Celestial Navigation Plugin Redux

[stelian]

Originally Posted by P_Dub

I think that you have the wrong Sights altitude entered. After checking with other sight I am arriving with the correct result as calculated by an other program. So their is no reason why you get this. Check your value with the time set correctly.
Bye

You're answering to me ? The times and altitudes are on the screenshot...

[stelian]

Originally Posted by rgleason

Using "Find" and Altitude set for the Star and enter the Lat/Long above:
Star - Altitude Entered
Alkaid 80.397926
Kochab 58.133196
Arcturus 66.507224
Sun 36.888867

Alkaid is wrong: should be 79.501993

[stelian]

But even if Alkaid is corrected, we're still 7 miles of the target when using the 3 star sights, 29 miles off target when using the Sun as well.

[rgleason]

This is easier to read with integrated text and screenshots. If you see something that needs to be changed please advise.



Armchair testing of Celestial Navigation

[rgleason]

Stelian. As I wrote at the top.

"Short Answer to my problem: I think the problem was the default setting of "Clock Offset: -10000 seconds"! This should be set at Default=0 when the plugin is installed! And I guess any new setting should be saved -IMHO"

[stelian]

Originally Posted by rgleason

This is easier to read with integrated text and screenshots. If you see something that needs to be changed please advise.



Armchair testing of Celestial Navigation

Originally Posted by rgleason

Alkaid 80.397926

should be 79.501993

Originally Posted by rgleason

Right click on WP, pick Properties set lat 41.35 long -71.10 Boston,Ma

Should be 42.35

Originally Posted by rgleason

move boat...

I don't think this is needed

[rgleason]

Ok, but notice in the sequence posted that I took a FIX and it was 42nm off.
Then I added a fourth Sight Alkaid, and then a FIX was 31nm.
Then I decided to turn off SUN because that was the outlier FIX=8nm!

Originally Posted by stelian

Alkaid is wrong: should be 79.501993

[stelian]

Originally Posted by rgleason

Stelian. As I wrote at the top.

"Short Answer to my problem: I think the problem was the default setting of "Clock Offset: -10000 seconds"! This should be set at Default=0 when the plugin is installed! And I guess any new setting should be saved -IMHO"

This setting is not saved, but its initial value should be 0. Are you able to reproduce the default offset = -10000 ? (by removing the config file perhaps ?)

[stelian]

Originally Posted by rgleason

Ok, but notice in the sequence posted that I took a FIX and it was 42nm off.
Then I added a fourth Sight Alkaid, and then a FIX was 31nm.
Then I decided to turn off SUN because that was the outlier FIX=8nm!

I see the same FIX errors as you, look at my previous replies !

Now, the question is whether this errors are normal for one reason or another, or if there is a bug somewhere.

[rgleason]

Originally Posted by stelian

Alkaid is wrong: should be 79.501993

Stelian,

I agree Alkaid should be 79.501993.
I found the TIME was incorrect, set to some settings from earlier! This is a PITA, I checked each time and had to change the time each time, but it does not seem to be remembered, and I am guessing I missed this Celestial Body when I was doing it.

I think that if you change the TIME TAB to something like
10/10/17 1800 (UTC) It should be Remembered!

Also shouldn't there be something on the TIME TAB page that says this is UTC Time?

------
However when I check CN with thatchange to Alkaid, I get an error message, "Measurement certainly likely too small, sight may not appear" and the Sight no longer appears.

It worked the first time and it showed, but then I went back and everything else is the same, but it does not appear and has that message.

Unfortunately, I believe there are still some problems.

[rgleason]

Right Click - "Move Boat to Lat\Long"

If this were available, we would NOT have to type in LAT/LONG each time in FIND.

Originally Posted by rgleason
Right click on WP, pick Properties set lat 41.35 long -71.10 Boston,Ma

Stelian

Should be 42.35

Correct. Yes, it should be! Where did you see this? Ah ha Line #3 -A typo, corrected. Thanks.

Armchair Test of Cel_Nav

-That is one reason to Move the Boat to exact Position. - Why shouldn't we have that ability in Opencpn? It is kind of basic. Also it is needed for DR anyway.

[rgleason]

[QUOTE=stelian;2496382]I see the same FIX errors as you, look at my previous replies ! [/Quote}

Stelian:

Now, the question is whether this errors are normal for one reason or another, or if there is a bug somewhere.

Actually Stelian, the FIX seems to be working. The changes are a result of
1. First FIX ---Fix1=42nm
2. Adding one more Celestial Body Arcturus. ---FIX2=31nm
3. Turning off the Sun because it was an outlier. ---FIX3=8nm

[rgleason]

The problems so far, I think, are:
1. Sights = "Measurement to small, sight will likely not appear." When we are using the value given by "FIND" which should work. Why will it draw without this error and then next time it shows this error and does not draw?
2. Remembering TIME.
3. Make it clear that we should be using UTC time.
4. Allow us to enter a position Lat\Long for "FIND" that is remembered rather than using the boat position? Or alternatively give us a much needed function for Right Click - Move boat to Lat/Long.
5. Really confirm that it is working. How big an error should we be getting? With the correction to Arcturus, the first time, when it wrote ok, my FIX got worse, like 45nm. Then when I went in to the sight again and checked all parameters, it gave me the message "Measurement too small, Sight will likely not draw." - I should try to delete it and start the sight over.... or just make a new sight.

[stelian]

That is one reason to Move the Boat to exact Position. - Why shouldn't we have that ability in Opencpn? It is kind of basic. Also it is needed for DR anyway.

You are correct, but I found somewhat confusing mixing different issues here. We should concentrate of fixing real bugs first (crashes, obvious bad calculations), then explaining some results (like the altitude fix we're discussing here, it might not even be a bug but a limitation somewhere), and finally improvements (UTC, boat position etc).

After reading the code and some thinking, I think the altitude FIX issue is not really a bug: when you're clicking on FIND, it will give the true altitude of the celestial body, not the instrumental one you're entering in the "Degrees" text field.

In other words, the altitude doesn't take into account the limb correction, the refraction etc etc. For example, take your sun measurement and set it to the center of the sun, you'll have a much better fix !

Finally, I don't think that FIND was meant to be used like you do. It was probably meant as a help for pre-setting the sextant to the good values in order to find the star and do the real measurement.

[stelian]

Originally Posted by rgleason

The problems so far, I think, are:
1. Sights = "Measurement to small, sight will likely not appear." When we are using the value given by "FIND" which should work. Why will it draw without this error and then next time it shows this error and does not draw?

This is indeed a problem, and I don't see it in Linux. It would be nice if you can find how to reproduce it easily, then I can search.

Originally Posted by rgleason

2. Remembering TIME.

I'm not sure we want this, or at least not by default. Using the current date/time seems more useful to me.

But maybe we could add a button in the main dialog for "Duplicate" a sight. This would allow to copy every parameter, including the time.

Originally Posted by rgleason

3. Make it clear that we should be using UTC time.

Agreed, and this is just a matter of properly labelling the text field.

Originally Posted by rgleason

4. Allow us to enter a position Lat\Long for "FIND" that is remembered rather than using the boat position? Or alternatively give us a much needed function for Right Click - Move boat to Lat/Long.

Move boat to Lat/Long seems more useful indeed. But this is outside the scope of this plugin.