NavList:

   

Reply

Hello Robert,

thank you very much for your quick reply and your elegant solution. I didn't 
go "Duh", if only for the reason that I don't think I would have made your 
sharp observation.

Here's the rest of Mills deriviation, should the aforegoing have made you 
curious, that is, assuming you are not already familiar with it - admittedly 
a meagre reward for your effort, but I don't know what else I can offer right 
now.

Departing from where we left of at equation(4), quoting Mills:

"By the sine relationship:" (Mills is referring to the sine rule relating to pherical triangles)

"sin(HA)/cos(alt) = sin(Az)/cos(dec)

Therefore

sin x = {cos(dec).sin(Az)}/cos(dec)  (5)

From...we know that

cos(Az) = sin(dec)/cos(lat)  when the (alt) = O  (6)"

[Mills is rearranging cos(Az) = {sin(dec) - sin(lat).sin(alt)}/cos(lat).cos(alt)]

"so we can eliminate Az from (5) and (6) and so obtain an expression for x in terms of lat and dec.

..... by squaring and adding (5) and (6) we have

sin^2(Az) + cos^2(Az) = 1"    ( i.e. Mills applies trig. identities)

"and therefore,

{sin^2(x).cos^2(dec)}/cos^2(lat) + sin^2(dec)/cos^2(lat) =  1

therefore

   cos^2(lat) = sin^2(x).cos^2(dec) + sin^2(dec)

     sin^2(x) = {cos^2(lat) - sin^2(dec)}/cos^2(dec)

              = [cos^2(lat) - {1 - cos^2(dec)}]/ cos^2(dec)

           = {cos^2(dec) - 1 + 1}/ cos^2(dec)"

The above line appears to contain another mistype: I'd say it should read

           = [{cos^2(dec) - 1}/ cos^2(dec)] + 1

Mills proceeds:

"therefore

1 - sin^2(x) = cos^2(lat)/ cos^2(dec)"

In my opinion there is another mistype in this line. I am referring to 
"cos^2(lat)". It can't possibly have been put in deliberately. I'd say that 
the term should in fact be (1 - cos^2(lat) so that the entire line of the 
equation reads:

1 - sin^2(x) = {1 - cos^2(lat)}/ cos^2(dec)

Mills continues:

"cos x = sin(lat)/ cos(dec)"

I agree (most humbly).

Mills then closes the section with this observation:

"so the angle of rising or setting of a body depends only on lat and dec...."


Kind regards and many thanks once again,


Christian Scheele


















----- Original Message -----
From: berneckywr-mail@yahoo.com
To: NavList@fer3.com
Sent: Monday, January 25, 2010 2:09:24 AM GMT +02:00 Harare / Pretoria
Subject: [NavList] Re: H.R. Mills on setting/rising bodies



Hi Christian,
You are right, Mills did drop a minus sign in equation (3).
Getting to step 4 is not hard (I'm thinking you are going to think, "Duh"). 
But it's just not easy to always be clear when writing about math.

Here's one way to see what he did:

cos(alt).d(alt)= cos(lat).cos(dec).sin(HA).d(HA) (3)

Divide both sides of the equation by d(HA).cos(dec):

cos(alt).d(alt)/(d(HA).cos(dec))= cos(lat).sin(HA)

so that we can replace d(alt)/(d(HA).cos(dec)) on the left-hand side
with sin(x) (from equation 2). This gives us

cos(alt).sin(x)= cos(lat).sin(HA)

or, solving for sin(x):
sin(x)= cos(lat).sin(HA)/cos(alt)

and, by equation 2,
sin(x)= d(alt)/(d(HA).cos(dec))

so we have two different expressions for sin(x), and that
is what Mills is indicating with (4).


Robert Bernecky
Mystic CT



----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
----------------------------------------------------------------

   

Reply