I realize that there are many tools on this site that allow the user to characterize a star. However being a programmer at heart, I'm writing something that will help me prepare the data for AAVSO variable star entry as well as generate my own light curve from the data collected.
In trying to determine the magnitude of a variable star I use the published V magnitude of a comparison star, and the ADU values of the Variable and the comparison star. [ VM=Cv-2.512 * LOG10(Va/Ca) ]. That I have working just fine. The other pieces of the puzzle are the instrument magnitude of the Comparison star and a Check star. This is used to insure that the comparison star is not also a variable. I get all that.
My problem is calculating the Inst. MAG for the C and K star so I can include them in the AAVSO input.
I'm looking for some help ( formulas) in calculating this magnitude with the information I have ,
Star:
- ADU of the star
- The published V Magnitude of that star
Camera (QHY8L)
1. The gain of the camera (10)
2. Readout noise (6-10e @ 600000/second )
3. QE: 60% at Green
4. Pixel 7.8 um
5. Chip size 25.1mm x 17.64mm
6. Effective pixels 3010 x 2030
I don’t know if this is enough data to make the calculation.
Any help would be greatly appreciated.
Thanks
Nor
Hi!
Not quite...this is a frequent mistake, tho: The photometric magnitude system works using logarithms to the base : "fifth root of 100" , so that a step of 5 magnitudes is exactly equal to a factor of 100.
So let's define b=100^(1/5) as this base. Then
log_b(x/y) = log_10 (x/y) / log_10(b) = log_10(x/y) / (log_10 (100^1/5) )
= log_10(x/y) / (log_10 (100) * 1/5) ) = log_10 (x/y) / (2/5) = 2.5 * log_10(x/y))
So, the factor 2.512 in your formula is wrong, it has to be exactly 2.5 . This is somewhat confusing because, it happens that b ~= 2.512 , so a very similar number, but that factor needs to be 1/log_10(b) and not b.
The instrumental magnitudes can be computed as -2.5*log(ADU_star) + Z for a completely arbitrary constant Z, but of course the same Z should be used for check and comp star. If you want you can use a value Z that will put the comp star at the value of the published Vmag of that star, but since this is the *instrumental* magnitude it really doesn't matter. Also if you are doing aperture photometry, just to be clear, ADU_star should actually be the ADU of the star (inner aperture) corrected for the sky (background) ADU level (outer aperture "ring"), as will be output by photometry software (just writing this in case you try to do this yourself in software as well).
CS
HB
Hi HB,
Thanks for the reply.
OK I'll buy the 2.5 instead the 2.512. IF Z is truley arbirary then why use it at all and just leave it out of the equation leaving -2.5*log10(ADU). However when I do that I get numbers that doesn't make sense. As an example I have a star who's adu is 122. Plugging that in the formula give me an instumental mag of -5.2159 . Obviously I need to do somthing with that number to plug it in the AAVSO form for data entery. Do I combine that with the published Magnitude of the star which is 11.019 ?
Thanks
Nor
Hi Nor,
the CMAG and KMAG fields can have any arbitrary zeropoint. If you want to set Z=0, that is fine, as long as Z=0 for both CMAG and KMAG. For human readability, observers often add a constant to get close to something that looks like a real magnitude, such as something around 16 for your example. Whatever value is chosen is just there to make it easier to read and compare - humans have trouble subtracting two negative numbers!
Arne
Hi Arne,
I was hoping you would chime in
.
OK, that clears things up a bit
.
I guess my next question would be , the values in those fields are not really used for anything other than making sure that the two don't vary beyond a certain point , right? The number that counts is the magnitude of the Variable star that being observed?
So I think I have all the values I need calculated except the MERR field. Does anyone have a formula for calculating this, that can be understood by someone who never learned calculus? I've been reading various calculation explanations on the web but most are over my head.
THanks
Nor
Hi!
Indeed, this can be one the more complicated aspects of a measurement: quantifying the uncertainty. A good place to start is chapter 5.6 of https://aavso.org/sites/default/files/publications_files/ccd_photometry…; .
Don't let the formulae scare you, this section describes several alternative methods, and the easier ones require little more than understanding the concept of "standard deviation" https://en.wikipedia.org/wiki/Standard_deviation and making multiple measurements (at least 3) and then estimating the uncertainty from those values.
Let us know if this helps
CS
HB
I will delve into thestandard deviation and see f I can ge a handle on it. THanks for the info.
Nor
Thank you, HB, for pointing me to use the SDev as the MagErr. I had never really studied what Sdandard Deviation is and how it is used. After reading the wiki on SD it was a simple process to put it into a routine to calculate it.
Thanks Again!!
