Aircraft Impact Damage on Movable Concrete Block vs. Non-movable Concrete Block

[Bittersweet]

The following excerpt is taken from a Swiss Federal Nuclear Safety Inspectorate (HSK) publication[1]:

“In the year 1988 a large scale crash test was performed at the US Sandia National Laboratory in which a Phantom military jet with a weight of 19 tons was impacted at a velocity of 774 km/h into an essentially rigid reinforced concrete wall. This was accomplished by using a two-rail rocket sled facility. The test was monitored by the American safety officials of the NRC. The test results were used to evaluate the “Riera approach” for prediction of the impact force-time history of an aircraft impact on a rigid target. The Riera approach was developed at the end of the 1960's in order to determine the effects of an aircraft crash on nuclear power plants. The target consisted of a block of reinforced concrete 7 m square and 3.66 m thick with a total mass of 469 tons (almost 25 times the weight of the military jet). At impact a portion of each wing and tail was sheared off. The remainder of the aircraft was completely destroyed during the impact. Pieces were dispersed over a large area; the dispersion of the water which was to simulate the kerosene, however, was relatively small. Due to the fact that the reinforced concrete wall was mounted on top of an airbearing platform, it was only slightly damaged [my emphasis] – with concrete spalling at the front face of the target. The penetration depth caused by the engines was 60 mm and that caused by the fuselage was 20 mm.”

Video of test available here:

Researchers at the Sandia National Laboratory published a two-part review of the experiment[2][3], in which they noted the following: “Damage to the target was relatively minor indicating that the major portion of the impact energy went into movement of the target and not in producing structural damage.”

I am curious, if the concrete target had not been mounted on top of an airbearing platform but was instead imbedded into the ground, would the structural damage sustained have been significantly greater? If so, could anyone replying to this post please explain why? The fact that the F4 Phantom is much less massive than the concrete target leads me to believe that the difference would probably be trivial, but I am uncertain, hence this post!

I have included some basic information below, however, more detailed data and a comprehensive analysis of the experiment can be found in references [2] and [3].

Modified (some avionics removed, addition of five rocket casings and water to simulate fuel) F4 Phantom impact weight: 12.7 tonnes

Concrete target: 7 m square and 3.66 m thick atop an air-bearing platform with a combined weight of 469 tonnes.

Friction: “insignificant”

Force required to initiate movement of concrete target: 816kg (less than 0.2% of the weight of the target)

Impact velocity: 215 m/s

Penetration depth of engines: 60 mm

Penetration depth of fuselage: 20 mm

Horizontal displacement of concrete target: 1.83 m

References

1. HSK (2003). “Position of the Swiss Federal Nuclear Safety Inspectorate regarding the Safety of the Swiss Nuclear Power Plants in the Event of an Intentional Aircraft Crash”, http://www.ensi.ch/fileadmin/english/files/gus_03_04_03_e.pdf.

2. W. A. von Riesemann et al (1989). “Full-Scale Aircraft Impact Test for Evaluation of Impact Forces Part 1: Test Plan, Test Method and Test Results”, http://www.iasmirt.org/iasmirt-3/SMiRT10/DC_250400 .

3. K. Muto et al (1989). “Full-Scale Aircraft Impact Test for Evaluation of Impact Forces Part 2: Analysis of the Results”, http://www.iasmirt.org/iasmirt-3/SMiRT10/DC_250401 .

 

[AlephZero]

I haven't looked at your references and video, but this is a general comment about imapact damage.

The impact will create a "stress wave" of compression which travels through the block at the speed of sound (a few km/s, for concrete).

If the block was free to move, the energy in the wave will tend to be smeared out into the global acceleration of the whole block.

If the block was fixed, the compressive wave would not dissipate so much, and most of the energy would be reflected from the back face of the block and return as a tensile wave. The tensile return wave would cause most of the damage, since most materials, and especially concrete, fail at a lower stress level in tension than in compression.

 

[Bittersweet]

@AlephZero

Thank you for responding, your post is very useful—I really appreciate the help.

If the block was fixed, the compressive wave would not dissipate so much, and most of the energy would be reflected from the back face of the block and return as a tensile wave.

