| mathias |
| short answer: 100mm in diameter, but I think you can cut it down on the hight if you are going to use it with a widescreen lcd. |
|
|
| slize |
new idea for perfect reflector:
what do u guys think about a bike frontlight?
most of them are made of alu and have a good reflector type...
i'll check this out tomorrow
something like this;
not the price, but the form of the left one may be |
|
|
| slize |
i thin i found my favorite:
Philips ALUline... |
|
|
| The Alchemist |
Ok so the PAR desing is nice for the smaller single ended 150w CMD lamp. But what about the dubble ended one's or the single ended big one's (mogul based). 400w or 250w. Why design our own reflectors when these are used in reef holding! So i searched for a test on these reflectors and found a great article:
http://www.advancedaquarist.com/iss...003/feature.htm
I couldn't believe the results of one of the reflector.
Diamond Light Lumen Arc III:
Look at the even light output!!!!!:
Very great for people with a larger than 7" screen.
one drawback: The reflector is 19"x19". But maybe we can make a smaller design?
It can be bought for 149$ at http://www.coralfragz.com/drygoods.html |
|
|
| ace3000_1 |
| quote: | | one drawback: The reflector is 19"x19". But maybe we can make a smaller design? |
Second draw back, frensels need a point source, the only way that could be used is with one frensel and with the frensel reversed otherwise your wasting your light.
Trev |
|
|
| The Alchemist |
| I alway though that we are using multiple fresnel lenses to convert our point source light into a parallel beam. with this design it is already a (all most) parallel beam so do you still need the fresnels? Maybe i see it all wrong, correct me then. |
|
|
| ace3000_1 |
2 frensels together make a big dual condenser lens, now inside of the double condenser structure cos u are using 2x plano convex lenses (frensels) we have a gap that gives us the paralelle rays and they are the rays that go through a lcd paralelle, this is ofcourse if u have the lcd inbetween the 2 fresnels, the other way is to have the lcd after the 2 frensels but in that way the fresnels have to be slightly bigger then the lcd as the rays are going through the lcd slightly offset to perpendicular of each pixel, depending on focal there are advantages and disadvantages in this system, also if u use paralelle rays before the frensel good luck in finding a condenser the same size as the reflector, as u know a condenser can intensify the lights intensity by 3x (dependant of the power), this will yeald us a brighter more intense light, its the same with using 2 frensels, 2 frensels are always brighter then just 1, if u wer to use the system above the light would be difused scatterd light, and also it wont have much intensity with just 1 frensel and just 1 condenser if u want to find one that big, the best way for projection and the eficentcy of using the majority of your light is a reflector with a focal point, the light reflected from the reflector will go back through its source, the frensel with also a focal point will gather the light directly from the focal point of your condenser while the condenser will gather all light from the point source of the bulb, an improperly matched condenser can loose us asmuch as 60% of our light, its just as important as our reflector, typically with the right matched condenser and reflector combo using a spherical reflector and a condenser u can achieve 80 - 90% from the source, ive been talking to a few optical engineers latley and im telling u that the majority of u guys in here are wasting upto 60% of your light by using the wrong configerations in your light engines, there are a few calculaions to take in mind, and these are what u need to know before u can even start to get any efiecentcy from your source, a spherical reflector with a short focal is the best for a ohp type system, this has to be calculated for size and focal by the specs on the light pattern generated by your bulb, also the condenser has to be calculated for the light pattern generated by the bulb so that u are gathering all of the light from the front of your source, also there is power thats an issue and condenser focal, with this system in mind it would far out do the parobolic reflector design in a reflector system no matter what lenses or configeration u used in projection, there is only 1 other thats better then a spherical system and thats an eliptical reflector, the reflector its self can gather 90% of the sources light where as a spherical can gather upto 65%, also an eliptical reflector has 2 focals, 1 for the bulb and 1 for the condenser, also the light doesnt go back through its source, so realy when u think about it with an eliptical reflector and a properly matched condenser its not that hard to get the 99% of light from your source, and its exactly what the pro projectors of today use.
Trev |
|
|
| neededandwanted |
To do a test for parallel light rays on a light source, shine it on a wall. If it is a true source of parallel light, the light won't be bigger that the source. That is, a lamp and reflector (or a lamp, reflector and the first fresnel) will make a bright image on the wall that it the same size as the reflector (or fresnel).
If you move the light closer to the wall and further away from the wall, the image on the wall will NOT get larger and smaller. It will not get brighter and dimmer either.
Image a square hole in the wall of your house that sunlight comes through and you see a bright square of light on the opposite wall. This is VERY nearly parally light from the sun being masked off by the wall (and coming through the hole in the wall). If you used a movable wall (or a screen) and moved it closer to the wall with the hole and further that bright light square on it would not change in size or brightness.
Another good test while you are holding the light is to see what sort of shadows can be cast by the light. If you hold up an object between the light and the wall, you should get a VERY sharp shadow. The shadow should be the same size as the object and not get smaller or larger when you move it.
So that's the test for perfectly parallel rays of light, once you have a light source to play with.
