Celestia Forums

Is it possible to start or end an object at a time which is not a whole second:

eg. Beginning "2006 01 01 12:00:15.45"

I don't think it's possible using YMD HMS format. If you use Julian dates you may have better luck.

Example:

Beginning 2451545.1234567890

Chuft-Captain wrote:Is it possible to start or end an object at a time which is not a whole second

No. This is not possible. I have investigated it thoroughly. Neither with julian nor gregorian dates.

-rthorvald

bdm wrote:I don't think it's possible using YMD HMS format. If you use Julian dates you may have better luck.

Example:

Beginning 2451545.1234567890

Thanks guys.
Since posting this question I think I have managed to do this by using julian date format (as you suggest).
http://celestiaproject.net/forum/viewtopic.php ... ght=#66980

rthorvald: was your testing in 1.4.x or 1.3.x? Perhaps it's a new feature.

Chuft-Captain wrote:rthorvald: was your testing in 1.4.x or 1.3.x? Perhaps it's a new feature.

It was one of dirkpitt??s OSX pre 1.4 versions, i don??t remember which one as i don??t use it now - it was when i designed my Titan scenario last year: i needed fractions of seconds to get a precice landing sequence for Huygens...

-rthorvald

rthorvald wrote:It was one of dirkpitt??s OSX pre 1.4 versions, i don??t remember which one as i don??t use it now - it was when i designed my Titan scenario last year: i needed fractions of seconds to get a precice landing sequence for Huygens...
-rthorvald

How did you eventually solve it, if you couldn't break down time into smaller intervals?
Also, I assume you were using XYZ's, so how did you go about setting the frame of reference and generating the trajectory? Did you use XYZ Builder?
Are XYZ coordinates always heliocentric, or can they be relative to the 'parent' planet or object?

This post (if you're interested): http://celestiaproject.net/forum/viewtopic.php ... highlight=
describes my original approach which I gave up on as I couldn't get the XYZ trajectory to work.

I'd be interested to know if you have any insights as to where I went wrong.

Cheers

Chuft-Captain wrote:How did you eventually solve it, if you couldn't break down time into smaller intervals?

I didn??t. So the final seconds of the flight aren??t too accurately portrayed; the descending probe model lives a little longer than it should, going through the surface instead of ending where the landed version appears.

Chuft-Captain wrote:Also, I assume you were using XYZ's, so how did you go about setting the frame of reference and generating the trajectory? Did you use XYZ Builder?

Jestr wrote the XYZ, i only tweaked it so that it intersects the actual landing coordinates (it was more than a hundred kilometres off, due to limitations in Celestia: i only managed to bring the two points together by cutting the flight time - hence, Huygens lands three minutes earlier than it did in the real world).

Chuft-Captain wrote:Are XYZ coordinates always heliocentric, or can they be relative to the 'parent' planet or object?

You can write it with anything as the origin: just specify the SampledOrbit relative to the parent in the SSC.

As for the Orion: if you want it to go from the Earth to orbit around the Space Station and then into docking, you will need to write two separate XYZ files - one from the surface and up, relative to Earth, and one from orbit around the space station and into the dock.

Since there are limitations on the coordinates one can express in an XYZ path, it might be neccecary to move the Space Station??s MeanAnomaly so that it intersects a possible XYZ point (there seems to be a kind of grid that determines where a point can exist).

-rthorvald

You can write it with anything as the origin: just specify the SampledOrbit relative to the parent in the SSC.

Does this mean the parent's position is considered [0,0,0] ie. origin

...one from orbit around the space station and into the dock.

This is what my simple goal was, just the second part, but as you know XYZ's are not accurate enough for this sort of small journey.

Since there are limitations on the coordinates one can express in an XYZ path, it might be neccecary to move the Space Station??s MeanAnomaly so that it intersects a possible XYZ point (there seems to be a kind of grid that determines where a point can exist).

Interesting. The grid's probably based on multiples of the smallest unit of movement "expressible" in the XYZ format. It's all far too complicated for the simple goal I was trying to achieve.

In the end I came up with a way to generate a straight line journey of about 600m ending with Orion landing in the docking port at Space Station V (accurate to the millimetre).
This avoids using XYZ's altogether, but has a few of it's own limitations.

Chuft-Captain wrote:Does this mean the parent's position is considered [0,0,0] ie. origin

Yes.

- rthorvald

Thanks for the comments Runar,

I'm sure this knowledge will be useful the next time I need to use an XYZ trajectory.

Phil