Bird Strike Tests

I was recently introduced to a subject of which I had previously been blissfully ignorant. Sure, I knew in general about bird strike tests. What engineer hasn't chuckled at the tale of a chicken passing effortlessly through an aircraft windshield because the testers forgot to thaw it out?

Lucky me, I recently had the good fortune of watching several shots using freshly "euthanized" chickens. No fear of these birds being frozen. It was so hot there was an evaporative fan keeping them cool. No such luck for the test observers. The reason for the fans was not only to keep the chickens alive, but also to make sure they didn't lose too much weight in the hot southwest sun. That's because they had to be within a certain weight in order for the test to be valid, and you can only make minor adjustments by injecting fluid or clipping off toes. These tests can also be performed using gel packs, but not all certifying agencies will accept them (the argument is that the consistency is not enough like a bird).

There is also a velocity requirement for the bird hitting the target. What is the target speed? Well, it depends. A lot. On what does it depend, you might ask? Good question. Depends on the certifying agency and the intended use. If that sounds vague, then you are doing a good job keeping up with me. So through some less-than-exact science, a target weight and velocity are determined. It is not clear exactly how the OEM is supposed to insure that no over-weight, or fast flying birds stray into the flight path of the craft once it has been certified to a lesser threat.

And since we have requirements, we have to have tolerances. The birds are weighed to the gram and adjusted appropriately. The velocity tolerance is normally around 2-percent of the target speed. So, now we are ready to perform a test. We have a recently deceased chicken that weighs maybe anywhere from 3 - 7 lbs (depending on your requirements), but is weighed to the gram. We then attempt to push this chicken through a 30' tube using compressed nitrogen (think "punkin chunkin") and hit a target within a few inches at a speed within a few MPH (or knots). If your strike is too slow, it doesn't count. If it is too fast, and you break something, your test is worthless. If you happen to be outside, your shots will be affected by the temperature AND you will have to maintain your test article at a specific temperature. Even if you are inside, chicken chunkin is not a highly repeatable process.

Which brings me to my point (Ha, I bet you thought I didn't have one!) As a person new to this process, I'd just like to say, "Are you people crazy!" What bunch of eggheads dreamed up this process, and have you not stepped back in the last 3 or 4 decades to see what a ridiculous house of cards you have built? One of the first things we learn as engineers is the concept of significant digits. If you are using data with only one decimal of accuracy, your answer can never be any better than one decimal. But the aerospace industry is spending millions of dollars every year shooting chickens out of a cannon trying to meet requirements that all started with someone's best guess about what size bird will hit a plane at a predetermined velocity. With that kind of fidelity as the input, we are wasting a lot of resources trying to introduce chickens to windshields. Develop a standard impact test that can be performed in a lab. Use energy calculations. Come on people! And to the guys that think a gel pack can't be made to adequately represent a bird, I say let's talk about Neal Armstrong walking on the moon. Or let's talk about the fact that I have a 99-cent App on my iPhone that emulates my HP 41CV. Seriously, we aren't smart enough to do better than chicken chuckin? I don't believe it.