|
Post by Ron Walker on Jun 13, 2022 14:17:49 GMT -7
Posted by: Philostopher1 Sep 16 2017, 11:02 AM QUOTE(moonmagic @ Sep 16 2017, 09:51 AM) * You guys are ALL just AMAZING! What a project.
Someone mentioned the original Spitz Model A was plastic. Before going to metal dodec's they were plastic. I have one of those in the "collections." It was much thinner plastic that I would have first imagined, with no lenses used for the brighter stars as they improved when they went to the A-1. I will be forced to eventually find a good way to re-glue or re-join some of the panels as it is having age issues. (Don't even go there!)
Meanwhile, with all this effort expended on making your own, wouldn't you still be looking for one that has become surplus? Every once in a while one still becomes available if you are willing to get a truck and go take one out. mm
Hi,
What kind of plastic is the projector made out of? At least with ABS or PVC you can make a slurry with acetone or M.E.K and fill in cracks for a solid bond. I forget which solvent for which plastic at the moment.
As far as building a projector vs acquiring one, I did have the "spare" GOTO M1, that was used in the NY worlds fair at the NASA Pavilion, then it went to Jeffco Public Schools, then found it's way to Omsi (Oregeon Museum of Science and Industry) then I bought it for $1200. I don't wish to discuss it's current fate, as that is unknown and I feel sick just writing about it. So as part of my healing I am building one.
I have built others starting with an Old Zeiss "Ikomat" slide projector when I was a kid. Then punched holes in a globe and put that on a barbeque rotisserie with a tiny xmas light inside. No slip rings so only so many turns before the wires had to be untangled. Later when I was doing laser shows I built a 24' geodesic dome and sewed surplus bedsheets into a dome. I made a 1\2 starball for effect rather than education with a motor and belt to rotate an aluminum hemisphere around a Stinger Streamlight lamp. That worked well and eventually was sold to a magic act. I like making them, and I have always wanted a serious machine with an accurate sky. This is really the culmination of years of skills coming together and I am pushing my skills to the limit to do this. This is kind of my Mt. Everest...
|
|
|
Post by Ron Walker on Jun 13, 2022 14:18:07 GMT -7
Posted by: Philostopher1 Sep 16 2017, 11:28 AM QUOTE(charles jones @ Sep 12 2017, 10:48 PM) * Jamie, I only uploaded the drawings of the Spitz patent. Would you like the worded description? Let me know.
I have no idea the size of the silver ball in the Model B. I don't think the patent mentions that as most patents tend to be broad and non - specific from my point of view. I know from my early experiments, the ball should be as small as possible as I mentioned before. I am thinking 25 mm or less.
There must be something out there, off the shelf, that could be used. That means intensive searching.
It is certainly an interesting approach to pinpoint projection, and obviously worked for the Model B. Is it practical for a small projector, I don't know, but I wanted to try it. It worked for the Spitz STP (? whatever). BTW Spitz only made a couple of these and I had a chance to see one in operation in San Diego years ago. Very impressive star images for pinpoint projection. But then again, so many of the stars were individually lensed and the star globe was massive--maybe 4 to 5 feet in diameter from memory.
Just remembered: Didn't Brent Sullivan (Phoenix) get one of these? Ron?
I think I remember reading that in this forum. If so, he could tell you the diameter of the mirror ball used.
Charles
Thanks! I'll see if I can track him down (sounds ominous). I was thinking about the mirror ball size in terms of light gathering. I downloaded the full patent 2827827 if the diagrams can be trusted, then the sphere-mirror called out as fig 52 is about 1\20th the diameter of the starball.
As I mentioned (and I don't know enough optical theory) is that if there is more reflective surface area there should be more light gathering power like a telescope. My Christmas ornament is on the way so we'll see what a 2" ball does. In the meantime I'll proceed with the mount.