Is it possible (without having to resort to FEA or something) to quantitatively describe the impact this difference would have on the damage sustained by the concrete target? After an internet search, I stumbled across this discussion:

Whether the impact block in the 1988 Sandia F-4 Phantom crash test moved or not makes very little difference. I will tell you exactly how little difference. The Sandia test was performed much the same way one would do the classic Ballistic Pendulum experiment for measuring the impact speed of a bullet: a block of wood is suspended by strings, so that there is no external force, such as friction, acting along the line of impact -- the same was achieved (nearly) by putting the reinforced concrete block atop an air-bearing platform. Quoting from the report,

"The target consisted of a block of reinforced concrete 7 m square and 3.66 m thick mounted atop an air-bearing platform with a combined weight of 469 tonnes (almost 25 times the weight of the F-4) Ten air bearings were installed in "pockets" in the lower surface of the air-bearing platform. After inflating the air-bearings, a force of only 816 Kg (less than 0.2% of the weight of the target) was required to initiate movement of the target."

When the bullet in the Ballistic Pendulum experiment hits the block of wood, it stops within it and thus transfers all its kinetic energy to the combined block-with-bullet mass. The movement of the block-with-bullet mass can then be used to calculate the initial speed of the bullet, or -- and this is the important part -- if you know the speed of the bullet, it can be used to measure the amount of energy expended in different ways in the collision. It turns out that in the case of such an inelastic collision, its easy to demonstrate that the fraction of energy going into destruction of the colliding objects is simply the total energy (i.e. initial kinetic energy), minus the ratio of the mass of the bullet (or the F-4 Phantom) versus that of the target-with-bullet [my emphasis]. For example, a 5-gram bullet hitting a 2000 gram block of wood will result in ( 1 - 5/2005 = ) 99.75 % of the total impact energy converted to destructive energy (mechanical crushing, heat, shrapnel spray, etc.). Similarly for the Sandia impact test, where the ratio of airplane mass to concrete block mass was 1-to-25, the amount of impact energy converted into destructive energy is ( 1 - 1/26 = ) 96.2 % of the total. This may be compared to the case where the concrete block had been fixed perfectly to the ground, which is exactly analogous to having a block with infinite mass. In this case the amount of impact energy converted into destructive energy is ( 1 - 1/10000000... = ) 100 % of the total. Note that there is a difference of only ( 100% - 96.2% = ) 4 % between the case where the block is fixed to the ground (infinite mass equivalent), and where it is floating frictionlessly. That of course is because of the large mass difference in both cases.

Source: http://health.phys.iit.edu/extended_archive/0111/msg00463.html

Is the above method for calculating the % transfer into destructive energy accurate? Is there only a 4% difference in impact energy conversion to destructive energy, or is the above method inaccurate or inapplicable? Any further help would be much appreciated.

 

[AlephZero]

It is correct so far as it goes, but it isn't the complete story.

The damage is not caused by the amount of energy tranferred to the block, but the rate of transferrring energy at each particular point in the block. That rate depends on how the "stress waves" propagate through the block.

As a thought experiment, consider dropping the same concrete block onto the ground from a height of 2000m. Ignoring air resistance, it will take about 20 seconds to reach the ground at a speed of about 200 m/s (i.e. similar to the plane's velocity in the test).

It should be clear (from common sense experience) that there would be no damage caused during the fall, even though the total energy in the block would be much greater than in the impact test. The damage would only occur on impact, when there is a suddent transfer of energy from the block to the ground.

Also, the ballistic pendulum calculation ignores the amount of energy consumed in "completely destroying" the plane. In a ballistic pendulum experiment, the bullet should be completely contained by the wood block, so that all the momentum is transferred, but that does not apply to this crash test. I would guess the damage to the plane would be similar whether or not the block was fixed, but in both cases the ballistic pendulum calcalation would over-estimate the energy transferred to the block.

 

[Bittersweet]

It is correct so far as it goes, but it isn't the complete story.

The damage is not caused by the amount of energy tranferred to the block, but the rate of transferrring energy at each particular point in the block. That rate depends on how the "stress waves" propagate through the block.

As a thought experiment, consider dropping the same concrete block onto the ground from a height of 2000m. Ignoring air resistance, it will take about 20 seconds to reach the ground at a speed of about 200 m/s (i.e. similar to the plane's velocity in the test).

It should be clear (from common sense experience) that there would be no damage caused during the fall, even though the total energy in the block would be much greater than in the impact test. The damage would only occur on impact, when there is a suddent transfer of energy from the block to the ground.

Also, the ballistic pendulum calculation ignores the amount of energy consumed in "completely destroying" the plane. In a ballistic pendulum experiment, the bullet should be completely contained by the wood block, so that all the momentum is transferred, but that does not apply to this crash test. I would guess the damage to the plane would be similar whether or not the block was fixed, but in both cases the ballistic pendulum calcalation would over-estimate the energy transferred to the block.

Thank you very much for your help, I appreciate it.