Re: Error was GSC 748-1618 period

[Tom Droege <tdroege2@earthlink.net>]

Remember that this is engineering data.  I consider it useful to point to 
stars that are probably varying.  I think this has been proven to be useful 
in this particular case.

That is as far as I will go with this data.  I was changing a lot of 
things.  There is no guarantee of night to night consistency.  I don't 
think the things I was doing had a big effect, but a lot of things were 
being changed.  That is what you do in an engineering run.  I was searching 
for the best operating conditions.

Perhaps I now have later data.   I will take a look now that I have a 
couple of scripts to test.

Tom Droege

At 09:34 AM 2/17/03 -0600, you wrote:
>My errors on the order of 0.01 in terms of internal consistency (not in an 
>absolute sense). In the past, the TASS observations and my observations 
>matched up very well in an absolute sense. See 
>http://www.konkoly.hu/cgi-bin/IBVS?5327
>
>In the past our zero point has been very close. This star is fainter so it 
>may not be as close. My Tycho-2 derived comp star mags are also not great. 
>I'm more worried about the shape issue that Arne mentions. I'll dig up the 
>TASS V-I on the var and comps today.
>
>Michael
>
>On Monday, February 17, 2003, at 08:28 AM, Tom Droege wrote:
>
>>Michael worries why he can't match up the two curves more precisely.
>>
>>To determine whether the miss-match is expected or not, we have to 
>>consider the errors.
>>
>>I now know a little about the tass error.  At 11.5 the internal tass 
>>error is of order 0.05 mag for this data.  This is just the rms scatter 
>>of points taken in V and run through the pipeline.  We don't not know 
>>about the tass error when compared to standards.  I would be surprised if 
>>it were zero (with some error attached).
>>
>>We need also to know something about the Koppelman error.  One can tell a 
>>little about the internal error by looking at the scatter of the measured 
>>points.  We know nothing about the Koppelman error when compared to 
>>standards?  I would again be surprised if it were zero (again with some 
>>error attached).
>>
>>Just for tass the 0.05 mag number is just a "most of the time" 
>>number.  You can be unlucky and see 3 or even more times this error.
>>
>>So thinking about the known errors, the match looks OK to me.  I would be 
>>more suspicious if it were closer to zero.