I never got a chance to see the STS at Ruben Fleet. When I was 13 Spitz sent me a promotional packet (see pic) and I was fascinated by that machine. I was in San Diego 20+ years ago and was planning to see a laser show there. Unfortunately I was too tired because I was at the San Diego Zoo all day 😔
|
|
|
Post by Ron Walker on Jun 13, 2022 14:20:52 GMT -7
Posted by: moonmagic Sep 16 2017, 04:02 PM Again, I am amazed at the talents and skill on here. I certainly understand and commend you on seeing your dream to fruition. While I have the love for them, I've never had the technical skills required to undertake such a project. It is fascinating to read about what you are doing. Please keep up your work. mm
|
|
|
Post by Ron Walker on Jun 13, 2022 14:21:28 GMT -7
Posted by: Philostopher1 Sep 21 2017, 07:42 PM Pursuing the Spitz Model B mirror
I got my 50mm silver glass Christmas ball today. I was so concerned about light loss that I didn't take into account the surface aberrations in blown glass.
First the spherical mirror is excellent in principle. It solves two problems with the led I am using. 1. The stars are smaller - better than the fisheye lens. 2. The stars do not become dramatically ellipsoidal on the wide angles.
Problems:
1. More light loss than the fisheye lens but acceptable and new led technology will come. 2. The surface aberrations create bands of varying light and shadow -extraordinary optical noise. This is a problem; shadows will fall on pinholes and cause noticible uneven illumination.
I spoke with Paul at Spectrum coatings and he will coat a couple glass spheres with enhanced aluminum for a reasonable price. There are 50mm and some smaller k9 (Chinese BK7) glass spheres on amazon for decoration. The ones made by Neewer look promising. They make a lot of cheap camera accessories, lenses etc and their stuff is pretty good. These are optical quite clear indicating that the scratch and dig surface quality should be very good to recieve an aluminum coating. Btw, 50mm ball lenses with 40/20 surfaces are $300! Too rich for my blood and probably unnecessary. I asked about surface quality but I don't expect an objective answer.
I couldn't get a photo of stars as I used a cree t6 flashlight for the test. My Nichia led I am going to use is still mounted under the fisheye lens. It is a much brighter led.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:21:48 GMT -7
Posted by: Philostopher1 Oct 12 2017, 03:42 AM It's been a while. I've been cobbling together a little shop in our tool-shed. It's about 10 x 5 feet. I have a small bench top drill press , and my 'Bob's CNC'. Spending a lot on tools...
Updates on light sources:
I am backing away from the Spitz Model B arrangement. Instead, I am going with four fisheye lenses with leds in each hemisphere to cover 90-degree quadrants. Ron and another friend suggested this. By clustering them on the same mount they can be raised and lowered effectively offsetting the star field for the split hemispheres. The lens is much more efficient than a mirror and cheaper. I have also upgraded the lens to a 238 degree fisheye from "Miao Lab" from Amazon. This one is made for the 8mm sensor on an iphone 6 yielding a much clearer and brighter center portion of the lens. The "Mozeat" lens was for a smaller sensor and cost $14.99 this one is $16.99.
Also, concerning the led. CREE makes an XQ-E hi intensity led. This has no dome lens, just a flat emitter. The emitter is square, but can be masked to cut the corners off with a large pinhole. It's binned lumen value is about 250. The beauty of this led is that the device is 1.6 x 1.6 mm. The actual emitter is smaller. The fisheye reduces the apparent spot size about 70% thus effectively creating a 1mm point +\-. I need to get the actual devices. I am going off the data sheets... This is getting close to the xenon arc size of the lamps Spitz used in the 512+ series.
This arrangement also solves the problem of elipsoidal stars at higher radian angles and most efficiently distributes the brightest portion of the led's distribution. Corrections for magnitude balance can easily be compensated for with minute drill size changes but maybe unecessary.
Below is some pics of another star test with the new lens pointing arrangement and some of my ABS pieces for the fork and Yoke.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:22:08 GMT -7
Posted by: moonmagic Oct 17 2017, 12:12 PM Sorry for the delay since your last post. It is all still fascinating. I enjoy hearing about the project you are working on. Please don't think we don't if we don't reply to a post right away. "Life" keeps getting in the way of real fun. mm
|
|
|
Post by Ron Walker on Jun 13, 2022 14:22:26 GMT -7
Posted by: Philostopher1 Oct 17 2017, 02:56 PM QUOTE(moonmagic @ Oct 17 2017, 12:12 PM) * Sorry for the delay since your last post. It is all still fascinating. I enjoy hearing about the project you are working on. Please don't think we don't if we don't reply to a post right away. "Life" keeps getting in the way of real fun. mm
No worries. I started a blog to try to organize the project in one place but I'll post here first. I am struggling with carbon fiber this week...
|
|
|
Post by Ron Walker on Jun 13, 2022 14:22:47 GMT -7
Posted by: Philostopher1 Oct 24 2017, 08:46 PM Carbon fiber starball
I've been working with carbon\ kevlar fiber to make my own starballs. The aluminum one below is too thick at 1\8" and heavy. I wanted a modern high tech aesthetic so...
