The Teeter Bolt and NDT

[Bill Clem]

The recent posts regarding fractured teeter bolts brings up an issue that I have had to address. During a recent annual inspection I removed and inspected my favorite teeter bolt, after more than 500 faithful hours. It looked a bit worn, at least the coating, but otherwise looked okay. It was a Grade 8 bolt and I thought that it was time to replace it but when I tried to buy a similar bolt I found that the bolt had been cut down to 5 inches from a larger size, so the grip (the nonthreaded portion) was sized for the rotorhead and bearings. The option was to replace it with a 3/8" AN bolt. The expert at the aircraft hardware store assured me that AN bolts and Grade 8 bolts were identical. This assurance began a chain of events which led to this post. The two bolt types are very different. AN bolts are MIL spec bolts ("Army-Navy") that have specific material, construction and inspection criteria. The minimum tensile strength is 125,000 psi. Grade 8 bolts are commercial bolts, also designed as high strength bolts, but are not held to the same construction and inspection standards. The ASTM standard is for a minimum of 150,000 psi tensile strength. There is a higher commercial bolt, L9, which is rated to 180,000 psi. So, the question arose, do I use the higher rated Grade 8 bolt or the higher spec'ed AN bolt?

The answer requires a bit of math. The psi tensile strength must be recalculated for the size bolt used. In this case, a 3/8" bolt has a cross-sectional area of 0.1104 sq in. So a 3/8" AN bolt has a rated tensile strength of 13,800 lb (125,000 x 0.1104). And a 3/8" Grade 8 bolt has a rated tensile strength of 16,560 lb. Plenty strong, right? Well, not so fast. Tensile strength is the force needed to induce failure along the long axis of the bolt, along the length of the bolt. Most forces in aircraft are applied across bolts, referred to as a shear force. And the shear strength of a bolt is usually about 60% of the tensile strength. So a 3/8" AN bolt has a shear strength of about 8280 lb and a Grade 8 bolt about 9936 lb.

How much force can a gyro develop? Chuck Beaty and I once had a discussion and decided that in almost all cases the maximum G force a gyro can generate is about 2 G's. My gyro, with me and 48 gallons of fuel would weigh a little less than 1000 lb. So I need to figure about 2000 lb.

In many things I do, I like a 5:1 safety factor. And the above calculations do not figure in wear and tear. So the numbers weren't looking too good for the AN bolt. But wait!!! The 60% shear force calculation is based on a SINGLE POINT shear. And the rotor head generates a double point shear, the force is applied simultaneously to two places on the bolt. In short the G forces are split between two points. So the safety factor for the AN bolt jumps from 4:1 to 8:1. Much better.

But how do I know that the bolt is safe? NDT stands for Non Destructive Testing. It can be quite complicated and can involve many different techniques including Xray, Ultrasound, Dye Penetrant and Magnetism. But for most purposes Dye Penetrant and Magnetic testing are used. Dye testing involves cleaning the bolt, for this discussion, and soaking the bolt in a penetrant which will be absorbed into any small cracks by capillary action. The residual penetrant is removed and a developer is placed on the bolt which draws out the penetrant from the crack, creating a pattern that is wider than the crack so it can be seen by the human eye. Visual dye systems such as Magnaflux's Spotcheck are cheap (about $28 for the Spotcheck Jr.) and can be used easily. Fluorescent UV dye systems such as Magnaflux's Zyglo system use a black light and a fluorescent dye which is more easily seen and therefore is more sensitive to small cracks. Both systems only work on surface cracks. There is a more complex method involving magnetism if the material tested is ferrous, ie. made of iron, like our bolt. The part is magnetized and sprayed or soaked in small particles. If there is a crack, either at the surface or even just below, then the magnetic field in the bolt "jumps" over the defect causing a clumping of particles at the surface, which can be seen. The main advantage to this system is that subsurface flaws can be seen.

So, what am I using? Since the magnetic system requires a huge investment in the yokes or coils used to create a magnetic field, I have opted for the UV system. I only have to pay for the chemicals and the black light. It will find very small cracks and allow me to sleep better when I'm flying.

