From: "cosmicrock2001" <ursamajor_1@mac.com>
Date: August 29, 2010 11:33:50 PM MDT
To: Astro_IIDC@yahoogroups.com
Subject: Re: My First (partial accidental) Variable Star
Thanks for all the info, just one or two questions to followup - you mentioned the "size of the inner annulus of the star"... so you are putting the star(s) slightly out of focus? I think I've read this in other sources as a technique for photometry.
Besides that I use AstroIIDC, I'm particularly interested in your methods and results as we have similar poor seeing conditions and I've not found much info on doing comparitive photometry with poor or not great seeing, mostly just comments that it can be done. Are you using a standard suite of filters or just the V?
Ron
--- In Astro_IIDC@yahoogroups.com, Milton Aupperle <milton@...> wrote:
Hi Ron;
On 29-Aug-10, at 9:18 AM, cosmicrock2001 wrote:
This is a really neat bit of work Milton. Given your SN error is
slightly greater than the expected dip, its no surprise it doesn't
clearly show up in the data. I presume you used an uncooled TIS or
similar 'video cam'.
Nope. I was using the Grasshopper 16 (14 actual) bit and cooled to
9°C, which was about 10 °C degrees below ambient that night. That was
before I did my tune up up on the Grasshopper cooling system. For each
10°C drop, I get a reduction in background noise by about 50%.
Do you think a cooled ccd camera with better signal, less noise
would do better under the conditions you had?
The issue that caused the high S/N was turbulence (which blurs the
star erratically) thin haze and a full moon. My sky background for 90
second exposures that night was 10,000 to 12,000 out of 65,535. With
no moon, it should be about 1,000 to 2,000 out of 65,535. I was using
a IDAS Light Pollution Reduction filter too on the C8 at 1100 mm focal
length.
The S/N error is basically a comparison of signal to noise and since
the moon and city illuminates the haze, this raises the background
noise and that decreases the signal ratio. Also the size of the
inner annulus you use for the star makes a difference too. To big and
your brightness estimate is including Sky background which raises the
variability. To small and your excluding the light from the star,
especially if you live under the "Jest Stream" and turbulence moves
things all over the place.
In indoor star bench mark tests, my S/N error is +/- 0.002 mags (+/- 2
millimags), which is pretty much raw CCD error and in the same ball
park as SBig gets. Out on the scope on a darker night with moderate
turbulence, I get errors in the +/-0.03 to +-0.015 mag range. The only
way I can probably do any better is to likely add a Adaptive Optics
unit which should reduce the FWHM error by about half.
And what source do you use to select your 'transit star' and others
for comparison? I haven't tried the diff photometry feature in
AstroIIDC yet, but hope to in the future.
For finding exoplanets to image:
http://var2.astro.cz/ETD/index.php
is a phenomenal resource. It has forecasts for future events that is
really good.
For selecting your comparison stars, I use Starry Night Pro 4.5 and
then download the stars to around 19th magnitude. However what is
missing in SNP is the ability to look at the star color ratio (B-V) to
figure out what star colors your comparing too. You want to use stars
of a similar B-V so that the air mass light drop doesn't cause more
changes than your target actually has or you have to adjust the
brightness for them afterwards. Many stars do not have B-V measured to
19th mag, so I don't know if this is a limitation of SNP or other
software.
Exoplanet milli magnitude measuring is very tough to do. Asteroids or
brighter variable stars only need 0.05 mag precision in most cases.
HTH..
Milton Aupperle