Polish HEMS Agusta 109 incident

[Ptkay]

The only Agusta 109 of Polish HEMS was destroyed beyond repair in
probable ground resonance incident today about 14:30 CET in Warsaw (EPBC).

It was an annual check ride for one of the experienced pilots with
check examiner/instructor on board.

After landing and starting the shut down procedures the helicopter
"started to shake violently", according to witness, rotor blades hit and
cut of the tail.

The crew was taken to hospital for examination. One of the pilots
has been already released, the other with suspected spinal injury
still remains hospitalized.

Some images and details (in Polish) under the link below:

?mig?owiec LPR rozbi? si? na lotnisku Bemowo - Stolica - TVN Warszawa - 20.11.2009 (http://www.tvnwarszawa.pl/-1,1629672,0,,smiglowiec_lpr_rozbil_sie_przy_lotnisku_bemowo, wiadomosc.html)

http://m.onet.pl/_m/dd51e21ec63555f48390b3ee8b24c684,34,1.jpg

http://m.onet.pl/_m/de732c0ef2339d89752dda0fdf96ce64,34,1.jpg

http://m.onet.pl/_m/3fcab8030ece24cd82eed8fd721a9a0e,34,1.jpg

[Ptkay]

Good by, SP-HXA...

You will be missed.

:(

http://img.airfoto.pl/my_img/b/48299.jpg

http://img.airfoto.pl/my_img/b/49028.jpg

http://img.airfoto.pl/my_img/b/49605.jpg

[Lt.Fubar]

And people were wandering what LPR will do with it when they get the EC135s... well, the problem solve itself :hmm:

The accident is weird though. Haven't heard of Ground Resonance incidents in A109s. BTW I heard it was straight from factory overhaul...
Just like that AW139 that "lost" its tail... hmmm...

[Ptkay]

This one was back out of maintenance in Belgium in June.
Enough time to show all the glitches...

But this one obviously also "lost it's tail". :}

[Tango and Cash]

Not the first A109 ground resonance incident (if that's what this turns out to be): DFW06LA057 (http://www.ntsb.gov/ntsb/brief.asp?ev_id=20060130X00132&key=1)

[ShyTorque]

Strange. I'd be surprised if this was a straightforward "ground resonance" accident.

I wonder what the wind was doing (apart from going down trouser legs)..

[Stuck_in_an_ATR]

Not a particularly windy day - westerly winds @5-8kts and probably no surrounding obstacles affecting airflow...

[Ptkay]

Hi res image of the machine today:

http://lukasz.eu.org/zdjecia/agusta.jpg

[ShyTorque]

Obviously, the main gearbox support appears to have failed in some way. Was this ground resonance in the true sense, seeing as the aircraft remained upright? I hope we get to know the results of the investigation of this one. :eek:

[blakmax]

OK guys, I have waited a while to see if the discussion progressed, principally because I don't want to be seen as an AW basher. I assure you I have no axe to grind, I simply comment on what I see irrespective of the OEM.

I am suspicious of the appearance of the blades. Do you notice the absence of the structure aft of the spar on several blades? I have seen this in a number of other helicopter accidents, and I have investigated failures on structures with a similar construction such as fixed wing aircraft rudders.

Now I am not familiar with the AW blade structure but usually the skin is bonded to the spar and also bonded at the trailing edge. Usually there is honeycomb bonded inside the skins to provide shear resistance and stiffness. That core is also bonded to the rear of the spar using a foaming adhesive to fill the gaps in the core. Now the analysis of the foaming adhesive bond between the core and the spar is usually trivial and often ignored because the adhesive usually has a strength above 1000psi and the core has a shear strength of 100-200psi, so the core should always fail before the bond.

Despite the trivial nature of the analysis, this bond is of importance to the structure because it transfers the shear loads from the core to the spar. In the rudder failure I mentioned above, the bond beteen the core and the mast of the rudder had not formed correctly at manufacture and an extensive injection repair was performed. Now if you have read any of my postings on other threads you will know that injection repairs have a zero chance of bonding anything. As a consequence, all of the shear loads had to be carried by the bond between the skin and the mast. The high stresses in the skin caused a fatigue crack which led to the rudder departing the aircraft after a high load event. In this case the bond between the skin and mast was backed up by fasteners, otherwise the adhesive would have peeled much earlier, probably even before the fatigue crack initiated.

