Inflight Breakup of a New Zealand Van’s RV-7 [Archive]

View Full Version : Inflight Breakup of a New Zealand Van’s RV-7

capngrog

8th Sep 2020, 23:20

Pilot DAR

9th Sep 2020, 02:00

That report is a good read! Off the top, I find it unlikely that the pilot was allowing the passenger to fly anywhere near those unusual attitudes. Sure, the passenger might fly some more or less straight and level, but not quasi aerobatics. I agree that this type of aircraft can build up speed very quickly when pointed down, particularly when carrying power down. I haven't flown the RV7, but I have flown the RV4, and they are very capable, and clean planes. They are easily capable of pleasing rolling, and getting too far over. A helicopter pilot would probably quite enjoy the solid feel of this plane in unusual attitudes, and let it get too far. Further, a helicopter pilot will be naturally adverse to controlling into higher G. I don't see any reference to the airplane being equipped with an accelerometer. If the G information were not available to the pilot, and his helicopter flying had not refined his "seat of the pants" sensation for G, he might not have applied enough, early enough in the dive, to recover before the speed built up. A spiral dive certainly makes sense in this situation. During flight testing, and intended dives, I've had to apply more than 2.5 G to prevent exceeding Vne (Cessna Grand Caravan). I had temporarily installed a G meter, so I could apply the required G confidently, to prevent exceeding Vne, and it worked. Without the G meter, I expect the G aversion would have seen me exceed Vne. And that was not a spiral dive, simply a spin recovery dive. The spiral would make it that much more difficult, and the moderate power on the way down would have made it all happen really quickly!

The relationship to the US and Canadian accidents appropriately shows that this was not the first time. This is a lesson worth learning! Build up unusual attitude flying skill in low drag planes with great caution!

3wheels

9th Sep 2020, 23:56

The accident happened extremely quickly.

The time from the commencement of the manoeuvre at 4500 feet to the final reported altitude of 1560 ft was only about 20 seconds. A descent rate of about 8,800 fpm. The time to the accident was only 27 seconds.

it seems that these aeroplanes can run away from you very quickly.

Pilot DAR

10th Sep 2020, 02:14

it seems that these aeroplanes can run away from you very quickly.

One must be careful with any slippery plane, this type would not be unique that way. During the aforementioned Caravan testing, spin recovery dives reached 9600 FPM down, and that was not out of control, it just built up speed really fast, pointed straight down for a few seconds! On the opposite end of the scale, I've had to fly Vd tests in a few floatplanes, and it's nearly impossible to get them to go down fast enough - too much drag!

20 seconds is lots of time to recognize something is going wrong, and apply recovery control inputs, though I agree the latter part of the 20 seconds was probably no longer opportune for any recovery effort. I helicopter pilot was probably surprised by unfamiliar attitudes, and control responses for such an agile small plane.

megan

11th Sep 2020, 00:46

One point that many may not be aware of is the significance of Vne being presented as IAS and the influence of TAS. The RV has a Vne of 200kts IAS, at 10,000 feet in ISA conditions that equates to a TAS of 233kts, in ISA+20 241kts. Flutter responds to TAS. Recently read a C210 accident report where the aircraft succumbed to an inflight breakup while making a high speed descent from altitude in calm conditions and this IAS/TAS disparity was given as the likely reason. DAR may wish to expand.

Pilot DAR

11th Sep 2020, 11:55

Flutter responds to TAS.

Interesting point. I've looked up the design requirements and the FAA flight test guide, and they do not specify which speed (IAS/TAS) to use for these tests. The flight test guide does mention "consideration of calibrated airspeed...", but no mention of TAS.

I'll be test flying with a Transport Canada Test Pilot, and Flight Test Engineer in the next month, including dive tests, so I'm going to discuss this with them. Thanks for introducing the thought Megan!

djpil

11th Sep 2020, 12:18

AC 23-629 is the reference. Done in conjunction with appropriate analysis so the crew have some idea of the critical condition. We (myself associated with analysis so on the ground) used to estimate the start altitude so the dive would get as close as it could to what we wanted.

I recall we had a chart of achievable max IAS vs altitude in the report. The reason why the FAA warns of modifying aircraft with more powerful engines is because you’re going to get outside what had been demonstrated originally.

