# [ntp:questions] Google and leap seconds

Chris Albertson albertson.chris at gmail.com
Thu Sep 22 23:27:44 UTC 2011

```The problem that they tried to solve when they designed the GPS orbits
was to have even coverage world wide such that the number of
satellites in view has about the same in any location world wide and
that this remain true most of the time. And to do this at minimum
cost.

If you just start thinking of random orbits you will have the problem
that something bad happens, like one a month all of then are on one
side of the Earth or something dumb like that.    Quickly you find the
orbits must be an integer fraction or multiple of the Earth's rotation
period.  a factor of "1" is is very high and takes a lot of energy to
get there 1/8 is so low that it's footprint is to small and you'd need
way to many satellites.   Turns out 1/2 works well.

There was talk some time ago about a pair of molniyan orbit (12 hours
also see here: http://en.wikipedia.org/wiki/Molniya_orbit )  GPS sats
for Tokyo so they would have coverage between tall buildings.
Circular orbits are not required.

On Thu, Sep 22, 2011 at 3:13 PM, unruh <unruh at wormhole.physics.ubc.ca> wrote:
> On 2011-09-22, Terje Mathisen <"terje.mathisen at tmsw.no"> wrote:
>> unruh wrote:
>>> On 2011-09-22, David Woolley<david at ex.djwhome.demon.invalid>  wrote:
>>>> The orbit plane is fixed relative to the fixed stars, so, for the orbit
>>>> to cover the same ground each time, it has to have a period that exactly
>>>> divides the sidereal day.  The period is between successive maximum
>>>> North points.
>>>
>>> It is the sattelite you see, not the orbital plane. The sattelite goes
>>> around its orbit in a much shorter time period than one siderial day.
>>>
>> David is correct, the GPS sats are in fact using orbits that have a
>> period of almost exactly half sidereal day, or about 11 hours 58 minutes.
>>
>> This is not a requirement though, afaik the upcoming Galileo sats will
>> have significantly higher orbits/longer periods, but still a rational
>> factor of the sidereal day, so that they will return to the same point
>> after a number of days.
>
> You and he are right, I am wrong. Although there is no particular reason
> why that needs to be true. The GPS system would also work if the
> sattelites were not in a siderial day orbit.
>>
>> Terje
>>
>
> _______________________________________________
> questions mailing list
> questions at lists.ntp.org
> http://lists.ntp.org/listinfo/questions
>

--

Chris Albertson
Redondo Beach, California

```