Bell 407 in-flight tail boom separation

[Cyclic Hotline]

Not a good thing to experience!

https://www.ainonline.com/aviation-news/general-aviation/2022-11-19/bell-407-tail-boom-separation-prompts-faasteam-caution

GENERAL AVIATION (https://www.ainonline.com/aviation-news/general-aviation)Bell 407 Tail Boom Separation Prompts FAASTeam Cautionby Mark Huber (https://www.ainonline.com/mark-huber)
- November 19, 2022, 10:07 AM
https://www.ainonline.com/sites/ainonline.com/files/styles/ain30_fullwidth_large_2x/public/uploads/2022/11/screen_shot_2022-11-19_at_8.42.37_am_copy.jpg?itok=cFvaLfiQ&timestamp=1669044535 (https://www.ainonline.com/sites/ainonline.com/files/uploads/2022/11/screen_shot_2022-11-19_at_8.42.37_am_copy.jpg)
Bell 407 operators are advised to torque check tail boom attachment hardware following a June 8 in-flight separation accident. (Photo: National Transportation Safety Board)A June in-flight tail boom separation on a Bell 407 has prompted the FAA's FAASTeam to issue a special notice reminding all operators of the standing FAA Airworthiness Directive (AD) 2012-18-09 to torque-check the boom’s attachment hardware every 300 hours. The NTSB’s investigation into the Hawaii air tour accident that seriously injured three of the six aboard is ongoing.

On June 8, the helicopter—registration N402SH and operated by K&S Helicopters doing business as Paradise Helicopters—crashed into a lava rock field near Kalea. The tail boom was recovered about 762 feet to the northeast of the main wreckage site.

The 407’s tail boom is attached to the intermediate fuselage with bolts, washers, and nuts that connect the four aft fuselage longerons to four corresponding longerons located at the tail boom’s forward end. Post-accident the operator elected to replace the tail boom attachment hardware on its remaining five Bell 407s. Before hardware removal, the operator torque-checked the attachment hardware and found no evidence of loose hardware, but eddy-current nondestructive inspection on the aft fuselage longerons uncovered cracks on two of these five 407s.

Several bulletins and ADs have been issued since 2007 related to Bell 407 tail booms with regard to the aft fuselage top skin and bulkhead, replacement attachment hardware and torque checks, installation of external strap doublers on the upper-left longeron, and repair instructions for cracked longerons.

[SASless]

Bell seems to like the four bolt method of attaching Tail Booms....perhaps there are lessons to be learned yet.

We went through some major issues with UH-1's in the mid to late 1960's over Tail Boom separations in flight.

Perhaps the Bell Engineers should look at the BO-105/ BK-117/135/145 family for some ideas. Just saying!

[helispotter]

This relates to the same accident as the thread "Bell 407 Crash in Hawaii" started by KiwiNedNZ on 9 June 2022 with most recent post by SASless on 24 June 2022. By that time in that thread it had been reported the tail boom had separated and one attachment bolt was missing. But prompted by the comment by SASless above (#2) and the photo of the tail boom (#1), I started to look for more info on this case, for example: https://www.westhawaiitoday.com/2022/06/22/hawaii-news/ntsb-tour-helicopter-experienced-violent-upset-before-crash/
It was the upper left attachment bolt that was missing and apparently lower left bolt was showing signs of fatigue. But aside from the upper left bolt the remaining three were still in place. So it appears that having only one bolt fail is enough to cause such a catastrophic failure of the entire structure. As SASless has already implied, this would be a complete lack of redundancy.
Note that both the upper left and lower left attachment bolts would be those that are predominantly in additional tension (beyond the initial pretensioning) due to the tail rotor torque reaction. So if any of the bolts would have failed due to fatigue, those two were the most likely to go first.
While reading material related to 407 boom attachment, I also noticed initial manufacture and maintenance manuals for the 407 had the bolts tensioned (torqued) to beyond their recommended range. That probably had no influence (in the meantime) on this crash, but it is a worry that this could have happened to such a safety critical attachment. Reminds me of the other case I read here of the Bell 212 which crashed as it was fitted with pins attaching main rotors that had been manufactured using the wrong type of steel (from a Bell subcontractor) and which failed as a consequence, killing the pilot.

