Welcome to

SPECTRAL OPTICS

Web Page

The Patent Pending DG Spectrum Details Page

Our DG Spectral Filters Cost Four Times (4x) LESS Than what Competitors Charge per stellar magnitude, and at Spectral Optics, if you already have a Telescope, a CCD Camera and its control software, You can get Our Entire Package with Everything You Need to Produce Professional Level Spectral Images the First Night Out and ALL for LESS than the Cost of an Eyepiece!

All of the following pictures were taken using a Type A1 - DG Filter fitted to a Meade LPI CCD camera. All of them, except where noted, were processed using PhotoShop 7. The general steps taken during processing were covered in the DG Spectrum Page under - Step Seven: Processing your pictures.

Unless otherwise noted all of the pictures here were taken with a Celestron C-8 (8") telescope using a Meade LPI camera and a TypeA DG Spectrum Filter.

The first group of pictures show the various stages of images from the raw original to the finished picture showing the spectral lines for star Acubens - SAO 98267. This is a star with a magnitude of 4.3. While not dim by telescopic standards it is a dim star in the arena of amateur spectrum photographs.

It is generally difficult to photograph, or look at visually, the spectrum of stars dimmer than the 3rd magnitude without expensive equipment. We have a five prism GOTO Star Spectroscope and 3rd magnitude stars are about at the limit of its detection. The GOTO Spectroscope is strictly a visual spectroscope.

The DG Spectrum Filter will allow several magnitudes of dimmer stars to be photographed as will be shown in the accompanying photographs.

The Images: The Proof of the Pudding.

The first image on the left is the raw image of a 4.3 magnitude star designated as SAO 98267 (the exposure was 4 seconds and 15 images make up this image), followed by its processed image in the middle (all of the details about the photograph are listed on this image - this data applies to all three images), and then the final image on the right shows its spectral content.

The next set of pictures are from the Jupiter family. First are two pictures of Jupiter. The first one on the left shows the planet and four of it s moons along with a full disk spectrum of the planet. This spectrum extends from 406 nm in the blue to 753 nm in the near IR. The IR Blocking Filter had previously been removed from the camera to allow more bandwidth. The next picture on the right is Jupiter again but this time the right edge of the planet had been blocked out prior to illuminating the CCD chip. This had the effect of creating a large slit. It was sufficient to show some of the methane absorption bands in the upper atmosphere.

The next two pictures of Jupiter's spectrum (full disk using a High Resolution DG Filter) show the virtues of taking the IR Blocking filter off of the CCD camera before doing any spectrum photography. The photograph on the left had the Meade IR Blocking filter in place on the LPI camera, while the one on the right had the IR Blocking filter removed. The IR Blocking filter cut off all light above 659 nm while the other spectrum extended out past 800 nm.

The next two photographs show the spectrum of one of Jupiter's large moons Ganymede. This moon had a magnitude of 5.3 on the night it was photographed. The image on the left was the processed image and the one on the right had the spectral enhancement processing that revealed the spectral lines. As Ganymede has no appreciable atmosphere we would expect its spectrum might be somewhat reflective of Jupiter's spectrum.

Fifth magnitude objects like this have long been outside of the realm of most amateurs due to the prohibitive cost of the equipment required to capture such faint objects.

Jupiter's Spectral Lines Without using a Slit

These two pictures are composites of the very first picture of Jupiter, shown above. Producing a spectrum of an extended body, like Jupiter, without a slip complicates matters. Generally, unless a slit is used to produce the spectrum the standard tools and techniques don't work. However, some useful data can still be obtained.

The picture on the left is shown in three separate sections. The top section is the first picture of Jupiter, from above, untouched. The middle section is the same picture of Jupiter only this time two slices have been removed. The missing parts are shown as two black horizontal bands extending across the planet through its spectrum image.

The reason for the slices is to reduce the size of the object producing the spectrum. If we just used the very top piece and the very bottom piece of the planet then the object would tend to produce a spectrum approaching that of a slit spectrum. A slit spectrum is one where the light from the object passes through a very small slit before it illuminated the grating. By cutting the object up into smaller pieces its spectral image starts to approach one taken through a slit.

Next, the three slices were then processed together, as covered in the DG Spectrum Page under - Step Seven: Processing your pictures. This way they would stay lined up horizontally so they could be checked for common spectral lines between them, after the processing was finished.

The bottom section shows the results after processing the three slices; Top slice, Center slice, Bottom slice. Additionally the Top slice was copied and moved vertically down to align right under the Bottom slice (hence the reason for four slices in the bottom section instead of three). The Top slice was copied and moved down so the spectral lines in the Top and Bottom slices could be compared side by side, as could be done with the Center slice as processed.

The picture on the right is the same as the one as seen in the upper left except one extra step of processing was done to make the individual spectral lines stand out better. Right away one of the spectral lines was seen to match up in all three of the slices. The wavelength for this line was 518 nm +/- 2 nm. It is marked in the photograph by two small yellow markers.

The image on the left is how VSpec analyzed the Top slice of the planet Jupiter. This is a two part graph; a flux graph is on the top and a spectral line graph is on the bottom. This is the very same slice that was used in the two previous images above. The wings of the spectral graph were enhanced to show the spectral lines that are visible in the flux graph but were too weak to be seen in the initial spectral graph. Compare VSpec's spectral graph, on the left, to the spectral image made with Photoshop 7, on the right. The red vertical line in the flux graph, and extending into the spectral graph as a white line, is simplay a marker line that was added to make it easier to see where the 518 nm point was located.

Now, we'll show a series of star spectrums photographed using our DG Spectrum Filter, from bright stars down to dim stars. The first is Arcturus which is a 0.2 magnitude star. The next is Spica which shines at magnitude 1.1, then Denebola at magnitude 2.2, Chort at magnitude 3.3, Asellus Borealis at magnitude 4.7, Tegmen at magnitude 5.2, and lastly Messier M44 with 6th magnitude stars.

The feat here is that the DG series of Spectrum filters will show detailed spectrum information for stars fainter than most amateur equipment will even show a dim line spectrum for much less anything else.

Line Spectrum Images:  Spectral Line Images:

Arcturus - 0.2

Spica - 1.1

Denebola - 2.1

Chort - 3.3

The remaining three objects in addition to being very dim stars were photographed when they were very close to the horizon. Generally astrophotography is done at altitudes above 30 degrees.

Asellus Borealis - 4.7

Asellus Borealis was the second lowest star that was photographed for this illustration. It was just 14 degrees above the horizon when photographed.

Tegmen - 5.2

Tegmen was the lowest star that was photographed for this illustration. It was just 8 degrees above the horizon. This is why its spectrum deviates from a straight line and looks so badly.

M44 - 7.6

The spectrum of these two dim stars will be examined in detail in the section on Spectral Lines and Flux Graphs.

NOTE: All of the above pictures were taken using Meade' LPI CCD camera with the internal IR Blocking filter removed. The LPI camera was attached to a Celestron C-8 (8") telescope. Dark Frames were subtracted as the images were made. The night time temperature at the time the pictures were taken was around 80 degrees F (this was in Phoenix Arizona) and they were all taken during the first part of June 2005. The 7.6 Magnitude star in the two M44 images, above, is at the detection limit of the LPI camera on an 8" telescope.

Meade's DSI camera operates a little differently than does their LPI camera. Their LPI camera is a GREAT color camera for capturing colorful spectral images while the DSI does not produce color spectral images no where near as good as the LPI camera does but the DSI camera is many times more sensitive than is the LPI camera. We cover this in a separate instruction manual included on the CD.

The three pictures shown above were taken with Meade's DSI camera operating in B/W mode. The first picture on the left is the raw image before any processing. The center image was processed using Photoshop 7. The picture on the right has been enhanced to show the spectral line spread.

The three pictures shown above were taken with Meade's DSI camera operating in Color mode. The first picture on the left is the raw image before any processing. The center image was processed using Photoshop 7. Notice how much longer the spectrum of TYC 4392 670 is in the B/W image, in the above set, then the one taken in color even though the color photograph's exposure is twice as long as the B/W photograph. As was done above, the picture on the right has been enhanced to show the spectral line spread.

The picture shown to the left is the line spectrum of TYC 4392 670 as processed by VSpec. The raw image file that VSpec used was the middle B/W picture shown above.

What can You Expect to Photograph with Your Telescope?

Every situation is different as are the night time conditions so the best we can do is to give you a general idea. The photographs above were all taken from Phoenix Arizona. Phoenix is a big bright town so we never have a dark sky unless you go 50+ miles from town. With that in mind and help from J.B. Sidgwick's book Amateur Astronomer's Handbook we have put together a table of what various sized telescopes should be able to photograph using one of our DG Filters with Meade's LPI camera.

Objective   Visual Mag.  B/W Spectrum   Dimmest Limit    Rainbow        SBIG SGS+     For Visual Use
Diameter   of Telescope  Pics. DG $160  Pics. DG $100   Optics $250    Camera $6700  Using a D3 Filter
in inches                Meade DSI)     (Meade LPI) (SLR Film Camera) (Using a ST-7)  Limiting Mag       
   2           12.0          6.6             4.6              1.1             5.5         4.8
   2.4         12.4          7.0             5.0              1.5             5.9         5.2
   3           12.9          7.5             5.5              2.0             6.4         5.7
   3.5         13.2          7.8             5.8              2.3             6.7         6.0
   4           13.5  $19.75  8.1      $16.39 6.1       $96.15 2.6     $957.14 7.0         6.3
   5           14.0          8.6             6.6              3.1             7.5         6.8
   6           14.4          9.0             7.0              3.5             7.9         7.2
   8           15.0  $16.67  9.6      $13.16 7.6       $60.98 4.1     $788.24 8.5         7.8
  10           15.5         10.1             8.1              4.6             9.0         8.3
  12           15.9  $15.24 10.5      $11.76 8.5       $50.00 5.0     $712.76 9.4         8.7
  20           17.0         11.6             9.6              6.1            10.5         9.8

NOTE: Individual results may vary depending on the camera and telescope used and the individual processing steps used to enhance the spectrum.

The first column, in the above table, list telescope sizes by objective diameter in inches. The visual magnitude values shown in column two, in the table above, were derived from Sidgwick's book. The next two columns indicate what star magnitudes can be photographed using our DG Compact Spectrum Filter with various sized telescopes. These magnitude numbers were determined from our Phoenix photographs and scaled using Sidgwick's information for other objective sizes.

The next to the last two columns show what can be expected to be photographed using our competitors products. The retail prices are shown at the top of each column so you can see how much light you can capture per dollar using the various equipment. The competitor's magnitude information came from their web sites, and additionally, for Rainbow Optics, the Sky and Telescope review of their product versus the GOTO prism unit.

The last column shows what to expect when using a D3 Spectrum Filter for Visual Observations, when screwed onto an eyepiece (no camera is used in this last column). These values were determined from visual observations where the sky darkness was only a magnitude four due to the lights and sky glow from Phoenix.

The dollar values shown for the various telescope sizes indicate how much it costs, in dollars per magnitude, to photograph stars with the indicated magnitudes using the three different manufacturers products.

For a particular sized telescope, say 8" on a fairly dark cloudless night, you should be able to take good spectrum photographs on stars brighter than magnitude 5.3. A magnitude 7.6 star's spectrum was captured by a 16 second photograph but it was to dim to be seen on the original photograph. Its spectrum was brought out using Photoshop processing. The spectrum of its 6.4 magnitude companion was visible on the original photograph. These two stars are shown in the M44 photographs shown above.

Taking pictures of extended objects, other than stars, is more difficult to predict as the camera responds to the individual patches of light rather than the over all brightness. M44 has brightness of 3.9 but very few of the stars in it are brighter than 6.3.

Nebulas cover a much larger area of the sky then does a star so the unit brightness is very low compared to a star. A 6 magnitude nebula may only have the unit brightness of an 8 magnitude star.

A rule of thumb is that if you can get a decent image on your monitor with the LPI camera then you should be able to photograph its spectrum, even if you can't see it on the monitor the spectrum may be recovered during image processing.

Identifying the Spectral Components

For this illustration we will use the spectrum we took of Arcturus using the Low Resolution DG Spectrum Filter. The processed spectrum was shown above but it was so bright that the fainter spectral lines were lost. So the same image was reprocessed cutting down on the brightness of the spectrum.

However, even this didn't solve the problem of loosing the weak spectral lines. So we used another feature of PhotoShop to enhance just the spectral lines as the broad bands have already been found in the image on the left. All that is left is to identify the weak spectral lines in the spectrum that are lost in the bright image shown on the left.

To do this we use the Stylize Filter in PhotoShop 7. Then we select Glowing Edges. This will highlight just edge intensities after we have adjusted the parameters. The result is shown in the image on the right. This tool just works on edges, which is what the weak narrow spectral line appear to be.

We identified 10 of the weak spectral lines in the image on the left and 26 spectral lines from the image on the right. The hard part comes next. Now that we have identified that a spectral line exists we need to identify what caused the line. A few of the more easily recognizable lines have been marked on the two images.

Spectral Lines and Flux Graphs

In the above images of M44 there are two stars whose spectrums were photographed. The first is SAO 98021 which is a Type G5 star of magnitude 6.4.The other star in the image is SAO 98020 which is Type A0 star of magnitude 7.6.

The spectrum of the 98020 seems rather short and all in the Blue. At first thought it might seem that the Red portion of that stars spectrum was missed because it was too dim for the DG Filter to capture (even though it had captured 4 magnitudes better than anything else on the market for under $1500) but this assumption is partly correct and partly incorrect.

A Type A0 star, like 98020, emits light mainly in the blue with its primary spectral response between 360 nm and 540 nm. The spectrum captured in the M44 photograph covered the range of 447 nm to 525 nm. The blue portion of its spectrum is several times brighter as the red portion of its spectrum. Its flux output for this range is shown in the graph on the left. This is in good agreement with the spectral flux of a Type A0 star. While there is still a red portion of SAO 98020' spectrum it is below the detection threshold of the LPI camera when used on an 8" telescope.

The horizontal scale is in linear units, #, starting at 0 on the left. To convert the horizontal scale to a wavelength use this simple formula: Lambda (the wavelength) = # (the horizontal number of interest) * 1.62 + 447. The results will be the wavelength in nanometers (nm). For example the tall peak at # = 41 would be 513 nm.

The graph on the right depicts the flux of the other star in the M44 photograph, SAO 98021. As this star is a Type G5 star (similar to our sun) we would expect a lot more red light in its spectrum than that of SAO 98020, and this is what we see. In order to convert the horizontal numbers into wavelengths, for the graph to the right, we use another very simple formula: Lambda = # * 1.61 + 436. For the dip in the graph at # = 80 the corresponding wavelength would be 565 nm.

These flux graphs were made using the Profile tool in Astroart 3.0. The formula for each flux graph had to be calculated individually from two known points in each graph. While this process may not be very easy we have come up with another novel solution. While flux graphs are interesting it is not necessary to have a flux graph to measure the spectral line wavelengths as we have developed an automated tool that will take all of the drudgery out of it for you.

Also included in this package is the commercial tool VSpec. It is used my many observatories and colleges around the world to analyze their spectrum photographs. The picture at the right shows our Arcturus spectrum as analyzed by VSpec. In addition to producing Flux Graphs it also spreads out the spectrum into a spectral line format where all of the individual spectral lines are shown. Detailed instruction are included in our DG Filter packages on how to do this using Photoshop as well as VSpec.

We have developed a tool that allows our customers to very quickly, and accurately, determine the wavelengths of the spectral lines in the spectrums they photograph. It is a special UCC Copyrighted spreadsheet. This particular tool will only work with our DG Compact Spectrum Filter. It will not work with spectrum filters that you purchase from other manufacturers. Consequently we do not sell or distribute or tool except to our customers. We also have developed a very similar tool for use with our DG High Resolution Spectrum Filters, so those customers won't be left out.

After you have photographed a spectrum using one of our DG Compact Spectrum Filters and finished processing the spectrum all you need to do is to view your finished work on the computer monitor and using a small ruler (in inches are mm it doesn't matter) and measure the distance from the center of the object to the yellow (or other known portion of the spectrum) - this will be the reference distance. Then measure from the center of the object to a spectral line you want to know the wavelength of - this will be the unknown distance. Be sure to use the same scale for both measurements (don't measure one distance in inches and the other in mm). Then enter these two measurements in the spreadsheet and the wavelength of the unknown line will be displayed.

In the spreadsheet example the reference distance is 4.18 inches and the distance to the unknown line is 3.9 inches. The wavelength for that line would be displayed as 554 nm on the spreadsheet. While these measurements were made using the spectrum image shown to the right they could have been made from any one of the more conventional spectral images shown above.

It is also not necessary for the spectrum image to be completely horizontal to use our tool. A number of tools on the market require that the spectral image be B/W and completely horizontal, etc. Not so with our tools. They work as acurately with angled spectrum images as they do with horizontal images and they work with color or black/white images equally well.