Jump to content


Photo

Sketching close double stars

This topic has been archived. This means that you cannot reply to this topic.
34 replies to this topic

#1 Jeremy Perez

Jeremy Perez

    Vanguard

  • *****
  • Posts: 2139
  • Joined: 12 Aug 2004

Posted 20 September 2005 - 11:42 PM

Hey all,

Eric G. (Cildarith) contacted me recently and raised some excellent questions about a few sketches that I posted in the double star forum last week. They touched on concerns that I've been mulling over since I posted them. You see, the Astronomical League double stars I've been working on are filled with a lot of pretty wide doubles that can be scaled easily into the sketch circle. But when I got around to a particularly interesting double, Zeta Aquarii with a separation of 1.8", I realized that my attempt to sketch the reality of a tight double star wasn't as straightforward as I had deluded myself into thinking. To quote Eric regarding his own tight double star struggles, they "look like two little balls squashed into one another". So from an aesthetic standpoint, it doesn't represent the eyepiece view very well. But what about overall accuracy?

Well I was pretty much already convinced that the Zeta Aq. scale was way off. The text I included with the report was "...The sketch is deceptive. To indicate brightness, I drew the star circles bold. This made me have to push the stars apart to keep them separated. So they were actually tighter and closer in the 240X FOV than they appear in the sketch." It lead me to take a look at some other reasonably close doubles I had sketched. 94-Aquarii was one of those. The separation for this double is supposed to be 12.7", but after measuring my sketch, I drew it as 28". I'm not saying I want to bring micrometer accuracy to my sketches, but to the extent I can either avoid blunt inaccuracies, or at least be aware of them, I want to know what I'm up against.

So I decided to quantify the situation. I fired up Adobe Illustrator and drew a 1.75 inch sketch circle. I then added a hatched scale divided into 720 units across the circle to represent 1" divisions in the 720" diameter of my 240X view. I don't suppose I should have been surprised at how tiny those divisions are, but I was still a bit overwhelmed. I next used the scale to draw a .9" diameter circle to represent the size of the airy disc for stars in my 6 inch scope. Then I drew a second star and positioned it 1.8" from the first to represent the view of Zeta Aquarii. When I printed the sketch circle, I was treated to a slightly elongated speck. Hah. Nice. So I doubled the size of the sketch to a 3.5 inch diameter circle and printed again. This time, there were two tiny specks touching each other. So I moved up to a 5.25 inch diameter circle and printed again. Now there were two distinct specks separated by a tiny speck of white paper. In all cases, the size of these specks was tinier than I could reliably make with even a very sharp pencil, mechanical or otherwise. And forget about scaling the star weights to indicate brightness :grin:

For the heck of it, I imported my original Zeta Aquarii sketch and measured it against the scale I created. In that sketch, the diameter of the stars themselves works out to 14" each, and they are separated by 21". Heheh. So I've pretty much come to the conclusion that if I don't want to sketch on a foot-wide sketch circle, I'm not going to be able to produce "true" field-of-view double star sketches for close companions. Not to mention how indecipherable they would look when uploaded to the web :D

So far, I've come up with a couple options. 1) I can shake on the delightful salt of impressionism and just acknowledge in the observation notes that the close double is "not to scale" compared to other stars in the field of view. Or 2), I can imagine a one-twentieth scale circle in the middle of my eyepiece when making an observation and sketch to that. The thing is, those tight doubles sometimes work nicely with other widely spread stars in the field, and option 2 would exclude those. I think I'll work with option one when necessary and see how it goes. But I'm interested in other thoughts on the approach to sketches that run past the limits of pencil and paper.

One thing is for sure, I'm going to be a lot more conscious of how fat I scale the width of bright stars in a close double.

I'm attaching a pdf (260K) of my double star scale sheet in case you're interested in zooming in at 1600% or printing it out.

Thanks for your thoughts.

Jeremy

#2 Rich (RLTYS)

Rich (RLTYS)

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 5125
  • Joined: 18 Dec 2004

Posted 21 September 2005 - 05:27 AM

Jeremy

Excellant posting. I just downloaded your PDF file to see if I can use it with some of my double star sketches.

