[time-nuts] Sidereal time

[Ken Duffill]

Ok so you really are NOT needing the EOP Data that deals with the 
irregularities of the Earth Rotation.
In fact it sounds to me like you are not really looking for Sidereal 
time at all. Sidereal Time is not a uniform time scale and so sounds 
like it is inappropriate for your needs.

If you get a source of Sidereal Time, from somewhere that claims to be 
able to give it to you to microsecond accuracy, there is a high 
probability that they will be taking things like nutation into account. 
Certainly if you use the SOFA routines I alluded to earlier they will.

Again it sounds to me like what you are really looking for is an 
independent source of uniform time based on the PPS produced by the 
divider in the chip TVB will get for you, which is then very close to 
the MEAN sidereal second.

The way Astronomers used to do it, they would have clocks regulated to 
the mean Sidereal Second, but have to recalibrate them every day 
(actually every night!). And they were not interested to the 
microsecond. It was the search for that level of accuracy that forced 
them to move away from this non-uniform timescale and produced the 
ICRF/ICRS and ITRF/ITRS that they now use.

HTH

Ken

On 16/06/12 10:57, Neville Michie wrote:
> It started with trying to run two long case regulators on one brick wall.
> Although the wall is founded on bedrock they interfere with each other.
> As I want to study their performance I tuned one to sidereal time, now they are independent.
> I run a TBOLT and a LPRO to maintain a mean time clock for the clock analysis.
> I am hoping to get a chip from TVB to divide the TBOLT or rubidium 10MHz down to PPS at sidereal rate to observe the
> performance of the sidereal regulator. Now I want to be able to set the sidereal time standard so, if I lose power on my
> rubidium, I can reset it so the longterm record of the sidereal long case will have no phase jumps.
> Also it seemed like a good idea, and the more it seems difficult, the more it needs to be done.
> cheers,
> Neville Michie
>
>
>
>
> On 16/06/2012, at 5:51 PM, Ken Duffill wrote:
>
>> Hi,
>>
>> First of all why would you want Sidereal Time to that level of precision?
>>
>> I know this is the time-nuts so 'because I can' is a perfectly acceptable answer.
>>
>> These days Sidereal Time is only used to display to humans in a recognizable format an old and outdated approximation to the current ITRF<->  ICRF transformations that the professionals would use to find or track a celestial object.
>>
>> See IERS (http://www.iers.org/IERS/EN/DataProducts/data.html) and SOFA (http://www.iausofa.org/index.html) for the details and sample code in FORTRAN and 'C' for these transformations.
>>
>> I suspect if you want microsecond accuracy you will have to use the SOFA routines, and have access to the IERS EOP Data.
>>
>> Cheers
>>
>> Ken
>>
>> On 16/06/12 07:20, Chris Albertson wrote:
>>> On Fri, Jun 15, 2012 at 9:49 PM, Mark Sims<holrum at hotmail.com>   wrote:
>>>
>>>> Lady Heather can do sidereal time.   Specify either the LMST, LAST, GMST
>>>> or GAST time zone (for Local/Greenwich Mean/Apparent Sidereal Time).
>>> I think the question was how to get Sidereal time to the microsecond level.
>>>   A computer display screen only gets refreshed roughly 60 to 100 times per
>>> second so a screen can be tens of milliseconds off.
>>>
>>> How is this done professionally.   Basically they don't.  What you do is
>>> record a UTC time code on a track parallel to the data.  Or now that
>>> everything is digital, the time code is sampled and multiplexed with the
>>> data.   Later the display software can convert the time to whatever format
>>> is desired.
>>>
>>>
>>
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>