DownWest

11th Sep 2020, 18:51

Back in '74, we has a Comanche 260 in and while putting her in the hanger, I noticed a ding in the stabilator. It had been filled and faired, but I thought to check... Nothing in the logbook, so checked the static balance...It was out by a bit, so corrected that with ballance weights and passed it.
Thing was, not long before, a similar a/c had the same problem and it rang a bell. In that case, the test flight was done by the owner. To make the weight, he got some friends along...Doing the Vne part of the test, he stuck the nose down and reached the Vne while still in the dive, rather than gently pull up and reaching it in level flight.. Result was, he exceeded the Vne and the stabilator fluttered and departed. Sudden pitch down, wings folded and that was it.
I think that was the case that resulted in all C of A test flights to carried out by comercial rated pilots. The investigation found that the stabilator was out of balance due to 'hanger rash' and had been repaired without thinking it through.

capngrog

11th Sep 2020, 21:53

The investigation found that the stabilator was out of balance due to 'hanger rash' and had been repaired without thinking it through.

GOOD CATCH, DownWest, in finding the improper repair to the stabilator of your Comanche!

Attention to detail is a large part of what makes Aviation safe. Ignoring those details can be fatal.

Regards,
Grog

megan

12th Sep 2020, 04:05

Russell Gulch

12th Sep 2020, 17:35

I think it's apparent why the aircraft crashed, sadly.

capngrog, the Original Poster here, quotes the report by the New Zealand CAA , which has apparently not chosen to present its data in the usual ICAO format. Would there be a reason for them not doing so?
Russ

capngrog

13th Sep 2020, 01:22

Examples of flutter in real life and wind tunnel.

https://www.youtube.com/watch?v=egDWh7jnNic

An article on why TAS is so important with regards to flutter.

https://www.vansaircraft.com/wp-content/uploads/2019/01/hp_limts.pdf

Thanks for the article on flutter. Although I thought I had a fair understanding of flutter lurking in the deep dark recesses of the remains of my mind, this article explains the phenomenon in an orderly and coherent manner. Every day, I learn that I'm not too old to learn sumthin'. It seems that my understanding of flutter was a bit muddy ... at best.

This article allowed me to understand why at an altitude of 70,000+ feet, the separation between V/S (stall) and V/NE (never exceed) of the U-2 is only around 5 knots. I can't imagine cruising around in the days of nuthin' but steam gauges and INS (no GPS) for hours in a U-2 flying in that narrow of an envelope. I'd be shakin' like a puppy passin' peach pits in a hail storm.

Cheers,
Grog

wdew

13th Sep 2020, 01:30

djpil

13th Sep 2020, 04:45

In the report it states the cause as flutter due to exceeding the Vne. However in the graphic of the sequence of events and location of aircraft parts the IAS of the RV was not close to Vne. The Vne was only exceeded in the later phase of the break up.
IAS at first break up location was 144kts . It was not above Vne in the preceding manouvres ?Breakup at approx where it got to 244 KIAS. I don't think they bothered with debris analysis so you'll just have to look at the wind and ponder. Witness statements mentioned in the executive summary but nought in the body of the report, they would have been interesting.

wdew

13th Sep 2020, 09:05

The graphic shows the break up started before the 244kts that was reached at the end closer to impact .

Genghis the Engineer

13th Sep 2020, 09:51

The New Zealand CAA has released its final report on the 01 January, 2018 fatal crash of a Van's RV-7.

Here's a link to the report:

https://www.aviation.govt.nz/assets/publications/fatal-accident-reports/2020-08-24-Vans-RV-7-ZK-DVS-New-Zealand.pdf

Regards,
Grog

Coming to this new, as my friend DAR asked if I'd take a look. What a dreadful report.

Firstly the repeated obsession with whether the passenger was flying or not? Why is this even relevant? Experienced pilots let pax fly all the time - and with an ATPL(H) this pilot was very experienced with nearly 400hrs FW and over 4000hrs RW. It also says that he was known seldom to relinquish control anyhow. He is the captain , and should be more than capable of handling the manoeuvers in question, or recovering if his pax got into them. So why is this even being mentioned?

