What can cause wrong results with eclipse finder?

[timcrews]

Hello:

If I run Celestia right now (1/30/2003, 5:32 UTC), go to Earth, track Jupiter, and zoom in, I can see both Ganymede and Io casting shadows on Jupiter. The shadow of Ganymede is about to leave Jupiter.

I get the same results using the Eclipse Finder, of course. It indicates that two eclipses are happening now.

So, the thing is, my telescope is pointed at Jupiter right now, and no such thing is happening. And what I _do_ see through my telescope precisely matches what SkyMap told me I would see.

So, my question is, what kinds of things can I have messed up that might have affected these phenomena? I started with Celestia 1.2.4, and added a lot of data to my solarsys.scc file, so when I upgraded to Celestia 1.2.5, I had some trouble merging the two. Here are the pertinent sections for Earth, Jupiter, Ganymede, and Io:

"Earth" "Sol"
{
Texture "earth.dds"
NightTexture "earthnight.dds"
SpecularTexture "earth-spec.dds"
Color [ 0.85 0.85 1.0 ]
SpecularColor [ 0.5 0.5 0.55 ]
SpecularPower 25.0
HazeColor [ 1 1 1 ]
HazeDensity 0.3
Radius 6378
Oblateness 0.003
RotationPeriod 23.9344694 # 23.93419
RotationOffset 280.5 # offset at default epoch J2000
Obliquity -23.45
Albedo 0.30

Atmosphere {
Height 80
Lower [ 0.5 0.5 0.65 ]
Upper [ 0.3 0.3 0.6 ]
Sky [ 0.3 0.6 0.9 ]
CloudHeight 9
CloudSpeed 100
CloudMap "earth-clouds.dds"
}

CustomOrbit "vsop87-earth"
EllipticalOrbit {
Period 1.0000
SemiMajorAxis 1.0000
Eccentricity 0.0167
Inclination 0.0001
AscendingNode 348.739
LongOfPericenter 102.947
MeanLongitude 100.464
}

}

"Jupiter" "Sol"
{
Texture "jupiter.jpg"
HazeColor [ 0.4 0.45 0.5 ]
HazeDensity 0.3

Radius 71398
Oblateness 0.064

CustomOrbit "jupiter"
EllipticalOrbit
{
Period 11.8622
SemiMajorAxis 5.2034
Eccentricity 0.0484
Inclination 1.3053
AscendingNode 100.556
LongOfPericenter 14.7539
MeanLongitude 34.404
}

RotationPeriod 9.92425
Obliquity 3.08
LongOfRotationAxis 0 # Not correct, but necessary with current
# calculations of satellite positions
# old values: 337.68, 157.68
Albedo 0.51

Rings
{
Inner 92000
Outer 221000
Texture "jupiter-rings.png"
}

}

"Io" "Sol/Jupiter"
{
Texture "io.dds"
Radius 1821

CustomOrbit "io"
EllipticalOrbit
{
Epoch 2443000.00038375
Period 1.769138
SemiMajorAxis 421600
Eccentricity 0.0041
Inclination 0.040
AscendingNode 312.981
LongOfPericenter 97.735
MeanLongitude 106.724
}

Albedo 0.61
}

"Ganymede" "Sol/Jupiter"
{
Texture "ganymede.dds"
Radius 2634

CustomOrbit "ganymede"
EllipticalOrbit
{
Epoch 2443000.00038375
Period 7.154553
SemiMajorAxis 1070000
Eccentricity 0.0015
Inclination 0.195
AscendingNode 119.841
LongOfPericenter 188.831
MeanLongitude 121.206
}

Albedo 0.42
}


Are there any other values that I can affect that might be causing this problem?

Thanks for any hints,

Tim Crews

[t00fri]

Hello:

If I run Celestia right now (1/30/2003, 5:32 UTC), go to Earth, track Jupiter, and zoom in, I can see both Ganymede and Io casting shadows on Jupiter. The shadow of Ganymede is about to leave Jupiter.

I get the same results using the Eclipse Finder, of course. It indicates that two eclipses are happening now.

So, the thing is, my telescope is pointed at Jupiter right now, and no such thing is happening. And what I _do_ see through my telescope precisely matches what SkyMap told me I would see.

So, my question is, what kinds of things can I have messed up that might have affected these phenomena? I started with Celestia 1.2.4, and added a lot of data to my solarsys.scc file, so when I upgraded to Celestia 1.2.5, I had some trouble merging the two. Here are the pertinent sections for Earth, Jupiter, Ganymede, and Io:

"Earth" "Sol"
{
Texture "earth.dds"
NightTexture "earthnight.dds"
SpecularTexture "earth-spec.dds"
Color [ 0.85 0.85 1.0 ]
SpecularColor [ 0.5 0.5 0.55 ]
SpecularPower 25.0
HazeColor [ 1 1 1 ]
HazeDensity 0.3
Radius 6378
Oblateness 0.003
RotationPeriod 23.9344694 # 23.93419
RotationOffset 280.5 # offset at default epoch J2000
Obliquity -23.45
Albedo 0.30

