Paul Foster <pfoster@gdi.net> wrote:
> This caused the times to be quite erratic and nonlinear. The best example was
> Chuck Sample - then a member of the SEB. His 3 runs varied by 8 seconds
> although he swore that they should have been within a second of each
> other. He protested and lost. But he was right. But he was also wrong
> because there was nothing that could really be done about it...
So what if he had had a datalogger in the car (like the Edelbrock QwikData, or a
GEEZCube) that could have shown otherwise?
This is related to an issue that has come up in my datalogging adventures. It's
really very difficult to determine _exactly_ where a run starts or stops,
especially on the typical Nationals course that has a tight turn before the
start lights. (ProSolo is a little easier, the run starts when acceleration
starts + reaction time)
I've been tossing around the idea of setting up one of those road race timing
beacons at the start and finish lights, putting a sensor on the car, and hooking
it up to a channel in the datalogger. It then puts a mark in the data that
corresponds to where the run starts and ends.
To help defer the cost of the beacons + tripods, I'd share. Throw a couple of
bucks at me, I tell you the frequency, and from then on you get to use my
beacons at every race I bring them to (and it saves having a huge cluster of
beacons at the start/finish) Or maybe we convince the National Office to bring
them around to every National Tour/ProSolo and make it part of their setup.
Or if everybody brings their own beacons, maybe we set them up to do segment
times. :)
There are safety and performance issues to work through, but I'm pretty sure
they could be solved. These beacons work across great big road race tracks;
there's more than enough range in them to put the transmitter in a safe place.
(Just gotta watch out for cross talk)
The nature of the receiver is simple enough that I'd bet GEEZ! customers could
convince Byron to add a jack to the GEEZCube and the software changes to support
it.
Ok, so now I have a record of my run. It won't be _exactly_ the same as the
official timers (for one, the beacon sensor needs to be placed as close as
possible to the part of the car that actually breaks the timing beam, for the
other, the beacon tends to be more of a "cone" than a "laser", so depending on
how far away the tripod is from the actual finish, the part of the cone that
trips the reciever may be a few feet away from the timing beam) but I would
expect it to be close, and repeatable for each car (although not necessarily
from car to car)
So if I get 4 runs, and the offical timing system records 50.100, 50.100,
50.100, 50.100, then I'd expect the logger to record something like 50.000,
50.000, 50.000, 50.000. Even something like 50.012, 50.023, 50.045, 50.032 would
be OK - I'd accept a little noise, up to a tenth or two.
But let's say we're in Chuck's shoes. The logger says 50.012, 50.023, 50.045,
50.032, but the timer says 52.000, 54.000, 58.000, 56.000. Is that enough
evidence to protest the results?
If so, where do you draw the line? Can I protest over .001 second's difference?
.050? .1? .5? 1?
Comments?
DG
|