Jump to content


Photo

using a green trifilter for photometry

  • Please log in to reply
13 replies to this topic

#1 Bernie Skerl

Bernie Skerl

    Sputnik

  • -----
  • Posts: 31
  • Joined: 24 Jul 2010
  • Loc: newburgh indiana

Posted 03 January 2013 - 05:54 PM

Can you use a green trifilter for photometry until i come up with funds for a v filter?

#2 Ed Wiley

Ed Wiley

    Apollo

  • -----
  • Posts: 1006
  • Joined: 18 May 2005
  • Loc: Kansas, USA

Posted 03 January 2013 - 08:14 PM

It is my understanding that a photometric filter is required for most variable work except eclipsing binaries.

Ed

#3 NJScope

NJScope

    Ranger 4

  • *****
  • Posts: 378
  • Joined: 08 Mar 2004
  • Loc: NJ

Posted 03 January 2013 - 10:12 PM

Bernie
You can use a clear filter for minor planets but should use photometric (UBVRI) filters for stellar studies if you want to publish or compare your data to other work. If you are just trying to get some experience with collecting variable data, then conventional RGB filters for color photography should be fine.

#4 Hubert

Hubert

    Mariner 2

  • -----
  • Posts: 248
  • Joined: 26 Sep 2006
  • Loc: Belgium

Posted 04 January 2013 - 02:28 AM

It is my understanding that a photometric filter is required for most variable work except eclipsing binaries.

Ed


Timings for pulsating stars like High Amplitude Delta Scuti stars are also done without filter.

There are enough objects that you can observe without any filter.

#5 rutherfordt

rutherfordt

    Ranger 4

  • *****
  • Posts: 395
  • Joined: 07 May 2006
  • Loc: Northeast Tennessee USA

Posted 04 January 2013 - 07:49 PM

The AAVSO does have "tri-green" as one of the filters that you can report data with. Ideally, you would want to transform it to "v," but they will take your data just the way it is-- be sure that you do pick "tri-green" as the filter, though.

#6 Ed Wiley

Ed Wiley

    Apollo

  • -----
  • Posts: 1006
  • Joined: 18 May 2005
  • Loc: Kansas, USA

Posted 05 January 2013 - 08:57 PM

Geez Tom, I missed that. Many thanks for the clarification.

Ed

#7 rutherfordt

rutherfordt

    Ranger 4

  • *****
  • Posts: 395
  • Joined: 07 May 2006
  • Loc: Northeast Tennessee USA

Posted 07 January 2013 - 10:47 PM

Ed:

I think that they have added it fairly recently since people are now trying photometry using DSLR cameras (I haven't tried that yet, but will sure open up things to a lot more observers).

Tom

#8 Ed Wiley

Ed Wiley

    Apollo

  • -----
  • Posts: 1006
  • Joined: 18 May 2005
  • Loc: Kansas, USA

Posted 08 January 2013 - 10:33 PM

Hi Tom:

I have been following the Citizen Skies initiative but have not tried DSLR work given that I have photometric filters. I agree with you, this would really open up variable work to observers who have a DSLR. One of these days I will give it a try.

Clear skies, Ed

#9 groz

groz

    Surveyor 1

  • *****
  • Posts: 1954
  • Joined: 14 Mar 2007
  • Loc: Campbell River, BC

Posted 09 January 2013 - 12:33 PM

A few years ago, my wife and I were asking these same questions, and I did a fair amount of research in this area. To do precision photometry, turns out it's all about the camera. To generate results that are directly comparable with others, it becomes important to match filters. But not all results require filtered data either, some are about generating event timing, which is independant of filters. Other projects are about long term data, where it becomes important to compare 'apples to apples' when introducing data from multiple sources.

Our interest homed in specifically on doing differential photometry, for exoplanet detection. In the end, the final result is analyzed mostly for event timing, and to a lesser degree, depth of transit. Neither of those results are dependant on which filter is used, but very dependant on the consistency of the data from shot to shot.

When we first started, it was an 'interesting direction' for our astro addiction, so, we bought a minimal bit of gear to experiment with the concept. I was already considering an upgrade from dslr, to ccd, so I grabbed an sxv-h9 off the used market. The very first light curve we generated, was done on a relatively fast variable, with a fairly large swing, just to prove we could actually get a result. GSC 3074-0114, great starter target for differential photometry first attempt, almost a full magnitude of swing, over a short period. I got two full cycles first time thru. I'll split this over multiple posts, because I dont know how to post multiple images in a single post. So, here is the result, our first time out, SXV-H9 with C8, using no filter (I didnt have any filters yet at that point).

