DSLR/AIP4WIN Help

Christian, At what zenith angle was Del Uma during the observation? At the moment our reduction spreadsheets aren't valid for zenith angles over 35 degrees. We're presently working on a reduction method that accounts for air mass, but getting it in an Excel-friendly format is a little difficult. Also, where did you get your B-V correction terms for the stars? Brian

Hello Christian-- I am fairly new to Citizen Sky and doing photometry with DSLR cameras. Most of my experience has been with cooled monochrome astronomical CCD cameras, where AIP4Win routinely cranks through hundreds of images automatically with no problem. Digital SLR images are trickier to work with because the raw file is a Bayer array of interlaced red, green, and blue pixels. Doing photometry requires splitting out and using only the green pixels, so right away, you have to open a color image and you have to take it apart to get at the green. Since I live in Oregon where it rains from October to May, I haven't had much chance to shoot DSLR sky pictures. But on April 2, I shot a series of nine eight-second exposures of epsilon Aurigae using my Canon 350D at ISO 200 with a 50mm f/2.5 macro lens focused on infinity. I extracted the magnitudes for 11 comp stars from each image using AIP4Win. Standard deviations for the comp stars ranged from 0.02 to 0.10. The 0.02 is too good to be true with a DSLR, and 0.10 would be totally unacceptable with equally solid exposures in an astronomical CCD camera. I didn't know about the Reduction-beginner.xls spreadsheet, but I found Tom Pearson's old spreadsheet (Pearson.xls) and modified it to use Brian's epsilon/Z method to transform the magnitudes.The transform graph looked pretty grotty to me (y=0.2491x-0.1484), but it yielded a believable magnitude and I submitted the observation. Now it seems to me there are two problems here. First is that your photometry isn't coming out right and that is frustrating. Second, is that with AIP4WIN, it is tedious extracting magnitudes from your images, at that is doubly frustrating. I haven't run the tutorials yet, so I'm not sure what they want you to do in the exercises. Anyway...let's look at the whole system from camera on down. The Nikon D40 is a nice camera. (I have a D70.) What ISO setting are you using? You have a 50mm lens, what focal ratio? Are you focused on infinity or purposely slighty out of focus? Raw format? Dark frames? Flat fields? With AIP4Win, what version of the program are you running? The current release is v2.3.1. When you open an image, does it come up in color? What do the monochrome images look like? Smooth? Crunchy? Do they have vertical or horizontal stripes? Which of the photometry tools are you using to get your magnitudes? Which steps or procedures are most frustrating to use? With the spreadsheet, are you doing a cut-and-paste from AIP4Win or save-to-file-and-import? What do the transfrom curves look like? does the fit to the points look good or kinda grotty (like my transform curve did)? --Richard

Thanks for reposting here and your reply, much appreciated. Well, Del Uma was within ~25 degrees. The other two were within 5 to 10 degrees. But some of my comp stars on Del Uma were almost 10 degrees away, so I assume I was wrestling with the air mass thing. I've been using the II/168 VizieR and the Hipparchos catalogues. Played with both to see if my issue was there. I get slightly different values at the end, but I'm still off as much as .2. Which one is better / more current catalogue to use? Thank you. -Christian