I am using an AN bolt at this time. I figure that the trade of 16% shear strength is made up by better documentation and testing of AN bolts. And well within the safety margins that I've set. One other factor plays into the equation. AN bolts tend to bend and elongate prior to fracture. Grade 8 bolts tend to fracture quickly. I am much more likely to pick up deformation on inspection.

What about that 500 hour teeter bolt? Stay tuned.

[thallett]

Thanks, Bill. I,m on the edge of my seat waiting.............
Tinkerin Tom in Port Orchard WA ::)

[ToddP]

I know that single bolt has caused me more concern than any other. Your research is appreciated. I'll be looking forward to your test results.

[quadrirotor]

Did you think at X-ray imaging?
What do you think of stainless bolts?

[Doug Riley]

Of course there are lies, there are damned lies and then there are statistics. Nevertheless, I like the statistic that, in the nearly 35 years I've been hanging around gyros, I have never heard of a 3/8" teeter bolt failing.

In that time, it's almost certain that somebody somewhere has used a bolt that is not an AN 125 KPSI bolt, but rather a 55 KPSI junk bolt. There are many. many fake AN bolts out there, including those sold by big mailorder houses such as Wicks.

My 1985 Air Command gyro was assembled mostly with suspect AN bolts. These bolts looked nice, but had an "X" with no manufacturer's I.D. on the head. Any bolt lacking the manufacturer's I.D. and/or which has the "X" stamped into the head rather than raised, is suspect.

The spindles on the old Bensen spindle head used to fail once in awhile, with deadly results. I had mine Zygloed a couple times and it passed. The failures may have been the result of the changes in section along the length of the spindle. A smooth radius at each shoulder was necessary to prevent cracking. This type of part is much better if forged, rather than cut out of simple round stock, but ours were merely cut on a lathe.

Gimbal head spindles are simple bolts, with no changes of section, so that danger is gone.

[Chuck Irby]

Doug, I really appreciate your post. However, you're scaring the hell out of me. The bolt on the left is the one I am currently using. The one on the right is a new "spare". I can't remember for sure, but I either got the new one from Aircraft Spruce or maybe Wicks. Do they both appear to be authentic?

[Harry_S.]

Chuck;

My opinion--- both are good bolts.

[Chuck Irby]

Thanks Harry. For several years I was an Inspector and Inspection Supervisor during the construction of a Nuclear Power Plant. So I feel that I am pretty well qualified to determine the integrity of a piece of metal. But this is the first time I had ever heard of counterfeit bolts.

[Udi]

I went through a similar process with my spindle (Jesus) bolt. I am completely rebuilding my new-old Air Command, and since I don't know the history of this aircraft, I am replacing all the critical parts, as well as parts which are not critical, but are worn beyond my level of comfort.

Not so long ago, there was a discussion on Norm's forum regarding AN vs Grade 8 bolts. The general consensus was that AN bolts are not really better than Grade 8 bolts because many AN bolts are fake, possibly made in a cave in Azerbaijan. AN bolts are supposed to be manufactured as per military specification but since the military is not using these bolts any longer, the bolts and their manufacturers are not being audited for quality by anyone.

Needless to say this scared the crap out of me, so I started a small investigation. I called Wicks and asked them to fax me a certificate of origin for this bolt (they did, without a problem). Then I found an AN Bolt manufacturer database on the FAA web site and, in that database, I found the manufacturer that made my bolt (according to the stamped markings).

Still suspicious of cracks, I took my bolt last week to an aircraft engine shop to get it magnafluxed. They were nice enough to check it for free. The guy who checked my bolt was surprised that I want to check a new AN bolt because, "all AN bolts are checked for cracks by the manufacturer". Anyway, he said the bolt looked great.

The moral of this story - in my experience, the AN bolt industry is not as loose as some may believe. Buying an AN bolt still assures tighter tolerances and a higher level of inspection than hardware store bolts (whatever grade they are). When you order bolts (or any other part for that matter), make sure to ask for a Certificate of Origin (C of A). It's free. You may look up all you’re AN bolt markings in the FAA AN bolt manufacturer database. I have posted the URL on the old forum, but I can look it up and post it here if anyone is interested.