Thanks
Rich (RLTYS)

#3 Jeremy Perez

Jeremy Perez

    Vanguard

  • *****
  • Posts: 2139
  • Joined: 12 Aug 2004

Posted 21 September 2005 - 06:03 AM

Hi Rich, thank you.

I don't know how useful that pdf would really be for sketching purposes, other than to demonstrate how exceedingly tiny an accurate double-star sketch would have to be for something in the 2" separation category. Let me know what you think.

#4 Sol Robbins

Sol Robbins

    Vanguard

  • *****
  • Posts: 2017
  • Joined: 01 Dec 2003

Posted 21 September 2005 - 07:12 AM

Hi Jeremy,

That's really cool! Interesting, the method you've figured out for getting a more accurate way to get DS's rendered.

Thanks,

#5 desertstars

desertstars

    ISS

  • *****
  • Posts: 42763
  • Joined: 05 Nov 2003

Posted 21 September 2005 - 10:15 AM

Jeremy, whatever the outcome of this discussion, it provides considerable reassurance to those of us who have so far limited ourselves to muttering, "Oh, that's not quite right," in the dark, and to making similar (albeit more eloquent) comments on our log sheets when we sketch double stars. (Nice to know it isn't just because we aren't capable of sharpening our pencils pointy enough. ;)) I lack the software (and the skills) to address this matter directly myself, and wouldn't have known how to begin (other than pointing out that this isn't how the two stars really look :smirk:). Fortunately, you do not share these limitation. Thanks for starting this off. :bow:

#6 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 21 September 2005 - 11:40 AM

Here are the results of some calculations I did on my own setup. My highest 'useful' magnification under normal circumstances is 240x with a Barlowed 7.5mm Plossl. This gives me a true field of view of 13.2' (=792"). I make my sketches on template 84mm in diameter. For pencil sketches I've found the minimum 'resolution' for any kind of accuracy is about 1mm which corresponds to an angular separation of 9.4" which is a pretty healthy separation for a double star. Anything less than that is going to be very difficult to render accurately, particularly when bright stars are involved and you need some way to indicate their brightness - which usually means drawing the star as a disk.

Ideally, I would like to sketch double stars at a scale of about 1mm/1"; of course this would require a 792mm circle (over 2.5 feet(!) in diameter) with a star like Zeta Aqr a pair of tiny dots separated by 1.8mm at the center.

No, I don't think so, either! :)

Here is my recent attempt at Zeta Sagittae (separation a healthy 8.6")... :ohgeeze:

Attached Files



#7 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 07 October 2005 - 10:12 AM

I've noticed that on the rare occasion that Sky & Telescope or Astronomy publish articles on double star observing, the sketches included seem to be extreme close-ups showing the two (or three +) stars and nothing else. Aesthetically, however, I like seeing the field stars as well, to give the object some kind context, rather than just two dots on a page.

Here is my attempt to have my cake and eat it, too!

The main drawing shows the field of view, while the inset represents a circle about 1 arcminute in diameter, centered on the primary.

Attached Files



#8 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 07 October 2005 - 10:14 AM

Here is a template for those of you who would like to try this and don't mind creating sketches using your software.

Attached Files



#9 EdZ

EdZ

    Professor EdZ

  • *****
  • Posts: 18820
  • Joined: 15 Feb 2002

Posted 07 October 2005 - 10:43 AM

There's a couple of pieces of information that make a double star sketch immensly more valuable.

Frankly, I think the discussion of scale is not as important as should be a discussion of rendering the accuracy of the image. I sketch double stars in my observing log books. They are grossly oversized. I might draw a 2" double with it's companion in a drawing that ends up measuring 1/2 inch across, just to represent the two stars accurately. But what makes the sketch accurate is the position of the secondary and the distance placement of the secondary in relation to the first diffraction ring. Or it could be the components are so close that they overlap and in this case accuracy is determined by precisely sketching the percentage overlapping disks. What's important here I think is that the accuracy of any sketch can be determined by the mathematics used in the physics of optics. There are some very precise measurements that can be determined about the view of stars in any scope and this comes in handy when viewing doubles. And magnification or field of view have nothing to do with it.