Secondly the conclusion of rudder flutter is far from conclusive to me in the report. The rudder was still attached to the fin, and one wing had come off. So torsional divergence failure of the wing, and/or overload due to incorrect application of rudder above Va are both in the frame. These aren't discussed. I'm not saying it couldn't have been flutter - I am saying that the investigators seem to have been a bit eager to shoot for a single explanation.

That said, that the aeroplane failed structurally in flight is clearly the case. Which brings me to another major omission in this report - any investigation of the handling characteristics and whether they (pitch stability, spiral stability being things I'd be looking at) have characteristics likely to cause inadvertent overspeed. Is there anything in this aeroplane that makes it hard to recover from a spiral dive? Is there a tendency over Vne to lock into a dive? They really should have had a test pilot looking at a similar aeroplane.

Here's another point not discussed in the report. The *correct* recovery from a spiral dive is power-roll-pitch IN THAT ORDER. I have certainly as an instructor seen a lot of pilots incorrectly applying pitch and roll together, which as an engineer I know creates a high potential for airframe overstress and breakup of the nature that happened here. Is this being adequately taught to FW PPLs? - I have some doubts, and the report missed the opportunity to ask that.

Also not mentioned in the report, is the possibility of mis-diagnosis of a spiral dive as a spin, and incorrect application of spin recovery actions to recover from the spiral dive. That would be totally consistent with the evidence presented.

Also the statement that "Flight at any airspeed over Vne, however, exposes the aircraft to the possibility of flutter." misses that flutter can be experienced in a maladjusted aeroplane below Vne. I've seen it test flying a couple of aeroplanes, even in level flight, with incorrect adjustments in the elevator and pitch trimmer circuits, one was certainly heading for an in-flight breakup if I'd not known what I was doing.

I do like the recommendation for verbal overspeed warnings in light GA EFIS. At the same time, most light aeroplanes provide loads of warnings (pitch attitude, wind noise, the ASI itself) of overspeed, which isn't mentioned in the report.

Not a happy read, on multiple levels.

G

megan

14th Sep 2020, 02:35

Genghis the Engineer

14th Sep 2020, 10:06

The military rule we were taught was only pull two thirds of permitted "g" for a rolling pull out. Bit of a mute point when few civvie aircraft have a "g" meter.
Speaking as somebody who used to certify military aeroplanes for a living before switching to the civilian world, and have also trained as a pilot in both environments. You can, to some extent, provide very specific advice like that for military aeroplanes, because the pilots are (relatively) high ability and disciplined. There will be much less of a tendency to freelance on how to fly manoeuvres, or forget the approved way of doing things amongst military pilots than you have to assume with civilians.

The military pilot training and airworthiness sides are also much more joined up with each other than in any civilian environment, so it's much more possible to ensure a truly joined up aircraft approval, and pilot training syllabus than you could ever do in the civilian world. That allows you to do the sort of thing you're describing here - g-meter or not.

G

India Four Two

18th Oct 2020, 19:05

The military pilot training and airworthiness sides are also much more joined up with each other than in any civilian environment,...

I still remember (and use) the unusual attitude/spin recovery training from my UAS Chipmunk days - "throttle closed, roll to the nearest horizon and pull".

We didn't have G-meters so to this day, I rarely look at the G-meter when doing gentle aeros. I use my calibrated "seat of the pants"!

Miles Magister

19th Oct 2020, 19:39

The *correct* recovery from a spiral dive is power-roll-pitch IN THAT ORDER. I have certainly as an instructor seen a lot of pilots incorrectly applying pitch and roll together, which as an engineer I know creates a high potential for airframe overstress and breakup of the nature that happened here. Is this being adequately taught to FW PPLs?

What a refreshing statement to hear. I have said this exact thing so many times and generally civilian trained pilots do not accept the technique because their simulator instructor did not teach it. To me and those lucky enough to have been trained by the Queen's finest it is basic technique which we could not ignore if we tried because it is tattooed on the inside of our brains; Height Speed G Roll Pitch in that order and all as separate actions, never together. It is rarely taught at all in the civilian world, OK shoot me down if you wish, but I have spoke to many pilots who have returned from UPRT and Advanced UPRT who have not learnt or appreciate this basic technique.