Now there is an analogous situation with rotor blades because if the bond between the core and the spar is ineffective, the shear loads must be transferred thriugh the bond between the skin and the spar, resulting in high peel stresses. Such a condition may lead to failure of the bond between the blade and the spar, and that would almost certainly lead to a loss of lift, the blade(s) out of balance and consequent impacting the fuselage as shown in this example.

Now before anyone jumps down my throat, I have no evidence of this being a cause of the example shown. I am only suggesting that the investigators should look for this type of failure. One characteristic would be parts of the blade without evidence of fusealge strikes which would be located well away from the fuselage.

Regards

Blakmax

[9Aplus]

http://cdn-www.airliners.net/aviation-photos/photos/7/8/4/1612487.jpg

Take look Bmax, it seems that wire cutter have some impact on blades during
wild dance....:confused:

[blakmax]

Ok, I remember. Old dogs, new tricks. I finally worked out how to post these photos.

This one shows a spar to adhesive failure which was NOT the cause of the crash. However, given that the adhesive is 10 times stronger than the core we should be seeing core attached.

http://i1012.photobucket.com/albums/af241/adhesionassociates/hamiltonbladesparfoamadh.jpg?t=1259063811

Now here is a lower magnification of the matching surface.

http://i1012.photobucket.com/albums/af241/adhesionassociates/corewithfoamingadhesiveattached.jpg?t=1259063974

Got a nice carpet in my house, yeah?

Note the adhesive has completely separated from adjacent to the spar. This is PROBABLY because of moisture (humidity) contamination during the bond curing process. Moisture at the interface has prevented adhesion.

Now given the evidence I have seen on the AW139 boom failures, it appears that at least some of the AW bonding is performed in less than optimum controlled environments which has probably led to moisture contamination of the bond and subsequent disbonding due to weakened core to adhesive bonds caused by micro-voids. Now I have worked out this photo-linking stuff, here is a photo of a good bond which fails by fracture of the core and tearing of the fillet bonds. Note the absence of micro-voids compared to the next photo.
http://i1012.photobucket.com/albums/af241/adhesionassociates/Goodbnd.jpg?t=1259064925

Here is a photo of an AW139 boom disbond.
http://i1012.photobucket.com/albums/af241/adhesionassociates/skincloseuphi-res.jpg?t=1259064833
Note the number of small bubbles, and the number of shiny surfaces in the fractured core areas. These are also micro-voids. As I said on the AW139 thread, there is published data which shows that similar micro-void levels typically cause a 50% strength loss in overlap joints and I would expect a similar loss of flatwise tensile strength for core bonds, See my postings on the AW139 tail boom failure thread.

Hence, if similar core to spar bond failures as shown in the first two figures are evident in the AW109 event, there may be a common link. I hope that the investigators read this posting before assigning the event to "ground resonance". The possibility of a structural failure should not be excluded from consideration just yet. Again, I make no definitive conclusions. I simply suggest that this failure mode be investigated and if appropriate, eliminated.

Regards

Blakmax

[blakmax]

Hi 9Aplus, your response came in as I was posting the photos.

If the wire cutter impacted the blades, then there would be damage on the leading edge of the blades and I see no evidence of that on the two blades visible on that web site. Further, I suggest that a sideways swipe on the wire cutters by a blade at even moderate rotational speeds would have at least produced some significant bending of the wire cutter in the direction of rotation. I don't see that either.

I don't think that the wire cutters played any part in this event. It is pure coincidence that the blade came to rest against the wire cutter.:sad:

Just to be clear, I am not saying that my theory is the cause. I just request that investigators eliminate it by appropriate engineering assessment of the failure surfaces. I'd be happy to discuss the matter with them (at no charge) if they wish to contact me.

