Crashworthiness

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
13,080
Location
Port Townsend WA
Also, the NASA testing found that the sharp bottom firewall digs into the ground as shown in the later part of the video. The firewall should have a rounded bottom to slide like a sled instead of digging into the earth.
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,894
Location
Evansville, Indiana
Spine is the least of your problems-Organ displacement & inner bleed out.
Quite a high number die from cervical or high thoracic spine injuries. Very few die from "organ displacement" to the exclusion of all else (often there are two or more possible causes of death in aviation fatalities and it comes down to just picking one or listing them all).

Even the classic attribution of aortic injuries to the heart being tossed around the chest cavity is not entirely clear in all cases. There are at least a half dozen different identified mechanisms involved with that particular problem (including thoracic spine fracture with anterior displacement since the aorta is rather firmly secured in the descending thoracic segment to the anterior surface of the vertebral column). That said, the same measures that will reduce spinal injury will also reduce the forces on the internal organs that might lead to massive internal hemorrhage (the heart, kidneys, liver and spleen) since they are often more sensitive to loading along the vertical axis than the horizontal/longitudinal. Anything that increases the time (and distance) over which the deceleration occurs will reduce the load on both the organs and the associated bony structures.

The other reason why spinal injury should be a primary concern is the evidence supports a high correlation between traumatic incapacitation (either physical disability or unconsciousness) and post-crash fire-related death as a result of smoke inhalation, hypoxia or the direct impingement of the fire on the victim. A fairly substantial number of folks who survive the crash only to die from these factors are found to have spinal injuries.


If you built up the cockpit enough to largely hold shape in a serious nose down crash, the airplane may well become unflyable for weight, and the decels will still be huge and thus tough to make survivable;
It's actually doable but you wind up with something that looks like a cropduster since that's one reason they are shaped the way you are thanks to Fred Wieck and his colleagues at Texas A&M along with John Swearingen and Stan Mohler (FAA CAMI) and Hugh DeHaven (Cornell Medical College).

Cockpit has to be sturdy enough to hold shape fairly well - failing here means direct damage to the human;
Seat has to protect against vertical crash pulse - the seat and (perhaps the floor structure) are the only things that can do this and we lose ALL of our vertical velocity over a very short travel;

Harness system has to keep you in place - failure here means secondary collisions and direct damage to the human;
The acronym we use to teach folks about occupant protection is: CREEP
Cockpit- no more than a 10-15% volume reduction or reduction in any of the primary directional measurements
Restraints- keep you in place and do not fail under survivable loads. Honestly, that last point is where GA as a whole is sorely lacking.
Energy Absorption/Dissipation/Direction- expend the energy before it gets to the occupants or minimize it by maximizing how long it takes to bring the aircraft to a stop (which is why the bottom of the firewall should be angled/canted rearward at least 20-45 degrees to avoid "plowing")
Escape/evacuation- make sure the doors do not jam shut and keep the occupants from escaping a burning or sinking aircraft
Post-crash environment- Minimize the product of toxic smoke (by careful selection of materials), reduce the risk of a post crash fire by use of crashworthy fuel lines and tanks with breakaway valves that seal in the process, etc.

There is a lot of moment arm in a high seat that is bolted only from the bottom. Is there a way to secure the seat from higher ?
Yes, but then you get into other issues like reduction of usable space, changing load paths so far as the seat occupant is concerned, etc. It's a lot simpler and more practical in many cases to increase the strength and/or number of seat attachment points especially if you're designing 'clean sheet'.

Also in that Air France crash, there was some indication that the seat attachment points had been subject to fatigue cracking which reduced their strength. I don't recall if that was mentioned in the final report or not.
 

topspeed100

Banned
Joined
May 4, 2009
Messages
4,063
Location
Oulu/Finland
Also in that Air France crash, there was some indication that the seat attachment points had been subject to fatigue cracking which reduced their strength. I don't recall if that was mentioned in the final report or not.
Ok roger that. I found that pretty strange..since the airliner was still moving but slowly...the mass of the pilot could have played a role too ?
 

topspeed100

Banned
Joined
May 4, 2009
Messages
4,063
Location
Oulu/Finland
Always nice to review real data. Even nicer to have it well presented!
This is great vid thanks for sharing it.

I was just wondering how much the weariness of the planes has to do with the results..or were these kites new on the tests ?
 
Last edited by a moderator:

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,894
Location
Evansville, Indiana
I found that pretty strange..since the airliner was still moving but slowly...the mass of the pilot could have played a role too ?
Possibly. It would depend on how the shoulder harness are attached on a particular seat. That said, for the most part with aircraft seats, it really doesn't take much to break them loose especially in GA. The failure points (in terms of load) are often well below the human threshold for surviving without any significant injury let alone not suffering fatal injuries.
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,894
Location
Evansville, Indiana
I was just wondering how much the weariness of the planes has to do with the results..or were these kites new on the tests ?
There have been some done (especially the helicopters and the composite aircraft) with new and sometimes purpose built birds. The Langley crash test series goes back into the 1960s (reports available online) and cover a fairly broad swath. They even did one using a Beech Starship.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,684
Location
Saline Michigan
The test series on film was with new airframes. If all you do is bash tin and bolt it together, without worrying over systems and interiors and such, they cost a lot less. Same for engines, if you just use production scrap crankcases, and fill them with concrete to get to weight, the engines are much cheaper.

