[time-nuts] Setting clocks 100 years ago

[Hal Murray]

>> When did clocks get good enough to notice the analemma?

lists at rtty.us said:
> My guess would be the 1600's. They certainly had everything they would need
> before 1620. 

What happened in 1620 to indicate that people could measure the analemma?


One way to "discover" the analemma is to have a good clock and take a year 
long sequence of measurements of the location of the sun at noon.  So how 
accurate does that clock need to be?

I found a web page that said:
  For instance, on November 2nd the Sun is 16 minutes fast according to
  clock time and on February 11th it is 14 minutes slow. Mean solar time
  match on June 20th, April 14th, August 30th, and December 20th. 

If the clock was off by 1 second per day (number pulled out of the air), that 
would be 5 minutes per year.  That wouldn't close the path cleanly, but it 
would be good enough to show the idea.

When did clocks become good to 1 second per day for a year?

The wikipedia article says pendulum clocks were invented in 1656.  Harrison's 
H5 was good for 1/3 second per day in 1772.  That probably brackets things.


Another way would be to notice that the Earth's orbit wasn't a circle and do 
the calculations to figure out the path of the analemma.  I'm not sure when 
astronomers were good enough to do that.  It was probably a long time ago.  
Some of those old-timers would be right at home with a time-nuts discussion.  
 One of Harrison's competitors was using Jupiter's moons as a clock, and then 
had worked out the speed of light correction for the Earth's orbit.


Here is a recipe for constructing one:
  http://www.analemma.org/constructanalemma.html
I don't understand what's going on.


-- 
These are my opinions, not necessarily my employer's.  I hate spam.