Regards

blakmax

[9Aplus]

Shiny tip on wire cutter leads me to consider in that directions, please
take care that leading spar is something like titanium and we can not see
possible marks from bellow of both remaining blades, tip can jump over
and remain soft structure than ripped off. This naturally can not be considered
like primary problem...

[RVDT]

This is not unusual. The blade has pitch so the trailing edge is lower than the leading edge. The blade is not strong in this area as it does not need to be. As you can see by the black stuff showing through the paint the skins are carbon and the core is only foam in this area. Touching trees with composite blades has varying rates of success depending on the blade manufacturer. i.e an AS 350 blade will survive more than an EC 135 and a alloy Huey blade is serviceable until you see the dings coming through on the TOP skin in the outer areas! The centrifugal force far exceeds the lift forces involved, work it out.

blackmax - I think you are barking up the wrong tree.

My guess is a possible gearbox strut failure leading to resonance. It has happened before. Problem now is trying to work out which happened first as the struts will be pretty sad.

How about the rusty old shed in the background with "Eurocopter" above the door? Class!

[ShyTorque]

I can't see a fourth blade, or its cuff.

[RVDT]

Or a third either! But what's left of it is in the previous pix. Both damper rod ends you can see, appear to have broken as well. The "sleeves" for want of a better term are carbon as well. If you pick one up when it is off the aircraft you wouldn't believe how light they are.

[blakmax]

Guys, I am not saying that this IS the cause. All I am saying is that not many investigators are aware of this form of failure and it needs to be considered in cases where the core and skins separate from the spar, even if only to eliminate it from the list of potential causes.

blakmax

[makrider]

I can't believe it...

I had wrote a long post to answer you guys, but when I've finished I decided to delete it because was not appropriate, like yours expertize comments on this accident, based on a couple of pictures....

AUTOCENSORED !!! :mad::mad::mad:

:suspect:

[skadi]

RVDT
Or a third either!

Then have a look at the pictures in Post No1! There its visible.

skadi

[RVDT]

But what's left of it is in the previous pixmeans - in the previous pictures you can see it. It is not visible in the high resolution photo.

[cayuse365]

There is a history of ground resonance back to 1979 in New York.
NTSB NYC79FHI15

[blakmax]

Hi Svenstron

Thanks for the firm but polite posting. I am only to happy to acknowledge that this may be a case where the goose gets away.

I have been involved in a crash investigation on another aircraft type where it appears that the blade disbonded and the resultant loss of lift on that blade and the concurrent lift on the other blade resulted in the head tilting and the damaged rotor spar impacting the empenage. I have no firmly held position on this. All I ask is that the investigators consider my theory and if necessary dismiss it.

They may wish to look at this paper for reference on adhesive bond failure forensics:
M.J. Davis and D.A. Bond, The Importance of Failure Mode Identification in Adhesive Bonded Aircraft Structures And Repairs, 12th International Conference on Composite Materials, Paris, July 05-09 1999. If they wish to contact me I'd be only too happy to talk to them.

Regards

blakmax

[blakmax]

Hi Svenstron

I assure you I have no agenda to criticise AW. I would have made the same observation if it was any company. The concentration on AW is due to the pure coincidence of the failures at DOH and Poland being on aircraft by the same manufacturer and is not of my choosing. There is certainly an element of guilt by association in my positngs because of the bonding issues I have observed on the photographs sent to me from actual disbonds from other AW139 tail booms. The characteristics there are indicative of a systemic problem associated with the integrity of their bonding processes.

Indeed I have made similar observations in relation to another manufacturer on other threads such ashttp://www.pprune.org/rotorheads/363066-robinson-r44-accident-panama-19th-feb-09-3-fatalities.html

It will take a short time to inspect the blades using the Mark I eye ball to see if my suggestion is worth further investigations, because there will be fractures through the foaming adhesive used to bond the core to the spar. There may also be failures at the interface between the adhesive and ether the core or the spar. For my theory to have any relevance, there must be coincidental separation of the skin from the spar immediately adjacent to the bond failures either by disbonding or by fracture of the skin. Now it will take a few minutes to assess these conditions and all I am suggesting is that it would be worth eliminating that possibility. If my theory is not relevant in this case, only a small amount of time has been wasted. If I am right and the theory is not investigated, then there remains a risk to continuing airworthiness which could have been addressed if only a few hours of visual inspection had been expended.