In retrospect, I didn't have to have these hemispheres made. A large excercise ball over-inflated would have made an excellent form! Whoops!
I think I will reflect on this carbon fiber as being the most difficult part. There is an art to this that I have to acquire as I go along. I was going to do this with gores and with more skill this would have been more elegant. Getting the cloth down, and edge-matched on a sphere is difficult. So I am going with a truncated Icosahedron pattern to keep each panel smaller so it will lie flat. Even with that the panels require a relief cut up the middle to take out the warps and folds. Flat surfaces don't map to spheres directly...
There are a total of 32 panels to do.
After some failed attempts with mold release using a gore, i had to peel off the cloth that was adhered with epoxy resin, then sand off the resin from the aluminum. That didn't work.
So far what does work is cutting an aluminum foil pattern to match the pattern you wish to skin with. This must be the thinnest foil that is practical to work with. I am using some from Dollar General.
Use 3M-75 or a similar repositionable spray adhesive and lightly coat the foil. Position this on your form. Smooth from the center out with a soft cloth and it will take on the shape very well. The wrinkles are pretty negligible in terms of the surface quality for the cloth.
Next, prepare your carbon and decide how you want to position the weave, paying attention to the front and back side. On mine the back has a hairy kind of appearance. Spray the back of the cloth with 3M-77 or similar. This is like spray on contact cement. Don't let the glue pile up it needs to be even or air will get trapped and work up through the resin. There are a lot of YouTube videos on Carbon fiber techniques.
Place your panel when tacky and secure the edges with masking tape. The tape will peel off later without too much trouble.
Wet out the cloth and work into the fibers. Leave it alone until the next day.
There is potential to get your pieces stuck. Especially in my situation with the aluminum. Epoxy resin may not stick to an excercise ball. It doesn't like plastics. Especially LDPE. I suppose one could just do one hemisphere at a time on an excercise ball then cut them in half and peel out the remains of the ball.
My plan is to loosen the panels as I go around and then lift the shell off the form when ready. At that point the inside of the carbon shell must be reinforced with fiberglass cloth.
Once I am really moving Ill take some video and link to it. That will probably happen with the second hemisphere when I am more confident in my skills.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:23:05 GMT -7
Posted by: Ron Walker Oct 25 2017, 12:36 PM I have never worked with this stuff and find it interesting. I wonder about drilling star holes in the material. First how brittle does it become and will a small drilled hole tend to "shatter" out from the center making a larger hole. Also if it is a weave kind of material would a star hole drill on one of the treads tend to "slide off" down between treads of material placing the star in the wrong place.
I like the idea of making the star ball out of gores as was the Spitz "B". You could use aluminum flashing to cut the gores from and drill the star holes on a flat surface. The material is about 1 mm thick which makes it easy to cut, drill, and bend into shape. The gores could be made a half inch wider to allow for a quarter inch overlap on each side. Put the overlap on say the right side and drill the proper star hole size on the lower gore. A larger eighth inch hole would be drilled in the overlapping gore to clear the underlining star hole. The gores could be glued, riveted, or screwed together but I think small number four screws would be the best bet. They could be placed where there are no stars. Three should be enough.
Anyway, keep the posts coming as they are very interesting.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:23:25 GMT -7
Posted by: Philostopher1 Oct 25 2017, 05:19 PM QUOTE(Ron Walker @ Oct 25 2017, 12:36 PM) * I have never worked with this stuff and find it interesting. I wonder about drilling star holes in the material. First how brittle does it become and will a small drilled hole tend to "shatter" out from the center making a larger hole. Also if it is a weave kind of material would a star hole drill on one of the treads tend to "slide off" down between treads of material placing the star in the wrong place.
