17 inch 16:9 assault on high end project

[Guy Grotke]

It's an abrasive disk that is about 1 mm thick and about 1 inch in diameter. The Dremel has a built-in speed control, but I ran it at full speed. You can cut out some paper to fit 1/2" back from the cutting lines, then use the masking tape to fasten it to the smooth side of the fresnel. The paper will protect the surface as you move it around on your bench, etc.

I think you could use a thin abrasive disk with a fine grit on a die grinder. You just need a steady hand so you don't turn the grinder out of perpendicular and catch the edge of the wheel on the fresnel. It would be a good idea to put the fresnel on a clean bench (with the paper-protected side down), so the line you want to cut is just over the edge of the bench surface. Then put a clean cloth, a piece of wood, and something heavy on top of it to hold it in place. Use the Dremel, grinder, etc. to cut with the leading edge of the disk moving down into the cut. That way the bench will support the fresnel and stop it from flying around.

[Hezz]

Guy,

Thanks for the info. Today I bought a package of five 3 inch abrasive cut off saws. They are about 1.5 mm thick and are fiber reinforced. Although they are for steel and aluminum they should work for plastic as long as they don't load up too much.

I think I may chuck them in my drill which has more power and will limit the speed to around 1000-1500 RPM. Instead of 8,000-10,000 RPM of the die grinder.

[Guy Grotke]

Good idea. They are probably not rated for 8000 RPM. You don't want one of those flying apart when you are in the same room!

The Dremel disks do fly apart at the least bit of sideways stress, but they are so light that the air resistance slows them down by the time they hit anything. I just wear safety glasses usually, so I don't get a piece of crud in my eyes. I went through five or six cutting a stainless steel quiche mold for my reflector. I think I cut all four sides of my two fresnels with just one disk.

[Hezz]

The variable speed hand drill worked great with the cut off wheels.

I noticed that they were rated for 24,000 RPM so they are made for high speed but since I didn't have a full face safety mask I didn't really want to run at such a high speed.

Also, I thought it might be easy to slip at higher speeds and scratch up the fresnel. With the 3 inch fiber reinforced cut off blades it only took me about five minutes to cut out the fresnel. I now have it mounted in the LCD mount and in the frame.

I may have the projector ready to try out tonight except that now I need to order a few parts for my HTPC which has been down for six months since I robbed the CPU and memory out for another computer. I haven't been using it since it is a dedicated use HTPC for the projector.

I will order a CPU on e-bay if I can find one for it.

Also, I have a lot of work to do on the light engine polishing the reflector and mirror. I am going to need to get another smaller right angle die grinder and a collection of abrasive pads in finer grits.

I will post another picture when I get it all assembled with the necessary guts for projecting a picture.

[Guy Grotke]

I have serious doubts about your mirror. Rays from each pixel will strike an area on the mirror surface about 1/2 the diameter of the projection lens before being reflected into the lens. The lens will then refract all of those rays to a single pixel of the screen image. If your mirror is not perfectly flat over that area, then the rays will not converge to the same point on the screen.

I know that polishing a lens surface to be perfectly flat is just as difficult as getting it to a particular curve. I don't see how you are going to get your mirror surface to be all that flat without a harder perfectly flat surface to grind against it.

You can always just buy a first surface glass mirror to attach to your metal mirror surface.

[Hezz]

Guy,

I don't know, it can be gotten very flat but it will take a lot of time. At this point though I am more worried about the parabolic reflector. Like you said it is easy enough to glue a mirror to the aluminum.

Last night I fired the thing up and the LCD is working just fine. But the light through the system is dismal. I knew it wouldn't be good because I have yet to condition the reflectors. At this point the big reflector is far more diffusive than reflective. But I thought that I would at least get a dim image. I did but it was too dim to make out the outer edges of the screen.

Also I am going to have to shim the lens mount out about 3/4 of an inch. Because my final LCD position in it's mount is somewhat closer to the lens than I had originally planned.

It isn't columnating the light as I had anticipated but then again it is hardly even polished. So I may line it with reflective mylar to test it and see if the increased reflection will prove out my concept.

Also I left the thing on for about an hour and a half and the bulb housing got a lot hotter than last time. Not too hot to melt wires but too hot to touch. For some reason the ballest was humming and the light was making noise that it did not the first time.

The first time I fired up the light and ballest were dead quiet now there is too much noise and the tranformer is getting too hot. As far as I can tell it is wired up the same. I even tried reversing the bulb polarity thinking that is was possible there was more resistance one direction than the other. It didn't make any difference.

I guess this is why guys are spending money on the electronic ballasts.

The LCD mechanism stayed reasonably cool as far as I can tell so this is positive and makes me think that the silent passively cooled goal is possible.

At this point it is the light engine that worries me.

[ramkishan]

Dear Hezz,
You are doing wonderful work. There are many experienced and knowledgeable persons who will solve your problems. In case the problem is not solved to your satisfaction, you may think of adopting our approach. We concentrated on the final image right from the beginning and created the set up backward. Every stage we kept provision for adjustments and fine tuning and now our results are excellent, even appreciated by experts in the field. The optical circuit in the projector is quite tricky though appears simple. Whenever you find free time, please go through the content of our web page.

[Hezz]

Dear Ramkishan,

Thanks for the positive feedback. I have seen your projector and it looks really good. At this point I feel like I have been punished for trying a different light engine concept but hopefully I may get it to work.

If not I can always redesign with a more standard setup.

[Hezz]

I have been working at the lower bulb housing reflector and have the uppermost part polished. It is much more reflective and shiney than before.

Here is a picture of the reflector. Though the surface is quite reflective I don't believe that I will realistically be able to get this curved surface really true. You can see a bit of the blurriness in the reflectied image of the bulb.

I am quite certain that I can eventually get the large parabolic reflector about this quality. I hope that it is enough to allow for the projection of an image.

[Nads]

Nice lampholder. If it gets too hot, I am sure you can manfacture a larger add on heatsink to dissapate some more heat and not have to go down the route of forced cooling and stay passive. If it gets too hot then you might want to put a heat guard on it!