Bill, I got my bolt magnafluxed at Firewall Forward at the Fort Collins-Loveland airfield, in case you ever want to mangaflux a part.

Udi-

[Chuck Irby]

Udi,

I appreciate the info you just posted, and yes please post the URL for the FAA AN bolt manufacturer database.

I feel that this will be of extreme importance to most of us here on the forum.

Thanks again,

Chuck

[Bill Clem]

Udi, the term "magnafluxed" is tossed around a bit without knowing exactly what it means. It has confused me more than once. "Magnaflux" is a corporate name for the company that first came up with this sort of inspection. Magnaflux (the company) sells visual dye penetrant systems, fluorescent dye penetrant systems and the MPI or magnetic particle inspection systems. The latter only works on ferrous metals, the first two work on all metals and plastics, glass and ceramics. TriCounty Propellers also does inspection but at about $50 a pop. Which of the methods did they use on your bolt? I don't know enough about the subject and the type of failure to expect to know for certain that one method is better than another. The Zyglo system kit is expensive, about $620 for 4 cans of chemicals and a $384 black light. I opted to buy the chemicals individually for about $35 and to buy a separate black light for about $100. I think that this is an issue that all of us have thought about but didn't have a reasonable way of inspecting. If we get a bunch of people out there checking bolts with either the Spotcheck kit ($28) or the Zyglo kit then we might get more info. Is it a problem? Don't know. But then we didn't think that repeated pressurizations caused a problem in an aircraft cabin until the Aloha flight popped.

[Udi]

Bill - they've used a dry magnetic particles process. "Magnaflux" is also a trade name for this specific process, which is owned by the Magnaflux Company... Kind of like New York, New York...

Is it a problem? Who knows? I personally have not heard of any cracked AN bolt. My local FBO aircraft maintenance shop (Gerald Gates) is having complete confidence in them. If they are using AN bolts in certified aircraft, who are we to doubt them?

On the other hand, inspecting a critical part, which has no redundancy, is cheap life insurance… I don’t want to meet you in your place of work… :o

Udi-

[Chuck Irby]

Bill is correct in that Magnaflux is a Brand name. The term is used by most people for the "Magnetic Particle" inspection process (MT). Also, like Bill stated, this inspection process can only be done on ferrous metals.

It can actually be done with a strong permanent magnet and steel dust that you may find under your grinding wheel or belt sander after you have been grinding on a piece of steel. Place your magnet on the bolt and sprinkle it with the steel powder. If there is a crack, the dust will accumulate along the crack.

Bill, we are glad you have joined us here on the forum. Thank you, and WELCOME!!

[Doug Riley]

Chuck Irby: "CS" AN bolts are about as genuine as they get. I believe the initials stand for Cleveland Screw.

The ones without any initials are the ones to worry about.

[Aussie_Paul]

I believe that the spindle/bearing bolt is a bigger concern than the teeter bolt. It is in tension, and a lot of them are threaded into a steel insert in the torque tube. These ones are usually 5/8".

In tension there is no redundancy and that bolt is subject to a lot of shake induced bending pressures.

Once again they don't fail, although Mceagle quoted our one and only failure here in Oz.

I never give a second thought to the teeter bolt other than normal maintenance. A 3/8" bolt in double shear is a fine way to do it, BUT you don't want the head to come off and the teeter bolt to work out. We only tension the teeter bolt to approx 13 foot lbs so it is unlikely to "pop the head".

Just my thoughts. Aussie Paul.

[mceagle]

Udi, - re your statement

"Buying an AN bolt still assures tighter tolerances and a higher level of inspection than hardware store bolts (whatever grade they are)."

As I understood it quality inspections on AN bolts no longer exists any better than it does on hardware bolts and the metallurgy compliance tests are no longer undertaken. I hope someone can prove me wrong on this.
As far as accuracy is concerned, a standard AN bolt is inferior to our "hardware" Ajax brand bolt. When selecting an AN 6-56 Teeter bolt I have to roll them along something flat to pick out a straight one. Also the bolt is thicker at the shank just below the thread, meaning that you have to slightly oversize the hole to fit the bolt.
However, I still use AN bolts, despite the sometimes rediculously high prices.

