|
Post by Ron Walker on Sept 6, 2022 14:14:36 GMT -7
Posted by: Ken Miller Jul 25 2007, 02:57 PM QUOTE(Ron Walker @ Jul 25 2007, 02:21 PM) * If I remember my physics correctly a hole smaller then the light source size will project the image of the light source more clearly then when the hole becomes larger then the light source. Once the hole becomes larger the the light source the projection becomes softer. Thus for star projection you would always want the light source to be smaller then the holes your projecting through. That's an interesting point. But as a matter of practicality, aren't the incandescent (as opposed to arc lamp) bulb filaments larger than the holes you find in the typical pinhole planetarium projector?
Is it true that the images produced by the smaller holes are also smaller images of the filament, and thus harder to recognize as not being starlike?
Is it true that smaller holes may refract better, but allow less light through because of the smaller aperture?
|
|
|
Post by Ron Walker on Sept 6, 2022 14:15:06 GMT -7
Posted by: Ron Walker Jul 25 2007, 05:43 PM QUOTE(Ken Miller @ Jul 25 2007, 02:57 PM) * That's an interesting point. But as a matter of practicality, aren't the incandescent (as opposed to arc lamp) bulb filaments larger than the holes you find in the typical pinhole planetarium projector?
Is it true that the images produced by the smaller holes are also smaller images of the filament, and thus harder to recognize as not being starlike?
Is it true that smaller holes may refract better, but allow less light through because of the smaller aperture?
And that is the problem with most bulbs (used because they are cheap) with large filaments. The larger holes project more of a elongated "blob" while the small holes project a very clear image of the filament. Neither project very good stars.
When the holes get to small not only will the stars be dimmer, but the projected image will also get larger, There is a limit as to how small a hole one can project through and still get a smaller projection.
Think about the pin hole camera. the smaller the hole, the sharper the image but the more light needed to do an exposure. The image is sharper because smaller detail is projected through the system.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:15:52 GMT -7
Posted by: Ken Miller Jul 26 2007, 09:53 AM I kind of feel in retrospect that some of my questions and comments can be interpreted as being snide or derisive. It's hard to tell what the poster intends in this kind of format. Please assume that I'm just trying to learn more, and there is never ever any intent to convey any other message. Also don't be put off if I ask seemingly stupid and uninformed questions. I'm just trying to put it all together in my head.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:24:12 GMT -7
And here all along I just thought you hated my guts!!!
No, seriously, I'm so used to snide and derisive comments from my clients that I just don't notice them at all anymore!!!
No, really, and I am serious now. I, for one, have never thought that at all of you. I'm personally glad that all of you out there have found this little corner of the world and are all contributing to its continued growth.
Hopefully any of my answers are useful and clear to all. If not, please ask again. I have a habit (learned [that I had the habit that is] during my teaching years) of assuming that those that I'm talking with have the same basic understanding of the subject that I have and I need to be reminded of this from time to time. If I'm confusing and unclear, ask.
Remember a lot of the time my explanations are based on "my understanding" of the science behind them as well as results of "my experiments" and then all tempered by "my level of acceptance" of any given result. What this all means is that my posts might not always be 100% correct to current scientific fact. (But then I remember that the Earth was flat was at one time scientific fact.) However, I believe them to be close enough to be very useful in the building of planetaria. The only real variable is "my level of acceptance" like I might consider a projection of a star "acceptable" while others may not. I will temper that last comment with the admission of my "perfectionist" attitude!
My biggest problem is when my left brain fights with my right brain. I want the best most accurate sky as possible, but then just seeing stars projected makes me melt, even if some of them do look a little like filaments.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:25:20 GMT -7
Posted by: Ken Miller Jul 26 2007, 11:19 AM QUOTE(Ron Walker @ Jul 26 2007, 10:30 AM) * My biggest problem is when my left brain fights with my right brain. I want the best most accurate sky as possible, but then just seeing stars projected makes me melt, even if some of them do look a little like filaments.
Amen to that!
|
|
|
Post by Ron Walker on Sept 6, 2022 14:25:43 GMT -7
Posted by: Ken Miller Jul 27 2007, 08:19 PM This is my 200th post. I am no longer a bunch of galactic dust. I am now an asteroid.