As far as looking at a picture of a light source and trying to determine (before buying it) if it will put out parallel light, consider this:
Can you see the filament of the bulb when the light source is not pointed at you? Can you see reflections of the filament in the facets of the reflector when you are off-axis? If so, some of the light will be aimed at you when it is powered on.
As far as a perfect reflector goes, it is possible to use a point source or a line source. As long as light is coming from an exactly known location and is coming out in a known direction is can be focused to do anything you want (including parallel rays). A point source is better for this because all of the light is is going in a totally know direction. When the light hits a surface (like a mirror, curved or flat) you can control exactly where it is going to go because you know exactly what angle it is striking the surface and can design the surface (mirror) to bounce the light exactly where you want it to go.
With a line source, there is less known about the angle. When a bit of light hits your mirror, you don't know if the light came from one end of the filament or the other or near the middle some where, etc. The longer the filament is the higher the uncertainty is.
With a planar source of light, there is the unpredictable on both an x and y axis. The light ray that hits the mirror could have come from anywhere on the light source. |
|
|
| The Alchemist |
Ok thanx both. Sorry for my late reaction. I am getting more and more confused. I don't believe a "normal" reflector (spherical, parabolic) will do the trick with dubble and mongul based lamps.
But which will???? I will dig into it deeper or just do some tests.
Oh, i found some pdf on the optic of theater, film and disco spots.
it is in german, but look at it, because the pictures also explain a lot.
http://www.revolux.com/Wissenswerte...rfersysteme.pdf
Ps. I have a question. I bough a wide converter lens van chinon (~1$) can I do anything with it? It realy looks like a projector lens. the diameter of the bottom is 4 cm and the diameter of the top is 5 cm. I know it is used to lower the focal lenght of an other lens, but can it be used on its own. |
|
|
| The Alchemist |
I found the design for the dubble ended bulbs!!!!!!!
It's is called a Dual paraboliod reflector and can be used for larger arced lamps (as is shown in the articel).
It is not easy yo make but, we have to try.
Here is the article
http://www.wavien.com/scientific/re...bruary_2003.pdf
And this is the reflector (build out of two half parabolic reflectors and one spherical)
 |
|
|
| ace3000_1 |
Thats very much the same as the later model car's head lamps but its not offset, the shperical in the center while the parabolic or eliptical behind, u can do that system with a dual eliptical reflector system too thats even more eficient and has a better light control providing u can find the right focals, that system u have posted above would have a long focal, it would be good for fiber optic light guides and not so much projection unless u had a small lcd and room to spare, its a nice design though.
Trev |
|
|
| Dwiel |
I finally have some pictures of my incomplete projector:
Find them here (under the diy section): dwiel.no-ip.com.
Let me knwo what you think.
There will be new pictures uploaded soon hopefully. I have been very busy recently, but I am hoping to get some better picture of it in action. I didnt realize that my version LOTR ROTK was so bad :(.
Let me know what you think |
|
|
| slize |
| i think i hit the jackpot, with the AluLine Pro in combination with the cdm-t 150W from philips, i will post some pics today... |
|
|
| slize |
here are the promised pics:
this is the package: Philips AluLine Pro 50W |
|
|
| slize |
| this is the bulb; as you can see it is very small, only 56mm in diameter |
|
|
| slize |
| it is also very easy to take apart, because the reflector is aluminium |
|
|
| slize |
| that is the disadvantage of an aluminium reflector: the whole heat will be beamed forward. but i think with a cold mirror and a heat blocking glas i will use, it will be no problem... |
|
|
| slize |
| with that reflector i will make a nice beam spot and with a condensor i will bring it on the fresnel; to be continued... :smash: |
|
|
| ace3000_1 |
| quote: | | but i think with a cold mirror and a heat blocking glas i will use, it will be no problem... |
Heya slize, nice find buddy, i think u will find the cold mirror will be all that u need, ive got some here also and its good stuff, removes about 90% of the heat, also are u in europe bud? just courious as to where u got your reflector from.
Trev |
|
|
| slize |
hey ace,
thx man, yeah i am from europe; i have 8 of this bulbs (Aluline 56) left here, if you are interested... normally they cost about 12 euros one bulb, but i got it on ebay for a very good price. the exact print on the package says: Philips Aluline Pro 50W and it has B15d socket
there a three types from Aluline: 111; 56 and 37
these are the diameters of the bulb in mm
just mail me if u are interested ak2kde@yahoo.de
i would need 2 cold mirrors with 85x65mm in size each...
do you have such mirrors and what will they cost? i am defently interested, may be we can trade bulbs with mirros or i think we will find another solution... what do u think bud? could you mail me some more infos about your cold mirrors? size price etc...
or do u know a good source in europe for cold mirros
as i only find firms and they don't sell to endusers :bawling: |
|
|
| ace3000_1 |
hehehe sure buddy, ill send u a mail now.