[Udi]

Here is the web site I promised:

http://www.uspto.gov/web/offices/tac/fqa/active.pdf

This is the US Patent and Trademark Office, where bolt manufacturers have their insignia recorded.

Udi-

[Chuck Irby]

Udi,

Thank you for that website address. I stored that info in a good place.

I went to the site, and was amazed at how many manufacturers there are worldwide. If I read the info correctly, the two bolts I pictured above were manufactured by Wurth/Service Supply in Indianapolis. Is that a proper assessment Udi? (I found it on page 11)

[scottessex]

How about NAS bolts? That is what Boeing uses on most of thier stuff.
Also just an FYI all AN and NAS bolts have rolled threads, not cut threads like comercial bolts. The rolled threads maintain the grain of the material and are much stronger that cut threads.

[Doug Riley]

Scott, NAS bolts are fine. They are stronger than AN bolts. They're also a pain to work with, however, because they have extremely short threads.

Properly-made Grade 8 bolts have rolled threads, too. A rolled thread is easy to spot: at the point just above the beginning of the thread, the shank of the bolt tapers down below the nominal diameter, and then below that point the material has obviously been "squished" back up to the nominal diameter. Cut threads look just like the threads you make with a hand die... because that's what they are.

[Udi]

Chuck - I couldn't find your bolt insignia in the list. Who did you buy them from? You can call your supplier and ask who makes these “CS” bolts. Then you can go back to the list and search for the manufacturer. I looked for a "Cleveland Screw", as Doug has suggested, but could not find them.

Udi

[Doug Riley]

Sorry, wrong company. Cleveland Screw is a screw-machine outfit. The "CS" trademark belongs to California Screw, www.calscrew.com.

[Aussie_Paul]

My interpretation was that AN bolts were the equivilent of grade 5 but with closer tolerances, inspection processes, rolled threads and cad plated.

I thought that grade 8 are more designed for tension aplications and grade 5 and AN more for shear applications.

Do I have the wrong understanding?

Aussie Paul.

[raghu]

But then we didn't think that repeated pressurizations caused a problem in an aircraft cabin until the Aloha flight popped.


I realize this is a wee bit pedantic but just in case someone is interested the danger of repeated cabin pressurization in fact pre dates the aloha incident by decades. The credit goes to the deHaviland Comet ( the first jet passenger aircraft) crashes ( i believe three of them where lost) in the fifties due to fatigue caused by repeated pressurization. The incident was a watershed in tackling issues of fatigue in airframe design. The folks at ALOHA had no excuses.

[Edwin Doetzel]

A couple quick points.

An AN6 bolt is actually rated for only 10100 lb. in tension. The bolt will fail in the threads and not the cross-section. Tension loadings should be avoided where ever possible. An ideal joint is in double shear with a minimum of 4 fasteners with the load passing directly through the fastener pattern. The cited value for an AN6 bolt in single shear at 8280 lb. is correct (source: Bruhn Table D1.1). The 60% derating for shear however is an innappropriate rule of thumb imo. I would recommend the homebuilder stick to published data on hardware rather than making calculations from the material properties. In fact, this is what engineers almost always do. You still need to check the lug material for failure in bearing and in lug tear out. Also, make sure the joint is tight (no large gaps) as gaps create bending loads. Consult Bruhn "Analysis and Design of Flight Vehicle Structures" section D for more detail.

Also, my understanding is that while cyclic pressurization played a role in the Aloha incident the larger problem was incorrectly installed fasteners. Engineers specified rivet diameters too large for the sheet material. Drilling for countersinking went through the thickness of the first sheet and into the second sheet. This created a "knife edge" on the outer sheet, producing a stress concentration in the rivet and thus greatly reduced the fatigue life.

[Udi]

I had a serious problem torquing my new spindle bolt. The bolt is AN8 working in tension (holding the whole ship), and John at Air Command recommended to torque it to 59 ft-lb. I have consulted an AN bolt torque table and the maximum torque specified for this size bolt is 57.5 ft-lb, and I decided to use 57.5.

