The two photos below were taken during the first set of qualifying runs at the NHRA Gatornationals in Gainesville, FL in March 2002. They were taken with my (cheap) digital camera by using a feature that takes four pictures in a row at 0.5 second intervals. The originals were recorded in the camera's medium quality setting (287 kB file size for 1.2 megapixel) so that I could get a large number (about 170) of pictures on a single flash card. This pair of pictures resulted from lucky timing of one of quite a few photo sequences I took; see below for comments on that timing.
It was pure luck that one photo put the rear tires right on the start line, whereas it was pure physics that put them close to the 60' line in the photo taken 0.5 s later. I'll add that the next photo is empty.
What you see below is a result of cropping and annotation of the image with Photoshop while preserving the original pixel density. Since the camera was handheld, the images cannot match perfectly. However, it does seem that these two images are "registered" at the key points on the start line and 60 foot line to about 1 pixel, and the two lines shown on the photos are at exactly the same places (and, most importantly, the same distance apart) in both pictures. The lines are about 60 feet apart (see note below).
Photo taken at time T:
Photo taken at time T + 0.5 s:
Clearly the nearest vehicle has traveled almost exactly 60' in 0.5 s, which is an average speed of 120 ft/s = 82 mph. Normally we would associate this speed with the time midway between these two photos, which is when the nose of the car has traveled about 60'. Interesting?
We can learn more. You can also see that they have traveled about one car length (about 30') between time 0 and time T, since they started with the front wheels just short of the start line. This gives us x(T) = 30' and x(T+0.5) = 90'. If we make the incorrect assumption that the acceleration is constant, we have two equations in the unknowns T and a (since v is initially 0). Solving these gives an acceleration a = 129 ft/s2 and T = 0.683 s for photo 1. This acceleration is 4 times the acceleration of gravity.
This result, an acceleration of about 4 G, is consistent with other estimates of the acceleration of a top-fuel dragster. Another internal check is my estimate of the 60' time. My result says it should be about 0.96 s, which may be a bit high. I did not hear 60' times for the run I photographed, but they tend to be around 0.9 s (they can be as small as 0.871 s for a good run, with an acceleration of 4.9 G). However, the run you see was not very good; both drivers shut down soon after the start with tire shake, so my results could be accurate.
Dragstrips have more than just start and finish lights to measure the elapsed time (ET) over a quarter mile (1320 ft). There are timing lights 60 feet from the start line and also at the "half track" (660 ft) distance. The 60' time is used by the teams (and spectators) to evaluate the initial acceleration of the vehicle, while the 660' line is there for use in 1/8 mile races. [In addition, there are also lights that provide a speed trap at the finish and at the 1/8 mile point. IIRC these give an average speed over a 100' distance.] Here I am using the start line (best seen in photo 2) and the 60' line (best seen in photo 1) to provide a common horizontal distance scale on the photos.
There is a significant time delay due to autofocus and exposure functions in the camera. These can be minimized by pre-setting them by holding the shutter button partway but not taking a photo. There is still a problem with camera response and reaction time. Tripping the shutter when I saw the yellow lights come on was often too late (but did give the pictures shown above), whereas tripping it when I saw the second stage lights come on seemed to give a nice sequence with a good chance at getting the moment of the launch. (If I happened to push the shutter at the same time the starter hits his, then the second photo might be just 0.1 s into the green light.)
if you have any questions.
My TCC home page.