Having gone through

1. LCD panel arrived cracked; no response from seller.
2. LCD monitor arrived DOA; not worth shipping back for returns
3. LCD panel reported to be compatible with controller turns out to not be compatible

You'd think I'd give up on this fantasy..

Instead, I just dropped ~$50 on one of those NEC 6.5" XGA monitors (NL10276BC12-02). I think that's an amazing deal, although googling these forums reveals that I might be in for some serious FCC challenges.

Anyhow, assuming I can get past that, I have yet another challenge; I've got the USHIO 400DD with an appropriate electronic ballast. Standard fresnels and acrylic plus the S15 lens. I'm wondering what contraints I should expect on minimizing my box size? Usually the LCD is the limiting factor here, but in this case it will clearly not be. Should I switch out the S15 for a varifocal? Can I use the standard LL fresnels? Is heat going to be a major issue with the 400W?

I've scanned the forums but to no avail. Wondering if someone can point me in the right direction.

You do know that you will need a controller for your Lcd ? And they are hard to find?

Just for clarity, I bought a monitor. At least the implication is that it is such, with pictures of the video inputs, operational screen and the inclusion of power supply and "all the electrical components". Its not just a panel. Sorry for the confusion.

Sounds like bad luick getting a screen.

Well, for a 6.5" LCD, there shouldn't be much problem with width or height.

You'll need at least the 320mm from LCD to triplet, however. The only way around that is to get a shorter FL triplet. (The varifocal that LL has would be of some use, here, but really, you probably want 220mm or shorter) You could also use a shorter FL collimator lens, and you'll likely want AT LEAST a precondenser in there or your brightness will suck.

If you can find it, I'd recommend a 110 or shorter FL collimator lens.

If you can find it, a 220mm triplet (Plus the 220mm fresnel you already have), and the 110mm collimator should reduce the overall projector length to about 16".

Very interesting LCD.
Where did you buy it?
There was a guy selling LVDS controllers on Fleabay a few weeks ago.
his auction said he will Write controller BIOS for your LCD.

Ebay was the one; the seller was moving whole units, with metal cases and basic VGA and audio inputs. Looks like they were removed from some sort of security or other monitoring application. That was the angle that the seller was pushing them on, but I figured a fully working 6.3" XGA monitor out of the box was one of the better deals you're going to find.

Supraguy, you've been a great help in here and in our back and forth PMs. I've read your 'understanding lenses' thread and its also great. What I'm having a hard time understanding are the following questions:

1. How is the FL of the fresnels determined? Another way to ask this is: why is it possible to use fresnels with smaller FL for a smaller panel? I assume this has mostly to do with the LCD panel size and the angle of light coming from the source (corners versus center) that needs to be collimated into the panel?
2. How do these factors affect the type of precondenser that can be used? Will the commonly used 6.5" surplus shed precon work in this setup?
3. Same question as #2 but for the reflector (will a hemispherical reflector just larger than the bulb suffice for this purpose?

Thanks again!

What FL you can use for a fresnel is all about inverse square law.

I should really keep this somewhere that I can copy/paste it, 'cause it's a pain to work out.

Inverse square law states that the further away you get from a radiating light source, the intensity (lux) of that light diminishes as the square of the distance increases. This means that if you are 2 times as far away, the light will be 1/4 as bright. 1/2 as bright happens at sqr(2) the distance, or 1.414 times the distance.

1/2 as bright is about as much as we ever want the corners of our projection to be, since more than this is easily visible to the human eye. This much or less, however, we can get away with.

Now say we have a 15" LCD, and we have the collimator fresnel at 220mm from the lamp. The CENTER of the fresnel is 220mm away, but the corners are further. We can use pythagorean theorum to find the distance: 7.5" (1/2 the diagonal size of the LCD) from the center is 190.5mm. The distance therefore is sqr(220*220+190.5*190.5) = 291mm. This is still under the tolerable maximum. (The ratio is 1.32:1)

Say we used a 155mm fresnel, though. The distance from the lamp to the center of the fresnel would be 155mm, but the distance to the corners would be sqr(155*155+190.5*190.5) = 246mm. This is an inacceptable ratio (1.58:1) where the corners absolutely MUST be less than 1/2 the brightness of the center.

But, with that same 155mm fresnel and a 10.6" panel, the corners are only 205mm away, for a ratio of 1.33:1 -- which is again acceptable.

Work it out for a 7" or smaller LCD, and a 110mm fresnel becomes acceptable as a collimator.

Incidentally, this is also how commercial projectors get high brightness from small panels.

Now for the front of the box, this is simply a matter of determining the desired throw of the projector.

A 320mm triplet gives a short throw to a 15" panel, a decent throw to a 10.6" panel, and a very long throw to a 7" or below panel. Thus for small panels, a shorter FL triplet, and matching collector lens becomes possible.

Actually, I think I'm going to pin this topic, so that I can just point to it in the future.