Trev |
|
|
| zkuger |
Hi,
I am kind of new here, so I apologize in advance
for this simple question, but I have just got to ask:
If you have a point 'light source', and you put it behind
an LCD panel, would it not just project the LCD image
over the wall? (like hand shadows)
I don't have an LCD panel yet, so I can't test it. But
I think that if a light beam pass the LCD strait, then
it is possible. On the other hand if a light beam that
pass the LCD is scattered in all directions then it isn't
possible.
If it is possible but the light source is not a point, than
many images would be cast on the wall from the many
virtual point light source, and so the image would not be
clear.
You can test it with a Laser beam or with a light beam
or with a point light source if you know how to make one. |
|
|
| mathias |
Because nobody else have mentioned this, and Iīm sorry I didnīt mentioned this along time ago. There is one important rule with all reflectors:
Big arc size = Big reflector size, small arc size = small reflector size.
There are many example of this, in pro-projectors they use a 50mm reflector and the arc size is 1-2mm, 50mm halogenreflectorlamps 12V have a filament size of 1-2mm, PAR30 with a reflector size of 100mm have a filament size of 10mm. The arc size in a cdm-t lamp is 9mm, so to get the best resultat you need a reflector size of 100mm or bigger.
I have self tested the cdm-t lamp with a 50mm halogenreflector, not good at all. |
|
|
| slize |
hi mathias,
what defines the size?
is it the wire in the bulb that defines the size or is the body of the bulb as the wire only brings the halogen to lighten.
so then the body of the 50W bulb you posted is nearly the same size like the light chamber of the cdm-t.
and may be you had to cut away too much from the 50mm reflector...
i have a 56mm reflector... the problem is i want to build a compact pj so i can't afford to use a 110mm reflector or even bigger...
i had very good results with this reflector as the beam was VERY intense...
i also tried a 110mm reflector and the bulb was as big as the 20W bulb in your post
and i had a PAR 20 bulb from Osram (Halopar 20) which had the same bulb in it like the the PAR 30 bulb and a reflektor with only 57mm...
only differences between those 2 was the angle of the light, and i want as less angle as possible the make a paralell beam for the condensor...
so my question is:
is there a specific rule that defines the size of a reflector for every lamp?
i thought:
the smaller the reflector the smaller the light angle the intenser the light...
what were the problems with the 50mm reflector?
i am trying to build a light machine like ace did, and i thought the intensiest light would be best; to be continued... :smash: |
|
|
| slize |
| u are talking about a Spherical Reflector of a 50mm halogen bulb |
|
|
| slize |
but i am talking about Parabolic Reflector
it is 30mm long, so i don't see any problems |
|
|
| slize |
| here you can see the difference between a 50mm and a 56mm reflector, the hole fits perfect for the cdm-t 150W |
|
|
| slize |
| here is a side shot of both in comparison; i think the aluminium one is a parabolic and the 50mm reflector is a elliptical one, but i may be wrong... |
|
|
| slize |
| here is the reason i don't want to use 110mm reflectors, it is just too big |
|
|
| slize |
i agree with you mathias that there may be too small reflectors for certain bulbs and that the 50mm is too small, because you have to cut away so much to make the bulb fit, but i think the 56mm reflector is enough
we will finally see when i start my first projected pics with the new light unit i am planning. may be i am proofen wrong then... |
|
|
| mathias |
| quote: | what defines the size?
is it the wire in the bulb that defines the size or is the body of the bulb as the wire only brings the halogen to lighten.
|
The filament size in a halogenlamp is the size of the rolled up thin wire in the glassbody, that is the only thing that give light, 1-2mm for 20-50W 12V Halogen and 10mm for 75W 230V Halogen.
Of course a big reflector with a small arc is not wrong, thats why you can find 110mm reflector with bulb big as the 20W bulb in my picture.| quote: | | and i had a PAR 20 bulb from Osram (Halopar 20) which had the same bulb in it like the the PAR 30 bulb and a reflektor with only 57mm... |
You can only get 10 degrees PAR20 lamp with 50W and I think that that filament is smaller than the PAR30 filament. 30 degrees PAR20 version goes up to 100W, with a bigger filament size.
| quote: | thought:
the smaller the reflector the smaller the light angle the intenser the light...
|
Yes this is true, smaller area more light, but with a big arc size you will not get parallel light beam out from the smallreflector and this was the problem with my 50mm lamp. I donīt know if there is a mathematical rule for this, but the distans to the reflectingmaterial should be as big as possible with a big arc size, because the angle difference gets smaller. So this rule will be true till somebody find a mathematical rule.
Big arc size = Big reflector size, small arc size = small reflector size(Can also be big). |
|
|
| ace3000_1 |
| quote: | | The filament size in a halogenlamp is the size of the rolled up thin wire in the glassbody, that is the only thing that give light, 1-2mm for 20-50W 12V Halogen and 10mm for 75W 230V Halogen. |
Ah huh, but its also horizontal, when u put a lamp in a parabolic reflector and if the arc is running long ways u will get unwanted results, this is cos the arc is longer then the focal point, some of the bulbs arc will be in the focal point and some wont, this will result in non true paralelle light being reflected from the reflector.
