NavList:

To Peter, that one sight is off by 2 minutes or so perhaps because the sextant wasn't perpendicular, or i just didn't get the limb on the horizon correctly; in essence, operator error.  I really can't remember that particular sight, but if it doesn't fit the model, i usually discard it (being fully aware that it might not be wrong at all).  I have a good gut feeling for when a shot is spot on or a off a bit.  I can't really explain that feeling well however.

 

As for which sights I chose, and how I chose them.  I am far from scientific with these selections other than to say that if there was one that had an overly different intercept, I tossed it out (usually greater than 2.0 minutes) and used the next one on the list.

 

For the away from transit numbers I used the sights with the following times

 

1) 17:50:54

2) 52:46

3) 55:37

4) 57:21

5) 59:51

6) 18:01:00

7) 01:29

8) 02:05

9) 02:34

10) 03:04

11) 04:26

12) 05:52

13) 06:28

14) 07:37

15) 08:13

16) 09:33

17) 10:03

18) 11:01

19) 11:29

20) 12:05

21) 12:38

22) 13:20

23) 13:49

24) 14:22

25) 15:01

 

For transit the sights at the following times

1) 18:39:30

2) 40:25

3) 41:05

4) 41:37

5) 42:08

6) 45:18

7) 45:58

8) 47:06

9) 48:15

10) 49:24

11) 50:00

12) 50:34

13) 51:19

14) 51:45

15) 52:28

16) 52:59

17) 53:48

18) 54:52

19) 55:19

20) 55:53

21) 58:37

22) 59:23

23) 19:00:00

24) 01:35

 

Jeremy

Aug 09, 2009


Dear ALL, and especially dear Jeremy and Peter,




... A bit surprised here and there by some of your comments, I came to the conclusion that I had missed some important data as earlier submitted by you Jeremy.

And that is true ... I have based all my computations and results on exclusively the 33 shots performed within the time frame UT 9:39:40 - UT 10:01:35 as detailed in the "Moon near LAM" file.

So, I had not noticed - until a few minutes ago - the file "Moon Away from LAM"...

     **** which explains why I could not understand some comments about an observation totally outside the time frame indicated hereabove, which apparently is considered a "faulty one " for some reason. I did not process such observation which is in the file "Moon away from LAN" which I have not studied until now.

      ***** This also explains why I was surprised at Peter's comments about processing "all 30" shots, while I did process "33 shots" ... simply because Peter was referring to this "Moon away from LAN" file which does have 30 shots while the file "Moon Near LAM" carries 33 shots,
     
      ***** It also explains why Peter is wondering why we have so different SDEV in Longitude ( you have over 40 ' while I have 3.8 ' ) ... simply because we do not speak about the same set of measures.

---------

So, thank you Jeremy for your reply and results. Before I could comment in detail about what seems significant differences between your GPS values, Peter's results and mine, I would first notice the following.

First

Jeremy you indicate that, out of the 33 shots you have posted, you processed only 20. In order to have all results compared on the same basis, would you be so kind as to indicate which exact 20 shots you processed so that I can re-run my computations on your exact same set of values.

Second,

Peter please be so kind as to confirm that you did process all the 33 shots of the "Moon near LAN" file, and only these ones.

Last but not least :

     **** for those of us who use a "parabolic fitting" type of software, I would surmise that your software first identifies culmination time and height, from the curve shape, and then computes LAM as well as associated Height at time of LAM. Please be so kind as to indicate time elapsed in seconds between Culmination Time and LAM time. I get 2.9 seconds for this result.

     ***** when running Marcq Saint Hilaire method on 33 running fixes using Oberved position at LAM as DR position, I am getting back exactly to the same result within 0.1 NM (while giving all 33 observations exactly the same weight ). I would therefore think that, all your 33 data having been processed by 2 entirely different mathematical approach, this would somewhat reduce the risk of error since in both cases I get the same results.

So I would offer also that you rerun all 33 sights from your time of Transit + your position at Transit and confirm that your final result exactly matches your position found through a "parabolic apporach method". If such is the case, then we have a very interesting case to study since all three of us would get exactly the same own results through 2 different methods, with the only BIG problem being that our various results would not currently seem to match. WE would therefore need to dig further and I will offer all intercepts and azimut values recomputed from any of your LAM posiiton and time used as a basis for running fixes.

Hopefully I am clear enough on this topic ...

I might be late to reply to your answers since I am flying again in a few days, but I will certainly follow up on this subject.


Best Regards to you all

Antoine


Antoine M. Cou?tte


PS : George, thank you very much for your extremely kind word of "welcome aboard " . Well ... I have to confess the following : before being an Airline Pilot, I was an Aircraft Carrier Jet A/C Pilot ( French Navy and US Navy ), and before being an Aircraft Carrier Pilot, I was a Naval Officer on a destroyer for a number of years in many areas on our Planet, and at a time when ONLY Celestial Navigation was available.