Small full-spectrum consistent light source?

[Serafino]

I need a small battery-powered full-spectrum light source with consistent output both in terms of color bias and light level to fit into a small device which will be used for color-referencing.

I had thought LEDs would be the answer and I was about to buy the Luxeon utility module, but the more I learn the less sure I am that LEDs are my best choice. I gather that LEDs have gaps in their output spectrum, and am getting conflicting views on whether they change color temperature over their lifetime. I don't need a particular white balance, just full color vision and a consistent bias.

Perhaps I should use a more traditional flashlight bulb on a regulated circuit and just be careful to use the bulbs for a certain number of hours to avoid major color shifts?

Any advice or ideas much appreciated.

[JSWrightOC]

I would begin with a halogen lamp that's voltage regulated. Some units will perform better than others, mainly due to operating temperature--a critical factor in the Halogen Cycle Tungsten Lamp. A properly designed lamp will give you longer operating life and better lumen maintenence. I have replaced halogen lamps after their rated life and the envelope is still crystal clear.

Another option would be a miniature tri-phosphor fluorescent light source or HID lamp. The CCFL lamps are small, rugged, and bright, and offer good color rendering once they warm up. The HIDs are expensive, but offer even higher CRI--but not quite matching a halogen incandescent.

An incandescent body (regular tungsten or halogen) will give you your best CRI as compared to natural sunlight, so that might be your best choice. Hope this helps!

[JSWrightOC]

"White" LEDs have blue-emitting dice that have a yellow-emitting phospor gooped all over them. This is why the spectrum output is so choppy.

Also, welcome to the forum! [img]/ubbthreads/images/graemlins/smile.gif[/img]

[paulr]

Go to a camera store and buy one of those little battery powered light boxes that photographers use for examining slides. The better ones have color adjusted fluorescent light sources. www.bhphoto.com is a good place to buy this kind of stuff, if you want to buy online.

[Serafino]

Thank you JSW and PR for your comments! Sounds like I should look into Halogen and Xenon flashlights. Then I'll need to figure out how to make a voltage regulator for one of these (unless there are some that already have this?), and track down specs for their luminosity and color temp. vs. voltage over their lifetime.

Fantastic forum!

[Doug Owen]

There is (of course) a traditional solution to this used in old style pyrometers. These compare against an internal lamp, perhaps something similar to what you've got in mind?

Anyway, the classic solution is to actually read the light output of the bulb and adjust the drive (current) to return it to the proper (i.e. 'calibrated') output. In your case I suspect it's simple enough to include a photo resistor near the lamp. This could either be referenced in the calibration phase of usage or included in the feedback loop of the regulator driving the lamp. I suspect getting a handful of lamps and screening them at the start would yield matches within a percent or so making a closed loop system very practical.

Any phosphor based system (LED or CCFL for instance) is going to have flat spectrum problems since such are quantum deals.

Or so it seems to me.

Cheers.

Doug Owen

[Illuminated]

How's the CRI of the *regulated* SF A2?

If that's not suitable, you could build you own using an MR11 or MR16 halogen lamp assembly and one of Willie Hunt's LVR's for regulation.

MR16 are manufactured in various color temps/beam patterns and widely used for retail display lighting where color rendition is very important.

I'd sugest contacting Sylvania, Phillips, GE, etc. for application information using one of their lamps. They *might* be of some help to you.

Once a lamp is chosen, you could select a power source with a voltage higher than the lamp design voltage, and then have Mr. Hunt program the appropriate model LVR for exactly the output voltage you need. He'll likely even help you select the optimum power source if you email him your application info. He's been very good with me in that regard. Highly recommended by this LVR user.

Hope this helps - John

[paulr]

I think if you really want anything like daylight balance, you're going to have to use an incandescent source with filtration that throws away a lot of light, or else a xenon flash tube. Again, photographers deal with this kind of thing all the time, so see a camera store about the right filters.

Can I ask what the application is? That might make it easier to suggest things.