Now I have in my office a section of a blade from another type which exhibits exactly the conditions I describe. In this case it was not the cause of the crash but resulted during the event. I have also a segment of a structure manufactured using similar construction methods where the same conditions led to in flight separation of a component on a fixed wing aircraft. I have also observed the conditions I describe on a number of other crashed blades. This type of failure actually does occur. Surely it is worth a few minutes of the investigators time to check this out.

With regard to the possibility that failure of the blade skin could cause such an event, if you couple a loss of lift on one blade with additional significant drag caused by the separated skin, I suggest that the resultant imbalance could cause sever loss of tracking of that blade. It may even be a driving force in setting up resonance.

I agree with you that there are other causes and it may be that one of these causes are correctly identified in this instance. Like you say, let's wait for the report.

Regards

blakmax

[captain_m]

Gentlemen,

Any updates about the investigations or probable cause of this accident?:ugh:

[captain_m]

Question:

Is it possible that the "torque plate" had failed in this accident?

Torque Plate is the plate that the transmission sets on.

[ShyTorque]

After some initial speculation, I think most of us are now waiting for the official verdict. ;)

[blakmax]

Svenstron

All I have done is to alert people to a form of blade failure which happens to present itself in a form similar to that displayed on these blades and suggested that it be given a cursory assessment. I have never said definitively that this was the cause. I have alluded to bonding issues with the tail boom as an indication that there are papparently some issues involving at least part of the AW system and the photographs I posted show that this is the case.

Now put yourself in my position. I am aware of a particular failure mode which is not widely known within the investigative community. Do I raise the issue, or do I just shut up as you apparently suggest I should have done? Now if I am wrong, all that is lost is a small amount of the examiner's time. If I am right and I did shut up and the examiner missed the defect and then the same thing occurs in flight, what then?

The type of skin on the blade (composite or metal) is not an issue. The bond to which I refer is between the core and the spar, which I am sure is metal. The usual cause of failure of this bond is exposure of the core to spar foaming adhesive to high humidity.The problem with the tail boom bonds as displayed in the photos is also due to exposure of the film adhesive to high humidity. Hence the connection.

Time out. Let's wait for the report and not get too aggitated about this. Life is too short.

Regards

blakmax (there is no "c" in blakmax).

[ShyTorque]

The bond to which I refer is between the core and the spar, which I am sure is metal.

I think it might be better to check facts before making this assumption.

[blakmax]

Hi Shytorque

I based my assumption that the spar was metallic from posting 14 of this thread by 9Aplus: leading spar is something like titanium

blakmax

[ShyTorque]

Not all things on this website should be treated as facts!

[makrider]

Not all things on this website should be treated as facts!

I agree 100% :ok:

[9Aplus]

No not at all....:=
but:

2005. from some CAA Bulletins
Several cases of main rotor blade cracking have been found in service, including one which failed catastrophically. Fatigue cracking has been initiated by chordwise scoring, introduced during manufacture, adjacent to spar mass balance weights.ROTOR&RESCUE 2006/2007
– The rotor blades, made of composite materials,have what Agusta
describes as a “vibration reducing mass distribution along the M/R blade span”.
This new design allowed removing the little spoilers typical for AgustaWestland’s
rotorblades, and drawing a new aerodynamic profile of the blade roots that significantly reduces the rotor’s drag.for me strong indication that metal parts are there...
(Because of well known reasons must use data from public sources only......)

[ShyTorque]

All I can say is: My A109E type conversion technical notes "workbook" (written by the Agusta factory training center) state:

Blades are made of composite: the blade consists of a 'D' shaped composite spar and the trailing edge is made of nomex core and carbon fibre skins. Blade aerofoils at the root are of the high lift type (AG8021).

Blade airfoils at the tip are of the high velocity type (AG8026).

The leading edge is protected against wear by a bonded stainless steel abrasion strip. Each blade has a bendable trailing edge trim tab adjustable for rotor dynamic tracking. The blades are designed with a "droop snoot" airfoil section and the thickness is tapered from rotor (root*) to tip to optimize the lift coefficient, the blade pitch range and to reduce noise level.