Attached Files



#10 groz

groz

    Surveyor 1

  • *****
  • Posts: 1954
  • Joined: 14 Mar 2007
  • Loc: Campbell River, BC

Posted 09 January 2013 - 12:36 PM

The inspiration that got us going on this, was a posting right here in this forum, Brian had posted a thread titled "I think I saw an exoplanet", and our first reaction was, we have to try that. The above post includes our first light curve headed in that direction. A few days later, I bumped into a bargain for a filter set, and bought a BRVI astrodon set on a blowout, picked it up from a vendor enroute to a star party. A couple weeks later, we finally got our first successful exoplanet transit, taken at the saskatchewan summer star party in cypress hills. This was again done, with the H9 in the C8, this time using the V filter. Result attached.

Attached Files



#11 groz

groz

    Surveyor 1

  • *****
  • Posts: 1954
  • Joined: 14 Mar 2007
  • Loc: Campbell River, BC

Posted 09 January 2013 - 12:40 PM

Well, after that success, my wife wanted to try it with her setup too, but we really were not sure how much could be done with a DSLR, so, we did an experiment to find out. One nice clear evening, we set up on the back deck here at home, both kits side by side, shooting at our favorite 'test' star for testing photometry results, GSC 3074-0114. His was once again the H9 in the C8, but hers was a canon 350D attached to the williams 110. Both telescopes pointed at the same star, side by side, at the same time. After we had all the frames, I extracted the green channel from the dslr images into fits files, and we ran the same photometery on both of them, using aip4win. I think the result speaks loudly.

I leave it as an exercise for the reader, figure out which plot came from the CCD, and which came from the DSLR data.

Attached Files



#12 groz

groz

    Surveyor 1

  • *****
  • Posts: 1954
  • Joined: 14 Mar 2007
  • Loc: Campbell River, BC

Posted 09 January 2013 - 01:00 PM

Our conclusion from that exercise, precise photometry with the kind of precision we wanted, was not going to happen with the dslr. Granted, it is an older one, and, something much newer may (or may not) do a lot better. But we knew for a fact, the ccd would do at least an order of magnitude better than the dslr. Quite by chance, I was browsing an online used equipment site the next day, and another H9 popped up that fit our budget, so, we bought it, and both had H9 cameras moving forward.

I dont have copies of the plots handy, from our first side by side test after we got matched cameras, but suffice it to say, they looked very similar.

How this all goes back to the original question, is this. Before you get excited about what filters are being used for photometry, first you have to define what you want to accomplish. In our case, the goal was precision differential photometry, to measure transits. For that application, it's really filter irrelavent, and, for the really dim ones, running by selecting the empty filter slot, will get more light into the sensor. The important information from a transit, comes from the timing, and edges of the plot, and not so much from the color spectrum. Since all of the relavent data comes from a single series, what is important, is the same configuration from start to end of the series. To get the precision we wanted, it was all about the camera, and not so much the filter.

There is a lot you can do that is 'filter irrelavent'. Particularily if you want to dabble, and, filter not available, you can still take data, reduce it, and generate valid results. they just aren't directly comparable to other folks results, done with different filters. That can come later, it's really just a minor part of the process.

#13 Hubert

Hubert

    Mariner 2

  • -----
  • Posts: 248
  • Joined: 26 Sep 2006
  • Loc: Belgium

Posted 09 January 2013 - 02:59 PM

Photometry of variables is possible with a DSLR. I have done it when the Epsilon Aurigae campaign was running. It's not the same as timings of eclipsers etc but is good for single magnitude measurements.

I took 50 images and processed them to get one single measurement of a magnitude. Error was usual 0.05 mag.

#14 brianb11213

brianb11213

    Cosmos

  • *****
  • Posts: 9047
  • Joined: 25 Feb 2009
  • Loc: 55.215N 6.554W

Posted 09 January 2013 - 07:35 PM

Photometry of variables is possible with a DSLR. I have done it when the Epsilon Aurigae campaign was running. It's not the same as timings of eclipsers etc but is good for single magnitude measurements.

I took 50 images and processed them to get one single measurement of a magnitude. Error was usual 0.05 mag.

Same here, I deliberately allowed the stars to drift across the sensor so as not to use the same pixels all the time, slight defocusing was used, I used the green layer only, taking 15 exposures & reducing them seperately I was able to obtain the standard deviation of the measures & hence estimate the error of the mean: this varied quite a lot with sky quality but under good conditions 0.005 mag or better was obtained on many occasions.

I found a ND 0.9 filter was useful allowing the aperture to be left fairly wide & the exposure kept reasonably long to reduce the effects of scintillation, which can be quite noticeable with small aperture lenses & short exposures.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics