A few worthwhile suggestions to date, however one thing that might have been overlooked I think.
How white are white LEDs? and how consistant is the colour temp over a spread of output levels. The circuits discussed so far seem to be concerned only with one type of LED or did I miss something?
I'm thinking of constructing a LED illuminated Light Box myself, and I have been tentatively thinking of running R, G, B, Leds bunched to hopefully give a white mix by the time light gets to the diffuser surface. I must confess I do not know how a white Led gets to be white, but if it is some sort of tri colour Led oscillating at a relatively high frequency the it would probably be unsuitable for an Astro-Imaging Light Box. Jerry Lodregus denounces the use of LEDs, but I have not read of his reasons. It might simply be that Red LEDs are not white. In any event I'd suggest that each led needs to be adjustable so that the final diffuser illumination can be tuned to be flat accross the whole usable surface or it just wont work; we don't want bright spots, and we don't want an artificial gradient.
Dennis, connect the "wiper" of the pot to one of the ends.
Thanks Erick - will update the diagram later. Still got my "Learner" plates on!
Quote:
Originally Posted by Doug
A few worthwhile suggestions to date, however one thing that might have been overlooked I think.
How white are white LEDs? and how consistant is the colour temp over a spread of output levels. The circuits discussed so far seem to be concerned only with one type of LED or did I miss something?
Doug - Thanks for the additional input; interesting stuff. Have a look at Msg#9 (Bojan) in this thread re spectrum of LEDs.
white LED is white because basically it emits blue light into a luminiscent material, which partially converts blue light into yellow and red. But the "temperature" is not the same as daylight, though...
Blue, Green and Red LED's combination are better idea, IMO :-)
Would the "white" colour of the LED really matter if taking flats through rgb filters? It may be naive on my part but if there is red, green and blue light emitted eventually, wouldn't the only difference be the length of exposure for each colour to get the correct histogram placement?
Would the "white" colour of the LED really matter if taking flats through rgb filters? It may be naive on my part but if there is red, green and blue light emitted eventually, wouldn't the only difference be the length of exposure for each colour to get the correct histogram placement?
This is quite valid remark. However the length of each exposure could be very different, possibly too much to be practical
Of course, for BW camera I think it does not matter because flat calibration done with one colour can be used for others, only difference would be a fixed correcting factor.
Also, if RGB LED's are used (for colour camera), I think it may be a good idea to have Red, Green and Blue grouped in such a way that the colour balance is adjustable.
Denis, thanks for the link to the other thread. How did you get on with the RGB LEDs? we are obviously thinking along the same lines, (I'm just at the thinking stage).
Paul, for one-shot colour work, I think taking individual R G B frames is not really practicable. Either individual colour frames would need to be combined, or a bulb exposure with a timer and switch for each colour.(plus absolutely light tight setup)
To reduce complexity, I’m going to proceed on the basis of just using White LED’s initially, and basically suck-it-and-see. I don’t have the smarts or software to model an LRGB system, nor do I have any prior knowledge or experience to base future results upon.
So, I’m going to prototype a White LED unit and take Clear and RGB Flat Fields through the ST7E & CFW-8A and just see what happens.
When I expose the ST7E through the Clear and RGB filters respectively, I’ll make a note of the exposure time required to gather around 20,000 ADU’s (my full reading is around 58,000). I’ll post the results when completed and this may lead to modifications or a complete re-design. Oh for some empirical evidence of the correct path to follow!
Cheers
Dennis
PS - thanks to everyone who has taken an interest in this post and is providing help and guidance, it is much appreciated.
one-shot colour work, I think taking individual R G B frames is not really practicable.
Hi Doug, your right about the one shot colour, I was thinking of mono camera through filters for LRGB composites.
Quote:
Originally Posted by bojan
However the length of each exposure could be very different
True, but I find that anyway with my ATiK. It has greatest sensitivity in the green range and poor in the blue and red ends so I have to adjust my flats exposure times.
QE of ST7 (may be a later chip than my old parallel model?).
SBIG filters in CFW8A.
I think the paucity of blue photons is due to me operating the bulbs at a low level of illumination. They glow with a dim yellowish colour due to me supplying less that the rated volts/current.
After languishing a short while, the Flat Field project is now back on, using a white plastic bucket from Bunnings.
Today I made the LED brackets (x6) from 20mm x 10mm Alum Angle. Each bracket will be (blind) riveted (via the small hole on the long face) to the inside wall of the bucket at 60° spacing, then a plastic LED holder will be fitted in the large hole on the short face. The 90° angle of the Alum Angle has been eased out to an angle of 110° to ensure that the light cone from each LED fully illuminates the bottom of the bucket, before the photons continue their hopefully even journey to the white, opaque Perspex sheet to be fitted at the top of the bucket.
After that, I will wire the LEDs in Series with a 120Ω Resistor – oh, just remembered, must go out and buy a 20KΩ - 50KΩ Potentiometer.
Whilst cutting my last off-cut of opaque Perspex for the light box, I accidentally scratched one surface and I figured this might show in the illuminated flat field.
With the scratched surface inside the white paint bucket, (facing the light source), the scratch did not show up visually, but when I flipped the Perspex so that the scratch was on the outside, (facing the ‘scope), it was plainly evident. I didn’t really want to but some more Perspex and cut another circle. What should I do?
Remembering my Notebook Red Gel filter, I managed to obtain a sheet of theatre lighting grade (read tough) diffuser filter from The Production Shop in Brisbane. It was easy material to work with and I cut a couple of discs to fit on the front and rear faces of the opaque Perspex. The scratch now appears to be completely masked, although I won’t know if this has proven a successful fix until I try the fully assembled light box out later this week.
The diffuser sheets will protect the easily marked Perspex and can be replaced cheaply and quickly if they become marked in use.
Dennis, if you know of a machine shop, close by, then you can get the perspex polished on a buffing wheel. They should have some 'buffing soap' which is used as a lubricant/polish.
Can also use 'Brasso' for the final polish.
'We' did this years ago (Electrical Apprenticeship), when making 'foreign orders'....Errr, table lamps/lamp stands etc etc out of 1/4" perspex, (painted in diff colours, of course), interleaved with 1/16th" brass sheet on a 3/8" brass tube.(for the cable).
Also made knife handles n such.
Came up very nicely.
HTH. L.
Thanks Rajah - I'll pursue the current strategy with the sandwich of Diffuser => Perspex => Diffuser and see how it works out. I finished all the soldering today and ran the first visual test and to the eye, the field looks very flat and evenly illuminated indeed.
All I need now is a clear night to test the unit then I'll post some pictures. It is interesting to note that the white LEDs project a light blue annulus onto the 240 grit sanded aluminium sheet at the base of the bucket, which acts as a reflector/diffuser. Looks like a blue, 6-ringed, modified Olympic Flag, or a sickly blue ring octopus!
The 50K linear pot works a treat, all for the princley sum of $1.76
Well, here is the finished prototype. I haven’t tested the flatness of illumination of the light box yet, just that is powers on and works! The next phase will be to capture some images and analyse the field, checking for gradients, uneven illumination, hot spots, etc. That however, will have to wait until I next take the ST7 out for an imaging session.
Thanks to all those that helped with this project, I really do appreciate the suggestions and ideas put forward.
Note the professional looking potentiometer dial on the outside of the box, Laurie!
![[1ponders]'s Avatar](../vbiis/customavatars/avatar45_9.gif){kind=avatar}
You're not wrong Dennis.
..
L.