The composite main rotor blades due to the static balancing of each blade to a master blade as a final manufacturing process, they are completely interchangeable.

*root - my edit. Everything else copied verbatim.

Other A109 models are different.

[blakmax]

This raises a question of terminology which has bugged me for years. The FAA discusses "composites" meaning both composite construction where a structure is fabricated from details (fibre composite or metal) which are bonded together and they also use the same term for fibre composites. (See AC20-107B) I would be surprised if EASA and many OEMs did not have the same mixed terminology. I personally prefer to reserve the word composite to mean fibre composites and describe structure made up of bonded details as bonded structure. No confusion.

If my theory is correct, the only difference the spar material will make is the visible part of the back of the spar will be black for composite or metallic for the alternative materials. Irrespective of the material actually used to make the spar, the strength of the bond should always be greater than the strength of the core. Hence, if there is core still attached to the rear of the spar, then my theory is not applicable, move on. If however the bond to the rear of the spar was weaker than the strength of the core, there will be an absence of core residue at the rear of the spar and there may be a problem. A simple five minute look at the rear of the spar is all that is initially required.

Circular/fragmented arguement. Wait for the report.

[ShyTorque]

My final observation is that (from the larger photo) the damaged blades seem to have the top skin broken / smashed off the spar, rather than disbonded from it. You can see the line of the spar near the tip of the near blade.

Wait for the report.

Good idea.

[Rocket2]

Agusta Blades: Just so there's no doubts left folks, Agusta use different blades on the 109 series. The A109A/AII use the 109 series blade that are of metal construction. The 109C series onwards (C, E, K, K2, S etc & the 119) all use the 709 series blades that are of composite construction (including the spar) but of course have the metal leading edge strip bonded on to fight against erosion. I doubt that debonding of one of these blades was the cause of this incident but I'll stand corrected.
Safe flying
R2

[captain_m]

Blakmax,

How would you describe the strength of the "torque plate" on the A109E?


Thanks

[blakmax]

Sorry Captain, but my areas of expertise is in composites, adhesive bonding and adhesive bonding forensics.

Regards

blakmax

[RVDT]

How do you tell an Aussie?


Ya can't!

[blakmax]

I gather by the tone of the last few messages that there is some conclusion to the investigation of which I am not aware. If as it seems is implied by Captain M 's message that there was a failure of the torque plate, then I am relieved that my suggestion did not in fact occur. I have said all along that all I wanted was to have the blades examined to prove/disprove the theory in this case. As I also said, the only reason I raised the issue was that I have seen these failures in blades and other structure and they presented similar characteristics. Because this is not a widely known failure mode, I believed I had a professional duty to raise the matter so that it could be considered during the investigations. Some of the responses evoked were less than professional, but I would like to commend Svenstrom for his detailed and well stated explanation of the event, even if there was a bit of rancour added in places.

As for RVDT's last remark, that is a poor representation of a very old joke: "You can always tell a **** but you can't tell him much." Substitute for **** any noun describing the rank, mustering, trade, sport, race, religion, nationality etc.

I suggest you contact some Kiwi's. They have much better Aussie jokes.

"What is the most significant geographic fault in Australia?...It is above sea level."
etc. etc.

Regards

blakmax

[captain_m]

Blakmax,

Thanks for the answer. The reason I asked is that I heard some operators found cracks or corrosion on that particular plate.

Regards

[captain_m]

I guess no news yet!

[Phoinix]

Maybe a report was issued in the mean time?

[ARRAKIS]

Only in Polish.

http://www.transport.gov.pl/files/0/1792161/20090995RK.pdf

Ground resonance.

Arrakis

[Phoinix]

Thank you!

[Ready2Fly]

Maybe this is far-fetched but seriously: Can you pass a checkride with such an incident given that they were already shutting down?

I.e. if it was pilot error, no doubt(?) but if it happened due to a mechanical failure (the report is in Polish only so i don't know what the conclusion was besides ground resonance).