MM

P.S. Most of the UPRT graduates I have talked to also fail to appreciate that the angle of attack of both wings increases as aileron (roll) is applied and that pulling to a high AOA and rolling at the same time will inevitably lead to uncommanded roll, or incipient spin. I have unfortunately witnessed with 100% consistently civilian pilots apply aileron as the first action when experiencing a turbulence (wake) induced upset in live flying.

capngrog

20th Oct 2020, 03:41

Rudder use seems to be a lost art these days. I've encountered quite a few pilots who seem to fly with their feet firmly planted on the cockpit floor.

Cheers,
Grog

sablatnic

20th Oct 2020, 09:36

Rudder use seems to be a lost art these days. I've encountered quite a few pilots who seem to fly with their feet firmly planted on the cockpit floor.

Cheers,
Grog
True, I was told by my first instructor, that "we use rudder on the ground only"!

cats_five

20th Oct 2020, 11:14

Rudder use seems to be a lost art these days. I've encountered quite a few pilots who seem to fly with their feet firmly planted on the cockpit floor.

Cheers,
Grog

Obviously not glider pilots

olster

5th Jan 2021, 11:33

Genghis your points are very well made as usual and we have to respect your engineering and flying credentials. In the airline world backed up by manufacturers advice we always teach in a nose low recovery scenario to roll the wings level prior to pitching. The thrust issue is semi optional dependant on aircraft type etc. I don’t know of any instructors who teach differently. However to reinforce that everyday is a school day I was asked to describe rolling g in my FI renewal last year. To my eternal shame I was not even aware of such a thing after 47 years flying, mainly airliners. The examiner, bless him, took the opportunity to educate rather than berate my lack of knowledge.

Big Pistons Forever

5th Jan 2021, 16:01

Coming to this new, as my friend DAR asked if I'd take a look. What a dreadful report.

Firstly the repeated obsession with whether the passenger was flying or not? Why is this even relevant? Experienced pilots let pax fly all the time - and with an ATPL(H) this pilot was very experienced with nearly 400hrs FW and over 4000hrs RW. It also says that he was known seldom to relinquish control anyhow. He is the captain , and should be more than capable of handling the manoeuvers in question, or recovering if his pax got into them. So why is this even being mentioned?

Secondly the conclusion of rudder flutter is far from conclusive to me in the report. The rudder was still attached to the fin, and one wing had come off. So torsional divergence failure of the wing, and/or overload due to incorrect application of rudder above Va are both in the frame. These aren't discussed. I'm not saying it couldn't have been flutter - I am saying that the investigators seem to have been a bit eager to shoot for a single explanation.

That said, that the aeroplane failed structurally in flight is clearly the case. Which brings me to another major omission in this report - any investigation of the handling characteristics and whether they (pitch stability, spiral stability being things I'd be looking at) have characteristics likely to cause inadvertent overspeed. Is there anything in this aeroplane that makes it hard to recover from a spiral dive? Is there a tendency over Vne to lock into a dive? They really should have had a test pilot looking at a similar aeroplane.

Here's another point not discussed in the report. The *correct* recovery from a spiral dive is power-roll-pitch IN THAT ORDER. I have certainly as an instructor seen a lot of pilots incorrectly applying pitch and roll together, which as an engineer I know creates a high potential for airframe overstress and breakup of the nature that happened here. Is this being adequately taught to FW PPLs? - I have some doubts, and the report missed the opportunity to ask that.

Also not mentioned in the report, is the possibility of mis-diagnosis of a spiral dive as a spin, and incorrect application of spin recovery actions to recover from the spiral dive. That would be totally consistent with the evidence presented.

Also the statement that "Flight at any airspeed over Vne, however, exposes the aircraft to the possibility of flutter." misses that flutter can be experienced in a maladjusted aeroplane below Vne. I've seen it test flying a couple of aeroplanes, even in level flight, with incorrect adjustments in the elevator and pitch trimmer circuits, one was certainly heading for an in-flight breakup if I'd not known what I was doing.