First would be the skecth has got to have NESW so an estimate of the position angle can be made. It's helpful if the observer makes that estimate and shows it in the data. Sometimes we read reports from people that attempt to observe really relly close doubles and they make an estimate of the PA that is off by 30-40-50 degrees. That raises the question did they really see the secondary or not.

Second is along with a larger bright spot for the primary and a fainter bright spot for the secondary, the sketcher should at least make an attempt to sketch the first diffraction rings. This single piece of information tells you so much about the scope used and the separation of the double that it is equivalent to having a measuring scale inside your scope view.

For instance, for a sketch that shows a 2 arcsecond double as viewed in a 100mm scope, if the observer shows the secondary outside the diffraction ring, we know either the sketch is inaccurate or the scope was larger than 100mm or the double was wider than 2" and maybe the observer was looking at the wrong star.

How do we know all that from a simple skecth? Because based on diffraction formulas, for a 100mm scope, the center of the first diffraction ring is located at almost exactly 2 arcseconds and the secondary should be dead center on that diffraction ring.

Of course, since diffraction is dependant on wavelength, any information the observer could list on his sketch about the spectra of the components will help veryify the accuracy of the sketch.

For doubles with separations on the order of 3" 4" or more, the sketch may be showing the secondary at some distance outside the first diffraction ring, and it should be fairly easy to determine the distance of the double in the sketch based on a knowledge of the aperture and difraction. In fact, this holds true for any scope size and any double separation, although its usefulness soon drops off after separation get larger, ie. 8", 10".

Of course, sketches should include all the data like scope, magnification, star mags and sep, etc.

Just a few thoughts, after having seen lots of sketches of doubles stars posted, but many sketches that really don't give you much information. I'm only trying to provide suggestions that would help confirm the accuracy of any sketch. There may be many observers whose concern is simply rendering an artistic skecth of the view. But if accuracy is what you're after, these are considerations you probably need to keep in mind to make double star sketches more accurate.

FWIW, I've been able to make some estimates of separtions of sub arcsecond doubles based on my sketches. Also, I've been able to accurely determine vigneting in some small scopes and determine effective aperture was much smaller than the primary lens, all based on the placement of the secondary in relation to the difraction ring. And those skecthes show two stars about a half inch across.

edz

#10 Jeremy Perez

Jeremy Perez

    Vanguard

  • *****
  • Posts: 2139
  • Joined: 12 Aug 2004

Posted 07 October 2005 - 11:08 AM

Eric, that format is brilliant! I think that would work perfectly as a means of combining the overall view with the deeper look into the actual separation of the double/multiple. Especially considering Ed's suggestions.

Ed, those are all excellent points, particularly about including diffraction rings to lock in the scale of the view to the separation of the stars and the diameter of the scope's primary. On my observing sheets, the sketch is not placed conveniently with regard to the data section to scan the whole thing and encumber folks with an unnecessarily large jpg. So I've been including that data as html text along with the sketch. Do you think (from the standpoint of people browsing the web) that this could be a potential handicap to the value of a sketch to have the sketch as a smaller jpg and the data as html text? I'm already considering revamping my sketch template to work with the inset style demoed by Eric, so maybe some other considerations are in order.

#11 Ron B[ee]

Ron B[ee]

    Tyro

  • *****
  • Posts: 4720
  • Joined: 27 Apr 2003

Posted 07 October 2005 - 01:07 PM

Dear Master Sketcher Jeremy,

Those are awesome, so realistic sketches, complete with diffraction spikes and all :bow:!

Ron B[ee]

#12 Ron B[ee]

Ron B[ee]

    Tyro

  • *****
  • Posts: 4720
  • Joined: 27 Apr 2003

Posted 07 October 2005 - 01:08 PM

Wow Eric, that's one of the most realistic sketch of double star (in your case Zeta Sge) I've ever seen :bow:!

Ron B[ee]

#13 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 24 October 2005 - 02:26 PM

I've uploaded 14 double star sketches to my gallery and modified several existing ones, most of these using the inset field method. This seems to give reasonable results but suffers a little breakdown at the narrowest separations (cf 36 And). Also note that these sketches are intended only to give an impression of these objects and should not be used to try evaluate my optical system (LOL) per Ed's excellent and informative post (thank you, sir :)).

Sketches: STF 2445, STF 2455, STF 2457, Albireo, STF 2540, Zeta Sagittae, Theta Sagittae, STF 2653, STF 2769, 1 Pegasi, STF 2841, STF 2877, STF 3007, STF 24, 55 Piscium, 65 Piscium, 36 Andromedae, Psi-1 Piscium.

Comments and suggestions are most welcome.

#14 Jeremy Perez

Jeremy Perez

    Vanguard

  • *****
  • Posts: 2139
  • Joined: 12 Aug 2004

Posted 24 October 2005 - 03:26 PM

woahhhh....

This is like walking in to the party just as the piñata gets busted open. Those are beautiful. That 36-And looks like it was a real stinker. Very nice work capturing the 'impression' :grin: of those doubles.

#15 FJA

FJA

    Sketcher Extraordinaire

  • *****
  • Posts: 6426
  • Joined: 17 Nov 2004

Posted 24 October 2005 - 04:43 PM

Wow, SUPERB sketch of Zeta Sagittae, Eric. As someone who likes double stars and who has a bit of bother with accurate placements this thread is invaluable information as ever. Cheers Jeremy and Eric! :bow:

#16 Rich (RLTYS)

Rich (RLTYS)

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 5125
  • Joined: 18 Dec 2004

Posted 25 October 2005 - 05:20 AM

cildarith

I havn't looked at all your sketches but from what I've seen they are absolutely beautiful.:bow: Even the dark gray background looks quite realistic. You should post them on the Double Star Forum.

Clear Skies
Rich (RLTYS)

#17 HAC

HAC

    Viking 1

  • -----
  • Posts: 534
  • Joined: 07 Apr 2005

Posted 09 November 2005 - 08:45 AM

Hey Eric,

What method are you using that allows such a realistic rendition of the doubles (color, halo etc.)?

Thanks for any information you can provide.

Take care,
john

#18 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:34 AM

Hi John, I'll do my best to explain exactly what I do, though the precise method will probably depend on the photo-editing software you have at your disposal.

At the eyepiece I sketch the field and label the stars in order of decreasing brightness, estimating their magnitude and their color. Generally I limit myself to the brighter field stars, the ones I use to anchor the sketch. From this field sketch I create a finished sketch indoors after the session:

Attached Files



#19 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:35 AM

After the sketch has been scanned into the computer I flatten the field by increasing the brightness by 20%. This wasn't really necessary for this particular sketch, but if your scanner produces a band of shadow on half the image, or your page has been adversely affected by dew this is an effective way to eliminate these artifacts:

Attached Files



#20 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:38 AM

The next step involves decreasing the brightness of the image by 20%, giving the whole thing a light gray tint:

Attached Files



#21 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:39 AM

The next step is to place a white matt around the image:

Attached Files



#22 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:40 AM

And invert. For many this would be an appropriate place to stop, but for some reason I always feel compelled to keep going, though this is the stage that probably best represents the view through the eyepiece:

Attached Files



#23 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:41 AM

Next (and I finally get around to answering your question, John) I add the halos around the brighter stars in white (a "primer", if you will). I use the airbrush setting, white "paint" and the transparency set at 60% for the inner ring, 70% for the second ring, and 80% for the third ring. For the first ring I select my third largest brush size and click on the center of the star image once for an 11th magnitude star, twice for a 10th magnitude, three times for a 9th magnitude, and so on. For the second halo I select the second largest brush size and click on the center of the star image once for an 8th magnitude star, twice for a 7th magnitude star, and so on. A third halo may be added to stars of fifth magnitude and brighter (there are none in this example) using the largest brush in the same manner. For stars of 2nd magnitude and brighter I'll add the diffraction spikes that become visible at about that magnitude in my scope:

Attached Files



#24 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:42 AM

The next step involves adding color to the disks in exactly the same manner as outlined above, but with transparencies set at 70% for the first disk, 80% for the second and 90% for the third:

Attached Files



#25 cildarith

cildarith

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2668
  • Joined: 26 Aug 2004

Posted 10 November 2005 - 11:43 AM

Finally, I burn in the centers of each of the star disks, first with my fourth largest brush, white paint, 50% transparency and one click per magnitude above 12. Then I repeat the processes with the fifth largest brush (which at this point is working on a single pixel):

Attached Files








Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics