Celestia Forums

This seems to be a rather interesting and useful addon (available from the utilities section of the Celestia Motherlode), but I've got some questions/comments I'm hoping that the author (bdm) can answer.

1) All the research I've found indicates that our own solar system's main asteroid belt is defined entirely by Jupiter - i.e. that all the main belt asteroids are between the 2:1 and 4:1 resonances with Jupiter. But the Asteroid Maker seems to define the inner border by a resonance with the next planet in and the outer border by a resonance with the jovian outside it. I've tweaked it on my own PC so that this is the case (just so it agrees with the numbers I have for my systems), but I was wondering where you got your limits from.

2) Aren't Hilda asteroids in a 3:2 resonance with Jupiter, not a 3:4 resonance? That's what my solar system dynamics book says anyway...

3) Since the star mass is a user-entered number in A12, why calculate it in A8 and J8 and get an "error" in F8 when the numbers don't agree? (that said I'm not entirely sure why F8 and H8 are supposed to be there anyway)

4) Is there an easy way to locate (in terms of semimajor axis) (a) the innermost asteroid, (b) the outermost asteroid, and (c) any Hilda-type asteroids?

5) Is there any way to control the "tightness" of the belt at all? The borders are somewhat fuzzily defined at the moment, I'd like to make the boundaries of the belt I'm generating a lot more sharply defined.

6) When I generated the SSC I had "#VALUE!" at the top of each asteroid entry, what was that there for? I just got rid of it along with the #BLANKs and it still seems to work...

Otherwise... it's a rather fine little program. Thanks for that!

I keep getting errors when I try to insert values for the Deneb system. any idea why?

Did you write him a PM?
See his profile to post one.

Adirondack

Adirondack wrote:Did you write him a PM?
See his profile to post one.

Adirondack

Yeah, I sent bdm a PM with a link to this thread. Hopefully he'll be around soon, I know he's still here since he posted on the boards recently .

More confusion here... take a look at the system contained in these files (unzip them all to the extras folder):

http://www.evildrganymede.net/temp/sextet.zip

You'll have to go to the "S12" barycentre to see this - if you select "Goto Object" from the Navigation menu when you're there and set the target as S12, Lat = 90, Lon = 0, dist = 10 AU then you can see the system from above.




So if you look at the sextet.ssc file, you'll see I don't have any of the fancy orbital parameters for the planets (P1-P5) set, I just have Period and Semimajor axis for them and that's it). In the asteroidmaker excel file I have told it that the Inclination and Ascending Node of the reference plane to the ecliptic are also 0.000.

However:

(a) if I set all the 'fancy' orbital parameters for the planets to be 0, then when I start Celestia shouldn't they all be in a line? Or is this view (with the planets at various points around their orbits at startup) what happens when you don't set the Epoch explicitly?

(b) The outer planet's (P4) trojans are on the opposite side of the orbit to where the planet itself is.

(c) I also tried to sharpen the inner boundary of the belt by reducing the peak eccentricity for asteroid orbits to 0.06 and inclination of the asteroids to 2 degrees. However, while the belt boundary does look a bit sharper than with the default settings, there are still many orbits in the generated SSC file that have inclinations and eccentricities that are much higher than these limits. Why would this be the case?

(Note that P2 and P3 are just placeholder orbits, there aren't really planets there since they are in the belt. Also note that the belt's borders are defined by the 4:1 and 2:1 resonance with the outer planet (P4)).

Celestia uses Y2K as the default Epoch, so planets without a specific Epoch, MeanAnomaly or MeanLongitude will be aligned on Jan 1, 2000.

I can't comment on your other questions.

selden wrote:Celestia uses Y2K as the default Epoch, so planets without a specific Epoch, MeanAnomaly or MeanLongitude will be aligned on Jan 1, 2000.

OK, thanks. Well, they align on 1/1/2000, but P4 rapidly goes out of sync with its trojans (within a few orbits). So evidently something is still not quite right here.

Malenfant wrote:This seems to be a rather interesting and useful addon (available from the utilities section of the Celestia Motherlode), but I've got some questions/comments I'm hoping that the author (bdm) can answer.

1) All the research I've found indicates that our own solar system's main asteroid belt is defined entirely by Jupiter - i.e. that all the main belt asteroids are between the 2:1 and 4:1 resonances with Jupiter. But the Asteroid Maker seems to define the inner border by a resonance with the next planet in and the outer border by a resonance with the jovian outside it. I've tweaked it on my own PC so that this is the case (just so it agrees with the numbers I have for my systems), but I was wondering where you got your limits from.

The inner limit is defined so the user has more control over the inner limit of the belt. This could be set arbitrarily small if desired, so a belt could produce Vulcanian asteroids.

The definitions for the inner planet allows a creator to create belts defined by two planets. In the unreleased version 1.7 (unreleased due to unresolved speed issues), this functionality is extended to allow the creation of Kuiper belts.

Malenfant wrote:2) Aren't Hilda asteroids in a 3:2 resonance with Jupiter, not a 3:4 resonance? That's what my solar system dynamics book says anyway...

I'll need to check this, but you may be correct here. If so I will need to correct this.

Malenfant wrote:3) Since the star mass is a user-entered number in A12, why calculate it in A8 and J8 and get an "error" in F8 when the numbers don't agree? (that said I'm not entirely sure why F8 and H8 are supposed to be there anyway)

The values in the top of the constants page were originally pasted in from SSC files. I think I provided this as a check to see if the implied values for the star mass were all in agreement with each other. If the SSC file was incorrect for some reason or the star value was incorrect, then an error would be raised here. I intend to remove this in the forthcoming 2.0 incarnation of Asteroid Maker (which will be a more efficient Windows application that should run fine under emulation on Linux).

Malenfant wrote:4) Is there an easy way to locate (in terms of semimajor axis) (a) the innermost asteroid, (b) the outermost asteroid, and (c) any Hilda-type asteroids?

I assume you mean the ones with the smallest and largest semimajor axes. Not as written, but you could use the MIN and MAX functions to insert your own lines. Create a cell that's equal to MIN(all the semimajor values) and then have a column that prints "SMALLEST" next to the cells that match the minimum value. The same can be done to find the largest value.

For Hildas, Trojans and other special orbits, the orbit family of the asteroid is printed in the SSC file.

Malenfant wrote:5) Is there any way to control the "tightness" of the belt at all? The borders are somewhat fuzzily defined at the moment, I'd like to make the boundaries of the belt I'm generating a lot more sharply defined.

I'm not sure I understand the question, but I'll answer as best I can. There is a "forbidden zone" that can be configured for the inner and outer planets that outlines the borders of the belt. The "fuzziness" may be due to the eccentric orbits that are generated. If you turn down the eccentricity of the orbits ("Peak Eccentricity for asteroid orbits" on the Constants page controls this) then you will get a sharper boundary.

Malenfant wrote:6) When I generated the SSC I had "#VALUE!" at the top of each asteroid entry, what was that there for? I just got rid of it along with the #BLANKs and it still seems to work...

I remember encountering and fixing this bug some time ago, but unfortunately I cannot remember the exact cause. There was a typo or something in one of the comment fields in the output file. Deleting it is harmless because it's not essential.

I will post details on how to rectify this later.

Malenfant wrote:Otherwise... it's a rather fine little program. Thanks for that!

I'm glad you appreciate it. I hope the answers I have provided are useful. If not, feel free to ask more questions.

Malenfant wrote:More confusion here... take a look at the system contained in these files (unzip them all to the extras folder):

http://www.evildrganymede.net/temp/sextet.zip

You'll have to go to the "S12" barycentre to see this - if you select "Goto Object" from the Navigation menu when you're there and set the target as S12, Lat = 90, Lon = 0, dist = 10 AU then you can see the system from above.



So if you look at the sextet.ssc file, you'll see I don't have any of the fancy orbital parameters for the planets (P1-P5) set, I just have Period and Semimajor axis for them and that's it). In the asteroidmaker excel file I have told it that the Inclination and Ascending Node of the reference plane to the ecliptic are also 0.000.

However:

(a) if I set all the 'fancy' orbital parameters for the planets to be 0, then when I start Celestia shouldn't they all be in a line? Or is this view (with the planets at various points around their orbits at startup) what happens when you don't set the Epoch explicitly?

Try setting the date in Celestia to be the default epoch date. Do they line up then?

Malenfant wrote:(b) The outer planet's (P4) trojans are on the opposite side of the orbit to where the planet itself is.

Can you post the settings you used in Asteroid Maker? I would like to try this for myself.

Malenfant wrote:(c) I also tried to sharpen the inner boundary of the belt by reducing the peak eccentricity for asteroid orbits to 0.06 and inclination of the asteroids to 2 degrees. However, while the belt boundary does look a bit sharper than with the default settings, there are still many orbits in the generated SSC file that have inclinations and eccentricities that are much higher than these limits. Why would this be the case?

I use a Rayleigh distribution for the eccentricities and inclinations. The limit here is not the maximum, but the peak of the Rayleigh distribution. Some values could be much greater than the peak.

Adding a limit for eccentricity and inclination should be an easy enhancement to implement.

PlutonianEmpire wrote:I keep getting errors when I try to insert values for the Deneb system. any idea why?

Can you provide more information?

Malenfant wrote:

selden wrote:Celestia uses Y2K as the default Epoch, so planets without a specific Epoch, MeanAnomaly or MeanLongitude will be aligned on Jan 1, 2000.

OK, thanks. Well, they align on 1/1/2000, but P4 rapidly goes out of sync with its trojans (within a few orbits). So evidently something is still not quite right here.

