What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

RV-7/7A In-Flight Breakup Accident Information

... In Rockets, and several Reno-racing RVs, the fwd fin spar gets a doubler that runs part way up (to the middle rib IIRC) which stiffens the fin in torsion (and also somewhat in bending). With this added thickness from the standard fin spar plus the doubler, a thicker attach bracket should also be used. The rear fin spar also gets additional strengthening/stiffening.

I wonder if any of the Reno-racing RV modified the vertical stabilizer by making it shorter or smaller. It seems the large standard kit size is for aerobatic flying and spin recovery but I suspect it works against the high top speeds that are seen at Reno. I've seen some of the Reno specific racers (non-RV) that have very diminutive tails.
 
Critical errors are often fatal in aviation.

Testing to 10% over Vne is pretty standard in the industry and Van's has done that.

FAR Part 23 requires 15%.

Actually, for Part 23 airplanes, the airplane is to to flutter free to 1.2 Vd by analysis (20% above Vd). Part 25 requires only 15%.

Flight flutter testing is done to Vd, which is 10% above Vne (1.1 Vne), to demonstrate that adequate modal damping margins exist at Vd, and that there is no large and rapid reduction in damping as Vd is approached.

Of course, Experimental airplanes are not governed by Part 23, but Van's has stated they test to Vd.
 
Last edited:
Dive Testing

There is of course no requirement for EAB to test to part 23 or part 25 requirements.
The Wittman Tailwind in 1953 was the first EAB to be allowed to carry passengers. The CAA(FAA) requirements were a dive test to 110% of Vne and a 4G test while loaded to gross weight.
The first five Tailwinds were tested to this standard.
On one of those tests a sandbag slipped out of place and pushed against the stick. The resulting dive was near 300 statute indicated and the G meter showed plus 8.3 and minus 3.8. Minor bend in windshield support, no other structural damage.
The elephant in the room is the question: was the RV7 ever dive tested with the RV9 rudder???
Why has Van's and or the FAA not canceled the aerobatic certification for the RV7 with the RV9 rudder?? Why have they not reduced the Vne of the 7 with 9 rudder as a safety precaution?
RV4 and 6 zero structural breakups. RV8 one wing failure caused by G load far in excess of design limits.
 
After watching this thread grow to 16 pages with great interest, I've decided to chime in with a few thoughts, in no particular order;

if you beef up one area, you are frequently just transferring the initial failure to an adjacent area

slippery airplanes can reach Vne in a heartbeat if you're inattentive, especially when you are coming out of something that can't reach Vne unless you point it at the center of the earth.

flutter margins aren't something to be trifled with, nor are they something to be improved upon unless you have a big brain plus the ground and flight test equipment to back up your proposed fix.

A few of the modifications that are being talked about here aren't something that we would do in the certified world without a ton of modeling and static test articles, followed up by flight testing with an emergency egress system, pilots with chutes, and sometimes spin chutes on the flight test plane.

Yeah, I get that with a big enough statistical sample, you can predict X number of events over time, and this statistic seems to say that if you overspeed an RV7 you can rip the rudder off before the rest of the tail comes apart.

This set of statistical data would also seem to indicate that the builder has a choice; he can either install a big rudder and have smaller flutter margins (maybe) or install a small rudder and have poorer spin recovery characteristics. Pick one-live within the constraints of your choice. It's obvious where Van stands on this because back in the day they ponied up free rudders for everybody.

I'm going to take the stance that there isn't anything inherently unsafe about a well proven design, and be thankful that it's been in service long enough to identify potential danger areas to stay away from. Is it as forgiving as a 172? In this particular area, probably not. But you know what, it's not as forgiving of being left out in the rain either. That doesn't make it a bad airplane, just different considerations.

I could look at any two airplanes and pick them apart as to which one is better and which one is worse. A lot of you guys are old enough to remember when the first Citation came out and the Lear guys started saying that it was so slow that it was going take bird strikes from the rear. It wasn't very long before the Citation guys started saying that a Citation was doing everything it could to keep you alive and a Lear was doing everything it could to kill you.