Trev |
|
|
| ace3000_1 |
| quote: | | the smaller the reflector the smaller the light angle the intenser the light... |
Actually thats false, the bigger the reflector the more light it can gather, only when u gather that light to a point it will be brither then having a small reflector reflecting only part of the light. The closer the bulb is to the reflector, the brighter the reflected light will be.
Trev |
|
|
| slize |
thx trev.
another question:
i am thinking about using two 150W lamps. But before i go and buy that stuff i thought about a way to make one spot out of two.
the black things are the mirrors; you see what the problem is, either i have two spots or i loose light because not all the light hits mirrors and go into the one spot. i want to have both spots fully made into one for doubling the lumen numbers
I also thought about using prisms i can use but i didn't find a good solution.
What do you think?
here is a quick drawing of what i meant
slize |
|
|
| slize |
here it is from a commercial projector, but i don't really get how they make it...
do they use it like in my drawing on the left side?
if it is like that, that's not what i want to achieve. i wanna have one spot out of thes two? isn't there something like a beamsplitter? (not the regular one, i know it works different)
i guess their mirrors are not 90°
but where do i have to focus both spots then; on the condensor, or on the fresnel, i don't really know |
|
|
| mathias |
| quote: | | but its also horizontal |
Yea, that was maybe stupid to post a picture on that type of lamp, more expensive 50mm halogen reflectorlamps has the filament mounted vertical, but the filament size is still 2.5-3mm in size against 9mm for cdm-t. |
|
|
| ace3000_1 |
| quote: | | but the filament size is still 2.5-3mm in size against 9mm for cdm-t. |
Thats so the arc stays in the reflectors prime focal position so they dont get unwanted results, the best reflector for a cdm-t is the eliptical as ive said before.
Trev |
|
|
| ace3000_1 |
heya buddy, with that idea on the left it will work, a prisim is the best thing for us because the fs mirror needs to be beveled 45deg on both mirrors so the light is 1 continuos wide beam, u will not get it any brighter trying to get it into a narrow beam from the 2 bulbs, but u will get it brigter if the beam is concentrated to only cover half of the lcd per bulb as that beam on its own is more intense and is more concentrated covering a smaller area rather then covering a bigger area. With this ideas u will need only 1 frensel, 1 that goes over the front of the lcd to guide the light through the projection lens, the light that will come from these mirrors will already be paralelle. This system i dont see as being a big advantage in certain ways, but i have tricks up my sleeve for this as ive fooled with this mid last year and found enhancments while testing, cash flow was the thing that stopped me and damage to my eyes.
Trev |
|
|
| slize |
sorry to hear that,
so you don't think it is worth to go for as you already tested it? cause the problem is, i'd love to use this cdm-t bulb but 12000 lumens is not enough for 15" tft. i already tested it.
I am using a 400 HQI-TS with 36000 lumens but i think it is too hot too heavy too big, the picture is nice though but i am searching a way to use the cdm s, and i thought i could increase light output to let s say 24000 lumens with this way...
I cant even use a reflector with the 400W cause it is so big, so i think i loose light anyways, that's the reason i wanted to go for 300W and an efficent way to use light
slize
p.s.: here is the link i got the lumen numbers from:
http://www.strassenlicht.de/technik/lampen/typen1.htm |
|
|
| ace3000_1 |
Yeah its a goer buddy, but bear in mind your going to use 2 bulbs lol, just make sure the light is paralelle, this is very important, ill have another crack it this soon for ya, ive got a billion projects going now so time is abit short, (ive prety much designed my next 3 projectors ready to go).
Trev |
|
|
| slize |
3!!! :bigeyes:
i'm still working on the first to improve it :smash: |
|
|
| ace3000_1 |
heh yeah 3, once u have a good base point on one the rest are easy to design, no matter the bulb or projection lens, one of them will be a xga native, i may use the dual bulb design in that.
Trev |
|
|
| zkuger |
Hi,
I have done a little experiment with a 1.5 Volt flashlight inside
a toilet paper roll, and a slide on the end of that roll. I have seen
the slide projection on the wall, in a dark room.
Conclusion: Point light source can project a slide over a wall.
Advantages:
1. All the light from the slide goes to the wall = No light is lost after the slide.
2. It is easy to project the slide as a square i.e. side projection.
3. No need for a lens after the slide, actually no need for any lenses
beyond the point light source.
Chalenges:
1. Does it work with LCD?
2. Building a strong and uniform 'point light source'.
PS
The flashlight with its surrounding reflector wasn't a point light
source, and didn't produce an image. Only without it, and probably
also due to the fact that the lamp is so tiny, it was a point light
source, or close enough, and the image was projcected over the
wall.
| quote: | Originally posted by zkuger
Hi,
I am kind of new here, so I apologize in advance
for this simple question, but I have just got to ask:
If you have a point 'light source', and you put it behind
an LCD panel, would it not just project the LCD image
over the wall? (like hand shadows)
I don't have an LCD panel yet, so I can't test it. But
I think that if a light beam pass the LCD strait, then
it is possible. On the other hand if a light beam that
pass the LCD is scattered in all directions then it isn't
possible.
