spring assembly is the problem.

This has been a fun brain teaser.

[mike mcs repair]

.... When the arc is flattened the length of the assembly is limited by the lower (shorter) spring. While it may flatten the upper spring can't. Since it's longer it must maintain an arch and that arch is not the same as what it was bent to. At that moment the tensile load on the bolt must be incredible. And the load wouldn't be transmitted in a perpendicular plane to the upper spring since the resistance to the arch is greater at the bolted end than in the middle of the spring's length. .....

owwww!!!

you are making my head hurt!!!

[sierra bravo]

Wanna make it hurt more? Our favorite oversized tailwheels are heavier than the originals so the unsprung weight needs to be factored into the whole discussion. Add more weight on the unsprung end and the dynamics of the assembly changes.

I should have stayed in school. Fun discussion just the same.

[8GCBC]

This is how I look at it:

I am running before the wind at 10 kts in 10-12 foot seas and there are two places to clip in my safety harness as I leave the cabin to reef the main...

First choice is a one piece forged pad eye which has a breaking strength of about 6K# held to the deck with 4 bolts and a stainless steel backing plate...
padeye.jpgeyebolt.jpg
...Or I could clip into an eye bolt which by design could shear with as little as .25 tensile strength.

Darwin says take the pad eye with the four bolt pattern, not the eye bolt.

(pictures courtesy of Schaefer Marine and Home Depot)

[Steve Pierce]

Stewart, It looks like to me the eye bolts broke the head off. No evidence of stretching. That is why I think some of the bolt's are hard (brittle). Testing will tell. I hear Otter's are hard on the tail wheel assembly also but they have a whole different arrangement. Come to think of it it is an issue on the C180 as well.

[Darrel Starr]

According to Bill at Bushwheels, all eye bolts are from one lot.... So why not now publish metallurgical tests from a Sample in stock so all of the $40 bolt customers get some real info instead of speculations or alternatively, send out a Service Bulletin to scrap the bolts and give those who bought them a $40 credit at Bushwheels? Not sure why they aren't moving to stomp out this fire!

[behindpropellers]

Wup...

Did AFC make those for you???

[Grant]

Nobody made anyone buy these bolts for their airplane. The idea is solid. The issue is with one lot. For anyone to suggest that Bushwheels is acting in a manner that is less than honorable is disappointing. Bushwheels has supported this community over and above for many years. I dont think they are sweeping this under the rug. Just give them time to do the necessary research and get everyone an accurate answer. In the meantime put your basic AN bolt in and wait for the solution. Stop complaining that you spent 40 buck on a bolt. It was cool when you bought it, so be cool and wait for the solution.

[Darrel Starr]

I just looked at the eye bolt we took out of Doctor Randy's SC yesterday. Looking this over again, I see that the transition radius is much larger on the eyebolt than it is on an AN bolt. So large, in fact, that the AN960 washer binds up on the radius and will not seat properly. Here are some photos to try to illustrate this issue:
Showing the different size radius on the eye bolt vs the AN7-23 bolt.
P1170270.JPG

Trying to show here how the washer hangs up on the eye bolt radius and doesn't seat properly.
My eye bolt - washer not seated.
P1170267.JPG
Randy's eye bolt - washer not seated.
P1170263.JPG
I can visualize how this interference in the radius can lead to stress concentrations, bolt loosening, torsion loads etc. Looks significant to me.

[Steve Pierce]

AN washers are soft. I don't use a washer under the head of the bolt. I do use the Grade 8 washers under the nut because of previous experiences with the AN washers cupping and then allowing the bolt to loosen.

[Darrel Starr]

More pictures of Randy's eye bolt showing tool marks and washer grooving in the radius and some deformed threads and bolt shank where the bottom spring contacted the bolt. We forgot to check the torque before we took this one apart.
P1170246.JPG
P1170259.JPG
P1170259a.jpg

[Steve Pierce]

Picture from the bottom.
102_5924.JPG

[hottshot]

Sorry guys I have been tied up with things the last two weeks pretty heavy and with Bill on the road he is not able to get info either, so with out info I /we are not going to give you speculation until we know what it is for sure

The material is off to the metallurgical and they are in audit so we will not see results for a few more days at least, at this time if you pull your eyebolt off and send it in we will give you a $40 credit or refund on the part your choice. Give me a call or let me know and we will get you taken care of.

