EarthSatellite rise and set of next pass

[Mr.Peanut]

I am getting inconsistent results for rising and setting times for the EarthSatellite type. I am running the code from an Idle window. Each time I run it, I get varying results. In addition the setting the observer horizon is not working perhaps due to the rising/setting time weirdness. Please see my code below.

import ephem
import os
from math import ceil

def readTLE(TLEfile, satellite) :
    print TLEfile, satellite
    if os.path.isfile(TLEfile) :
# logging.info('TLE file: %s' % TLEfile)
        f = open(TLEfile,'r')
        for line in f :
            if line.strip() == satellite :
                TLE = [ line.strip() ]
                TLE.append(f.next().strip())
                TLE.append(f.next().strip())
                break
    else:
# logging.warn('TLE file: %s not found. Aborting.' % TLEfile)
        return
    f.close()
    return TLE

TLE = readTLE('C:/Project/meris_transfer/calc_orbit_number/resource.txt','ENVISAT')
#print TLE
envisat = ephem.readtle(TLE[0], TLE[1], TLE[2])
halifax = ephem.Observer()
halifax.lat, halifax.long = '44:39:00', '63:34:00'
halifax.horizon = '00:00'
np = halifax.next_pass(envisat)
#print np
#print envisat._orbit, envisat._epoch
orbit_number = int(ceil((np[0]-envisat._epoch) * (501/35) + envisat._orbit))
print orbit_number
rising_time, setting_time = np[0], np[4]
print rising_time, setting_time
envisat.compute(rising_time)
rise_eclipsed = envisat.eclipsed
envisat.compute(setting_time)
set_eclipsed = envisat.eclipsed
iseclipsed = rise_eclipsed and set_eclipsed
print iseclipsed

[Mr.Peanut (seamus-spam)]

(Latest pyephem version, Python 2.6.5, WinXP)

And deleting ephem, halifax and TLE in the Ide Python shell window, between successive executions from the Idle code window, makes no difference.

[Mr.Peanut (seamus-spam)]

In fact, sometimes the setting time precedes the rising time!

[Dan CaJacob (dan-wph05)]

I have also noticed this bug. Has there been any work on it? I will probably take a look at the source tonight, because I would really like to use this package for some projects I am working on.

[Kevin Warren (warrenkm)]

I'm seeing the same thing regarding the setting before rising problem. It's fairly rare but it does happen. It seems to be most likely when the satellite is in a pass or just leaving a pass (visible).

[François Steinmetz (francois-steinmetz)]

I have also encountered this bug :
1) I define an earth satellite using TLE and an observer, and find the satellite next_pass()
2) I set the new observer date to the previously found set_time, in order to find another next_pass()
3) In this case, this bug happens (I get inconsistent rise, transit and set times)
4) If in 2), I define the new date as "set_time + 5 min" (ie, obs.date = float(set_time) + 5./(24.*60.)), the results are consistent

[Brandon Rhodes (brandon-rhodes)]

The earth satellite rise and set logic is an independent part of "libastro" that I have never yet been able to improve upon, despite its inconsistencies in output — the problem of finding when a fast-moving object intersects the horizon are more complex than when a well-behaved slow-moving planet or comet does! I have opened a bug for this on GitHub to track my progress, and will close this old issue here on LaunchPad since I am not using it for PyEphem any more:

https://github.com/brandon-rhodes/pyephem/issues/2