metal halide lamps for projection?

[polyglot]

Hi all,

250W and 400W metal halide lamps are readily available (under $200 including ballast and luminaire) for use as floodlights, and they have a colour temperature of about 4000K. Has anyone tried using them as a homemade projector lightsource? Or know if the large ones designed for floodlight use are suitable for projection?

Their spectrum is not quite continuous but according to wikipedia, bulbs using this technology are commonly used in projectors, though I don't know if the halide mix is the same. Do I need to look for something specific like a HMI lamp?

I've just ordered two 250W units (each with similar lumen output as about 1.2kW of tungsten light) for use as floodlights and think I might have to try one as a projector bulb. While I'm sure they'll look nicely white as floodlights, I'm just concerned that the spikes in the spectrum won't necessarily play nicely with the dyes in E-6 chromes and that some colours may look artificially bright or dim depending on how the source and dye spectra align.

 

[scarletbegonias]

Polyglot- It sounds like an interesting experiment. MH bulbs create fairly clean light to the naked eye, but it is a bit green/blue, which may have an impact on your experiment. I grow orchids (yes, orchids) under MH bulbs and two issues that one might run into are warm up time and temperature. They take a good minute to reach capacity and they get VERY hot. I still have a scar from when I touched one 10 years ago and I can envision melting film under one quickly. I suppose as long as the film is at least a foot away and vented, one may be able to pull it off. Good luck, Larry, Colorado

 

[polyglot]

Yes, the film will be on the other side of a condenser so I don't imagine it getting too hot. I think I'll just leave the bulb in the luminaire for safety (explosion-proof glass front) reasons. That ought to keep the rest cool, at least I hope so.

I've no idea how much heat the slide will absorb due to it not being transparent but I'm sure I can find a black frame to test with.

Now I need to finish converting my LPL 6x7 film holder to take 85mm slides...

 

[samcomet]

Mate. A few things to consider...yes by all means keep it inside a quartz/glass housing as they emit a lot of UV radiation (i.e. melanomas and retinal burns). One should NEVER look at the light source directly of any arc lamp anyway. If you need to look at one, use a welding glass please. HMI's were developed by Philips specifically for the film industry and present with a Colour Rendition Index (CRI) in the upper 90's. That means when an object illuminated with an HMI and photographed with a film stock that matches it's "daylight" balance of 5500 (aprox) degrees K, it will appear as it should - CRI of 100 is, in the daylight example, the midday sun (i.e. perfect). A QI (quartz iodide) globe would have a CRI in the high 90's when photographing an object with tungsten balanced film (3200 deg. K). MH's do not necessarily have great CRI's. You should look for a lamp in the 90's to get accurate colour rendition. Because of the mercury content in all AC arc lamps (discharge lamps) they tend to emit a green spike or a magenta spike and DO in fact age over time - a few degrees K for every hour they burn. CRI is a standard industry reference so check out the globe's manufacturer's website. I would also consider investigating some IR (infrared) gel (heat absorption) just to be on the safe side too. It is readily available in film/tv lighting supply houses or some gaffers who may be in and around Adelaide. Good luck with the project! sam

 

[polyglot]

thanks sam - know about the safety stuff.

It looks like all of the lamps available for the luminaires I've got (E40 mogul base, 250W) have a CRI of 65-75. I have found a very few bulbs with a CRI of 90 and 95 but have no idea where I'd get an affordable ballast for them yet, let alone a good safe way of containing them.

I think I'll have to give it a go with the 250W CRI 65 bulb anyway just to get an idea of what the brightness and heat level are like and whether there are colour rendering issues. While the colour quality is lower, I suspect that the effect during projection will be less than during photographing since there are only a small number of absorption dyes in a developed chrome but a whole lot of sensitisation dyes. My suspicion therefore is that the spectrum should matter less for projection, but of course I may be about to prove myself wrong...

 

[samcomet]

Just a quick thought that if you are not happy with the colour of the projected image and can get your hands of a colour temp meter, you can gel the lamp to the proper colour temp (degrees K in the orange/blue range) and colour compensating temp (mired shift in the green/magenta range) to improve the image. good luck anyway!!!! cheers, sam