If it is possible but the light source is not a point, than
many images would be cast on the wall from the many
virtual point light source, and so the image would not be
clear.
You can test it with a Laser beam or with a light beam
or with a point light source if you know how to make one. |
|
|
|
| slize |
this is what i use right now 400W HQI-TS
bright as hell but not effecient at all, that's why i decided to change to the cdm s |
|
|
| slize |
look what i've found;
it is from this side:
http://www.led-lichtcenter.de/index...42.html&lang=de
they put a normal 150W hqi in a reflector that is 65mm in dia
i'd love to test this one; but the bulb costs 174
i wonder what they do with their damaged ones, may be i'll call the producer to ask if they sell their damaged ones |
|
|
| slize |
ok, today i tried to take a few pictures to show the difference between the the 110mm and the 56mm.
i aimed at a white paper which was about 23cm away from the lamp (my focus of the first fresnel).
but all i got were hot fingers and nonshowing bright pictures on the camera. sorry i wasn't able to take it in picture to show the difference, but it is defently there.
so the main difference with the big refl i saw the side wire of the bulb in the projected lamp spot. with the small one it wasn't visible because the light spot was so intense...
now my thought and aim is, with a condensor lense in front of that i can evenly spread it onto the fresnel; but i wasn't able to additionally handle that; to be continued... :smash: |
|
|
| slize |
| quote: | Originally posted by slize
i wonder what they do with their damaged ones, may be i'll call the producer to ask if they sell their damaged ones |
No chance, they won't sell it :bawling: |
|
|
| ace3000_1 |
| quote: | | so the main difference with the big refl i saw the side wire of the bulb in the projected lamp spot. with the small one it wasn't visible because the light spot was so intense... |
Slize get the light from the big reflector to a point with a condenser, then focus the frensel to this point, only then u will have a brighter image with a big reflector, the big reflector will capture more light but we need to get this back to a point to work in an eficient manner, otherwise we are just loosing intensity. With a smaller reflector it is more intense, but u will capture less light then the big reflector.
Trev |
|
|
| slize |
ahh now i see;
i finally decided to buy the stuff for the dual bulb projector, if i got all my parts together i will begin to build it
andy |
|
|
| ace3000_1 |
Nice one buddy, hey how much can u get cdm-t sockets for where u are? also the actual bulb.
Thanks in advance
Trev
btw i got your email. |
|
|
| mathias |
| quote: | another question:
i am thinking about using two 150W lamps. But before i go and buy that stuff i thought about a way to make one spot out of two.
|
This subject has been discussed alot on this forum over the years.
But I havenīt seen any good solution yet. But the last idea I got is from all the 3 panel lcd projectors, they use a prism to mix together the red, green and blue color to a full color projectingbeam. I think this will work on diy-projectors also.
And if I have known this 2 years ago, I had probably not used the cdm-t lamp, because of this: 100W 12V halogen give 2400lm if you overvoltage them a little they give 4800lm and with this prismdesign I can get 3 times more = 14400lm and thats more than what the 150W cdm-t lamp give. 100W 12V Halogen bulbs are dirtcheap. So this could be a nice thing for diy-beginners. |
|
|
| ace3000_1 |
mathias, u wont get 3x the light output, that takes a special prisim for this kind of combining, it take a dichric prisim, if the center lamp just shines the light into a normal prisim the light will get shinned strait back to its self from the other 2 reflecting halfs, so the 2 mirrors on the outside have to be dichoric so that light can pass through the prisim from the center lamp.
As for the cdm-t, if u get a phillips cdm-t they yeald 12500lm, a GE brand cdm-t will yeald 14000lm, this is also on the 4200k spectrum not the 3200, the GE also lasts 12000hrs.
Trev |
|
|
| mathias |
| quote: | | that takes a special prisim for this kind of combining |
Yes that is right, but this is the only lampmixer I have seen, and I donīt think there are any other way to do it. So if people can find special condensors they can probably also can find this type of prism and with it you will get 3 times more light.
And yes the Halogen thing is not as good as the cdm-t, but itīs much cheaper to start with and many people donīt want to start with HID-lamps. |
|
|
| ace3000_1 |
| quote: | | So if people can find special condensors they can probably also can find this type of prism and with it you will get 3 times more light. |
The prisim required is a 45deg prisim, it is also dichoric, the back lamp can only output 50% of its light without some sort of light recyler, even with that said only 50% of the light can be reflected by the 2 side bulbs, so realy we can only get 50% out of each light, so in reality we will get 1.5x the light. It woiuld be way better to use 2 bulbs on a normal 60deg prisim to combine the light into 1 beam and then we would have roughly 2x the light output. No special condenser needed either, 2 pcx condensers will do the job fine.
Trev |
|
|
| ace3000_1 |
Slize your a champ, ill be incontact with you soon.
