Dull red from a video projector

I have no idea if it's commonplace, but the cause is likely related to the effect of photographs and advertisements turning magenta under exposure to daylight.

The red filter is red because it absorbs green and blue light. Blue light is most-effective at breaking molecular bonds, as its photons have the highest energy. QED?

It might also be that the designers chose a poor-quality dye for the filter.

is

highest

filter.

Perhaps the dystuff centrifuges out to the edge ;-( I'm surprised these colour wheels hold together as they seem to be individual glass sectors super-glued into a disc.

Is there any easy way to look at the colour wheel and see if the filter really is the problem.

Could it be the lamp failing in some strange way? You can get a rough idea of the spectrum if there is some light spillage which has not come through the colour wheel (reflections coming out of ventilation slots in the projector, for instance). Take one of the clear CDs that is used to protect the bottom of a pack and, taking care not to get any fingerprints on the disc, hold it so that the tip of your nose is poking through the centre hole. Close one eye and move your head so that the light from the lamp appears above the top edge of the disc.

If the light source appears sufficiently small, you should see a clear spectrum with a good quantity of each colour.

computer

leads

different

What an intriguing test, I'll make sure no one can see me while doing it, they may phone up the funny-farm. (look, the loon has taken to playing CDs with his nose) The lamp spillage looks normal white but if such a spectrum test shows an absence of red it would be useful elimination test. Would your "CD" difraction grating disc yest work with any lamp souurce for comparison say car headlight or a xenon floodlight bulb just to see what a complete spectrum would look like with such a disc.

I once demonstrated it at a coffe morning of the Bath Royal Literary & Scientific Institution. After handing out half a dozen discs, I had a great time watching them trying it out (pity I didn't take a photograph).

I would have thought that a normal-looking white would suggest that the lamp is OK - although I suppose there is a faint possibility that some sort of filtering or correction system could be upset by the absence of some parts of the spectrum which are too narrow to affect the overall appearance.

It works with any source, but the image you view should be as small as possible if you want to see any detail of lines or holes in the spectrum. You can either mask-down a large source or stand a distance away from it.

I've just tried the test with my own projector and the lamphouse spill shows a fairly continuous band with a couple of strong lines in the green and one in the cyan. I have tried to photograph it for you, but the bands don't show up:

With the screen set to 'black', the residual light coming through the lens appears a bit stronger in the red region and the lines are not noticeable, so I suspect there is some correction going on.

The light seen through one of the other vents appears to be a very strange magenta-ish colour (too diffuse to analyse with the CD). I would guess that this is coming out of the back of the lamp and is the unwanted transmission from a dichroic mirror.

Had that intermittent problem with an Epson (?) DLP projector. Blue and green were fine, red was brown sometimes, cable seemed fine, but replaced it anyway. Problem disappeared.

Yes, an excellent point! Can this projector display any setup or splash page? Does that also have the bad red? If the splash page is good, then it very well could be the cable. I have several LCD monitors that sometimes shift colors when the cable is bumped, must be poor contacts on the connectors. This would be especially true with analog video signals.

Jon

replaced

signals.

The menu structure is that functions selected are designated vy changing blue to red lettering so are not that obvious as highlighted

computer

different

it

So you replaced the cable? As the monitor out seems fine I assumed it could not be a cable problem

Instead of

limits

for a

loose

filter

come

in

to

poking

it,

I suppose those discs are rejects from the pitting stage so never go on to the matalising stage.

I tried with some LEDs , the most convincing was "point source" non diffused blue, just a slight bit of green in the spectrum with the blue. Even orange lens over a neon was quite convincing. The more light the less convincing presumably because the eye is overloaded with the most prominent allowing the minor spectra components to be more prominent

Through one section of the fan louvres you can see part of the dichroic mirror and that is distinctly red when the lamp is on , I wonder if the dichroic plating? can break down over time and stop reflecting red and absorbing it instead, ie passing red and infra-red light.

If I get inside again I may try mounting a piece of mirror in the lamp light path and try reflecting a red LED laser into the light tunnel and colourwheel and see if that brings up the red component/s of the image

Don't risk damaging your eyes, a focussed arc lamp is a dangerous beast*. If you can see inside with the machine switched off, it would make more sense to just look at the red sector of the colour wheel and see if it has turned brown.