I like the idea of making the star ball out of gores as was the Spitz "B". You could use aluminum flashing to cut the gores from and drill the star holes on a flat surface. The material is about 1 mm thick which makes it easy to cut, drill, and bend into shape. The gores could be made a half inch wider to allow for a quarter inch overlap on each side. Put the overlap on say the right side and drill the proper star hole size on the lower gore. A larger eighth inch hole would be drilled in the overlapping gore to clear the underlining star hole. The gores could be glued, riveted, or screwed together but I think small number four screws would be the best bet. They could be placed where there are no stars. Three should be enough.
Anyway, keep the posts coming as they are very interesting.
Once the carbon kevlar is done the resin thoroughly impregnates the fibers. The fibers that form the yarn for the weave are probably in tens of microns. So once the resin is soaked in and hardened the material is solid all the way through and very strong and light.
You may be noticing that the surface looks rough and bumpy. This is the initial wetting of the cloth. The rest of the dome will be continued this way. Once this part is complete and it can move freely on the form, a process of sanding and layering epoxy ensues building up a glass smooth finish. This is an arduous process... If this goes well the shell should have a deep tiger-eye like finish. Then the back side will be reinforced with fiberglass mat and painted black.
Carbon fiber and fiberglass are hard on drill bits. I am not going to be drilling star holes in the traditional way because of small bits being chewed up quickly. Instead, each star hole will be drilled in thin aluminum (Marie Callendars pie tins -we have a lot) stacked in layers. Then deburred by sanding through three grits. This makes excellent pinholes...
This is madness, but I am drilling the holes in the domes to accommodate black glass jewlery beads of a few different sizes. The pinholes will be epoxied to the beads, then placed into their appropriate positions. Larger stars will have their own lens and a little tube similar to the Spitz STS. As we know, the thinner the pinhole the lower the diffraction problems. This overcomes any "noise" from stray fiberglass in the pinhole or little cracks and fractures. I really don't have anything better to do with my time lol.
Here is a pic of the fiber detail
Here is what it 'should' look like when finished
|
|
|
Post by Ron Walker on Jun 13, 2022 14:23:44 GMT -7
Posted by: Ron Walker Oct 26 2017, 12:18 PM So then I've got to ask why not just use the large aluminum spheres you now have? The STS has each star (even the pinholes) mounted in a threaded mount which is then screwed into the thicker four foot diameter globe. You could use the same black glass jewelry beads mounted in the 1/8 " thick aluminum hemispheres thus not dealing with the carbon Kevlar at all. Drilling light clearance holes on such thin material could be a problem. With my luck the plastic would shatter while drilling the fiftieth hole. I would probably save that work and just drill out 1/4 " holes for each star and then glue the pie tin with proper star hole over or under the main globe. Pie tin material is great for so many things, I love the stuff.....the pies aren't so bad either.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:24:06 GMT -7
Posted by: Philostopher1 Oct 26 2017, 08:09 PM QUOTE(Ron Walker @ Oct 26 2017, 12:18 PM) * So then I've got to ask why not just use the large aluminum spheres you now have? The STS has each star (even the pinholes) mounted in a threaded mount which is then screwed into the thicker four foot diameter globe. You could use the same black glass jewelry beads mounted in the 1/8 " thick aluminum hemispheres thus not dealing with the carbon Kevlar at all. Drilling light clearance holes on such thin material could be a problem. With my luck the plastic would shatter while drilling the fiftieth hole. I would probably save that work and just drill out 1/4 " holes for each star and then glue the pie tin with proper star hole over or under the main globe. Pie tin material is great for so many things, I love the stuff.....the pies aren't so bad either.
Hi Ron!
The two reasons are weight and aesthetics.
The hemispheres are a bit heavy at about 1\8" thick. Also, I am going for a bit of artsy-ness. The carbon fiber is very light, and has a nice high-tech kind of appearance when finished. I am driving the Diurnal axis with a NEMA 17 stepper motor and planetary gearhead and GT2 timing belt to keep the two domes close together while allowing room for all the little milky way lenses etc and I think the aluminum will be too stressful on the motors causing them to be noisier. I think I am going to use a NEMA 23 on the latitude axis. As far as drilling, aluminum is definitely more forgiving and precise. The epoxy and cloth actually drill pretty nicely, without chipping and fracturing. The carbon kevlar is tougher on bits though, and doesn't give a polished quality hole; it is a bit fiberous.
I'd love to make individual threaded barrels like the STS, but cutting the threads and tapping all those holes in composite is a bit farther than I want to go.