Trev |
|
|
| slize |
| quote: | Originally posted by mathias
So if people can find special condensors they can probably also can find this type of prism and with it you will get 3 times more light. |
Yep they will find it probably and will pay A LOT of money. A friend of mine had 3 panel lcd projector and this thing got damaged by a fall and this prism got damaged not alone, but to replacement of this special prism costs 2000!!!$ :bigeyes: so he decided to buy a new one after all and sold it his broken pj on ebay for about 400$ |
|
|
| ace3000_1 |
Was that a prisim near the light or near the 3lcd's? the one near the lcd's is the most common. I can get them cheaper then that. try about 1/18th the price lol.
Trev |
|
|
| The Alchemist |
Funny -----> click
Propably not useable, but still funny :D. |
|
|
| slize |
| it was next to the lcds... but 100$ is still too much for me... |
|
|
| slize |
| quote: | Originally posted by The Alchemist
Funny -----> click
Propably not useable, but still funny :D. |
i saw that on the german diy forum, and it started a VEEERRRYYYY long discussion if this will work...
but they ran out of it, so nobody could get and try it yet... |
|
|
| mathias |
| quote: | | It woiuld be way better to use 2 bulbs on a normal 60deg prisim to combine the light into 1 beam and then we would have roughly 2x the light output. |
Trev, I donīt understand how you can get that to work, so please explane that to me, because itīs not working for me.
| quote: |
Yep they will find it probably and will pay A LOT of money. |
Slize, how are you going to solve the problem then ? |
|
|
| ace3000_1 |
| quote: | | Trev, I donīt understand how you can get that to work, so please explane that to me, because itīs not working for me. |
Heya mathias, the 60deg prisim will invert the beam 10deg towards the center of the prisim to converge the light beam into 1, (you would think they would call it a 40deg prisim but its 60deg from the outer plane).
Trev
P.S i sent you an email did u receive it the otherday? |
|
|
| slize |
| quote: | Originally posted by mathias
Slize, how are you going to solve the problem then ? |
probably by using mirrors |
|
|
| mathias |
Trev, I'm sorry I still donīt understand it: What I can see, the only thing the 60 degrees prism is doing is changing place on the two beams they never get mixed. And if you change the angle on the lampbeam there will be rainbows.
Ps. I havenīt got any mail from you ? |
|
|
| slize |
| he was asking me if i got his mail |
|
|
| ace3000_1 |
| quote: | | he was asking me if i got his mail |
Actually i asked both, mathais, that picture is wrong ill draw u one up, u dont get rainbows with white light nor with the right optics. The picture u have drwn up is 1 way of doing it though, but with a dichoric prisim with a beam splitter.
Trev |
|
|
| ace3000_1 |
Here is my pic mathias, its self explanitory so i dont need to explain, sorry for the crappy drawing lol.
Trev |
|
|
| mathias |
Trev, what you have drawn is a mirrorprism, there is no way a glassprism can work in that way you have drawn. With a 60 degrees mirror prism the two beams will spread from each other. And the 45 degrees mirrorprism is not working either.
My conclusion is, there is no way to mix two beam with normalmirrors or prism. The only way to do it is with semipermeable material. |
|
|
| The Alchemist |
What about using curved mirrors? When looking at a telescope i can think about a system like this:
It's just a wild drawing in 2d, in 3d it will be not ok i guesh but maybe other non planer mirrors can be used? |
|
|
| ace3000_1 |
| quote: | | Trev, what you have drawn is a mirrorprism, there is no way a glassprism can work in that way you have drawn. With a 60 degrees mirror prism the two beams will spread from each other. And the 45 degrees mirrorprism is not working either. |
mathias, the angle u have thought is wrong, its measured from a outer plane as stated earlier and its why i said i dont know why its a 60deg prisim when it should be a 40, the angles on yours are the wrong way round, BTW i like everyones drawings lol (makes mine look better lol)
Trev |
|
|
| ace3000_1 |
| quote: | | My conclusion is, there is no way to mix two beam with normalmirrors or prism. |
Yup your right on that note, but depending on the grade we have our materials we can acheive things that the human eye can and cant see.
Trev |
|
|
| mathias |
| quote: | | The angle u have thought is wrong, its measured from a outer plane as stated earlier and its why i said i dont know why its a 60deg prisim when it should be a 40 |
Okej, I think I understand what you mean, but with this the beams only cross each other, so the problem still remain. |
|
|
| ace3000_1 |
Ya and thats where u use a lens to corect this, u dont have to use a 60deg prisim, u could use a 46deg prisim, it all depends on how wide and where u want your light to go, but to get a strait beam u would have to use a corecting lens to convert the light so its paralelle, probally 2 condensers could do this, i havnt fooled around in this that much with any condensers so its realy only a guess at this stage, it should work though, it would have to be a double condenser setup i would think. But with this idea we can also use a frensel at the point of where the light converges, this will give us a point source, also u could use this with a normal condenser setup.
Trev |
|
|
| mathias |
| After some thinking and calculation, I think I have a new idea of a lamp mixer with two 45 degrees glassprisms they are cheap and easy to find. The last prism is for colorcorrection of the first prism, otherwise there would be rainbows from one lamp. |
|
|
| ace3000_1 |
Mathias, a good rule to follow, what is on one lamp should be on the other, this will keep it eficient and help get the desired results, the last idea may work though, but u would need 3 prisims.