Hi Richard, Thank you for the reply! Camera.....I've used 55mm at F4.5 @200 ISO, and also, 80mm at F4.5 @200ISO, both seem to give the same results......ha, I think using an 80mm forces me to stay within acceptable degree limits for air mass :) It's the same lens, AF-S 55mm to 200mm. I use 200 ISO to keep the noise down. I don't use the "noise reduction" on the camera due to what Christian Buil's article addresses regarding the Nikon D70 vs Canon. But I don't go thru the trouble of getting a 'true RAW' on a Nikon that he describes using....seems even more tedious when one is shooting the number of exposures for this, but I digress. Yes, i defocus between infinty and a sharp focus. Using RAW. I shoot Darks right after shooting my star pics. I usually set the camera outside for about 30 mins before shooting to allow it to cool down. I've shot flats, using a white T-shirt over the lens, cool trick....forgot where I read that from, but works great. Although, I've been using the "Basic" calibration setting in AIP4WIN, and haven't seriously incorporated Flats for the photometry yet. Flats is something I've been playing with on the tutorials in the back of the book. I do have the latest version, 2.3.1. Following the tutorials here, I set the Red and Blue channels to zero in the DSLR & Bayer Conversion Settings window. Here's my first problem. I hit the save button, and the Settings window stays open. Also, if you close the window and reopen it, the Red / Blue have a default setting, but not the zero I entered. Gives the impression that it didn't save to the settings, however, it does. Anyways, Tom Pearson got me clued into that one (Thanks Tom). So, after I set the settings for Red & Blue to zero, I close the window and load up the pics......they load up with only the green channel. From there, I convert to greyscale, then save to FTS file. Oh, the monochrome images look ok. After that, I set up the calibration for "Basic". Then use the Multi-Image/Auto-process/Deep Sky tool. Here's where tediousness beats me up a bit, and another reason I've only used one set of images (15 pics) for now......I have to manually stack the frames. No matter how I set up the alignment tab, the stack doesn't stack the stars correctly and I get a huge enlogated target star. The comp stars are never on and it looks like I've got four extra stars for every comp star. Manual stacking has worked, but again, very time consuming. I then use the Measure/Photometry/Single Image tool. A very frustrating thing I find is that a pic window does not 'maximize'. That's for any window that opens up for a pic. So I scroll around on the window to find the stars I'm looking for. Once I click on a star, for target, check or comp star, and hit the 'star1' or 'star2' button on the tool window, the pic window re-centers itself and I have to scroll around to find the star I just placed the star aperture / annulus on. As far as the spreadsheet, at the moment, I use the ol' cut and paste method into the excel sheet provided here. Transform curve looked ok to me, but being pretty new to this, unless it was grossly out, I wouldn't know if it was good curve or off. Again, thank you so much for the reply Richard. Hope this has given ya something to go off of. -Christian

Hi Christian- Not showing the Bayer Conversion settings as saved is a known bug in AIP4Win 2.3.1, and has been fixed in an unreleased beta version. Sorry about that. Also, another conversion mode has been added that enables you to get a monochrome image directly. After doing the work of conversion to green monochrome, you're quite sensible to be saving the image as a monochrome FITS file. That way, you only have to do the conversion from NEF to monochrome FITS once. By default, AIP4Win opens images at 100% scale. However, you can change this to any zoom percentage you want. On the Image Display Control, select the Defaults tab, and set the Zoom to 33% percent or whatever is a convenient size to enable you to see the entire image. Click the Save button. All images will then open zoomed to your default zoom setting. The Measure>Photometry>Single Image Tool is an old tool, but you're apparently not having problems using it (aside from chasing the stars around) because you have measured raw instrumental magnitudes from your images and transferred them to the spreadsheet. Do you know how consistent are the raw instrumental magntitudes from one image to the next? Below I have pasted in raw instrumental magnitudes from a set of nine images that I took with my Canon 350D. The magnitudes for EPS AUR, and comp star 32, 38, and 43 are listed. --------------------------------------------------------------------------------------------------

AIP4Win v2.3.24 Magnitude Measurement Tool Raw Instrumental Magnitude in Text Format One line per image. Stars in columns. Target name = EPS AUR at RA=75.5 DEC=43.83333 Observer name = BYY at LON=-122.60957 LAT=44.79092 Import into spreadsheet using ';' as the delimiting character. Now = 4/3/2010 9:36:49 PM

