| my account | login-logout | resources | classroom help | support | catalog | home | get webcard |

 search | help desk | commons
 » Online Classroom   »   » Public Discussion of Cel Nav   » HBS - Fit-slope question

Author Topic: HBS - Fit-slope question
 Rich A posted March 21, 2017 11:16 PM                   Is there an expectation that we do a Fit-slope analysis for each set of sights? If so, I need some help understanding it. This is what I believe to be the process (i.e. for four sun sights):1) plot each sight's Hs and WT on graph paper (like Fig. A4, pg. 101)2) then compute the Hc for the 1st and 4th sight, and plot to establish a trend line (using the same DR for both)3) results -a. if none fall on trend line, then let your conscious be your guideb. if two or more fall on line, then average those (I have questions on that process, too)Regardless, it is mostly your judgment call on what sight to use (based on the perceived quality of the sight). Am I off on this?Also, an Hc computation was mentioned that can be found in Appendix A5. I could not find this computation. From: Lacey
 David Burch posted March 22, 2017 12:15 AM                   Message: Is there an expectation that we do a Fit-slope analysis for each set of sights? If so, I need some help understanding it. This is what I believe to be the process (i.e. for four sun sights):=== no, you do not have to fit slope on all sights, and in fact you would not underway either. This is just for times with poor data or times when you want the very best fix... often not much of an improvement... other times big improvement1) plot each sight's Hs and WT on graph paper (like Fig. A4, pg. 101)==== yes2) then compute the Hc for the 1st and 4th sight, and plot to establish a trend line (using the same DR for both)===== you can use any convenient time just before and just after last sight. ie round off the seconds. BUT must be at the right DR for each end. ie NOT from same DR, but from the DR at the times you use.3) results -a. if none fall on trend line, then let your conscious be your guide==== chances are no sights fall on the computed line as Hs you have plotted has no corrections applied. ====use parallel rulers to move the computed line to the line of data for the best fitb. if two or more fall on line, then average those (I have questions on that process, too)==== then once you moved the computed line to the best fit, almost certainly one will be on the line, and you can use it. If in fact none are on the line, just choose any Hs, Wt that is on the line and use it.Regardless, it is mostly your judgment call on what sight to use (based on the perceived quality of the sight). Am I off on this?=== we are not using perceived quality of individual sights for this. this step is just fitting a curve. If you have isolated sight or sights that look well off the slope, then chances are these are not as good as the others that have the right slope.==== back to your point 1). You would do this often enough to feel confident in the process.---------------------------------------------------Also, an Hc computation was mentioned that can be found in Appendix A5. I could not find this computation.==== please cite the specific reference you refer to ie page and column and we will look into this.There is a long section in the back of the book about this process, and there are numerous examples. also we have an article online about the process that might help http://davidburchnavigation.blogspot.com/2014/04/fit-slope-method-shortcut-to-manual. html From: Starpath, Seattle, WA
 Rich A posted March 22, 2017 12:30 AM                   thanks David ... and the computation reference is on pg. 100, 3rd para on right side From: Lacey

 All times are Pacific