Trev |
|
|
| mathias |
| quote: | | what is on one lamp should be on the other, this will keep it eficient and help get the desired results, the last idea may work though, but u would need 3 prisims. |
Iīdont know what you mean, you can always move the prism so the beams overlap each other and you can turn the last prism around. Why 3 prisms then I itīs work with 2 ? |
|
|
| ace3000_1 |
Sorry mathias i miss read your last post, but one thing that has got me curious is how can that work with one prisim having that degree of refraction while the other hasnt? wouldnt both sources of light have the same refraction difference on the 2 colour corrected prisims? they wont match up, and if only the prisim to the left is colour corrected, the prisim on the right has a big difference of refraction, its working as if its a lens.
Trev |
|
|
| ace3000_1 |
| quote: | | Why 3 prisms then I itīs work with 2 ? |
Why 2 prisims when its the work of one? lol use a correcting lens after the prisim. |
|
|
| mathias |
| quote: | | but one thing that has got me curious is how can that work with one prisim having that degree of refraction while the other hasnt? wouldnt both sources of light have the same refraction difference on the 2 colour corrected prisims? |
The first prism (that with the green rays) give refration in one direction on the red rays, to correct this the last prism is turned in the other direction, so the refration take out each other. The green beam has no refration because totalreflection.
But one very strange thing with this design is that it doesnīt work in the other direction (backwards) ? So if there are anyone good at optic is this magic or what ? |
|
|
| mathias |
| Sorry, with refration I mean dispersion (rainbow effect). |
|
|
| ace3000_1 |
Mathias, what about the blue spectrum? there is red green and blue to make up white light, the rainbow effects wont accur on a lcd, ive never seen this and about all u would see on a miss- matched light system is a blur, the rainbow effect could possibly be seen if the operating frequencty of the light is lower then the lcd's but even then with light combined already coming from each source combined into 1 u wont see this effect, the lcd filters this light with its own static colour filter. Dlp has this issue because the light beam is filterd into colour and operates at strict timming with a colour wheel, maybe u need to elaberate with this rainbow efect on a lcd because ive never seen it in my life.
Trev |
|
|
| mathias |
| quote: | | what about the blue spectrum |
Iīm sending in white light into the prism from the lamps, if that is what you mean. If the rainbow effect is not a problem than this thing will be even simpler, angle the second lamp so the rays comes out in the same direction as the first. |
|
|
| ace3000_1 |
| quote: | | angle the second lamp so the rays comes out in the same direction as the first. |
Thats all u need to do with using the 1 prisim :) |
|
|
| slize |
what do you think about this?
the first prism has a rainbow effect on the white light, the second reveals this effect, so one white light ray comes out. the black second light hits the second prism at a negative angle so this should be reflected and goes into the first light ray...
so i have one out of two beams...
based on this site and paint shop :D
http://www.uni-leipzig.de/~stann/java/prisma.html |
|
|
| ace3000_1 |
Guys, im talking about using simple mirrored prisims here, its all u need, what prisims are u talking about? because u wont get a rainbow effect on a normal mirrored prisim.
Trev |
|
|
| slize |
it's a normal prism, fully made out of glass...
but anyways, as i looked on my budget i decided to give the single 150W bulbed version a try, as i saw good results in the german diy forum where this guy claims to have a 70W MH bulb and got a picture bright enough on an acceptable size.
May be the 150W will be enough to lighten a 15" tft; i'll give it try...
P.S.: Did you receive my adress? |
|
|
| ace3000_1 |
heya slize ya i got it, im running out now to post it, btw i was talking about mirrored prisims, use those they are what we need for this aplication, the others will loose too much light.
Trev |
|
|
| mathias |
| quote: | Guys, im talking about using simple mirrored prisims here, its all u need, what prisims are u talking about? because u wont get a rainbow effect on a normal mirrored prisim.
|
What I can see the mirrorprism will not work as a light mixer, even with lenses, the result out of that is a beam that is double as wide as high. And that is logical, to mix two beam you have to place them over each other and that is only possible with semipermeable material.
Glassprism will not lose more light than a condensorlens, because a condensorlens is many small prisms. |
|
|
| ace3000_1 |
| quote: | | the result out of that is a beam that is double as wide as high. |
Hows this so? u could make this double wide depending of the angle of your choice, but not double high, its the same as having 2 mirrors but because with a mirror the peak cant be together Beacuse of the thickness of the mirror) you use a mirrored prisim, you have the sharp peak at the front so there isnt a gap in the lights center, noticable by the human eye anyway. You will find mathias no projector will use your idea, its impractical and runs costs very high.
Trev |
|
|
| ace3000_1 |
| quote: | | Glassprism will not lose more light than a condensorlens, because a condensorlens is many small prisms. |
A condenser lens is a lens, nothing to do with a prisim, prisims arent magnified unless your wer to order one that was. They are totally different.
Trev |
|
|
| mathias |
| quote: | | could make this double wide depending of the angle of your choice, but not double high |
Trev I donīt think you have tried your idea, you donīt need the mirror at all actully you can change the beam angle on the lamp instead but the problem is when you mix them together with the lenses. you can not change the angle because then you got a hole in the middle with no light.