[megan]

We went through some major issues with UH-1's in the mid to late 1960's over Tail Boom separations in flightThat was kept quiet SAS, never ever heard of it, learn something new every day. Reasons for separation?

[SASless]

The Huey had a four bolt mounting method....tail boom butted up against the stub on the fuselage and pick your choice of reasons...weak bolts, improper torque of bolts, weak structure, oil canning...bottom line it was just a weak design. The cure was beefing up the structure and requiring closer inspections of the bolts and witness stripes for indications of losss of torque on the bolts.

As time progressed the problem moved on out the tail boom to the 42 degree gearbox mounting and after that was fixed....then it was the 90 degree gearbox mount. Part of those problems were vibrations.....the running joke was in time the bad vibe would get to the end of the tail stinger then we could just saw off the tail stinger and be rid of the vibe.

Memory serves me that Fort Rucker lost two aircraft in one day to tail boom separations.

As it was over Fifty years ago....those are my fuzzy recollections from back then.

Found this while looking for information re tail boom failures on the Huey.

Forestry Logging with surplus Huey's has resulted in a lot of crashes due to mechanical failures with many being tail boom losses.

That has led to a monitoring system of the Tail Boom that has been approved by the US FAA.

https://helihub.com/2016/03/15/faa-approves-uh-1h-tailboom-motion-detection-system/



It was about the time I was in Chinook School so it would have been in mid-to-late '68.

John Dixson would have been at Rucker about that time too I am thinking....perhaps he has a better account.

[helispotter]

I have found the NTSB report in the meantime:

https://www.ntsb.gov/investigations/AccidentReports/Reports/AIR2210.pdf

Part of it reports: "According to Bell (the helicopter manufacturer), the upper-left tail boom attachment fitting has the highest tension loading of the four attachment fittings and is considered the most important of the four attachment fittings between the tail boom and the aft fuselage. The upper-left tail boom attachment bolt separated first, resulting in the redistribution of the load normally carried by the upper-left attachment fitting to the remaining three attachment fittings and the subsequent initiation of fatigue fractures on the lower-left attachment fitting. Based on preliminary analysis of the fatigue crack growth on the lower-left attachment fitting, it is possible that the fatigue crack grew over multiple takeoffs and landings..."

So it isn't a case of the tailboom separating as soon as the first bolt parted, as I was initially thinking was the case.

[wrench1]

The weak link on most Bell helicopters has been the tailboom. The proverbial Achilles tendon. Didn't matter the model they would fix one issue then it would move to a different place and crack it there. Throw in inconsistent maintenance practices across the spectrum and it would surface with some unsuspecting operator. For example, the upper LH tailboom mount area on all 206L/407 models has been cracking for eons with the 407 improvements initially moving the crack forces to the upper and lower TB skins. Is it a bad design... I don't think so but its not very tolerant of bad or irregular mx. Same reason behind the Bell 24 month flight control bolt inspections and lube requirements. One operator didn't remove the majority of those bolts for over 10 years and one broke. Every helicopter model has some sort of weak link, some just get more press than others.

[helispotter]

megan: Prompted by your comment, I see SASless already posted a link re logging UH-1's that have lost tail booms, but I also came across this item on the topic, see "background" in this for some cases:

https://www.uh1ops.com/single-post/2019/02/19/saib-18-29r1-rotorcraft-tail-boom-attach-structure-failure

[megan]

Many thanks guys, obviously a well known weak link, information certainly never filtered down to we pilots in our military or civil when I was employed in both, spent a lot of time in the civil sheet metal shop between flights helping reskin 205 and 212 tailbooms, me on the gun and the expert on the bucking bar, did a bit on a 76 tail boom once as well. Good practice for when I and two of our engineers when on to build three RV's.

[helispotter]

The NTSB report indicates there was evidence the upper left attachment bolt had previously been attached but that this bolt wasn't found with the wreckage. I could imagine that the bolt head end (that must have been lying inside the tailboom?) could have been flung out as the boom fell to the ground, and so it would be hard to now find. But would the nut end of the apparently failed bolt not have fallen down within the fuselage? Or could that likewise have been ejected once the fuselage started to spin on loss of the tailboom?

