SGP4 vs SDP4

[FIU-ELF]

Hi Brandon,

I noticed PyEphem has both SGP4 and SDP4 code in it. Does it select the best propagator for the problem or do we have to somehow select it in Python?

Thank you,
Chris

[brandon-rhodes]

It’s been a long time since I’ve looked at that part of libastro, so let’s do a quick search of the repository using the GitHub search bar at the top of the page:

https://github.com/search?q=sdp4+repo%3Abrandon-rhodes%2Fpyephem&type=Code&ref=advsearch&l=&l=

The only call that I see to the routine in the codebase is here:

pyephem/libastro-3.7.7/earthsat.c

Line 388 in 1bb605c

sdp4(&sd, &posvec, &velvec, dt); /* GEO */

It looks like it chooses based on the satellite's rate of revolution, without input from the user:

	if (se.se_XNO >= (1.0/225.0))
	    sgp4(&sd, &posvec, &velvec, dt); /* NEO */
	else
	    sdp4(&sd, &posvec, &velvec, dt); /* GEO */

I'm basing my interpretation of se_XNO a few lines up:

	se.se_XNO = op->es_n * (2*PI/MPD);	/* revs/day to rads/min */

Unless there's a way for users to manually change es_n without materially changing the satellite’s other attributes (and someone would have to do a deeper dive into the code to see if that’s possible or not), it looks like both routines can be called, and are selected based on one of the satellite’s own properties rather than a user option.

[FIU-ELF]

thank you for the explanation and for teaching me a search function I didn't know about! That is perfect (and astronomically accurate in terms of revs/day)! I really appreciate it.

[brandon-rhodes]

thank you for the explanation and for teaching me a search function I didn't know about!

I'm very glad it came across as educational (which was my goal) and not “you could have done this yourself”!

From the comments, I suppose NEO and GEO must be “near earth orbit” and “geosynchronous orbit”.