Seq_#; Julian_Day; Integr; Filt; Airmass; EPS AUR;; 32;; 38;; 43;; Image Name 00001; 2455288.52350; 8.0; ; 1.0002; 3.783; 0.022; 2.977; 0.012; 3.822; 0.022; 4.642; 0.045; _MG_1349.CR2 00002; 2455288.52367; 8.0; ; 1.0002; 3.617; 0.019; 2.997; 0.012; 3.836; 0.023; 4.730; 0.048; _MG_1350.CR2 00003; 2455288.52384; 8.0; ; 1.0002; 3.685; 0.020; 3.011; 0.012; 3.847; 0.023; 4.505; 0.040; _MG_1351.CR2 00004; 2455288.52404; 8.0; ; 1.0002; 4.131; 0.029; 3.000; 0.012; 3.739; 0.021; 4.747; 0.049; _MG_1352.CR2 00005; 2455288.52421; 8.0; ; 1.0002; 3.657; 0.020; 2.972; 0.012; 3.710; 0.020; 4.529; 0.041; _MG_1353.CR2 00006; 2455288.52439; 8.0; ; 1.0002; 3.574; 0.018; 2.944; 0.011; 3.763; 0.021; 4.793; 0.051; _MG_1354.CR2 00007; 2455288.52457; 8.0; ; 1.0002; 3.696; 0.020; 2.937; 0.011; 3.809; 0.022; 4.557; 0.042; _MG_1355.CR2 00008; 2455288.52475; 8.0; ; 1.0002; 3.715; 0.020; 2.956; 0.011; 3.812; 0.022; 4.725; 0.048; _MG_1356.CR2 00009; 2455288.52493; 8.0; ; 1.0002; 3.614; 0.019; 2.973; 0.012; 3.756; 0.021; 4.836; 0.053; _MG_1357.CR2

-------------------------------------------------------------------------------------------------- The standard deviation for EPS AUR is 0.167 mag; for 32, 0.025 mag; for 38, 0.048 mag, and for 43, 0.12 mag. If I leave image _MG_1352.CR2 out of the sequence, the standard deviation for EPS AUR drops to 0.066 mag -- still not as good as the photon statistics call for, but at least reasonable. When I computed the magnitude for submission, I included all of the values for all stars and got a V magnitude of 3.733. When I did this run, I actually measured 11 stars, and saw several similar discrepant values in other stars in other images. Since I shot my images in focus with a very high-quality macro lens, I'm inclined to attribute these discrepancies to the random walk of the star image across the Bayer array. Next time -- assuming that we ever get a clear night here! -- I'll take images a tiny bit out of focus. I can't tell from your description why there's a problem with stacking. The stacking software find the alignment stars and stacks automatically if the stars don't jump all over the place. If you have to handle the camera when you take each picture, that moves the starfield around, and then you have to stack manually. I use a remote shutter button so I don't touch the camera. I shoot the pictures one right after another as fast as I can. From your discussion of stacking, it sounds like you stacked all the images and did your photometry on the stacked image? If I'm wrong, what did you actually do? As you can see from the output above, I extracted my raw magnitudes from the individual images in order to judge how large the random errors were. Maybe you would attache a sample of individual images -- three would be enough -- so that we could take a look? --Richard

Hi Richard, First of all...Thank you very much for graciously giving your time to help me and work thru this with me. Ya know, I didn't even think of analyizing each pic before I stacked them. Yes, I stack first and then did the photometry. I use a tripod for my shots, and like yourself, I shoot as fast I can....4 sec shots. Oh, and I also use a remote shutter button. Given your last reply, and with some constraints with time on this end.....I did the following....I used three, single pics of Mu Leo. I'm using Mu Leo because I want to make sure my technique is good to go before I start working on variables, and then onto submitting those values. Anyways, these are photos I took on 4-2-10, at ~ 2145hrs (MDT).....that was the last time I had clear enough skies. Been snowing here since...got 8" lastnite here. But, I digress. So my target is Mu Leo, my check star is Eps Leo, and the comp stars, 22 leo, Kap Leo, and Lam Leo. I probably should use more comps, but those are the closest. For V cat, and B-V values, I used the Vizier II/168 catalogue. I loaded 3 darks I took at that time, and processed those in the calibration setup screen. Then I loaded each star pic, labeled s55, s56, s57. I used the Manual Calibrate/Subtract Dark from Image tool on each. Then saved those pics Pic 55, and so on. I used the Single Star Photometry tool on each three pics. I only included the Target and Check star values here. Here's what I came up with V mags after plugging in the instrumental values into the excel sheet.... # Mu Leo / V Cat Eps Leo / V Cat Mu Dev. Eps Dev. 55 3.92 3.888 2.991 2.978 .032 .013 56 3.933 " 2.999 " .045 .021 57 3.947 " 2.992 " .059 .014 Looking at these figures, I'm pretty impressed. Definitely within reportable range. So, now I'm sorta wondering if a couple of my pics were way out there due to some atmospheric issue. I admit, I did not, nor have yet to, analyze each photo to see what happened, again, didn't think of it till now. Perhaps this is why I should be taking several different series of photos. And thank you for getting me squared away on zooming out on the window. I zoomed out to 50% and that did the trick. I included the three pics I used, s55 thru 57, had to convert to jpg for the post. Also attaching the pic 55, which was the FTS with darks subtracted mentioned above, again converted to jpg. Not sure if converting to jpg is going to be of any help. Also, I'm attaching the worksheet that I use to convert the instrumental values to V mags I used for Pic 55. I did one for each, but figured just reviewing one sheet would be sufficient. I use the worksheet from the DSLR tutorials here on Citizen Sky. I did try the Magnitude Measurement Instrument tool, however, after I put in my Long, I got a error window that said, "Run Time Error '13': Type Mismatch".....and then it shut down the program. It did this three seperate times in a row. My lon is -105.76694 and lat is 39.88694. Hopefully I didn't confuse you on anything above. Again, thanks for your time. -Christian