My problem was that the nut is recessed inside the torque tube, and no socket or wrench would fit in. I tried to torque the bolt from the bolt head side, while holding a screw driver in the gap between the nut and the torque tube, but that didn't work. The screwdriver only managed to make some markings on the torque tube itself. I hope these markings don't affect the integrity of the torque tube. Any comments?

Eventually I added 2 more washers (a total of 3) underneath the nut, so the nut would protrude just enough for the socket to grab it. I was able to torque the bolt to 57.5, but then the cotter pin hole was too low to stick the pin through it. I have ordered a new bolt, one size longer, and I think that will work out fine.

My only problem now is that I don't know if I have damaged the torque tube when I was trying to grab the nut with a screwdriver... The markings look superficial, but I don't know whether the forces that I applied with the screwdriver affected the integrity of the aluminum bar.

Udi-

[Doug Riley]

Udi, what brand of rotor head do you have? Normally, the torque bar is counter-bored around the nut to allow a standard socket to reach in.

Beware of over-long spindle bolts. On some head designs, if the bolt is a little too long, the bottom end of it will scrape on the inside surface of the "U" block. No use stacking washers in that case.

[rehler]

Udi,

Many people have this problem. I took my socket (socket wrench) and put it in a lathe, held a file to it and turned it down so it was not as thick (smaller diameter). It worked very well. I'd rather do that than use a longer bolt and washers.

If you don't have a lathe you might figure a way to attach it to an electric drill, spin it and hold the file up to it as it spins. Or, use a belt sander to sand it down to a smaller diameter.

[Udi]

See my rotor head below. I assume this is an original Air Command. I have checked for clearance, Doug, but I will check again for sure.

Thanks for the great idea, Ken, I will buy a new socket and try to reduce it's OD. I better buy a cheapo (maybe a Walmart band...) - good quality sockets are impossible to file.

Do you think I should be concerned about damage to the tube?

Udi-

[rehler]

Don't know about "damage" to the torque tube, since I don't know how bad it is. But, the touqe tube is only as big as it is to give room for the bolts (vertical and horizontal) to cross and still have "meat" around them.

The rest of the tube (not around the bolts) can be chopped up a lot without any problem, except looks. Many people drill holes through it to make it lighter, about 3/4" diameter at 1" or more on center horizontally (left to right).

The only stress on the torque tube is around the bolts, so if you did something to weaken this area then it may be a concern.

[RHerron]

Bill and all,

I can understand having some concerns about the strength of the rotor head hardware.
However, you couldn't get a 3/8" "stove" bolt to crack in a teeter-bolt aplication. It might shear under exteme loading (not possible in-flight) but it would not crack. That requires cycles of repeated bending stress. A teeter bolt doesn't experience this kind of stress.
When this bolt is torqued, there is hardly any area for shearing to take place. All the material is "stacked up" with no space in between. This, plus it is in double-shear as already mentioned.
I have never even seen one shear after blades have been trashed in an accident.

It is the very least of you worries.

Bill, if you want me to do a magnetic particle test on yours
just send it over and I will, no charge, no problem.

Ron

[Bill Clem]

I agree Ron. I would characterize it as a curiosity rather than a concern. But I know that all of us have had thoughts about that bolt, or others like it, while motoring along. My intent was to outline the process and get people to think about forces, rated strengths and adding NDT to annual inspections, as I have. Who knows what might be found? Remember the welded aluminum stick? As an A&P mechanic and teacher the methods needed for inspection are drilled into you. As a repairman, I have only hands-on experience to guide me. Do I think that the bolt will fail? Nope. Will I run the bolt through my Zyglo kit? Yep. And probably the engine mount bolts as well. And I still replace the control push rod bolts periodically. I guess because we don't have enough consistency in design to have a large enough database to understand what might be a problem and what needs to be looked at more closely.

You have an MPI setup so you probably inspect some things regularly. Just piston rods? Anything structural? Have you found any surprises? Have you found anything by MPI that wouldn't be picked up by dye penetrant?