Even if my glassprism idea is not the best, it is in the right derection.
| quote: | | A condenser lens is a lens, nothing to do with a prisim |
Wrong! Have you been on a physics lesson ? |
|
|
| ace3000_1 |
| quote: | | Wrong! Have you been on a physics lesson ? |
Teach me otherwise. The things u are talking about mathias are on a different level, this is about afordable and efiecient projection, we could go on and on all day about this, but at the end of the day it will cost so much not only in time, but also money making it not worth it.
Mathias, a lens is a lens, a prisim is a prisim, if they wer the same they would not be named a different name, a condenser lens is not a series of small prisims, if has SOME certain properties of a prisim but by all means it has nothing to do with a prisim, AND WILL NOT BE THE SAME. This is being told to me from my gf who has a master in quantum machanics, if you want to argue further about it, then i sujest argue with a proffesor in this feild, im not qualified nor yourself to be arguing in such issues without the relivant experience.
As for testing my theories out, i have actually, i tried it last year in june with 2 mirrors and 2 light sources with relevant reflectors. Ive told my findings, ive given my advice on how to get it to work and thats just it, unless your going to use a rear frensel u dont need a point source light, ive given you my findings, my advice, what you need and now the rest is upto you.
Trev |
|
|
| mathias |
| quote: | | a lens is a lens, a prisim is a prisim |
Yes thats right, why I bring this up was because you said a prism is loosing much light what I mean is that a condensor lens loose the same light.
The only diffrens between prism and lenses are that lenses has a curved surface but they bend the light in the same way as a prism so there is nothing to say about that.
Anyway I have a simple solution to your mirror-lens design. It important to think in two demension X-Y, your design work if we use 4 lamps, to get it to work with 2 lamps we need two special lenses curved only on X-side and staight on the Y-side. X=--- Y=||| |
|
|
| mathias |
Here is a little tip to all diy:ers, a laser is a good instrument to measure lenses and control your setup with. Itīs really intresting to see what happen when it goes through a fresnellens.....more than you think.
Itīs friday night, so I show a cosy picture of a laserbeam through a prism. |
|
|
| ace3000_1 |
Mathias, our problem in here on this matter is comunication lol, your talking about 1 thing while im talking about another, none of us is wrong, we are just talking with different parts on our mind that the other dont know but asumes its the part they are talking about lol, ill get into this abit latter on with you, but right now i got to finnish my project, by all means keep the tech rolling.
Trev |
|
|
| zkuger |
Hi people!
I am thinking about building a solar LCD projector i.e. I want
to build a sun tracking parabolic mirror that will collect the sun
rays, and from there I will make it become a small strait beem that
I will insert into the house and the projector through a window.
What do you think?
I am also thinking about what to do in the night. If I would put the
light engine outside and pass it also through the window i.e. it can
be very big, could I use cheeper/bigger light source(s) ? |
|
|
| rmccoll |
I have a question about split fresnel lens, are they two regular fresnel lense put together or two special fresnel put together?
Can they be cut to fit smaller lcds? Will cutting it reduce the effect caused by cutting them?
Can I just use two plain fresnel lens? |
|
|
| diylabs |
The fresnel lenses that we use for DIY projectors are actually made up of two fresnel lenses that have been glued together. A split lens is one of the original lenses that has been un-glued. The reason this may be beneficial is that you can then place the LCD between the two lens halves, taking advantage of more ideal angle of incoming light rays.
You can cut down the fresnel to fit your LCD, but doing so does reduce the amount of light that you can collect. If you are using a 7" LCD for example, then I would recommend putting the LCD about 4" away from your fresnel lens (non-split) to get more of the light that comes out of your system. This will result in a much brighter image than if you were to cut the fresnel down to be the same size as your LCD.
Note: cutting and/or spliting a fresnel can sometime crack your fresnel lens. If you don't already have a fresnel lens and would like one that has been split, then you should purchase one that has been split for you. diylabs.org offers such lenses.
If you split your lens or buy one pre-split, and then you decide that you would like a single (normal) lens, then you can simply put the two lenses back together - they will function just as if they had never been split. |
|
|
| ace3000_1 |
| quote: | | If you don't already have a fresnel lens and would like one that has been split, then you should purchase one that has been split for you. diylabs.org offers such lenses. |
These can be cut with a router that wont crack or damage the lens, also will not scratch the lens if setup in the right way.
| quote: | | If you are using a 7" LCD for example, then I would recommend putting the LCD about 4" away from your fresnel lens (non-split) to get more of the light that comes out of your system. This will result in a much brighter image than if you were to cut the fresnel down to be the same size as your LCD. |
Yep that would work but it would be the same as having a huge ohp with a tinny lcd, would look prety darn silly imo, also there would be a heap of wasted light and light leaks. The better way is to get short focal frensels with the apropiate condenser so the light can be placed closer to the lcd and the light guided just to this size, it will far out perform then the above mentioned idea.
Trev |
|
|
|