The video that John posted is great. IIRC, the series got its start when a maker (Beech?) found themselves with several airframes (Bonanzas?) that were not saleable for some reason, and NASA ended up with them, doing these sorts of tests at that great gantry. After that, it was pretty straightforward to get a program to get results with a wide range of airplanes.

I read in Air & Space that the test facility was recently de-commissioned.

Billski
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,195
Location
CT, USA
There were a bunch of airframes (I seem to recall Piper, but I could be mistaken) that were damaged in a flood... still solid but could not be made airworthy)... that were donated for crash testing.

One amazing thing, which I also remember seeing in equipment shock testing when I was working for the Navy, is how far things can move during an impact, sometimes far beyond the material's normal yield point, then spring back straight so you'd never know it bent. You'd see parts were damaged by being hit by other parts that couldn't possibly move far enough without breaking off... but they did.

-Dana

Cause of crash: Inadvertent contact with the ground.
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
13,080
Location
Port Townsend WA
Yep, I was impressed how the Cessna spring steel main gear legs bent past the belly than came back to normal almost.
Of course, that much spring back probably is NOT good for the occupants. Some amount of permanent deformation would reduce the shock load, I think.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,825
Location
US
Actually it doesn't...Here's some good reading.
Thanks, that product would be very handy. Seems it would be fairly easy to design a seat with good crash absorption properties using this aluminum honeycomb. Back of the envelope:
Occupant weight: 180 lbs
Horizontal seat area: 144 sq inches.
Attach the seat with bolts/pins that shear at a total vertical load of about 2100 lbs (approx 12 Gs)
Affix seat pan to slider tubes on rods/cables so it can only move down, not forward.
Put this aluminum honeycomb under the seat. We could just use the Plascore 1.3 PCF Foil Alloy 3003 material (25 PSI crush strength) in columns with a total cross-sectional area of 100 sq inches so that crushing would begin at 14 Gs. We could also make the columns into cones/pyramids so that the crushing required progressively more force. This would help reduce the chance of bottoming-out at higher loads (but would somewhat increase the chances of injury at lighter loads due to the "stiffer" energy absorption and reduced effective stroke length).
The problem is compounded by the relatively short stroke possible in most experimental aviation aircraft (5" is typical, much less than the 16"+ in helicopters). Well, we should do what we can . . .

A good paper on the types of energy absorption historically used in helicopter seats, and the progress that has been made). : http://www.fire.tc.faa.gov/2004Conference/files/crash/S.Desjardins_Energy_absorption-helicopter_seats.pdf
 

viva_peru

Member
Joined
Apr 12, 2012
Messages
13
Location
Detroit, Michigan
Vigilant,

That paper on helicopter seat design made for some very interesting reading. The plascore honeycomb material is used to construct the deformable barriers used in automotive testing (i.e. for side and front impact testing by both NHTSA and the IIHS). The material works well and it has a very predictable behavior; however, after it has been crushed, it does not always present a nice smooth surface. In most cases, it will have jagged edges which could cut easily cut skin. Just something to keep in mind.

There is also a foam alternative. Frangile polyurethane foams are sometimes used within door panels to help absorb and dissipate the energy in a side crash (from the occupant hitting the side panel). Unlike the foams we are familiar with, these frangible foams will crush at a fairly constant pressure and will not spring back, thus absorbing energy. I am guessing that you would probably need foam in the 6-8 lb/cuft density range in order to have something which will not crush under normal usage. I believe that the trade name for one of them is enerflex. Another thing to keep in mind is that unlike the honeycomb, the foam is not too sensitive to the direction in which the load is applied; it will pretty much behave the same way regardless of loading direction.

At any rate, very interesing discussion.

Teo
 
Last edited:

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,748
Location
Rotterdam, Netherlands
Note that you need some elastic deformation. Even during 10 mph taxi-ing you will see absolute massive accelerations (albeit with very small deflections). I'd guesstimate that you need at least half an inch or so of elastic deformation before you start hitting the crush zone.
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,894
Location
Evansville, Indiana
There were a bunch of airframes (I seem to recall Piper, but I could be mistaken) that were damaged in a flood... still solid but could not be made airworthy)... that were donated for crash testing.
That's how the NASA Langley test series got started in earnest. There are a few NACA/NASA tests that predate that (the earliest I am aware of is a vague reference to a test in the 1930s for which no one seems to be able to find the results).

Of course, that much spring back probably is NOT good for the occupants. Some amount of permanent deformation would reduce the shock load, I think.
One of the complicating factors for seat design is the recoil rate of the foam used for padding. Small things can make a big difference.