My point is that like most things in life, you have to weigh the pro's and cons, mitigate the cons while being confident that the cure isn't worse than the disease, then live with the result.

This is an interesting discussion for sure, but as for me, I'm going to accept that my airframe has some well documented areas that need to be respected, with different considerations that a Cessna (or whatever). I'm building the airframe to it's well proven design, with the big rudder that Van's recommends/supplies and will try to live within those constraints.

edit- I have one final thought here; I've seen some comments about modifying stuff and the rebuttal has been "you're going into test pilot mode" followed by somebody else chiming in; "you're in test pilot mode when you're in phase 1" or something to that effect, implying that these two things are somehow equivalent.

I'll respectfully opine that no, in phase 1 you're in CONFORMITY pilot mode. You're confirming that your particular airplane behaves the way everybody is expecting it to, because that's the way it was originally designed and built to behave. If it's out of conformity with the rest of the fleet, then you try to figure out why and fix it, but it's extremely unlikely if you build to plan that something is going the be catastrophically out of whack.

In this case, it would seem to hold true whether you use the small rudder or big one you're in safe, well documented territory. They are both known quantities with known characteristics. But, If you go off the reservation with extra stiffeners or balance weights or otherwise start monkeying around with your own "improvements" without knowing what you're doing, you're truly in test pilot mode, which isn't where most of us want to be without significant training and safeguards in place.
 
Last edited:
There's an awful lot of talk about statistics and failure rates and whatnot, but you know what I haven't seen?

Actual statistics and failure rates and other numerical data.

How many RV-7s are there? How many with each rudder option? How many RV-8s?

Now, the claim is that there are 0 RV-8 rudder flutter-induced fatalities. Maybe, but let's accept that for now.

How many RV-7 flutter-induced fatalities? Below Vne? Above Vne? With which rudder?

I've seen nothing to convince me that there's any problem with the larger rudder other than when a pilot grossly exceeds Vne there's an increased risk of flutter-induced failure. No s**t, Sherlock, as we used to say.

I'm with the poster who asked to see any evidence of rudder issues on any aircraft flown within the envelope. Haven't seen it yet.
 
Quote:
The kit manufacturer had conducted flight testing of the RV-7A prototype. No indications of flutter were encountered at a speed of 217 knots. Additional theoretical flutter analysis was done where the flutter speed was calculated to be 300 knots for the baseline design


This is from a previous thread. It looks like the initial testing with the 8 rudder looked pretty good.
 
Arlington, AZ RV7

Some interesting highlights from the NTSB Probable Cause:
"cockpit canopy, vertical stabilizer and rudder were found about 1 mile from the main wreckage".
Probable Cause:
Inflight overstress separation of vertical stabilizer and rudder during flight-----Contributing-----was an inflight collision with a bird. (end)
"The RV7 is a two place, tandem seat,------(can't they do any better than this??)
Comments: No discussion of why canopy and vertical tail were all found together. No discussion of how bird feathers might have gotten under the passenger seat. No discussion of size/weight of bird which was identified as Rock Pigeon which is under one pound. No discussion of aircraft speed but a detailed discussion of ground track.
They do mention that the overstress could have been a startle reaction to the bird(s)
 
Looked it up myself: 23.629... "predict freedom from flutter, control reversal
and divergence must cover all speeds up to 1.2 VD."
 
Looked it up myself: 23.629... "predict freedom from flutter, control reversal
and divergence must cover all speeds up to 1.2 VD."

That is for the flutter analysis results. Flight flutter testing is done to Vd (i.e., 10% above Vne). Read all of CFR 23.629.

Again, Post #154 has details.
 
Last edited:
That is for the flutter analysis results. Flight flutter testing is done to Vd (i.e., 10% above Vne). Read all of CFR 23.629.

Well, there ya go...the two cases that we *know* were possibly caused by flutter demonstrated, via test if you will, that that requirement was well met. In one case, over *20* percent above Vne.
 
Reading some of the posters on this thread leaves me with the feeling that if I accidentally exceed VNe, then I deserve the consequences. The reason why I got a 7 kit was its ability for aerobatics. I may never do aerobatics but I like the fact that it can.