The Abbatantuono thesis on Spitz offers the following about pinhole sizes:
"Based on trial and error experiments, Armand Spitz discovered that the smallest pinhole which also produced an acceptable star image was 0.004 inch (Norton, The Planetarium and Atmospherium, p. 68). For the planetaria in his various Model A series, therefore, pinholes of this diameter produced the star images for the unit's minimum stellar magnitude. In the Model A, the dimmest of the roughly 1,200 stars projected were generally of the fourth magnitude, with variances made for completing certain constellations as Spitz saw fit; for the Model A-1 Spitz used stars as dim as the fifth magnitude (Schran, "The Age of the Spitz Dodecahedron", Phenomena, Fall 1993/Winter 1944, p 3.) With the extension in the magnitude limit, the .004 inch hole size was simply transferred to the dimmer stars."
Note that this was based on a Dodecahedron diameter of appriximately 18 inches.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:27:02 GMT -7
Posted by: Ken Miller Jul 27 2007, 08:21 PM For some reason the new designation of Asteroid makes me want to reach for the Preperation H.
Somehow that takes something away from the thrill of being honered with the new title.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:27:32 GMT -7
Posted by: Ron Walker Jul 28 2007, 11:42 AM QUOTE(Ken Miller @ Jul 27 2007, 08:19 PM) * This is my 200th post. I am no longer a bunch of galactic dust. I am now an asteroid.
The Abbatantuono thesis on Spitz offers the following about pinhole sizes:
"Based on trial and error experiments, Armand Spitz discovered that the smallest pinhole which also produced an acceptable star image was 0.004 inch (Norton, The Planetarium and Atmospherium, p. 68). For the planetaria in his various Model A series, therefore, pinholes of this diameter produced the star images for the unit's minimum stellar magnitude. In the Model A, the dimmest of the roughly 1,200 stars projected were generally of the fourth magnitude, with variances made for completing certain constellations as Spitz saw fit; for the Model A-1 Spitz used stars as dim as the fifth magnitude (Schran, "The Age of the Spitz Dodecahedron", Phenomena, Fall 1993/Winter 1944, p 3.) With the extension in the magnitude limit, the .004 inch hole size was simply transferred to the dimmer stars."
Note that this was based on a Dodecahedron diameter of appriximately 18 inches.
I should get a copy of the Abbatantuono thesis to add to my collection but I will elaborate on what I've based my findings on. Abbatantuono is essentially correct but he doesn't take his conclusions far enough (at least based on the above quote and the information he bases his findings on.)
When building any man made device we must consider not only the physics involved but also the actual mechanical limitations.
Abbatantuono notes page 68 of the Nortin book which I will quote now:
"The smallest pinhole for a point source is about 0.004 inch. For a smaller pinhole a curious phenomenon known as diffraction becomes the influential factor. Diffraction is the bending or spreading of light around an object, in this case the circular aperture, causing an increase in the emergent cone angle which results in a larger image. The linear size of the light source is another limiting factor. No light source is a geometrical point (with the exception of a distant star), and a source larger then this will produce a larger projected image."
Interpretation would suggest that, while a theoretical limit of 0.004 inch is possible, producing a light source small enough (and drilling the hole for that matter) would not be.
Various articles on the arc bulb used in the Spitz A3P projector put the actual arc size at anywhere between 0.015 and 0.008 inch. From actual examination of an arc bulb, I would tend to believe the 0.015 more. Since we would probably want to keep the hole larger then light source I would probably try and limit my hole size around this size as well. One point to remember is that actual projection of the arc would not at all appear as a filament would on the dome and thus this "rule" could be relaxed a bit. Now, I would not at all dismiss the 0.008 figure as this might be the apparent size of the light source after transmission through the Spitz lens.
Other collected articles also back up my conclusions. When the Flint planetarium installed the Spitz model "B" the following information was provided. The star globes at either end of the projector were 36 inches in diameter (twice that of the A3P and A1 at 18 inches). The 54 brightest stars of magnitude 2.0 and brighter are done with lenses and the remaining of the 3,083 stars are done with pin holes. A 5.8 magnitude star (the faintest shown) is projected through a hole 0.0135 inch in diameter. This is a number 80 drill.