I do like the recommendation for verbal overspeed warnings in light GA EFIS. At the same time, most light aeroplanes provide loads of warnings (pitch attitude, wind noise, the ASI itself) of overspeed, which isn't mentioned in the report.

Not a happy read, on multiple levels.

G

I fellow I know once asked me to do some basic aerobatic training in his RV7. I refused as I consider this aircraft completely unsuited for any aerobatic or near aerobatic maneuvering except in the hands of an experienced aerobatic pilot.

It has the trifecta of bad things in an aerobatic aeroplane

1) It is quite slippery and will gain speed very quickly on a downline

2) It has light elevator forces and so it is easy to pull a lot of Gee especially in the event of a panic pull

3) It has a surprisingly poor roll rate. Rolling to erect from the inverted or even wings level from a high bank angle takes a long time which is problematic if the airplane is well nose down and accelerating like mad

In addition the forward opening canopy is not jetsonable and so there is no way to abandon the aircraft with a parachute descent if it breaks up or is unrecoverable

The other challenge is of course how the aircraft was built. Even minor lapses by the builder can be catastrophic under high loads. The Canadian crash mentioned in the report was precipitated by the weight of too much paint on the rudder which unbalanced it and left it more predisposed to flutter.

I advised all my students to take a basic aerobatic course. Some did not really like aerobatics and did not pursue aerobatic flying after learning the first basic aerobatic , but all reported that they were much more confident they could control and recover the aircraft no matter what orientation it was in

Jamesel

6th Jan 2021, 03:33

I fellow I know once asked me to do some basic aerobatic training in his RV7. I refused as I consider this aircraft completely unsuited for any aerobatic or near aerobatic maneuvering except in the hands of an experienced aerobatic pilot.

It has the trifecta of bad things in an aerobatic aeroplane

1) It is quite slippery and will gain speed very quickly on a downline

2) It has light elevator forces and so it is easy to pull a lot of Gee especially in the event of a panic pull

3) It has a surprisingly poor roll rate. Rolling to erect from the inverted or even wings level from a high bank angle takes a long time which is problematic if the airplane is well nose down and accelerating like mad

In addition the forward opening canopy is not jetsonable and so there is no way to abandon the aircraft with a parachute descent if it breaks up or is unrecoverable

The other challenge is of course how the aircraft was built. Even minor lapses by the builder can be catastrophic under high loads. The Canadian crash mentioned in the report was precipitated by the weight of too much paint on the rudder which unbalanced it and left it more predisposed to flutter.

Since I have engaged in basic aerobatic training with various RVs, while respecting your position, I do have a slightly different opinion. While I agree up to a point with you, I wonder if you may have been exposed to a “rogue” aircraft?

Point 1). Agreed, especially with the coarse fixed pitch propellor usually attached, but the installation of a constant speed propeller greatly aids in controlling excessive speed.
Point 2). Yup! Be very careful, Fly with 2 fingers (& don’t get scared!:cool:)
Point 3). Every RV-7/7A (4+) I’ve flown had a roll rate at Va of at least 120 degrees per second, which in my mind is sufficiently rapid for recovery from an unwanted attitude. I would MUCH rather be in an RV whilst nose low and inverted than a Citabria or a Cessna Aerobat! Having also taught aerobatics in gliders (Blanik L-13 to DG-1000), I find this aircraft similar in rewarding an appreciation for when to roll, and when to push (& when to take over!). The only real criticism I have in the roll characteristics is the vibration that shows at large aileron deflections at cruise speeds and higher, (which isn’t all that bad of a thing when above Va).
Point 3A). A canopy jettison lever is standard (IIRC), if not, it certainly is available - at least 3 of the planes I’ve flown were equipped with one - a T handle in the upper centre of the panel. Besides it’s obvious advantage in emergency egressing, a lot of people install it as it is very handy to completely remove the canopy when working on the radios & instruments.
Point 3B). Totally agree. While I have also refused to fly certain certified airframes, I flat out tell prospective homebuilt aerobats on first contact that I will not fly any amateur-builts until I have conducted a thorough inspection of their particular aircraft. I will not do acro in amateur-built glassfibre airframes simply because I am not knowledgeable enough to do a meaningful inspection of the structure.

Regards,
Jamesel