Back when I was learning to fly in a 150, on my first solo I decided to practice some stalls. I probably had about 10 hrs total by that time. The power off stalls went fine but when I did the power on, it flipped right into a spin. I just turned loose and it kept spinning. (never reduced the power) I probably did about 4 or 5 turns before I remembered an article on uncontrolled flight... Opposite rudder. It popped right out but leaving the throttle wide open, I maxed out at 135 mph in the pullout. I climbed back up and did it again but recovered as soon as it started the spin. I decided to ask someone about that. I never asked anyone but reading about stalls that night told me that the instructor had been applying right rudder.

Unexpected things do happen and something as simple as a roll can be screwed up. I don't need more restrictions in life. After 60 years of flying, I find out that VNe is TAS. Why didn't I have a TAS gauge in my aircraft? Reading the Canadian accident report tells me that extra paint on my rudder decreases the margin of flutter significantly. An airplane coming apart is the worst possible penalty for screwing up. After reading the Canadian report, it was an easy decision to change to the 8 rudder. If you don't have one, the kit price for the 8 rudder is $375.
 
jask;1629964 After 60 years of flying said:
Vans are the only planes that I know of that VNE is TAS. All others that I know of are IAS. Please correct me if I’m wrong,
 
Vans are the only planes that I know of that VNE is TAS. All others that I know of are IAS. Please correct me if I’m wrong,

When something like a 172 has it's Vne marked in indicated airspeed, that number will keep you out of trouble at any corresponding TAS, i.e. the limiting corresponding TAS is incapable of being reached below that indicated airspeed. It would be redundant to publish them both

However, there are lots of airplanes that aren't that way, not just Vans.

Here's a quick screen grab of an old King Air airspeed indicator. The barber pole represents Vne and it moves by itself as your altitude changes to effectively show you Vne as a true airspeed.
 

Attachments

  • s-l500.jpg
    s-l500.jpg
    38.2 KB · Views: 145
That is for the flutter analysis results. Flight flutter testing is done to Vd (i.e., 10% above Vne). Read all of CFR 23.629.

Again, Post #154 has details.

I realize analysis and testing is to 1.2 X Vd, not Vne. In any case, Van's doesn't flight test to this criteria. GVT may have been done to this level in simulation.
 
Last edited:
When something like a 172 has it's Vne marked in indicated airspeed, that number will keep you out of trouble at any corresponding TAS, i.e. the limiting corresponding TAS is incapable of being reached below that indicated airspeed. It would be redundant to publish them both

However, there are lots of airplanes that aren't that way, not just Vans.

Here's a quick screen grab of an old King Air airspeed indicator. The barber pole represents Vne and it moves by itself as your altitude changes to effectively show you Vne as a true airspeed.
That is incorrect. King Air Vmo (airspeed) does not change until the crossover altitude between airspeed and mach number at approximately 27'000' (FL270) Below that altitude Vmo is a constant indicated airspeed with no correction to true airspeed.
 
Does it matter? If the aircraft designers and engineers tell you it's in IAS, then use that. If they (e.g., Van's) tell you it's in TAS, then that's what you should use.

FAR 91.103 Each pilot in command shall, before beginning a flight, become familiar with all available information concerning that flight.

I'd say knowing the airspeed limitations of the aircraft itself would fall in that category of information.
 
That is incorrect. King Air Vmo (airspeed) does not change until the crossover altitude between airspeed and mach number at approximately 27'000' (FL270) Below that altitude Vmo is a constant indicated airspeed with no correction to true airspeed.

Thanks for correcting me- a little counting on my fingers tells me it's been somewhere around 28 years since I've been in king air so my memory is a little fuzzy on the details.

Anyway, point being, Vne on a little airplane isn't the only gospel out there.
 
I realize analysis and testing is to 1.2 X Vd, not Vne.

Okay, now you're just pulling my chain, right? Your statement is incorrect (again).