In an article about the A3P the following information was obtained. Of the 1500 stars projected, 71 of the brightest are reproduced with lenses. The limiting magnitude is 5.0, however fainter ones are shown through pin holes as small as .0145. This is a number 79 drill. This information is for an A3P installed in 1963.
Improvements in the star globe apparently culminated in about 1968 when about 110 stars were projected with lenses (2.5 magnitude and above). At that time you could (and still can on the 1024) get a star globe with the basic 1500 stars or as many as 4000 stars. My guess is that they drilled that extra 2500 stars with a #80 drill. Basically (and Abbatantuono backs this up), the more of the brightest stars that are limited in size by lens projection, the more fainter stars can be reproduced by the smallest hole attainable.
I have on my own A3P star globe noted no hole smaller then a #80 (.0135 inch) but then I'm sure it is a standard globe with the basic 1500 stars. Of the fainter ones (drilled to fill in familiar constellations) again no hole is smaller then a #80 drill.
I do not have an A1 but I would bet that you will find no hole smaller then the #80 drill. I would be very surprised to find anything approaching that 0.004 theoretical limit.
Basically what Abbatantuono has said is technically correct but I believe one could be led astray in assuming certain conclusions. The availability of materials in the real world limits drill size to a #80 at 0.0135 and I'm assuming this is the limit Spitz used on all of his planetaria.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:32:01 GMT -7
Posted by: Ron Walker Jul 28 2007, 11:44 AM QUOTE(Ken Miller @ Jul 27 2007, 08:21 PM) * For some reason the new designation of Asteroid makes me want to reach for the Preperation H.
Somehow that takes something away from the thrill of being honered with the new title.
Keep posting and you'll soon be a planet. At least they don't have sharp edges. tongue.gif
|
|
|
Post by Ron Walker on Sept 6, 2022 14:32:35 GMT -7
Posted by: Ron Walker Jul 28 2007, 12:03 PM QUOTE(Ken Miller @ Jul 24 2007, 07:40 AM) * 33/64 inch sounds way too big. I've been concerned about the "cartoonish" or unrealistic look of stars that are projected with holes as large as 3/16 inch or slightly smaller. A lot does depend on the size of the dome, the diameter of the star globe and even the size of the lamp filament, and the thickness of the material that the pinhole is drilled through.
We need a good technical presentation on the details of pinhole projection. I don't feel qualified myself.
Some of the experiments that Spitz went through when designing the A3P was to limit the size of the stars with lenses. Since the threshold of the average eye's resolution is about 0.05 degrees or three minutes of arc, they designed the star globe so that no star would be larger then 3/16 of an inch on a 30 foot diameter dome. On a smaller dome the projections would be proportionately smaller and vise versa on a larger dome. To a viewer looking up would see stars the same size. The only place this would not hold true was someone sitting directly at the edge of the dome looking at projections of stars right next to him. Under these conditions thew would look like bigger circles no matter what projector was used.
From my limited experiments I will tell you that the stars look great from three feet to 12 feet away.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:33:37 GMT -7
Posted by: Ken Miller Jul 29 2007, 04:13 PM QUOTE(Ron Walker @ Jul 28 2007, 12:03 PM) * From my limited experiments I will tell you that the stars look great from three feet to 12 feet away. Thanks for all the clarifications Ron. I thought .004 inch sounded awfully small for the pin holes. I'm assuming that you have a used arc lamp, so the gap may have increased in length from use, and that may account for the light source being at the long end of the range. I'm guessing that those lamps don't actually "burn out", but that the gap gets too long at some point.
Since that projector of yours probably came from a dome that was on the order of 30 ft size, I'm guessing that the stars look ok at even more than 12 ft away. Did you notice any focal length on the lenses?
|
|
|
Post by Ron Walker on Sept 6, 2022 14:34:34 GMT -7
Posted by: Ron Walker Jul 29 2007, 06:03 PM QUOTE(Ken Miller @ Jul 29 2007, 04:13 PM) * Thanks for all the clarifications Ron. I thought .004 inch sounded awfully small for the pin holes. I'm assuming that you have a used arc lamp, so the gap may have increased in length from use, and that may account for the light source being at the long end of the range. I'm guessing that those lamps don't actually "burn out", but that the gap gets too long at some point.
