Although it's not in its final box, now that I have a pretty good projection I can afford to be more critical.

When I project a black screen such as a tv station with no signal or if it's a darker scene this effect is more visible but if I look closely it's there all the time.

This is an exaggeration of the fog/mist it's less noticeable than this but it's there and the same pattern as shown.

Click to view attachment

The dark areas seem to me to have slightly better contrast but it's hard to tell up close.

It is visible on every tv station and dvd.

There is no reflected light outside the box.

The inside of the current box is not flat black it is still a natural beigy brown.

I am projecting on a white roller blind but I held up a flat white sheet of paper and could still see the efffect

Is this stray light picked up by the triplet or is this common?

btw) For a long time I didn't notice it until I looked very closely at the screen and then I could see it easily.

Looks like a fresnel de-polarization effect that I've seen before. Or it may be a small light leak from the front of your projector. Some details of your set-up may help shed some more light.

DJ

I had a very similar issue with my first "projector"(a Sharp QA-1400 projection panel on an OHP). It certainly did look like fog in a dark scene or black screen. In my case, I believe the panel itself was the culprit. Since I was using an OHP, I was able to rotate the lcd atop the projector. The exact "fog" pattern moved when the LCD panel moved. Unfortunately, I wasn't able to solve the issue, even after disassembling and thoroughly cleaning the QA-1400, the pattern remained. Hopefully, this is not the problem with your projection, and DAZZZLA will have an answer for you.

just a stab in the fog here, have a look at your lcd's F.F.C's and make sure all connections are fully inserted.

I am having the same issue on my pj. I have eliminated the fresnels and stray light as being the culprit, which leaves the panel.

I read elsewhere that this might be related to light reflecting/refracting inside the panel, and can be corrected by darkening the edges of the panel with electrical tape or some such. I haven't tried that mod yet, though.

Another member clued me in via pm.

A search revealed this thread on glowies:

Glowies link

This pic was posted there which shows the exact problem I see on mine. I always thought glowies was something else.

I can't confirm the bleeding is bad like that around the text but the foggy pattern is very similar to what I posted above and I had never seen that picture before.

Click to view attachment

For those who asked or wondered, I am using a straight box, 15" lcd, split design standard LL lenses and an older fat 400 mh (4000 - 4200K) with coil ballast. No reflector with this bulb although I have bought and IKEA stainless steel bowl to try later.

I am also testing all kinds of bulbs ie.. 300 watt incandescent, various tube halogen worklights and a 26 watt cfl bulb. The cfl bulb is dimmer of course but no fog and beautiful contrast. As a side note I am looking for a brighter cfl locally and also one with a a higher kelvin rating ie... 5000 to 6500. Although dark at only 1600 lumens, I like the basic look on my 66" diag image but that's another story.

I bought the pj as is for a great price from another member knowing full well that it needed tweaking. I just wasn't sure what it needed. At first it is exciting just to see such a good projection coming from a home made box that it's hard to find fault. But of course after that the experimenter and tinkerer is awakened.

I think it may be reflected light inside the box. The lcd is not tight to the wooden frame so the edges of the panel are showing. The sled is not tight to the side of the box. I have a small piece of wood between the rear fresnel and the lcd to block the extra light but it's not perfect, when I look through the the triplet I can see a thin sliver of light all the way through the bottom edge. The top of the fresnels are held in plastic page/binder type holders but there is a thin line of light hitting the screen. The metal LCD controller is not covered and could be causing some reflected light also. The effect may not be noticeable with the cfl because the image is so much darker.

An afterthought:

The only inconsistency I see is that the pattern I see doesn't seem to match the light leakage in the box. The top and bottom edges of the fog might be explainable this way but the thicker broad area running though the center is harder to understand. I could believe a fresnel diffusion issue because that is where the maximum light would hit the center and spread outwards from there towards the edges. Is there anything in the design of fresnels that would cause a greater spread towards the wide side ie left to right along the center of the fresnel?

I believe there are two possible causes to this problem. One is the simple edge bleed problem the other is a complex total internal reflection inside the fresnels themselves. The first cause is easy to fix, simply blacken the edges around the LCD. The other cause isn’t so easy. Not all LCDs seem to suffer, probably because of different materials or manufacturing processes.
Here’s a couple of pics I took of a fresnel placed in front of an LCD monitor. The camera had a polarizer placed on its lens and rotated to extinct the polarized light from the LCD monitor. If you look at the LCD where the fresnel isn’t covering it you can see that it is blocked by the polarizer on the camera lens. But the fresnel is somehow changing the polarization of the light traveling through different areas. Does the second pic look familiar?
Click to view attachment
Click to view attachment


What I’ve found is that different materials alter the polarization in different ways. For example a sheet of plain paper placed hard against the LCD will change the polarized light to un-polarized light. A sheet of plain acrylic has little effect on the polarization when viewing light passing perpendicular through it but reflections from an angle are heavily polarized.
If you look at the second pics again you can see that the center doesn’t alter the polarization, it’s still black like the rest of the LCD but the edges are either un-polarizing or retarding (rotating) the light allowing it to pass though the polarizer on the camera.
So the groove added to the plain acrylic are somehow altering the polarization particularly at the edges.
I suspect that the light passing though the edges has an angle that interacts with the facet angles causing it to be total internally reflected towards the edge of the acrylic then reflected back again towards the facets where it again is total internally reflected and exits the fresnel. Somewhere in this action the polarization is being altered. The rear fresnel is probably the cause in susceptible LCDs.

A couple of test you could try: Move your lamp position forward and backwards and note any changes.
If you can, try offsetting your rear fresnel and lamp. Move it 20 or 30 mm from there center. I realize that doing this will mean that one edge of your LCD will no longer receive any light but it should show on the opposite edge a reduction in glowies because you are now effectively using a larger fresnel and moved this problem area out of view of the LCD, if my theory is correct.

DJ

Here’s another pic with the polarizer on the camera rotated so it is passing the light from the LCD. The edge dimness may indicate that the fresnel is retarding the light, but I’m not certain of this yet, or it could be just inefficiencies of the fresnels causing the light to appear dim.

DJ