So as to not leave other folks with incorrect information, this is the correct (and simplified) info (again):

Flutter analysis --> flutter free to 1.2 X Vd (20% above Vd).

Flight flutter testing --> to Vd.

Note that Vd is 10% higher than Vne.​


Scott M. ("rvbuilder2002") has stated that Van's normally conducts flight tests to Vd.

Again, Post #154 has details.
 
Last edited:
RV14 rudder on RV7

Can anyone confirm whether an RV14 rudder would fit without modification on an RV7? Has anyone done it? What level of confidence is there that this configuration would provide adequate spin recovery and better flutter margin vs the RV9 rudder currently supplied?
Thanks,
 
Can anyone confirm whether an RV14 rudder would fit without modification on an RV7? Has anyone done it? What level of confidence is there that this configuration would provide adequate spin recovery and better flutter margin vs the RV9 rudder currently supplied?
Thanks,

Can you define "better"? What particular type of behavior are you trying to fix?
 
Can you define "better"? What particular type of behavior are you trying to fix?

He had a great question, the -14 rudder...with the answers to your question(s) in his post. It's debate bait AndyWW. Nothing wrong with debates, you get to see both sides. Always question everything, that's how we improve.

Originally Posted by AndyWW View Post
Can anyone confirm whether an RV14 rudder would fit without modification on an RV7? Has anyone done it? What level of confidence is there that this configuration would provide adequate spin recovery and better flutter margin vs the RV9 rudder currently supplied?
Thanks,
 
Spins

I will break this down so the most uninformed person MIGHT understand. If someone offered to let me fly an RV7 with the small tail I would accept the offer and then do a lot of spins and form an opinion of the spin recovery. If someone offered me the keys to their RV7 with RV9 rudder I would politely say no than you.
I have taken a brief look at RV7 spin videos with what appears to be small rudder and I can't see any issues with spin recovery. No one has said that spin recovery is a problem with the 7 with small rudder, just that it wasn't quite what was expected.
I will be watching more videos and trying to establish exactly when recovery controls are applied.
 
https://www.jmbaircraft.com/safety

There are a number pilots that are used to VNE being in terms of IAS only and anything we can do to help educate the risk with TAS in our airplanes will help to save lives.

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

Part of the problem (but I don't believe it's the whole problem) may be that Van's has never issued a Service Bulletin or Service Letter for the older RV models (including the -7s) stating that Van's changed the definition of Vne from IAS to TAS. They did put out an SL for the RV-14, and they put out a POH revision for the RV-12s, but nothing for the older RVs. It's disappointing that Van's has not issued an SB or SL addressing their change of that critical safety-of-flight information. Many RV pilots are unaware that Vne is now in terms of TAS, instead of IAS. At higher altitudes RV pilots may operate above Vne without realizing it.

As you indicated in your post, Van's did publish the article, "Flying High and Fast", on high horsepower and the concern about IAS vs TAS, flutter, and operating at higher altitudes. But it is not an acceptable substitute for an SB or SL for each model stating the new definition of Vne is in terms of TAS, since an SB or SL would reach a much wider audience of pilots and builders.

It does seem that more and more folks are putting -8 rudders on their -7s, and that’s a good thing, IMO.
 
It's disappointing that Van's has not issued an SB or SL addressing their change of that critical safety-of-flight information. Many RV pilots are unaware that Vne is now in terms of TAS, instead of IAS. At higher altitudes RV pilots may operate above Vne without realizing it.
It appears to be irrelevant for most of these accidents... Where the aircraft were being flown beyond even the IAS limits. No SB would have stopped that.
 
Can you define "better"? What particular type of behavior are you trying to fix?

Seems fairly obvious in this context - "better" means that the rudder ideally doesn't fail with any greater frequency than those of the other RV aircraft. 8 accidents related to this issue in 12 years is definitely statistically significant, especially compared with the very good record of the other similar but different models. I'm not saying Vans is at fault, just that a quest for continuous improvement is a defining characteristic of the human race...
 