Seems it would be fairly easy to design a seat with good crash absorption properties using this aluminum honeycomb
It's not easy for a few reasons with the primary one being that many crashes have more than one vertical impulse involved (especially if the aircraft cartwheels or bounces). With energy-dissipating materials designed to work by crushing, one they've "shot their load" (the vaguely sexual term that gets used fairly often in discussions of this in the safety community; we're an odd bunch.....) then the occupant is exposed directly to whatever comes next. "Multi-shot" energy absorption is one of the goals of a lot of the research into seat design. This is one reason why landing gear and the subfloors of aircraft are often used to "isolate the cabin" so to speak from initial impact forces.

Frangile polyurethane foams are sometimes used within door panels to help absorb and dissipate the energy in a side crash
One of the problems with non-metal foams is the pyrolysis of their constituent materials into various nasty compounds (hydrogen fluoride, hydrogen cyanide, CO, etc). There are some that have very minimal toxic gas production but they tend to be quite expensive. Rohacell-S is the one that jumps to mind.

Note that you need some elastic deformation. Even during 10 mph taxi-ing you will see absolute massive accelerations (albeit with very small deflections). I'd guesstimate that you need at least half an inch or so of elastic deformation before you start hitting the crush zone.
...or you simply attach the energy-absorbing material to the bottom of the seat and leave a little clearance from the floor/walls. This negates that issue while still serving its intended purpose. ;)
 

viva_peru

Member
Joined
Apr 12, 2012
Messages
13
Location
Detroit, Michigan
Steve,

I agree with you that multiple impacts present a problem; you can build systems that will work when loaded the first time, but provide little protection if loaded a second time. An example of this, is seat belt spools: they can be designed to absorb energy as the belt is paid out, but once the belt is paid out, it is slack and of little use. Regarding the frangible foam, the one made by Woodbridge, known as enerflex, is used in vehicle interiors. I would assume that it vents a minimal amount of noxious gasses. If using it as a part of a seat cushion, you would still provide a soft and bouncy surface to sit on, so you will still a normal cushion over the foam.

Along with the possibility of multiple impacts; there is also the uncertainty regarding the direction of the impact. In this respect, having a BRS system can aliviate that issue. If properly deployed, the crash in a sense becomes more predictable and controlled, thus making the design of a "safety cage" easier. In the end, I don't think that a system can be designed to provide safety under all circumstances.

Teo
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,748
Location
Rotterdam, Netherlands
...or you simply attach the energy-absorbing material to the bottom of the seat and leave a little clearance from the floor/walls. This negates that issue while still serving its intended purpose. ;)
What would be the advantage of fitting it to the seat bottom instead of to the bottom? I'd think that, if attached to the bottom it's less likely to "tumble over forwards"?
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,894
Location
Evansville, Indiana
I would assume that it vents a minimal amount of noxious gasses.
You'd be surprised how lax a lot of the vehicle fire safety standards actually are.

n this respect, having a BRS system can aliviate that issue.
In theory. The problem is that most situations that lead to crashes are not amenable to simply including a ballistic parachute. This is one reason why Cirrus has a safety record that makes the Pinto look like a IIHS five-star crash test case.

In the end, I don't think that a system can be designed to provide safety under all circumstances.
It can't. But the circumstances that occur in real world crashes, while they tend to not lend themselves to crash testing as easily as the "let's belly in this plane that Piper handed us" scenarios the NASA folks have tested for the most part (not knocking the work done by my colleagues and friends there, just pointing out the limitations), are fairly predictable, reasonably straight-forward and can be designed for. You can shave 30-40% off of fatality rates simply be improved fuel tank design. That honestly is the "easiest" fix in all of these sorts of discussions that is likely to render a significant and noticeable impact (pun intended) upon the number of pilots and passengers killed annually. Once we get past that, we can start to focus on correcting the misguided notions that lead to pilots and builders underestimating the forces involved and the capability of the human body to survive those forces.
 

viva_peru

Member
Joined
Apr 12, 2012
Messages
13
Location
Detroit, Michigan
Steve,

I did not realize that you were concerned about the gases vented by the foam if ignited. Given that the pilot would be sitting on the foam, I think that he would be more concerned about the fire and the gases given out by the foam :grin:...

Before we go too far, I have always been interested in aviation and do a lot of reading on the subject (for the fun of it), but I am not a pilot (unless R/C counts) and I am not an airplane designer nor currently building one (although I have a set of plans for the Fly Baby and I have designed RC models). If it is of any help, I am an engineer by training and trade. At any rate, anything I say should be taken with a big grain of salt.

Regarding the BRS, I was under the impression that part of the poor record for the Cirrus was the reluctance of the pilot to pull the handle until it was too late. I believe that Peter Garrison wrote about it in Flying. At any rate, I can see that it would not work in many instances, but in those that it might, I believe it would make designing the structure to withstand the impact easier.

Teo
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,195
Location
CT, USA
Regarding the BRS, I was under the impression that part of the poor record for the Cirrus was the reluctance of the pilot to pull the handle until it was too late.
Part of it, maybe-- pulling that red handle is real expensive-- but you see a lot of cases where the pilot panics and pulls it too soon instead of flying the airplane. Another contributor is cases where the chute gives the pilot a false sense of security and continues on into a situation where he should have turned around.

-Dana

A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools.
 
2
Top