PPRuNe Forums - View Single Post - TACA aircraft crashed in Honduras
View Single Post
Old 6th Jun 2008, 16:03
  #175 (permalink)  
DingerX
 
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes on 0 Posts
Dear Simon,

I was merely giving the methodology for "how" those numbers were arrived at, not making any statement about their veracity; indeed I did note the problems with moving cameras and angular speed.
For the record, the "plane-lenth" measurement involved counting those lengths traveled with reference to the ground, in each frame, not taking a general measurement.
I also agree that as the aircraft moves away from the camera, it will look slower than what it suggests.
I do not find your initial arguments conclusive, namely that if it were going at 100 kts at that point, it would not have overrun. That's putting the horse in front of the carriage, or using analysis to drive the facts.
You did, however, inspire me to go back and take a close look at the data.

First, you are factually wrong. The tape is time-stamped to the thousandth of a second. Second, the camera's perspective is good enough that, even with the panning, you can overlay two images without noticeable problems from distortion, which I've done. Third, you are intuitively correct: at the point it comes on the screen, it is going notably faster than 100 kts.

I took the first and third frames where the aircraft appears. The time-stamp of the first frame ends with 18.928; the third frame is stamped 20.421, so the interval is .593 seconds. Matching the vertical section of the first image that contains the aircraft with the scene in the second image, it is clear that the aircraft has traveled on the runway over a plane-length.
Assuming conservatively that's forty meters, that would come out 243 kph or 131 kts. I'm sure greater precision (and perhaps an even higher ground speed) can be had.
If you'd like to play with the image, and make the necessary angular motion calculations, or just see for yourself or post for others, drop me a line and I'll mail it to you.

So: 1) camera motion can be compensated for, 2) The image gives us precise time measurements, and 3) you'd still need to calculate angular effects, but back-of-the-envelope shows it's at least 130 kts.
DingerX is offline