Jump to content


Photo

A few questions regarding IP program


  • Please log in to reply
1 reply to this topic

#1 jtubbs

jtubbs

    Member

  • Members
  • PipPip
  • 16 posts

Posted 06 July 2011 - 04:31 PM

I'm learning photometry, using the appearance of SN2011dh as a good excuse.  I do have some questions concerning Meade's IP which have vexed me for some time.
1.  When in the photometric analysis portion of the program, what information do I use to ascertain my S/N ratios?
2.  Does the value for sigma give me meaningful information, and if so, how do I use it?
3.  Perhaps the $64,000 question - I use the DSI III pro.  When taking images, I keep the gain slider at 100 and the offset slider at 50.  But in the window that brings up options such as 2x2 binning is also a hi gain toggle. Everything I've read said to basically run the camera in full gain.  Does that include check marking the Hi Gain box? 
4.  The best info I can find puts full gain at about .27,  but of course, there's no such thing as .27th of an electron.  That value would require 3.7 electrons to equal 1 adu unit, so it seems to me full well depth = 65535 x 3.7 = 242,479.  Am I on the right track here?   

#2 jtubbs

jtubbs

    Member

  • Members
  • PipPip
  • 16 posts

Posted 29 August 2011 - 05:56 PM

I guess I'll take a stab at answering my own questions in the hopes that others who've been wondering the same thing might find things helpful...

Let's take number's 1 & 2 together since they are related.
  S/N ratio will be determined by the size of the aperture and annulus in the Set Reference Magnitude box of IP's  photometry window.  The aperture and annulus sizes will be determined by the value contained in the FWHM field of the Determine Magnitude box.  Too bad that info isn't part of the reference setting window...sigh.  All the reading I've done suggested setting aperture 3-4 times the average of the FWHM value. So if FWHM is reported at values that average out to 3.5, set the aperture around 10-12.  Then I set annulus (as a general rule) at approximate 80% of aperture * 2.  IP uses these settings to determine how much of the collected photons are actual signal, and how much are noise.
  Now as to the ratio,  IP doesn't give that information, and as far as I can tell it doesn't provide the information required to calculate the S/N ratio.  However, it does give an alternative, and that is the Sigma value in the Determine Magnitude box.  It occurred to me if the programmers followed good design guidelines,  the grouping of the Sigma value with the Magnitude value indicates the Sigma is the amount of uncertainty expressed in magnitudes.  It's essentially the interpretation of the S/N ratio.

  As to question #3,  I'm concluding worrying about the amount of gain for the purpose of calculating instrumental magnitude is pointless.  In fact, what I have found more important was to pay attention to the histogram window in Envisage. It seems to me if the signal is too low, I lose linearity.  As a result,  I set the gain controls in Envisage to whatever is required for me to get the proper amount of signal per the histogram.  The Total Flux field in IP gives the calculated total ADU's (not electrons) when subtracting noise from signal * number of pixels in the aperture.  I have found as long as I do all my photometry with IP (for consistency), using Total Flux to determine instrumental magnitude yields consistent results.  Even easier,  I found that if I take care to set my aperture and annulus carefully,  then using my known comparison star to set the reference magnitude yields me consistent results in the Determine Magnitude window. 
    For example, my first night of viewing SN 2011fe the calculated Magnitudes of my known check stars closely corresponded (to the 1st decimal point) with the published values of AAVSO and were well within the uncertainty expressed in Sigma.  My only major complaint is IP writes the Magnitudes to the log using only 1 decimal place,  meaning I cannot use the program for photometry requiring .01 or higher resolution.  However,  IP is proving to be a very good package to for me to learn photometry.  I monitored SN 2011dh for a month and half, and for my purposes,  didn't need better than .1 accuracy.  I'm also monitoring SN 2011fe when clouds permit,  and for my own kicks and giggles,  measuring the distance and speed  Comet Garradd travels over time. 

    I've seen other occasional posts similar to mine.  I have no idea if the information I presented is completely correct.  It's the conclusions I have drawn by working with the program, doing a lot of extra outside reading, and good old trial and error.  But hopefully this will be useful to anyone else wanting to work with Mead's IP. 




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users