Attachment	Size
s55.jpg	7.96 MB
s56.jpg	7.74 MB
s57.jpg	7.67 MB
Pic55.jpg	7.63 MB
Mu LEO Pic 55.xls	27 KB

Hi! I also use AIP4WIN a lot for single image photometry, in the same way you mentioned: stack a few pictures, then take the magnitudes. Note that in the "Deep Sky" processing dialog you can choose the search radius for the alignment & tracking (left lower corner), I always set it to 50 (the max value) before selecting the 2 alignment stars. I really don't think that if you shoot the frames directly one after another, the stars will shift enough to fool the tracking <scratch head>. There is one potential trap to avoid, tho: If your picture files are named (say) p1.fit p2.fit ....p9.fit, p10.fit,p11.fit..p19.fit ..... and you select p1.fit as the reference frame, AIP4Win seems to order the files in lexicographic order: p1, p11,p12 ....p19,p2,p3,p4.... and this may well derail the tracking because the distance between star images on p1 and p11 and p19 and p2 etc might exceeed the search radius. Renaming the files to p01,p02 .... will prevent this. CS Heinz

By the way .... if you have made several shots of eps Leo with a few days delay in between, you will probably be able to see asteroid Vesta moving relative to the star field, as Vesta is lingering around there at the moment and quite bright (~7mag). It's just now going into EDITcoming out of (apparent) retrograde motion I think so a frame from a week or two ago or something like that would be best. A good opportunity to try AIP4WINs frame blinking feature maybe. http://en.wikipedia.org/wiki/4_Vesta CS Heinz

Thanks Heinz for the reply. Well, just ran some test stacks, and wow. Set the tracking radius to 50 and it made a big difference. Thanks. Haven't tried renaming yet, but will. Makes sense tho. I noticed it does that when I load about 10 RAWs to converter to greyscale, didn't think it might use the same sequence for stacking. I've been following Vesta's path the past month, but with Binos, not a DSLR. Good idea on the blinking tool. I'm definitely gonna try that in the next couple days. Thanks again. -Christian

Hi Christian-- Your raw instrumental magnitudes appear to be very consistent. That's good news. All software that creates files in numbered sequences ought to prepend zeros so that 0001.ext comes before 0010.ext and 0100.ext. It takes but a moment in the programmer's life to insure that files appear in time sequence when they are sorted into alphabetical order. After days of rain, last night was clear here! I shot another set of test images of the Epsilon Aurigae area. I'll be out of town for a couple days, but when I get back, I hope I'll have time to process them and post the magnitude extraction and data reduction steps and the final results. BTW, next week I'm speaking in the NEAIC conference that immediately precedes NEAF. I'd be glad to talk to any of you there. --Richard

Richard: I have just upgraded to AIP 2.3.X and I am experimenting with the Magnitude Measurement tool. I have watched your video. Here is my question, should I calibrate the image group for multiple image photometry (flat and dark subtract) prior to using the Magnitude Measurement Tool? I have always done reductions using other photometry software, but I didn't see it mentioned in your posting on your web site and it wasn't mentioned in the video. Thanks in advance for the reply. You have developed a really useful too!