[wrench1]

Or could that likewise have been ejected once the fuselage started to spin on loss of the tailboom?
Probably this scenario as they did not locate the bolt, nut, or washers. The docket gives more details and pics. Given the airframe tailboom ring is pulled forward and the upper L/H hardware missing, my guess would be that after the upper bolt failed in tension the remaining bolt shank captured the ring as the boom fell away. The fact they found exhaust traces between the faying surfaces of both mount rings at the upper L/H bolt hole would indicate to me it had been "loose" for a period time. The docket also mentioned the upper and lower R/H nuts were over torqued and the lower L/H nut under-torqued.

[RVDT]

As time progressed the problem moved on out the tail boom to the 42 degree gearbox mounting and after that was fixed....then it was the 90 degree gearbox mount. Part of those problems were vibrations.....the running joke was in time the bad vibe would get to the end of the tail stinger then we could just saw off the tail stinger and be rid of the vibe.

Similar thing with the first Agusta 109C's on introduction. Most of the frames inside the upper and lower vertical fin were found cracked and the remedy after repair was to hacksaw off the huge monstrosity of a stinger that was fitted.

[wrench1]

the remedy after repair was to hacksaw off the huge monstrosity of a stinger that was fitted.
And just the opposite they put a weight on the 407 tail skid to dampen vibrations. I always found it interesting that one could purposely move around vibrations within the airframe especially those with airframe mounted tuning weights like in the 350/355 or 135. A 4-per in one area and life is good, but move that same vib into the tailboom and rivets start to work or skin cracks.

[cavuman1]

When my family and I went down in a 206 in 1977 (discussed elsewhere on this forum), the main rotor was still carrying significant rpm as we autorotated to the beach. It severed the tail boom cleanly at the fuselage interface; that impact spun us 180 degrees. I clearly remember turning around as we ran from the crash site and seeing the fuselage separated from the tail boom by some twenty feet.

- Ed

[megan]

Wonder if the Huey tail boom attachment was helped by US Army folk jumping up on the tail stinger to give the tail rotor a close once over during preflight, 170 pounds on a fairly long arm. Not a practice I saw anywhere else.

[helispotter]

...Given the airframe tailboom ring is pulled forward and the upper L/H hardware missing, my guess would be that after the upper bolt failed in tension the remaining bolt shank captured the ring as the boom fell away. The fact they found exhaust traces between the faying surfaces of both mount rings at the upper L/H bolt hole would indicate to me it had been "loose" for a period time. The docket also mentioned the upper and lower R/H nuts were over torqued and the lower L/H nut under-torqued.

wrench1: Where did you read the observations that two nuts/bolts were over torqued and the other under torqued? I couldn't find this in the NTSB report I provided a link to. In any case, given the additional loading on the lower LHS bolt once the upper LHS had separated, it may well have deformed and so any post-accident torque check would no longer be representative of the torque this bolt had been set to when last checked, and it could be expected to then have a (much) lower torque. Also, the upper LHS bolt shank didn't exert any significant pull on fuselage side flange (if the shank was even still in place) as that was the one flange that still remained attached to the fuselage wreckage.

[helispotter]

Wonder if the Huey tail boom attachment was helped by US Army folk jumping up on the tail stinger to give the tail rotor a close once over during preflight...

Also makes me wonder how many times during the fast flared landings in Vietnam jungle (as seen in footage and photos) would have resulted in strike of the tail skid/stinger?

[SASless]

Remember the Civilian models, 205/212/412 have a baggage compartment just aft a bit from the attach point which had about a 400 Pound capacity.

The Arm of that compartment is far less than the base of the Tail Stinger for sure but does add to the load moment of the tail boom.