Atmosphere {
Height 80
Lower [ 0.5 0.5 0.65 ]
Upper [ 0.3 0.3 0.6 ]
Sky [ 0.3 0.6 0.9 ]
CloudHeight 9
CloudSpeed 100
CloudMap "earth-clouds.dds"
}

CustomOrbit "vsop87-earth"
EllipticalOrbit {
Period 1.0000
SemiMajorAxis 1.0000
Eccentricity 0.0167
Inclination 0.0001
AscendingNode 348.739
LongOfPericenter 102.947
MeanLongitude 100.464
}

}

"Jupiter" "Sol"
{
Texture "jupiter.jpg"
HazeColor [ 0.4 0.45 0.5 ]
HazeDensity 0.3

Radius 71398
Oblateness 0.064

CustomOrbit "jupiter"
EllipticalOrbit
{
Period 11.8622
SemiMajorAxis 5.2034
Eccentricity 0.0484
Inclination 1.3053
AscendingNode 100.556
LongOfPericenter 14.7539
MeanLongitude 34.404
}

RotationPeriod 9.92425
Obliquity 3.08
LongOfRotationAxis 0 # Not correct, but necessary with current
# calculations of satellite positions
# old values: 337.68, 157.68
Albedo 0.51

Rings
{
Inner 92000
Outer 221000
Texture "jupiter-rings.png"
}

}

"Io" "Sol/Jupiter"
{
Texture "io.dds"
Radius 1821

CustomOrbit "io"
EllipticalOrbit
{
Epoch 2443000.00038375
Period 1.769138
SemiMajorAxis 421600
Eccentricity 0.0041
Inclination 0.040
AscendingNode 312.981
LongOfPericenter 97.735
MeanLongitude 106.724
}

Albedo 0.61
}

"Ganymede" "Sol/Jupiter"
{
Texture "ganymede.dds"
Radius 2634

CustomOrbit "ganymede"
EllipticalOrbit
{
Epoch 2443000.00038375
Period 7.154553
SemiMajorAxis 1070000
Eccentricity 0.0015
Inclination 0.195
AscendingNode 119.841
LongOfPericenter 188.831
MeanLongitude 121.206
}

Albedo 0.42
}


Are there any other values that I can affect that might be causing this problem?

Thanks for any hints,

Tim Crews

Being an amateur astronomer, I am sure you took into account the fact that on 1/30/2003 5:32 UTC the light takes exactly 36 min 0 s to travel from Jupi to earth;-). That means when looking from earth at an event that Celestia predicted on Jupi at hh:mm:ss to occur, you will see it exactly 36 min 00 sec later in your telescope.

How about this?

Bye Fridger

[Kendrix]

And also be sure tu have a "good" 1.2.5 version installed... Your mix between 1.2.4 and 1.2.5 could give relly wrong results since orbit precisions have been significantly increased between the two releases...

And don't forget something... If your telescope has a to much small aperture or doesn't get a good magnification (zoom, sorry I'm french) you will not be able to see those eclipses !

[timcrews]

Hi guys:

Thanks for your answers.

The 36-minute light delay is not the problem. The event lasts 3 hours from start to finish (Ganymede eclipses Jupiter, Io eclipses Jupiter, Io leaves, Ganymede leaves). I checked several times over the next few hours, and never saw the event. Also, I can clearly see the four moons widely arrayed around Jupiter, in no danger of eclipsing it any time soon.

Aperture definitely isn't a problem. I spent quite a while last night observing the festoons on Jupiter's bands. I have seen eclipses on Jupiter several times before, but I have never seen a double eclipse, which is why I was excited about using the Eclipse Finder to plan an observing session.

So that leaves me with a hosed-up installation. I really hate to lose all of the work that I have done with my textures and solarsys.scc, which is why I really didn't want to start over with a new distribution. But I guess I will give it a try.

Can anyone say conclusively that the difference between actual and real is within the Celestia executable itself, and not due to an error in my solarsys.scc? Because even if I start with a new 1.2.5 distribution, I plan on copying my current solarsys.scc right over the one that comes with 1.2.5.

Thanks for your time,
Tim Crews

[t00fri]

Hi guys:

Thanks for your answers.

The 36-minute light delay is not the problem. The event lasts 3 hours from start to finish (Ganymede eclipses Jupiter, Io eclipses Jupiter, Io leaves, Ganymede leaves). I checked several times over the next few hours, and never saw the event. Also, I can clearly see the four moons widely arrayed around Jupiter, in no danger of eclipsing it any time soon.

Aperture definitely isn't a problem. I spent quite a while last night observing the festoons on Jupiter's bands. I have seen eclipses on Jupiter several times before, but I have never seen a double eclipse, which is why I was excited about using the Eclipse Finder to plan an observing session.

So that leaves me with a hosed-up installation. I really hate to lose all of the work that I have done with my textures and solarsys.scc, which is why I really didn't want to start over with a new distribution. But I guess I will give it a try.

