[Prev][Index][Thread]
Re: red light runners
In a previous article, gn1@umail.umd.edu () says:
>In article <1995Sep6.33449.30@tcs.wap.org>,earl.shoop@tcs.wap.org (Earl Shoop)
>wrote:
>>... Just how close must you be, at the limit, to proceed when the light turns
>>yellow? Two things might be helpful... 1) a yellow line (or mark) to
>>indicate that you should stop unless you have already passed this point,
>>and...
>
>I've always believed that the point at which the dashed lines that mark the
>lane change to solid is the indicator you're talking about.
>
>Can anyone confirm?
>
NO. That varies all over the place, and is NOT a standard.
The timing of the Yellow and All Red periods of traffic signals is one of
the most complex and misunderstood parts of traffic Engineering. The
Dilema Zone (ie Do I Stop or Speed Up?) is a real problem and has no real
good solution. Too many variables, such as the speed, perception time,
road conditions, width of the cross street, etc. make the calculation
difficult. What most engineers do is find a REASONABLE compromise.
In DC, when I was there in the 70's, all signals were set for a 3.5 to 4.5
second Yellow period, then an optional 1.4 to 2.4 second All Red. The odd
lengths were due to mechanical constraints of the timers (They had to be an
integral percentage of the overall cycle, 4% of 90=3.6, 5% of 80=4.0, etc.)
The cycle lengths varied from 60 to 120 seconds, with 80 being the norm.
DC now has solid state equipment at all of their signals, and the timing
can vary by 0.1 second increments for the Yellow, and, I think, 0.01
seconds for the All Red. They probably still use Yellows of 3.5 to 4.5
seconds, however.
Your Friendly Neighborhood Cyberspace Traffic Engineer,
References: