A very cheap and compact Lightbox and why flat field frames matter...

[Liliensternus]

Hello Everyone,

Since all work and no play makes for one sad amateur astronomer i decided to finally get my act together and construct a flatbox after realizing that synthetic flat frames will never cut it, especially under the light polluted sky of Bucharest with every kind of gradient immaginable.

After scouring the internet for various solutions that people have implemented and ranging from lightweight but bulky classical lightboxes to laptop tft screens and ending with electroluminescent panels (EL panels) i decided to go with the tft laptop screen ideea.

I obtained a broken laptop from work, gutted of essentials but with the screen still intact and in full working order and after dismantling it i discovered that at least that particular screen and i am assuming older models as well work by shining a high powered halogen lamp through a piece of plexiglass and are backed by an aluminium foil to aid in reflecting the light and by a number of polarized diffusers at the front.

The problem with this idea was the size of screen.......somewhat larger than an A4 size piece of paper and that meant a lot of wasted space as i only have an ED 80 for imaging.

Dismantling the screen proved valuable in that it furnished me with the mode of operation of these devices and gave me the idea which you see below.

In as few words as possible, laptop screens work by internal reflexion of a lightsource placed on one edge of a plexiglass sheet and remove any gradient by using polarized difussers.

Realizing this i odered a piece of 4mm clear plexiglass worth around 3 pounds and a strip of 15 white SMD 3528 LED lights for about 2.5 pounds bringing the total cost for this little project to a 5.5 pounds, dont think you can get any cheaper unless you already have those items.

So here are the actual stages of construction:

1 - I cut out a circular piece of plexiglass 90mm in diameter and used sandpaper to give the edges and the side facing the objective a matte finish whose purpose is to aid in diffusing the light from the led's.

2 - Onto the periphery of the plexiglass i superglued the led strip using a few dabs of glue inbetween the individual led's and connected the trailing wires to a 5mm power jack making sure the connections were tip positive (since all Skywatcher instrumentation seems to use this it makes sense to take the extra bit of time with a volt meter).

3 - Cut out and glued a piece of aluminium foil to the back of the plexiglass disc.

4 - Sanded the face of the plexiglass facing the objective.

5 - Glued a piece of black paper with an internal diameter of 85mm and an external diameter of 90mm to the side facing the objective (this serves as a sort of field stop for the led's).

6 - Glued circular cutouts of the laptop screen polarizing sheets to the face of the plexiglass facing the objective and added a couple of layers of tracing paper on top securing them with scotch tape (i know it's not that pretty but it works).

7 - Drilled a hole in the top of the cover of the ED80 and fitted the 5mm female power jack.

8 - Superglued the whole assembly to the inside of the ED80 cover and left it for 3 days to allow for any outgassing of the glue.

The result of this is a very compact, dual purpose lightbox/objective cover that does not take up any extra space.

For larger diameters where led strips would be farther apart from the center of the plexiglass disc and hence give light frames which would be brighter on the periphery and dimmer at the center one could use an EL wire glued in a spiral onto the back of the plexiglass starting from the inside out or the other way round depending where you would want your power jack to end up...

And since a picture is worth a thousand words here are some of the actual item and the results obtained with and without the flat frames.

Diagram illustrating the arrangement of the the plexiglass disc, aluminium foil, filed stop and polarizing screens inside the ED 80 objective cover.

Side facing the objective.

Opposite side with 5mm female power jack, tip positive.

With power on.

And now for the actual testing.

I took three images with a stock Canon EOS 500D and ED80 on the NEQ6 mount, guided with PHD guiding.

The lights were shot in RAW format and show a patch of stars near the constellation of Lacerta.

Lights were shot at ISO 400 and are 2 minutes long, 3 dark frames, 3 flats, 3 bias/offset frames.

Images were stacked in DSS and the output files were processed with PS 7 portable edition.

Flats were taken with the camera set to AV mode which resulted in exposures of 1/45 seconds at ISO 400.

All images were compressed quite heavily to fit them here so please bear that in mind when you see jpeg artifacts.

In photoshop, the central star has a value of 255, 255, 255 and the background and average of 20, 36, 40 (after normalization).

WITHOUT FLAT FRAMES:

DSS output without the addition of flat frames (central star value 245, 245, 245, background around 230)

Levels ajusted, central star 255, 255, 255, background around 30 (vignetting and various lp gradients from floodlights reflecting off buildings and window frame)

Synthetic flat applied to the previous image in PS7 from a blurred copy with a Median filter of 45.

Same as above but with the application of a Gaussian blur filter of 250 on top of the Median of 45.

CONCLUSION: In heavy light pollution in the center of a major city with numerous gradients, synthetic flats are of little use. It could be that with further work or with the use of the DBE tool in PixInsight the image could be improved, but for those of us using Photoshop for one reason or another synthetic flats under these conditions are of very limited use.

WITH FLATS:

DSS output with flat frames.

DSS Master Flat, Adjusted in PS7 with auto leves to emphasize vignetting.

PS7 output with central star values of 255, 255, 255 and background values of 30, 30, 30.

Same as above but with threshold levels of 40 to emphasize gradients which are very small in the original image.

With a Median filter of 35 applied to a copy of the original image and then subtracted from the original with an offset of 50.

Same as above with a Threshold of 30 levels to emphasize smoother gradients, at 40 levels gradients are not visible.

Final image with central star levels adjusted to 255, 255, 255 and background normalized to 20, 36, 40, star colors enhanced with match color.

CONCLUSION: The use of real flat frames brings dramatic results and corresponds to fewer operations necesarry to remove any leftover gradients.

I know that these conclusions are taken for granted by those of you with considerable experience but for someone like me it was a real wake-up call as to the importance of all calibration frames not only darks which are universally recognized as essential for a good image.

From a personal standpoint i am quite happy with my little gadget despite the fact that it will not win any beauty contests, it took about 2 hours to put together and saves many more hours in processing messing about with Median and Gaussian filters.

Of course, in dark sky areas, even without flats, the results are nowhere near as disastrous as those taken from light polluted areas, but gradients and vignetting are still present to the extent that i have to actually look for them in photos, but still there presence bothers me......no more after this little weekend project though.

All the best.

Mihai

[Gina]

That's very interesting - thank you for posting I've been weighing up all the options for making a flatbox for my ED80. I have a few old duff laptops lying around - I didn't realise the older ones used edge lighting for the back light. Must do some demolition work

[Liliensternus]

You are very welcome. This solution has the advantage of being more compact and simpler to build then the classical lightboxes and cheaper than EL panels. Besides that, i love the fact that it can do double duty as telescope cover and lightbox. .

All the best