ANd how the hell does the lamptimer keep drifting? You say KAZG never signs off
at exactly the same time.
I'd love an actual serious, technical response from someone.....
..Of course, the usual funny responses will do as well, as I'm sure there are some;)
Serious or funny? You be the judge...
From whenever I've tracked it over an extended period such as a week
or more, the lamptimer loses about :08 every 24 hours.
If one day's sign-off is at 5:31:00, the next day is 5:30:52, then back to
5:30:44, etc.
Similarly, a sign-on at 7:01:00 slides backwards over the next few days
to 7:00:52, 7:00:44, etc.
Depending on how (in)accurate the lamptimer is at the start of a month,
it can, and has, drifted enough by late in the month so sign-on can occur
5:00 or more earlier than the specified time. This error has even continued
into the next month--July's monthly sign-on time is 5:30, but on July 2
KAZG popped on at 5:24:02 (atomic clock time), nearly six minutes early.
I think it's reasonable to give a station a window of up to 2:00 early at
sign-on and up to 2:00 late at sign-off. I say that because, for example,
during a month with a TOH sign-on--such as November--a daytimer might
want to do a formal sign-on announcement (KAZG doesn't even do the bare
minimum of a legal ID!

), play the National Anthem, and hit a network
newscast on the hour. All of which could take up to two minutes.
Akbar's right. If Gumpdusky can't get a new, more accurate lamptimer,
then they really need to go the "clap on...clap off" route.

And have
the Clapper trigger a relay that overrides the music automation and airs
a legal ID.