Since that projector of yours probably came from a dome that was on the order of 30 ft size, I'm guessing that the stars look ok at even more than 12 ft away. Did you notice any focal length on the lenses?
Basically an arc lamp reaches the end of its useful life when the gap between the anode and cathode have reached a point when the high voltage starting current can no longer jump the gap. This is usually well past the useful (bright) life of the lamp. I have heard stories of planetarium projection from an A3P that looked very dim and the stars look like long dashes even with arc projection. This is probably because the arc bulb is being used long beyond its useful life. I've been told that the life of a Spitz arc lamp is around 1000 hours and my guess is that is its useful life in a planetarium projector. I'm guessing that such a bulb will actually burn more like 2000 hours before it can no longer be lit. New bulbs cost the better part of a grand, so I would expect most institutions to use the lamp until it will no longer light. I also have read that only the 20 watt arc lamp uses the Spitz lens and the 75 watt lamp does not. I'm guessing that the extra heat generated by the 75 watt lamp causes problems when used with the lens. Also the 75 watt lamp is used in the larger domes (probably 30 to 40 feet) and the stars should look fairly small in a dome of that size. While I do need to get a spare bulb at some time, I was told the bulb I received with my Star Globe has only 200 hours on it and I would believe that from looking at it. I work with a lot of arc lamps, both in projection films and in light sourced for the production of films and I can tell you from first hand knowledge that it is easy to see use on a bulb. It can show up as physical deterioration of the anode and cathode of the bulb and/or deterioration of the interior of the bulb glass envelope itself. The history of the Star Ball that I have states that it came from a school with a dome of 20 to 24 feet. This I would believe as the star images at 12 feet look very good indeed. I have actually only used the arc lamp for about a half hour (to make sure everything worked well) and the projection was just beautiful.
Again the lenses on my particular generation of Star Globe where of the collimation type which were introduced in the later 60's. Rather then actually projecting and focusing an image of the arc or filament onto the screen, they take a rather large hole and limit the normal expansion of the light rays emanating from the larger hole. Thus the lenses restrict the normal expansion of the light cone from a large hole and "concentrate" the light to a brighter small star. The lenses are of a Plano/convex type with a focal length equal to the distance from the light source to the star hole or radius of the star ball. If the star was projected by just a 1/4 inch hole say, the image on the dome would be three inches or larger in diameter. With the lens, the hole image is kept to a design limit of 3/8 of an inch or so on the screen.
What is interesting is that the projection from a lens-ed projector whether formed from an arc lamp or filament lamp appear the same. It is the projection from the smaller pin holes that make a difference on the dome.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:43:35 GMT -7
Posted by: Ken Miller Jul 29 2007, 07:13 PM Ron, thanks.
I keep picking up bits and pieces of knowledge, little by little.
I think you said before that these lenses were "slightly negative", which I assume means that they were slightly concave rather than convex like the lenses that I have.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:44:40 GMT -7
Posted by: Ron Walker Jul 29 2007, 10:25 PM QUOTE(Ken Miller @ Jul 29 2007, 07:13 PM) * Ron, thanks.
I keep picking up bits and pieces of knowledge, little by little.
I think you said before that these lenses were "slightly negative", which I assume means that they were slightly concave rather than convex like the lenses that I have.
The lenses for the Milky Way are very obviously concave. I'm thinking that the star lenses are actually slightly convex but the convex side is facing into the projector. Some may have a very slightly concave surface facing the outside of the globe but it is very hard to tell without taking the entire lens apart. I have not removed one of these lenses to find this out, but to do what they need to do optically would indicate this. They basically need to take the normal expansion of the light beam that goes through the larger hole and bend the light beam in such a way that there is little or no normal expansion. This is pretty much the case as the larger holes project a star about the same size from three or four feet away out to twelve feet. There is probably a slight increase in size and it would probably be worth doing some measurements at various distances to get some very accurate information.
|
|
|
Post by Ron Walker on Sept 6, 2022 14:45:12 GMT -7
Posted by: Ken Miller Jul 30 2007, 07:50 AM I hadn't considered ramifications of turning the plano-convex lenses around. I'm going to try some experiments with that (assuming that the flat side is facing inward now). Thats easy to check on the Model A projector. Those lenses disassemble very easily. Also, it's easy to turn the stalks inward on the A3 to see what difference it makes.
|
|