As far as matching beads to hole diameters, glass beads are not precise, but approximate. The small beads like in indian crafts, are called seed beads and are sized small to large, 15 (2mm), 11, 8, 6(6mm), among each size there is some variation of diameters, roundness length etc. So you can pick the ones for a tighter fit in the hole. Approximately 300 stars will have a little tube and a lens in a few sizes. These will get cut and tapped threads or maybe just cut with a die and held with a washer and flat nut from the back. We'll see...
I was a partner in a small manufacturing business and we did a lot of production work assembling small parts. I calculated that each star will take five minutes to drill assemble and mount. 5 X 3000 (approx) = 15,000 minutes. 250 hours...
I am wondering at what point I will lose my sanity, though I think I already have - so I'll press on.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:24:25 GMT -7
Posted by: Ron Walker Oct 27 2017, 11:29 AM The more I think the more problems I come up with. If the beads are round wouldn't they be so thick that they could cause alignment problems with the light source?
The aluminum spheres weight should be no problem if mounted on a central shaft on ball bearings.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:24:42 GMT -7
Posted by: mrgare5050 Oct 28 2017, 03:24 AM Sanity is overrated - I haven't had much to say on this thread because, this is a niche in home planetariums that is to my mind unique, have we ever seen such high end engineering in pinhole projection, fantastic! I've often eyed beads in Hobby Lobby as the source of very cleancut star holes. Or maybe I secretly want to wear them. As a person who has extension cords that run into walls and I don't know where they go anymore, I can't contribute much but encouragement here!
How many niches in home planetariums are there? They blend seemingly - the legacy commercial (large and small), the homemade, the so called toy's which really arent toys, the high end video projection. I BOUGHT some homemade projectors, which is another iteration, preserving AND using them. The hybrids, combining parts of commercial and homemade. Its a large umbrella seemingly. Multi-faceted indeed.
|
|
|
Post by Ron Walker on Jun 13, 2022 14:25:03 GMT -7
Posted by: Philostopher1 Oct 28 2017, 12:31 PM QUOTE(Ron Walker @ Oct 27 2017, 11:29 AM) * The more I think the more problems I come up with. If the beads are round wouldn't they be so thick that they could cause alignment problems with the light source?
The aluminum spheres weight should be no problem if mounted on a central shaft on ball bearings.
You're right they would be fine, and there is plenty of torque with a 27:1 planetary gearhead on the motor. Again, it's an aesthetic issue, and I want to prove the concept of making starballs in this way as another option. I got a great price on the hemispheres but they are generally very expensive. The average among all the quotes was $1500 for the pair!
Also, the idea here is to be able to reproduce this project with readily available parts and materials through Amazon.
Beads and alignment.
I made some crude drawings that may help share my concept.
The idea is to cut identical sheets of aluminum from pie tins, brass shim stock, or cinefoil. The sheets are stacked together, and the pinholes are drilled in bulk.
A grid is scribed on the top sheet creating a drilling guide. E.g. a 10x10 grid on 10mm centers. This way one can stack 10 sheets, drill 100 holes and have 1000 pinholes etc. This allows the drilling to be done on a drill press with more precision, and less bit breakage, especially for #80 bits.
As I learned from others the best way to make pinholes, aside from laser drilling, is to punch or drill the material and then sand the dimple down through 3 grits. This creates a more diffraction limited hole. So each sheet is sanded through 3 grits with an orbital sander and cleaned.
A jig is made to neatly receive the pinhole sheet without shifting around. The sheet is used as to mark the centers for the jig. From here the jig is drilled through with holes to allow a small finishing nail to be inserted. An extraction tool is made -think bed of nails- that will push up through all the holes in the jig simultaneously.
The jig holes are drilled on the receiving side just large enough for the desired bead to be inserted and stand a few thousandths above the jig surface. Once the beads are inserted spray glue is applied (TBD but probably 3M-90) the pinhole sheet is then laid over the beads and clamped for 10 minutes or so.
Then, the part is extracted and individual star holes with their beads are cut up for mounting in the starball.
From here, each star is individually placed in its location through a hole large enough to receive the bead and pinhole, and epoxy to secure them. The process ought to orient the clear aperture to the lamp. Any off axis errors will be miniscule.
Yes, this is a lot of work. It's not necessarily the most efficient method, but it is the one that will yield the aesthetic and quality I am looking for.
|
|