Seems fairly obvious in this context - "better" means that the rudder ideally doesn't fail with any greater frequency than those of the other RV aircraft. 8 accidents related to this issue in 12 years is definitely statistically significant, especially compared with the very good record of the other similar but different models. I'm not saying Vans is at fault, just that a quest for continuous improvement is a defining characteristic of the human race...

Fair enough - but the rudder failures are a symptom - not a cause. By masking the symptom you're just transferring the problem somewhere else. The root cause here is flying the aircraft beyond the flight envelope.
 
Statistically significant?

According to Vans there are 1906 flying RV7s. With 8 accidents that is less than 1% of the flying fleet (0.4). We don’t know the build quality of those rudders. Maybe it was excellent or maybe not. We don’t know if there was damage prior to the accidents, like what just happened at AirVenture. The damage may have been internal, not viewable during preflight, and hidden by the damage of the accident. We do know there was some issues with exceeding recommended speeds.

Somehow this thread has taken on the thought process that the RV-7/9 rudder should be condemned and replaced. I don’t believe the data supports that conclusion. But it’s your aircraft, order a RV-8 rudder and install it if that makes you more comfortable.
 
Let's look closer at the supposed data that shows 8 rudder failures causing accidents, based on the list in the first post:

1. Preliminary report only, cause undetermined yet.
2. Preliminary report only, in-flight break-up, weather involved, cause undetermined yet.
3. Exceeded Vne by > 20% (44 knots).
4. "There were no clear indications that any of the [empennage] components that fractured in overstress did so before ground impact or independently of the bird strike."
5. "damage to the horizontal stabilizers and elevators that was consistent with a downward failure in positive overload. The loads required to fail the horizontal stabilizers and elevators cannot be generated from normal flight or
control movements. Such failures would have required an abrupt pull back on the stick and corresponding movement of the elevator to a trailing-edge-up position, at speeds greater than the airplane's maneuvering speed."
6. Probable cause undetermined because wreckage was not recovered (sunk in 500' of water).
7. "At the time of the inflight breakup, the airplane was traveling 20 knots above the published never exceed speed."
8. Exceeded Vne by 34 knots.

Not one of those indicates to me that the rudder was the initial cause of the failure. Two of those specifically cite rudder failure, both at speeds far in excess of Vne.

Tell me again why the rudder is a problem here?
 
Let's look closer at the supposed data that shows 8 rudder failures causing accidents, based on the list in the first post:

1. Preliminary report only, cause undetermined yet.
2. Preliminary report only, in-flight break-up, weather involved, cause undetermined yet.
3. Exceeded Vne by > 20% (44 knots).
4. "There were no clear indications that any of the [empennage] components that fractured in overstress did so before ground impact or independently of the bird strike."
5. "damage to the horizontal stabilizers and elevators that was consistent with a downward failure in positive overload. The loads required to fail the horizontal stabilizers and elevators cannot be generated from normal flight or
control movements. Such failures would have required an abrupt pull back on the stick and corresponding movement of the elevator to a trailing-edge-up position, at speeds greater than the airplane's maneuvering speed."
6. Probable cause undetermined because wreckage was not recovered (sunk in 500' of water).
7. "At the time of the inflight breakup, the airplane was traveling 20 knots above the published never exceed speed."
8. Exceeded Vne by 34 knots.

Not one of those indicates to me that the rudder was the initial cause of the failure. Two of those specifically cite rudder failure, both at speeds far in excess of Vne.

Tell me again why the rudder is a problem here?

Because the accident reports state rudder flutter….yes with an over speed. Again and again and again…..easy to say don’t exceed any flight envelope….that states the obvious but accidents and incidents happen….you cannot assume absolutely no RV8’s have NOT exceeded VNE and possibly by quite a margin….yet no accidents attributed to rudder flutter….that begs the question of a rudder designed for a slower aircraft stuck on the back of a faster one and flight tested to be acceptable within the normal envelope is actually weaker and more prone to destructive flutter than the 8 rudder….doesn’t it?
 
Acccident #4

This is a very misleading statement by the investigators. The rudder and vertical stabilizer were one mile from the main wreckage.
In the NJ accident the lower half of the rudder and the fin were the first items in the accident path.
 
would suggest it’s the speed.
We all know speed is the killer.