Can anyone say conclusively that the difference between actual and real is within the Celestia executable itself, and not due to an error in my solarsys.scc? Because even if I start with a new 1.2.5 distribution, I plan on copying my current solarsys.scc right over the one that comes with 1.2.5.

Thanks for your time,
Tim Crews

I have just checked here in my office with XEphem: on your date there are NO moon shadows on Jupi in agreement with your observations. I have spent a lot of time in the past checking Celestia versus XEphem such that I am ready to bet that tonight when I get home, Celestia will not show an eclipse either!

I am convinced that you must have done something wrong, like e.g. messing up different parameter upgrades for the moons in solarsys.ssc...

Bye Fridger

[Calculus]

I think you don't have the latest version. I checked with version 1.2.5 (final) and Celestia is correct. There is no double eclipse on jan 30, 2003.
The eclipse finder gives 2 eclipses :Io (7:11) and Ganymede (12:38) and they don't overlap.
(If you reinstall celestia, make sure to uninstall the previous version).

P.S. here are all the Galilean events on Jan 30 from bdl , Celestia agree with them (beside light time delay)
7:51 I O.C. (begining eclipse)
7:56 I P.C.
9:33 II E.C.
10:08 I O.F. (ending eclipse)
10:13 I P.F.
12:35 II EM.
13:22 III O.C. (begining eclipse)
13:39 III P.C.
17:00 III O.F. (ending eclipse)
17:16 III P.F.

[selden]

Just a quick comment on "installing" and "deinstalling" Celestia under Windows: you don't need to actually do a "deinstall" using the "Add/Remove Programs" utility. Just rename the old Celestia folder so it has a different name before you do the new installation.

After you've installed the new copy of Celestia, copy that whole new Celestia directory tree to a different one with a different name. Always keep a copy of Celestia around that has not been modified in any way whatsoever.

I use names like Celestia124, Celestia1257, Celestia125Unmodified, etc.

For convenience, you can "drag a shortcut to the desktop" from each of the different Celestia.exe files using the rightmost mouse button.

I hope these suggestions help a little.

[timcrews]

Hello everyone:

Whatever is causing my problem, it is definitely in my solarsys.scc. If I start fresh with a 1.2.5 installation, the eclipses are shown correctly, the same as the reference values that selden provided. As soon as I copy over my solarsys.scc, the eclipses go wrong again.

I have very carefully compared the entries for Earth, Jupiter, and the Galilean satellites, and they match in everything except their textures and Jupiter's rings. Is there anyplace else I should be looking for mismatches?

Thanks,

Tim Crews

[selden]

Tim,

Actually, that was Paul "Calculus" who gave the times of the Eclipses.

You might want to consider doing a line-by-line comparison of the two files. You might be overlooking a typo. I know from experience that it's all too easy to read what's supposed to be there instead of what's actually there.

The Windows Command (or DOS) window provides the COMP command to compare the contents of two files. COMP /? gives a brief description of its qualifiers.

I hope this helps a little.

[Calculus]

I found your problem. It is Jupiter, you have to use its new CustomOrbit.

CustomOrbit "vsop87-jupiter"

instead of

CustomOrbit "jupiter"

[timcrews]

Even with the vsop87-jupiter orbit corrected, I get the wrong results. Different wrong results than before, but still wrong.

That's OK, for now I will just use an unmodified copy of 1.2.5 to perform my eclipse calculations, until I can carefully narrow down what the problem is with my modified copy.

In fact, I did exactly that to discover that a double eclipse would occur last night. I set up my telescope to take a look. What an anticlimax! Unfortunately, as you can see in Celestia too, Europa and Io were obscuring most of their own shadows on Jupiter, so that in effect I was looking at Europa's/Io's reflected light with Jupiter's reflected light as a backdrop. The seeing was bad last night in the Phoenix area, so both moons basically disappeared as they crossed Jupiter, with only the barest hint of shadows, only discernible with averted vision.

Maybe I'll try again when the eclipse shadows are a little more "out in the open".

Thank you all for your help.

Tim Crews

[Calculus]

In fact, I did exactly that to discover that a double eclipse would occur last night. I set up my telescope to take a look. What an anticlimax! Unfortunately, as you can see in Celestia too, Europa and Io were obscuring most of their own shadows on Jupiter, so that in effect I was looking at Europa's/Io's reflected light with Jupiter's reflected light as a backdrop. The seeing was bad last night in the Phoenix area, so both moons basically disappeared as they crossed Jupiter, with only the barest hint of shadows, only discernible with averted vision.

Maybe I'll try again when the eclipse shadows are a little more "out in the open".

Thank you all for your help.

Tim Crews

Tim, what you saw last night is in fact a pretty rare event. Io occulting its own shadow can be seen only when jupiter is at opposition.
It happens about only once every 2 years and since it lasts 2 hours, if the weather is nice, it is visible from your location (or any location on earth) only once every 4 years. If you take into account the weather, a 25% chance of clear sky, you might wait for another 15 years to see such event again.