Thanks, Bill. I'll keep your MPI offer in mind.

[Doug Riley]

Tops on my list of items to check is the pushrod bearings. I had the ones that AEROTEC used to sell Magnafluxed before sale at about $.50 apiece. None ever flunked. However, there was a crash (fortunately without injury) out west some years ago in which the banjo housing on a rod end failed. It was found to have an internal manufacturing crack.

[Bill Clem]

Isn't anybody still interested in the 500 hour bolt?

I did a dye penetrant test today on both the teeter bolt and the rotorhead pitch bolt. A drum roll please.....

Found a tiny porosity in one, just a tiny speck, but no cracks. So I guess I'll just slide that bad boy back in the teeter block for another 500 hours..... Naaaah.

But I was interested in the variety of opinions on a relatively esoteric subject. And I learned a few things. Like there is no general consensus about something as simple and common as a bolt.

I am reminded about an old Vince Lombardi story... the Packers were trailing at the half after a dismal performance. In the locker room, Lombardi ranted, raved and then suddenly stopped. Fuming and glaring, he picked up a football and turned to his players. "We'll start at the beginning..... This is a FOOTBALL!" From a corner, one of the interior linemen said softly, "Slower, Coach." Maybe somebody ought to say, "This is a bolt."

[Udi]

I have inspected the two bolts you are talking about, Bill, and I am more worried about the brass bushings that these bolts are rolling in, than the bolts themselves. Some non-steel parts in the gyro are more susceptible to ware than bolts. The gyro I am rebuilding has about 400 hrs TT. I have replaced many many parts - most of them are not bolts. I have replaced almost every "plastic" part, most bearings (incl. the spindle bearing), many corroded washers, and most of the locknuts. All the bolts look good, but I have replaced the critical ones, just for the heck of it.

You have said it before - we don't have enough statistical information to know which parts become fatigued and need replacement at X hrs. My favorite saying by Donald Rumsfeld:

As we know,
There are known knowns.
There are things we know we know.
We also know
There are known unknowns.
That is to say
We know there are some things
We do not know.
But there are also unknown unknowns,
The ones we don't know
We don't know
Happy flying.

Udi

[RHerron]

Hey Bill,

Yes, I'm still interested. I just lost this thread! Couldn't remember where it was.
Yes, I have found cracked hardware such as the Piper lift strut forks.
Mostly, I find crankshaft cracks, particularly in the small Continental engines.
I have seen the porosity you mentioned too. Usually in non-aircraft hardware like grade 5 and 8.
As far as rotor heads go, I have found brinnelling of the teeter bearing inner bushings in a couple of heads.
Ernie says that some of the early ones were too soft.

Ron

[Bill Clem]

Ron, I'll make you a deal.... I won't talk medical if you won't talk metal. Please define brinnelling. On inspection of my teeter bearing bushings, I found some small ridges from the roller bearings. Sort of like washboarding on a dirt road. I polished the bushings and decided that the indentations were a thousandth or less and figured that the bearing would accept a bit of asymmetry. I have had to replace the teeter bearings every couple of years or so, about every 200 hours. I'm convinced that some of the bearing damage was due to abuse during blade changes.

[RHerron]

Bill,

I'm sure you've looked up brinnelling by now. (hope I spelled it right)
What you have found (the washboard waviness) are the same thing I found.
On fully-articulated rotor heads where the individual blades are supported in tension by roller bearings (like the Hughes 269-300 modesl) this "furrowing" of the bushing is a deadly issue.
As an example, an indention of more than .002" is cause for rejection of the associated rotor blade!
In our application where there is only teetering action and the aircraft's loaded weight (no centrifugal loads) on the bearing, it probably only contributes to more shake as it takes more torque to freely flap the blade.
My "ridges" were not parallel by the way....sort of skewed as if the needles were twisted out of alignment and "stacking". I suspect that even if the needles were jammed, the blades would find a way to flap. It would induce quite a bit of vibration though, and potentially, some undue blade root stresses.
Trust me, I can't talk "medical".