[Darrel Starr]

I believe the AN960 washers are matched to other AN hardware so that if they are subjected to design loads, they will not yield. I could be wrong but can't think why AN hardware would not be matched. If the washers yield, I think that indicates that the joint is being subjected to loads higher than intended. So, if that logic holds, yielded AN washers should set off alarms that the joint is being overloaded.

[8GCBC]

Could we extend the "shoe" (mating surface) up the spring a little more and use (2) bolts? I.e. manufacture the TW "shoe" a couple of "inchs" longer, giving enough room for a second bolt?

[Steve Pierce]

Darrel, I am not redesigning the tailwheel, spring etc. I believe we are using these in a lot more abusive environment than Piper ever intended. The use of the Grade 8 washers has served me well after finding the tail wheel spring attachment washers and tail wheel assembly washers cupped. I find main gear bolts bent also after years of abuse. Like the tuner knob on my Dad's old TV, mechanical stuff wears out. Now I know why he made us look in the TV guide before we flipped through the 4 channels we got when I was a kids.

[sierra bravo]

Go back the the post where someone suggested enlarging the hole through the spring to use a AN8 bolt. While the AN8 is 15% larger in diameter than the AN7 the strength ratings are 30% greater. That idea has merit.

[8GCBC]

Respectfully...

If we are using them in an environment that they were not intended, time for a rework?

Just like TV's have changed.

[8GCBC]

Go back the the post where someone suggested enlarging the hole through the spring to use a AN8 bolt. While the AN8 is 15% larger in diameter than the AN7 the strength ratings are 30% greater. That idea has merit.

Seems reasonable too.

[sjohnson]

Respectfully... If we are using them in an environment that they were not intended, time for a rework?

Would it work to replace the grade 8 washer that Steve Pierce uses with similarly hard bracket that extends the loop to the side of the spring/tailwheel assembly? We use the original bolt, replace an existing washer, and move the loop so that it won't damage the rudder if the springs flex a lot or the fuselage/spring bolt breaks.

[Steve Pierce]

I have not had any problems with a AN7 bolt, grade 8 washer and the AK Bushwheel tail spring. Not worth trying to drill a hard spring and possibly inducing a stress riser in the spring to me. I modify if I can't solve the problem through proper maintenance. I deal with this stuff everyday, not just on my own airplane. Modify if you like but not worth the effort to me. Remember that everything you modify usually effects at least three other things and it is not all obvious at first. I quit needing the practice a long time ago.

[Steve Pierce]

Would it work to replace the grade 8 washer that Steve Pierce uses with similarly hard bracket that extends the loop to the side of the spring/tailwheel assembly? We use the original bolt, replace an existing washer, and move the loop so that it won't damage the rudder if the springs flex a lot or the fuselage/spring bolt breaks.

I think that could be done very easily.

[sierra bravo]

I have not had any problems with a AN7 bolt, grade 8 washer and the AK Bushwheel tail spring. Not worth trying to drill a hard spring and possibly inducing a stress riser in the spring to me. I modify if I can't solve the problem through proper maintenance. I deal with this stuff everyday, not just on my own airplane. Modify if you like but not worth the effort to me. Remember that everything you modify usually effects at least three other things and it is not all obvious at first. I quit needing the practice a long time ago.

Water jet. Easy fix. Makes no never mind to me. I tie around the spring. Don't need an eye. If I wanted one it would be exactly like spinner illustrated early in this thread. Simple. Cheap. Strong.

[Steve Pierce]

What will it cost to water jet a tail spring? Set up would be a pain from my experience with water jet. They are used to cutting out of a solid sheet, usually 4'x4' at my local machine shop, not a 1 3/4" part. $100 an hour minimum charge here. I could have it done on the mill easy enough but it takes the machinist's time, my time to take it by etc. Like I said, I have no problems with the tailwheels that I have maintained for years, they get looked at and proper maintenance at least yearly. Maybe someone needs to make a new tailspring with a 1/2" hole in it. As long as it took me to get someone to build one with the proper arch and temper I doubt that will happen anytime soon.

[8GCBC]

If drilling an oversize hole is desired...

Drilling usually anneals and if the hole has a slight chamfered edge and rounded corners it should not provide a stress riser.