Is there a argument that the -7 is slipperier than the -6 & -8, thus easier to bust Vne when distracted? If not, then the -7s are more likely to come apart flown by the same average group of pilots...and thats the point, not speed.

The second point, IMHO, is how do we make the -7 match the inflight break-up rate of the -6 and -8 (which is zero). Hence the talk of smaller rudders.
 
you cannot assume absolutely no RV8’s have NOT exceeded VNE and possibly by quite a margin….yet no accidents attributed to rudder flutter

Now you're just guessing. There's now way to know if, nor how many, nor at what speeds, any RV-8 pilots have exceeded Vne. Might be none, might be all of them. But there's no data, so any figure you have is merely a PIFMA value.
 
how do we make the -7 match the inflight break-up rate of the -6 and -8 (which is zero). Hence the talk of smaller rudders.

Training? Education?

3/8 accidents in the list involved flight well in excess of Vne. Remove those 3 and the rest appear to be overstress in reaction to possible bird strike, something maybe to do with flying into bad weather, and so on.

I don't now how much more you can teach people not to fly VFR into IMC, or to get aerobatic training if you're going to do aerobatics, etc.
 
Red corvettes are more likely to be involved in wrecks than blue ones. It must be the paint. We need to change the paint. Why isn't anybody changing the paint?
 
Now you're just guessing. There's now way to know if, nor how many, nor at what speeds, any RV-8 pilots have exceeded Vne. Might be none, might be all of them. But there's no data, so any figure you have is merely a PIFMA value.

PIFMA? Please define.
 
Well said Michael!

This thread reminds me of the old saying "You can lead a horse to water but you can't make him drink!"

Something else needs to be said also.
The 8 rudder just looks way better than that big 9 rudder:D

Mark


We all know speed is the killer.

Is there a argument that the -7 is slipperier than the -6 & -8, thus easier to bust Vne when distracted? If not, then the -7s are more likely to come apart flown by the same average group of pilots...and thats the point, not speed.

The second point, IMHO, is how do we make the -7 match the inflight break-up rate of the -6 and -8 (which is zero). Hence the talk of smaller rudders.
 
Now you're just guessing. There's now way to know if, nor how many, nor at what speeds, any RV-8 pilots have exceeded Vne. Might be none, might be all of them. But there's no data, so any figure you have is merely a PIFMA value.

Not guessing….assuming. It’s quite different. We are to think all 8 pilots are perfect and not all 7 pilots cannot fly within the envelope…let’s just think for a moment how likely that scenario is.

The 8 should be more slippery compared to the 7 just by way of frontal area. The difference in these two aircraft is the rudder….and it appears to be an area of concern in several accidents.

And yes to the guy that said the8 rudder looks cooler….yup….it’s better looking for sure. Fits the aircraft’s look better. :D
 
The 8 should be more slippery compared to the 7 just by way of frontal area. The difference in these two aircraft is the rudder
One of many differences.
- Frontal area, which you mention, is also different.
- Canopy configuration, affecting airflow around the tail, is also different.
- Landing gear configuration, which affects airflow around the wing enough that people put little flow guides ahead of the wing on the fuselage, something not needed on the -7.
I'm sure there are others.
 
Resolution

So how does this get resolved short of destroying two airplanes with flutter testing?

Can this be resolved by analysis? or does it turn into an endless debate like the primer wars?
 
Last edited:
Does that 8 have the same tendency for the rudder to get damaged in heavy winds from the elevator like the 7 does? I know ours had slight damage and had previously been replaced because of the same issue. I also saw one at Oshkosh was damaged in the storm in the same fashion.
 
Testing

The standard is ground vibration testing, not inflight testing with the risk of losing the airplane and possibly the test pilot. Ground vibration testing is very expensive.
 
The standard is ground vibration testing, not inflight testing with the risk of losing the airplane and possibly the test pilot. Ground vibration testing is very expensive.

Kermit Weeks has a great video doing this to his GeeBee..
 
Back
Top