I think I know what's going on here. You may have a disagreement with the mass of the sun as entered into the Mass of Central Star constant, and the Estimated Star Mass as computed from the orbital elements for that planet. This disagreement will result in different orbital periods for the Trojan bodies. If you adjust the Mass of Central Star to be in agreement with the Estimated Star Mass for the planet, the Periods for the Trojan bodies will be in better agreement with the period of the planet and they will no longer go out of sync. This error will also exist for all the other asteroids, but it is only evident with the Trojans.

bdm wrote:Try setting the date in Celestia to be the default epoch date. Do they line up then?

Yep, that's sorted now

Can you post the settings you used in Asteroid Maker? I would like to try this for myself.

http://www.evildrganymede.net/temp/bdm_mod.zip

That's my modified asteroidmaker file with the 2:1 and 4:1 resonance limits. The other files are in the sextet.zip file I linked earlier.

I use a Rayleigh distribution for the eccentricities and inclinations. The limit here is not the maximum, but the peak of the Rayleigh distribution. Some values could be much greater than the peak.

Ah, I seemed to be getting quite a few that were bigger though.

Can't talk much now, but will respond more later on... thanks for the answers so far!

bdm wrote:I think I know what's going on here. You may have a disagreement with the mass of the sun as entered into the Mass of Central Star constant, and the Estimated Star Mass as computed from the orbital elements for that planet. This disagreement will result in different orbital periods for the Trojan bodies. If you adjust the Mass of Central Star to be in agreement with the Estimated Star Mass for the planet, the Periods for the Trojan bodies will be in better agreement with the period of the planet and they will no longer go out of sync. This error will also exist for all the other asteroids, but it is only evident with the Trojans.

BTW, I thought of this too and manually entered the values of the Estimated Star mass so that they're the same as the Mass of Central Star. It didn't seem to do much good, they were still about 180 degrees away from P5 on 1/1/2000.

Why aren't the textures appearing on the asteroids?

PlutonianEmpire wrote:Why aren't the textures appearing on the asteroids?

I noticed that too - it seems that the ones listed in the asteroidmaker file aren't actually textures that come with Celestia.

bdm wrote:The definitions for the inner planet allows a creator to create belts defined by two planets. In the unreleased version 1.7 (unreleased due to unresolved speed issues), this functionality is extended to allow the creation of Kuiper belts.

I know that Neptune traps some KBOs in resonant orbits, but does it actually define the belt too? I guess it does to the extent of clearing out the stuff near its orbit, but the outer edge is undefined isn't it?

Malenfant wrote:I'll need to check this, but you may be correct here. If so I will need to correct this.

I'm pretty sure I'm correct... my SSD book says it's 3:2 , as do these sites:
http://www.rzuser.uni-heidelberg.de/~s24/hilda.htm
http://www.alcyone.de/POrbits/english/hildagroup.html

Wikepedia also has some articles on them, but they get the resonance the wrong way round - they call it a 2:3, it should be a 3:2.

I'm not sure I understand the question, but I'll answer as best I can. There is a "forbidden zone" that can be configured for the inner and outer planets that outlines the borders of the belt. The "fuzziness" may be due to the eccentric orbits that are generated. If you turn down the eccentricity of the orbits ("Peak Eccentricity for asteroid orbits" on the Constants page controls this) then you will get a sharper boundary.

That's what I ended up trying... it seemed to work but not as well as I'd hoped because of the Raleigh distribution of the objects going beyond the bounds that the user sets.

bdm - just wondering, I'm not sure how to interpret the silence here after your responses - are you looking at these problems or is this thread done? I'd like to get the Hildas into the 3:2 resonance where they belong, and to sort out the trojans at least - but I have no clue how to do that.

Malenfant wrote:bdm - just wondering, I'm not sure how to interpret the silence here after your responses - are you looking at these problems or is this thread done? I'd like to get the Hildas into the 3:2 resonance where they belong, and to sort out the trojans at least - but I have no clue how to do that.

This is easy to do.

On the constants page, find the cells with the text "Chance of 3:4 (Hilda) with Outer Planet" and "Chance of 4:3 (Hilda) with Inner Planet". To the right of these cells are cells that compute the semimajor axes of asteroids with these orbits. The formulae look like this:

=((3/4)^(2/3))*$A$3
=((4/3)^(2/3))*$J$3

Change the fractions 3/4 and 4/3 in these formulae to the correct values and you're done. The 2/3 is derived from Kepler's laws (kD^3 = T^2) and should not be confused with the ratios.

Don't forget to change the ratio in the explanatory texts so it reads correctly.

Cool, thanks - I'll give that a go.

done it, the correct way is:



in the "Chance of 3:2 (Hilda) with Outer Planet" cell.

I set the "inner planet" cells below that in my modified excel sheet to 0.00% and for some reason that seems to have fixed the trojan problem as well...