I am begining to think that the other respondents have got better suggestions and the problem is more likely to lie in the cables, the signal, or even the colour-handling software.

• These lamps are dangerous in other ways, too. I once had to replace one which had burst after a cooling failure, the explosion had bent a

light

Not just the bright light , but the cooling is disrupted by removing the cover. I drape silicone sheet over wonen glass mat over the active area to obviate shatter, intense light/UV, venting pitfalls.

The spill over light gives plenty of red on the "difraction grating". I see this "blank CD/DVD" has a big clear patch on its surface, so those discs are rejects from the first stage of production

Yes, replaced the cable, problem disappeared. Since it was used with a laptop, I suspect a bad pin or socket. When the problem stopped with a new cable, I stopped looking. BTW, the problem hasn't reappeared.

limits

loose

and

replaced

new

Tried another cable but the same absence of red

This thread is a bit confusing. To got one person talking about a color whe el and another talking about a dichroic mirror.

The twain shall never meet. there can be a DLP without a color wheel, but i t will be a three panel and it will have dichroic mirrors. Though I don't b elieve it has been done, there could be an LCD with a color wheel, but it w ould need one superfast refresh panel.

Why don't you just post the model number.

You have two completely different scenarios here, and one main point, is th e red dull in the OSD generated inside ?

If it is a three panel of whatever type, and the fault appears in the menus , you are looking at a fried dichroic. If it is a one panel it is most like ly the color wheel but a physical inspection is warranted.

If it only has a three section color wheel that's one thing. They just are n ot very likely to fail in that way. If they do you will see it, there sho uld be a band where the light was. If it has a seven segment color wheel, y es it could quite have darkened but it's not all that likely.

Incidentally one brand of these does get a fault in which the red section o f the seven segment wheel breaks free and then not only is the color screwe d up, the thing vibrates like all hell.

Why no model number ? Or did I miss it ? I do believe I read the thread her e. Hell, I might have the service manual on my harddrive. sure you have it, but I don't know what this thing is and we don't even know if it's a DLP o r LCD or lycos or what.

If it's a seven segment color wheel and the red looks good, it probably is in the signal processing. That is a whole different basllpark.

The twain shall never meet. there can be a DLP without a color wheel, but it will be a three panel and it will have dichroic mirrors. Though I don't believe it has been done, there could be an LCD with a color wheel, but it would need one superfast refresh panel.

Why don't you just post the model number.

You have two completely different scenarios here, and one main point, is the red dull in the OSD generated inside ?

If it is a three panel of whatever type, and the fault appears in the menus, you are looking at a fried dichroic. If it is a one panel it is most likely the color wheel but a physical inspection is warranted.

If it only has a three section color wheel that's one thing. They just are n ot very likely to fail in that way. If they do you will see it, there should be a band where the light was. If it has a seven segment color wheel, yes it could quite have darkened but it's not all that likely.

Incidentally one brand of these does get a fault in which the red section of the seven segment wheel breaks free and then not only is the color screwed up, the thing vibrates like all hell.

Why no model number ? Or did I miss it ? I do believe I read the thread here. Hell, I might have the service manual on my harddrive. sure you have it, but I don't know what this thing is and we don't even know if it's a DLP or LCD or lycos or what.

If it's a seven segment color wheel and the red looks good, it probably is in the signal processing. That is a whole different basllpark.

+++++++

The dichroic is the parabolic reflector behind the discharge lamp, letting through IR and reflecting visible light. 3 sector colour wheel. If I can rob a colour wheel from a defunct projector I will try its red sector in the light path of this one just to check the white areas go red etc in the output image. As putting a normal mainly white image through the projector from a graphics package, ramping up the red slider to 200 so the pc monitor image is redded out , the projector image is no change in the white areas , so I suspect signal processing area fault

This is a Mitsubishi XD200U projector. I've just taken the optical block apart from an InFocus LP420, RGB when viewing through the colour wheel but interestingly complements of cyan,yellow , magenta mirroring reflected when viewed at an angle