Survivability lesson from the past

[Machinbird]

Recently, I was researching a question on the MG 08/15 when I stumbled across a picture in my Spandau folder which started me thinking about Survivability in WWI replica crashes. There has been some debate regarding whether to make one’s replica aircraft just exactly like it was in WWI, or to make it as good as modern material, methods and knowledge could make it. I personally am of the second opinion but the picture I found underscores for me the value of this view.
The picture I found was of Heinrich Gontermann’s Dr.1 taken around the cockpit area



Here was an aircraft built exactly like they built them during WW1. Here was an accident that could have been survivable had it not been for a partial collapse of the cockpit during deceleration and the extreme hazard created by the butts of the guns. Gontermann survived for a day but succumbed to his injuries. Gontermann was not just any German pilot. He was a nice looking fellow who happened to be the leader of Jasta 15 and the number two scoring German pilot (behind MvR of course) at the time of his death.
An excellent description of the accident circumstances can be seen in the following link to |-:-:-:-|1916|-:-:-:-|Koeniglich Preussische Jagdstaffel 2 "Boelcke"|-:-:-:-|1918|-:-:-:-|Broken Wings - The tragical death of Heinrich Gontermann.
From the pictures of the wreck it appears that the accident was potentially survivable but that the side walls of the cockpit buckled during deceleration which reduced the pilot’s separation from his guns and allowed his head and (possibly) upper torso to impact the gun buts. The shoulder harness does not appear to have failed. The aircraft struck an open field in a slight dive at moderate to high speed. There appears to be a ground scar of at least the length of the aircraft extending back in a direction approximately 120 degrees from the aircraft’s heading in the post crash pictures. See photo’s taken from the right rear of the aircraft.



It would appear that the aircraft was under partial control in the moments leading up to the ground collision in that the pilot was attempting to pull up and the wings were almost level based upon the witness description and the damage pattern. Since the aileron system appears to have been destroyed by the in flight breakup of the upper wing, it is probable that the pilot was using rudder to level his wings and may have touched down in a yaw.
As an aid to estimating deceleration rate, I prepared a velocity versus stopping distance chart for various rates of deceleration (in multiples of one g) in a range of speeds and accelerations that would encompass potentially survivable WWI crashes. http://home.comcast.net/~shademaker/GvsStop.xls The ground scar length is not known precisely but can be roughly estimated from the picture taken at the right rear of the wreck.
If we assume a 30 ft long ground scar and uniform deceleration, then we are looking at deceleration of between 7 g at a 80 mph touchdown to 16 g at 120 mph touchdown.
Design conditions for survivability. Survival during crash situations requires that the human body be situated in a protected volume of space that does not collapse up to some maximum design load, and which is not violated by outside objects during deceleration. (It does no good to hold the pilot off of the guns if the guns are instead are pushed back into the pilots protected space during a crash). Aircraft tend to experience heavy frontal loads during crash deceleration and may also experience significant vertical loads. Side loads are possible but usually are of a lower magnitude if the aircraft is under some level of control leading up to the accident.
Skinny structures under compressive loads tend to collapse by buckling sideways. The longer a particular structural element is, the easier it will collapse under compression. In the case of cockpit design, the cockpit bay longeron length is generally longer than the bays immediately behind and in front of the cockpit. This is real bad news for cockpit occupants. Further since the rear cockpit cross structure is generally canted, The upper longerons sections composing the cockpit tend to be longer than the lower ones and thus they begin to collapse before the lower longerons. What you as a cockpit occupant want is for the cockpit to be even stronger in compressive strength than the immediately adjacent bays. This can be done by using heavier wall tube in the cockpit longerons, by increasing the cockpit longeron diameter, by providing additional cross structure, by reinforcing the cockpit tubes with sleeves and doublers, and by installing internal stiffeners such carbon fiber internal reinforcements. These modifications need not be obtrusive or weight prohibitive. We merely must apply survivability concepts to aircraft which originally were designed with minimal understanding of survivability. If done well, these modifications will not detract from the historical example we are attempting to provide, but we will probably have to test representative fuselage examples to be assured we have met our design survivability goals.
I was working on this article before the recent unfortunate events involving two WWI replica aircraft. The particular example is applicable to steel tube fuselage aircraft, but also has some relevance to other forms of construction. I present this article with the hope that it will bring more builders to my viewpoint and cause useful discussion. There is still more that can and should be done to enhance survivability. I hope we can get into a gentlemanly discussion of what more should be done.

[womenfly2]

In the debate on another thread concerning seat belt attachment (Jeff Brooks), I was more concerned about what you show above. The weakest part of the aircraft is in the fuselage side tubing. Long span with little column strength. Even without hitting the gun butts, one maybe crushed anyway.

Look at the picture of the D.VII that bellied in, very low angle of impact but you can see deformation in the cockpit area too. That is were the focus should be first in safety.

One needs to increase column strength in that area. short it, bigger diameter and/or thicker wall tube will help.

In wood aircraft (Fred's Camel) wood splintering is even a bigger concern.

WF2

[snipe]

Originally Posted by womenfly2

In the debate on another thread concerning seat belt attachment (Jeff Brooks), I was more concerned about what you show above. The weakest part of the aircraft is in the fuselage side tubing. Long span with little column strength. Even without hitting the gun butts, one maybe crushed anyway.

Look at the picture of the D.VII that bellied in, very low angle of impact but you can see deformation in the cockpit area too. That is were the focus should be first in safety.

One needs to increase column strength in that area. short it, bigger diameter and/or thicker wall tube will help.

In wood aircraft (Fred's Camel) wood splintering is even a bigger concern.

WF2

Here are some poor quality stills captured from the TV news report of the Great War Flying Museum's D-VII that crashed at Geneseo NY on 13th July 2007. As you pointed out, look at the deformation around the cockpit.





Authenticity is good, but these airframes must be engineered and built with safety the primary concern.

[YavorD]

Originally Posted by Machinbird

...
As an aid to estimating deceleration rate, I prepared a velocity versus stopping distance chart for various rates of deceleration (in multiples of one g) in a range of speeds and accelerations that would encompass potentially survivable WWI crashes. http://home.comcast.net/~shademaker/GvsStop.xls The ground scar length is not known precisely but can be roughly estimated from the picture taken at the right rear of the wreck.
If we assume a 30 ft long ground scar and uniform deceleration, then we are looking at deceleration of between 7 g at a 80 mph touchdown to 16 g at 120 mph touchdown.
....

Hello Machinbird!
Thank you for the detailed crash analysis. I would like to point out at some assumptions involved. The formula used and numbers given above provide lower estimate for deceleration. There are at least two factors very hard to take into account:

1. Only velocity component parallel to the ground (horizontal) is accounted for. Vertical velocity component acts to increase absolute value of deceleration.
2. Average deceleration is used. Instantaneous value can be significantly higher, albeit for a very short time.

Regards,
Yavor

[Taz]

Machinbird- Your thoughts on building what is essentially a roll cage for WW-I aircraft has merit. Carbon fiber internal rods would prevent the steel tubing from collapsing, no doubt. The big trick would be figuring out how to weld the steel tube without burning the carbon fiber. Possibly filling it with water and providing a drain for post welding drain. Similar to the way gasoline tanks in cars were welded in the old days. As mentioned in the earlier post, shear pins for the MGs in the forward direction only would also cut down on the odds of eating a gun butt.

Taz
Terry Phillips