[wrench1]

Where did you read the observations that two nuts/bolts were over torqued and the other under torqued?
As mentioned, in the Public Docket linked here below. I believe it was in line item 2 report.
https://data.ntsb.gov/Docket?ProjectID=105233
it may well have deformed and so any post-accident torque check would no longer be representative of the torque this bolt had been set to when last checked,
Doubtful. The fuselage side longeron fittings failed due to the fitting bolt torque holding them in place. In my experience either the bolt deforms/breaks or the structure deforms/breaks. In this case the structure lost so same torque should be on those remaining bolts. I will add, torquing these bolts is a pain as you have both high torque and thread engagement requirements in a very tight area.
the upper LHS bolt shank didn't exert any significant pull on fuselage side flange (if the shank was even still in place) as that was the one flange that still remained attached to the fuselage wreckage
FYI: the fuselage tailboom mount flange was ripped out and is still bolted to the tailboom minus the upper LH bolt. The nuts face forward so the attached broken fittings are the fuselage side. The Docket has better pics but you can see this in the OP pic.

[helispotter]

wrench1: Thanks for the link to all those NTSB documents. They make a lot of information available to the public, which is impressive from an openness perspective. The discussion about the torque measurements on the remaining nuts/bolts following the accident is indeed in the "Materials Laboratory Factual Report 22-071" along with further detailed examination. About the torque of lower right nut/bolt being measured at 240 inch-lbs vs 360 inch-lbs required minimum, be patient with me: Isn't it possible the bolt would have plastically deformed (stretched) slightly in the process of the tail boom pulling the entire longeron attachment fitting from the longeron? If the bolt stretched in that way, then the remaining torque when tested would be less than what it had been prior to the failure simply because the bolt isn't pulling the nut up against the washers as much as it did beforehand. However in any case, NTSB doesn't seem to raise their torque readings as a direct issue*.

I commented on your comment "after the upperbolt failed in tension the remaining bolt shank captured the ring as the boom fell away" as I think you are proposing here that the broken part of the upper left bolt was still in place and tore away the aft fuselage bulkhead (or fuselage tailboom mount flange, or ring). But isn't it a case of the other three attachment bolts, still being intact, tearing off their respective fuselage longeron attachment fittings and in the process, these then peeled the fuselage bulkhead off the fuselage?

* I had an indirect involvement in bolt pre-tensioning calculations for a non-aviation application years back. From memory, both too much pre-tension and too little were not good from a bolt fatigue life perspective if the application was subject to cyclic loading. So there is a 'sweet spot' that the fasteners should be torqued to.

[Jetstream67]

wrench1: Isn't it possible the bolt would have plastically deformed (stretched) slightly in the process of the tail boom pulling the entire longeron attachment fitting from the longeron? If the bolt stretched in that way, then the remaining torque when tested would be less than what it had been prior to the failure simply because the bolt isn't pulling the nut up against the washers as much as it did beforehand. However in any case, NTSB doesn't seem to raise their torque readings as a direct issue*.
Yes very likely, but some measurable distortion of the stretched thread dimensions vs unstretched section of the same bolt should be evident if that was the case. I imagine they know this ;)

[wrench1]

Isn't it possible the bolt would have plastically deformed (stretched) slightly in the process of the tail boom pulling the entire longeron attachment fitting from the longeron?
Sure. Without knowing the failure sequence anything is possible with the bolts and ring. For example, why didn’t the upper RH bolt stretch as well? But if there was any bolt stretching I would think it would be the lower LH bolt with the upper bolt missing or broken. But it was over-torqued as well. Regardless as you mentioned it’s a moot point. I just base my observations on experience and other tailboom failures I’ve been around.

[helispotter]

wrench1: Yes, good point about expecting other bolts, especially lower left to then also be subject to stretching if that was the case on the lower right. I just had tunnel vision about the single (seemingly) under-torqued bolt!

Jetstream67: Assessing bolt distortion via thread pitch measurements would certainly be a good check, but there is no reporting of this in the "Materials Laboratory Factual Report 22-071". They documented number of washers used and number of exposed threads (beyond nut end) on all three remaining bolts.

[helofixer]

Bell issued ASB 407-22-128 today.Basically a one-time torque check of the four bolts and nuts, visual inspection of the surrounding structure, and then fill out a bunch of paperwork and email that to Bell but I'm sure an AD will be following, since ASB not mandatory unless you're Part 135 etc.