I found this: (on the site: Image Brightness guide for projection )
Here are some reference figures:
Movie Theatre with emergency lights on approx. 50-80 LUX
Typical office theatrette 120-150 LUX
Ballroom 150-200 LUX (dimmed for presentations)
Training room dimmed 200-250 LUX
Training room full light 350-450 LUX
So if I only gona project 0.75 square meter and darken te room it will be enough?
Here are some reference figures:
Movie Theatre with emergency lights on approx. 50-80 LUX
Typical office theatrette 120-150 LUX
Ballroom 150-200 LUX (dimmed for presentations)
Training room dimmed 200-250 LUX
Training room full light 350-450 LUX
So if I only gona project 0.75 square meter and darken te room it will be enough?
There's one design that I've been considering lately. That's using a parabolic reflector (with the lamp in focal point) and a negative fresnel after that with the same FL than the lower fresnel have but with negative sign.
As reflector you could use PAR64 or PAR56 relflectors which are widely available, cheap (< $/€10) and can be fitted with 150/200/250W type MSD / HSD bulp with a RY 9.5 connnector (I've seen one with double-ended rx7s connector too). That way I think you could get first the collimated light from the reflector (well, almost, as the direct light isn't very collimated) and then widen the spot with negative fresnel for the fresnel & LCD. That way also a larger LCD (15") could be used, with good efficiency, too.
As reflector you could use PAR64 or PAR56 relflectors which are widely available, cheap (< $/€10) and can be fitted with 150/200/250W type MSD / HSD bulp with a RY 9.5 connnector (I've seen one with double-ended rx7s connector too). That way I think you could get first the collimated light from the reflector (well, almost, as the direct light isn't very collimated) and then widen the spot with negative fresnel for the fresnel & LCD. That way also a larger LCD (15") could be used, with good efficiency, too.
Java applet, optics
I found This programm.
it is a java applet where you can see what happens when you change the distance or size of you (spherical) reflector and wat happens when you add a lens.
As I can see it: your light source in the center of your spherical reflector is the best way.
I found This programm.
it is a java applet where you can see what happens when you change the distance or size of you (spherical) reflector and wat happens when you add a lens.
As I can see it: your light source in the center of your spherical reflector is the best way.
Check this site out sell fresnel lense. On the site you will see a 200 foot fresnel lense. I don't the prices or if the will sell them individuals http://www.ntkj.co.jp/products-e.html#Fresnel
Ah,indeed. The program only uses a (almost) half point source.
As i see it; you loose a lot of light when you place your lamp at the focal point. Or atleast the picture wil be blurred.
Place the light source in the middle of the sphere would be the right way i think.
But also then, the exact middle has to be found otherwise you also get unwanted results.
As i see it; you loose a lot of light when you place your lamp at the focal point. Or atleast the picture wil be blurred.
An externally hosted image should be here but it was not working when we last tested it.
Place the light source in the middle of the sphere would be the right way i think.
An externally hosted image should be here but it was not working when we last tested it.
But also then, the exact middle has to be found otherwise you also get unwanted results.
An externally hosted image should be here but it was not working when we last tested it.
It´s a very nice program you have found ! But it´s only use the half light from the point source, what about the light going up and down wards ?
What you are simulating is a double-ended bulb or a halogenbulb. A 1-ended bulb has the most light going up and downwards because you have turned it 90 degrees. As you can see in my diagram on page 117, a 1-ended HQI-bulb doesn´t spread the light very much directly from the bulb, the most light will bounce in the PAR-reflector. The light directly from the bulb will go in a straight beam to the fresnelllens.
What you are simulating is a double-ended bulb or a halogenbulb. A 1-ended bulb has the most light going up and downwards because you have turned it 90 degrees. As you can see in my diagram on page 117, a 1-ended HQI-bulb doesn´t spread the light very much directly from the bulb, the most light will bounce in the PAR-reflector. The light directly from the bulb will go in a straight beam to the fresnelllens.
mathias said:
So you think that when using a bulp inside the focal point the direct light will not be a problem? How about if you use HSD/MSD type MH bulps? I must order one PAR64 (or PAR56) reflector to try with it.
Maybe it's possible to get the source code for the Java app and modify the light source.
Origonal adress
A PAR reflector is a PARabolic reflector and that is diffent for the spherical reflector. And I agree with mathias that this program can only be used for dubble ended tubes. I found the orginal adress of the program including mail adresses from the makers. So go on and mail them, maybe they can add a parabolic reflector to.
http://webphysics.davidson.edu/Applets/optics4/intro.html
A PAR reflector is a PARabolic reflector and that is diffent for the spherical reflector. And I agree with mathias that this program can only be used for dubble ended tubes. I found the orginal adress of the program including mail adresses from the makers. So go on and mail them, maybe they can add a parabolic reflector to.
http://webphysics.davidson.edu/Applets/optics4/intro.html
I can´t see any problem with the bulp inside the focal point, I get a very sharp picture with my PAR30 reflector, the direct light beam from the bulb actully compensates the light that should bounce into the hole of the reflector.
I haven´t seen any spread-data for the HSD/MSD-bulbs, but I think they works the same way, because of the arc-tubes edges.
I haven´t seen any spread-data for the HSD/MSD-bulbs, but I think they works the same way, because of the arc-tubes edges.
Attachments
I think that is the absolute best way you can build a diy-projector. But use the coollight version of the PAR30 when using a 7" tft because you press much more light throu a 7" panel than a 15" panel. The 7" will get much more hot than a 15" tft. Of course you can build it like Ace do, with a ohp-setup with a strong condensorlens, but that will be around 40% darker than the PAR-design.
A PAR30-coollight version with a 7" tft will be very bright,very quiet, (we don´t need to cool the lcd so much) and small.
Can we wish for more ?
A PAR30-coollight version with a 7" tft will be very bright,very quiet, (we don´t need to cool the lcd so much) and small.
Can we wish for more ?
ahhhhhhhhhh more room? lol in my design i dont have a fan on the lcd so it wouldnt make it any quieter, the light would actually run hotter in your design being the reflector getting so hot from filtering ir and therefore u would need to cool that very eficiently, i do have a design for a par30 type system, ill bring that out later on if i keep going with projectors, the size of the projector will be the same mine is now and will use a cold mirror.
Trev
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.