Cheers
Nor
As suggested by HB, taking 3-5 exposures per filter and then calculating mean and standard deviation is always the preferred method. Such a set of images empirically takes into account all random error sources in your observation. There is a theoretical formula, called the CCD Equation, which calculates signal/noise taking into account typical noise sources, such as Poisson noise and sensor read noise, but it requires that you have thoroughly tested your camera and know all of the normal parameters. It also assumes that each image is "perfect"; that is, no external error sources such as clouds or tracking occur. The mean/SD method avoids that complexity, at the expense of taking multiple exposures and then combining them either by hand or in software.
Good luck! It sounds like you are well along the path to observing variable stars!
Arne
OK with the great help I got from the group I have my applicatio to the point ofdisplying the light curve as well as building the output file for entry to AAVSO. I have one question, though. The last run I made on CY Aqr consists of 73 - 3 minute exposured. I calculated the SD using the calculated Mag for all 73 shots of the Comp star 000-BLD-971, as suggested on page 48 & 49 of the manual.
WHen uploading the data sould I upload all 73 shots? I don't want to clutter up your DB. THe 73 images capture 2 nice curves. I've attached a screenshot of the curves
Thanks again for all you help.
Cheers
Nor
Hi Nor
Congratulations. You have made tremendous progress since those first discussions on the Astrometricia Forum.
Inj this case you want to up load all 73 images. This is a periodic variable with relatively short period (0.06 days) so there is a lot of info in the light curve. If you were observing a LPV with a long period (say 200 days) you would only upload a single observation.
Again congratulations on your pprogress.
Jim
Hi Jim,
Yeah It was a bit of a struggle but thanks to Keith Graham, Arne , HB and the others who provided invaluable information on how things are calculated, etc., I was able to put my program together.
I’ll attempt an up load this weekend, I need to run the data thru my program one more time to double check things and then send it up.
Now all I need is some clear sky and another target.
Cheers
Nor
I'll echo what others have said: great progress! I hope you are starting to understand the fun part of variable-star observing, in addition to the valuable scientific contribution.
Unfortunately(!) there are hundred of thousands of known variables. Your problem now is in deciding what else to observe. You could continue to go after some fast variables for now, and maybe add some longer-term projects.
There are a couple of new novae that might be of interest. You often see major changes over the course of a few days to weeks, but not huge differences during a given night. Those make good "back burner" projects, where you only need to visit them a couple of times per night while doing other things. These stars are:
AT 2018jro (Mon) 6:44:41.371 +04:14:38.26 roughly mag11
TCP J03005508+1802290 (Ari) 03:00:54.95 +18:02:28.9 roughly mag12
They are normally discovered near peak brightness, and decay a magnitude or more within a week.
One interesting field (well, to me anyway!) is NGC7790. This small open cluster has not one, but three 5-day cepheid variables around 11th magnitude. They change in brightness by about a magnitude over the 5-day interval in a repeating fashion. Two of the cepheids form a 2.7arcsec double, which is fun to try and resolve and which show a different appearance every night. One night one star of the pair is bright; the next night the other star is bright! There is also a small-amplitude variable just south of those two, and the cluster itself is part of our calibration program. It is a good one to get familiar with. Also pay attention to QX Cas, the bright blue star just to the SSW of the cluster. It was an eclipsing variable in the 1970's, but has stopped undergoing eclipses (apsidal motion has moved the line of nodes so we are no longer in line with the eclipses). It will start eclipsing again some day - maybe tomorrow!
As you triy a few fields, you will have the background to decide what kind of observing program you want - time series on cataclysmic variables, long term monitoring of Miras, whatever. Lots of choices out there!
Arne
Hi Arne,
I'm having a hard time getting into variable star observation, mostly due to the fact that I'm not sure what I'm looking at and what to do with the info about a particular star curve. I understand the technical part of the observation but don't have the scientific back ground to appreciate what I'm looking at.
I guess if there were a project that needed data collected about a star I would be more than happy to gather as much data as possible to aid the project.
The 10 years I've spent taking pretty pictures has been very satisfying and resulting in seeing the beauties of interstellar space.
At any rate I’ll go with the suggestion you provided and see how things play out.
Would you have the designations for the cepheids in NGC7790. Perhaps I’ll give them a peek.
By the way here is a link to an image of NGC7790 (http://njstargazer.org/pix.asp?id=583 ) I took back in 2016
Cheers
Nor
Hi Nor,
The three cepheids in NGC 7790 are CEa, CEb, and CF Cas. They are the two bright orangish stars in your nice color image of the field. The one on the right is the CEa, CEb pair, blended in your image but obviously elongated. QX Cas, the eclipsing binary that stopped eclipsing about 40 years ago, is the blue star SSE of the main part of the cluster. I usually observe the cepheids about once/decade to look for period changes, and about the same interval for QX Cas, waiting for the orbit to shift and the eclipses to begin again.
