NavList:

   

Reply

Richard M. Pisko wrote:
> On Thu, 05 Feb 2009 08:53:13 -0700, Nicol�s de Hilster  
>  wrote:
>
>   
>>  I still use one of the T2's for sun shots when
>> I need a reference azimuth in the field (for gyro calibrations).
>>     
>
> That's one that I have never done.  Unless you count checking deviation of  
> a compass rose for our aircraft, and I think I used another magnetic  
> device for that anyway.  Are you talking about the Wild gyro for  
> underground work?
>   
The gyro's I calibrate are gyro's used on board of hydrographic survey 
vessels. In general these are FOG's (Fiber Optic Gyro) nowadays. The 
ones used in surveying usually have an accuracy of 0.2xSEC(LAT) degrees.
> How many observations do you need for your desired accuracy of xx  
> seconds?  How many sun/time shots do you take each observation?  (Target,  
> sun, invert, sun, target being the minimum for one observation.)
>   
That is a good question. I usually take at least 8 sun shots (and a few 
extra to reject pairs of bad observations).
Personally I think the best sequence is (but who am I?): ref-sun-ref, 
transit, ref-sun-ref, turn horizontal circle, ref-sun-ref, transit, 
ref-sun-ref, etc. as this allows to check whether the instrument 
remained undisturbed between sun-shots. It does take however 
considerable time, so to speed things up I usually use the following 
sequence (or something like it):
ref-sun-ref-sun, transit, sun-ref-sun, turn horizontal circle, 
sun-ref-sun, transit, sun-ref-sun-ref.
So enclosing the whole observations between the references, but having 
more sun than reference observations. For the latter method one has to 
be a bit more confident about what you're doing, as a disturbing the 
instrument will render quite some observations useless.

I teach practical hydrography and land surveying at the hydrography 
course here in the Netherlands (Terschelling) and even my students have 
no trouble to get fairly good results. The accuracy of a multiple sun 
shot as described above is around 0.005gon and up to now the average 
error has been even less than that (we always compare the sun to a very 
well known 566m long baseline).

I have a graph on my web site showing the results of the second method 
(using that same 566m reference on Terschelling), scroll down to figures 
9 and 10 on: http://www.dehilster.info/instrumenten/theodolite2/index.html
> At least you don't need a neutral density filter for the objective to  
> avoid burning out the electronics with the T2.  
I use a Leica TCRA1101 for most of my measurements, but a T2 is always 
at hand for sun's azimuths. One mistake will seriously damage the total 
station. Before I acquired my Roelof's prism I used a home made lens cap 
of solar filter material which I bought at an observatory. It worked 
like a charm as well, but of course you will need to correct for the 
semi-diameter, like with the notebook method (it does however work 
easier than the notebook method). For those who never did this: this 
correction is something else than the one used for a sextant as you will 
now have to calculate the semi-diameter in the horizontal plane, which 
will increase with the sun's altitude.
> I bought a Wild Roelofs'  
> Solar Prism for determining the Azimuths, works a little easier and  
> quicker than the "sun's image on the notebook" methods.  
It is a very useful attachment, making life a lot easier. Above link 
also shows what this is for those who do not know.
> No need for it,  
> just a fun thing; and I can use it on the T1A (which has automatic  
> compensating for the vertical circle readings) too.  The T1A is good  
> enough to determine the alignment of the archaeological grids, and I  
> usually use three timings on the sun face left, and three face right  
> before going back to the reference object.  The little program in my HP  
> solves it all on the spot, if I enter the correct data and set the  
> semi-diameter to zero.  I can't find the 2009 Sokkia Ephemeris in the new  
> supply catalog, so I'll have to print off Solar data from a web source  
> now, I'm afraid.
>   
I have a basic listing from an old school syllabus that allows you to 
calculate the azimuth with an accuracy of about 10" (0.003gon). You can 
use it to create your own program from it, so let me know if that is 
useful to you.
> Thank you very much for the screen image of Google Earth.  I failed in my  
> last three attempts to find a driver that works, and was planning to visit  
> our public Library to see where the   
> lead, so you saved me a lot of time.
>   
You're welcome.
> I'm glad you annotated the view, as the towers would have been hard to  
> pick out unassisted, and I'm not very sure I'm reading the photograph  
> correctly.  As it is, the steam clouds are hiding some, I'm sure. 
Yes there are a few hidden.
>  The  
> "470m" dimension shows the space between between which two numbers of  
> towers?   
These are numbers 4 and 5 from the eastern corner, so one is missing 
there, although on Google Earth you can just see it between the clouds 
if you zoom in a bit further.
> Rough measurement between the towers on the East of the  
> peninsula range between 180m and 660m, while on the North there is one  
> stretching 1130m, unless one or two more are hidden by white clouds.   
> Possibly, some of these towers were not in existence when you were there.   
> I can't see where the towers go back to land, either.
>   
The towers continue westwards a few more then turn south to go to land 
again. On the east side they turn south, then there is one going west 
again of which I cannot remember it was there when I was in Nigeria (I 
remember the line going straight south).
> Take care,
>
>   
Thanks, mind the eyes! ;-)


--~--~---------~--~----~------------~-------~--~----~
Navigation List archive: www.fer3.com/arc
To post, email NavList@fer3.com
To , email NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---

   

Reply