You do not need to be an astrophysicist to enjoy monitoring variable stars. Very few AAVSO members publish research papers; most just take and submit data because it is fun and because it can be useful for some future researcher. That said, the AAVSO does post announcements every few weeks detailing current requested observations by some researcher. Your starting place is the About Us/Publications/Alert Notice page, or
https://www.aavso.org/aavso-alert-notices-for-observing-campaigns-and-d…
There is quite a variety of projects described there. Most of the individual Sections also supply lists of target objects, such as for exoplanets or long period variables. Finally, at least once per year (and I think more often than that), the AAVSO offers a CHOICE course
https://www.aavso.org/choice-astronomy
on developing a Visual Observing Program (they should offer a similar one for CCD Observing IMHO), and others on how to improve your photometric techniques. These are always helpful.
You've taken the first step - getting a light curve of a variable. Don't get disappointed if the next step is a little harder; there are rewards awaiting you!
Arne
Thanks Arne. This will give me something to work with. I'll look into that Alert Notice page and see what that is all about also.
Now I just need a hole in the clouds so I can gather some photons. Hasn't been easy here in New Jersey of late.
Cheers
Nor
Hi Arne
I assume that you observe the cepheids with a CCD. You indicated that you observe these about once/decade. How long is your observation session? Do you just take a quick snapshot or do you take a long series of images?
Nor
Hi Nor,
These three cepheids are very similar, kind of what you would expect from stars with a common birth history in such an open cluster. They have 5 day periods, with pulsation amplitudes of 0.5-0.7mag. Because of the low amplitude, they are not ideal visual observing targets, and besides, I'm a professional and so want multiwavelength observations. So yes, my monitoring is with a CCD.
One of my primary goals is to watch for period changes. For some cepheids, the change is slow and uniform, related to their internal structure changes as they evolve. With quality period determination, though, you can see these changes, even if they are small fractions of a second. For other cepheids, the changes seem more abrupt. Not quite sure what causes those, but more data can't hurt in understanding them.
To get a good period requires time series. With a 5-day period, it takes at least 10 days to cover the variability cycle completely, and often longer than that, based on weather. Like I said, I do it once/decade, so it is not a major impact on the rest of my observing.
One of the problems with CE Cas A and CE Cas B is that they are about 2.7arcsec apart. Usual aperture photometry doesn't work for them, so I use a special technique called psf fitting, where an analytical model of the star profile is simultaneously fit to both stars to determine their separate magnitudes. CF Cas, and other cluster variables like QX Cas, are done with aperture photometry and multiplex the use of the monitoring images.
Clusters, because they are concentrated groups of stars, tend to have multiple variable stars in them. At the same time, they are more crowded than field stars, and so you often have to use special techniques to measure their brightness. I tend to look for exotic stars, as the commonplace variables are usually well-studied by the various surveys. However, all variables are valuable, especially if you (a) use the best techniques possible to get high-quality measures; (b) conside using multiple filters to give spectral information; and (c) monitor them frequently, so that there are few gaps in the coverage.
Lots of ways to get involved - I'd try several projects, sorta like when deciding a college major, before settling in on any given path.
Arne
HI Arne,
Thanks for that Info. This gives me a better understanding of what's going on.
For now I'll play with CF and QX to see what I can get. I'm going to looks at the 5 subs I used to create that image I showed you. This was done back in 2016 . It will give me a starting point.
Have a Happy and Prosperous New Year!
Cheers
Nor
ONe more quick question then I'll stop bugging you all for a while.
Now that I'm able to programatically get the Chart data I noticed that in the XML response there is no sequence ID to put in the output file that gets up loaded to AAVSO. I do get the chart ID. Can I enter that instead?
THanks
Nor
Hi Nor,
You can just use the chart ID in the report. The chart ID uniquely identifies a sequence, so any researcher gets the right information. You can also get a sequence ID off of the "Photometry Table" of a requested chart. Either works for submission to the AID.
Arne
Thanks Arne,
I was hoping that would suffice. I know the sequence ID is on the print out of the data table but it doesn't seem to come thru on the XML data file. I'll go with the chart ID form now.
CHeers
Nor
OK so this is my first shot at Photometry, I took some readings of CY Aqr on 12/8 and again last night ( see attached). The curve in blue is the 12/8 and the one in purple is last nights section. The interesting thing about the curves is that front part is quite steep where the trailing part has a much more gentle slope. Does anyone know what the reason for this is and what could be causing the change in intensity anyway?
Nor:
There are many resources that help explain the mechanisms that cause such variations in variable stars. One place to start is under the observing tab on the main aavso web page. Take a look at the Short Period Pulsation Section under Observing Sections. Also, look under VSX for CY Aqr to find out what type of variable this star is.
John Percy's book "Understanding Variable Stars" (Cambridge Univ. Press, 2007) is a good, recent book for all variable star observers.
Ken
Thanks Ken for the info. Have to do some studying.
Cheers And Happy Holidays!!!
Nor