[Serafino]

Doug--I think the pyrometer is a different direction from what I'm after (more on that later in this post.) While reading up on it I found this interesting recipe for a build-your-own:

http://www.metalwebnews.com/howto/optpyro.html

A custom self-regulating circuit with a photo-resistor would be a very elegant solution. Hopefully something already available with or without that feature will be enough for my purposes.

John--I'm reading up on the LVRs and the A2 and will check out those bulbs, thank you! I don't need a whole lot of power as the subject sample will be within 2 inches of the light.

Paul--in essence I want to build an independently lit reference-color box. I've already gotten into several discussions about why it can't possibly be useful but I want to solve the internal lighting problem and see for myself. [img]/ubbthreads/images/graemlins/smile.gif[/img]

I don't need a particular color temperature, i.e. daylight, I do need a full spectrum so that a full range of colors can be represented, and I need consistency so that a given color and value looks the same inside the box in February of 2005 as it does in April of 2003.

Filtration as a means to wider range of colors and mechanical value-control may be something I look into for a Mark II.

[paulr]

In that case, I think your best bet if you need a full range of colors is still to use a photographic light box.
As mentioned, there are some nice ones available that are carefully calibrated, though expensive.

[Serafino]

The portable light-boxes for viewing transparencies are too large (the smallest screen size I've seen is 5 x 7), and if I understand flourescent light correctly, have many gaps in their output spectrum.

Are there voltage-controlled incadescent-based models?

John--looks like even the MR11 is rather more powerful than I need, but the LVR idea continues to look promising.

[paulr]

I've seen some 4x5 light boxes but don't know if that's small enough for you. Maybe you can remove the light source and repurpose it. The ones I've seen are fluorescent--maybe they used special full-spectrum phosphors like Vitalights or something. I'm not an expert and have never used a light box, just seen them in stores a few times.

[ QUOTE ]

The Just Transparency Viewer # 21873 is a super-flat, light and handy viewer that goes anywhere for successful presentations. Professional presentations of slides, proofs and copies and now possible anywhere - studio, ad agency or your client's facility. The Viewer includes a carrying handle that doubles as an adjustable stand. This model is equipped with a resting roll and an electronic dimmer. The special Just Normlicht fluorescent tubes with a unique color rendering index combine energy-saving high light strength with maximum color rendering.


[/ QUOTE ]

[paulr]

Here's links to a moderately priced 4x5" unit and a very expensive 7x10" unit. I have no idea whether either of them is any good, etc.

http://www.bhphotovideo.com/bh3.sph/...#goto_features


http://www.bhphotovideo.com/bh3.sph/...ID=F4E04A0EDD0

Edit: the quote in the previous post was supposed to go into this post. It's from the blurb from the 2nd viewer linked here (the expensive one).

[snakebite]

i would use a dc rv ballast to drive a ott or vitalite tube.
they are available with cri's in the mid 90's
this is done with a mix of up to 8 phosphors.

[Serafino]

Well I suppose I should have been specific at the start but the visible
area of color swatch in my design will only be around 1" square, so I've
been envisioning a light source much more like a diffused flashlight (or
the aforementioned Luxeon module). Also I'd like this to be comfortable
to hold in the hand for long periods, so I was thinking to power it with 2-3
AAs or the like.

Thank you for these suggestions and I will look into them. I appreciate
your taking the time to ponder this project.

Meanwhile if anyone has a recommendation for an incadescent flashlight
bulb with an especially good CRI and comparatively high color temp.
numbers please post it. [img]/ubbthreads/images/graemlins/smile.gif[/img]

[Illuminated]

Serafino,

Check with www.carleylamps.com or www.walamps.com

Contact sales or applications dept. and ask for help selecting a suitable lamp.

Carley lamps will sell lamps in stock with no minimum requirement.

John

Welch-Allyn has a $200 minimum order requirement, I think.

[Doug Owen]

Serafino,

Yup, that would build one, not exactly the sort I was talking about, but serviceable.

The idea, I think, is to start with a long life bulb so change is slow. Then drive it to the same output each time you use it.

If you decide to go ahead with it and need some help with the regulator, let me know.

Doug Owen

[Serafino]

Well I'm still trying to track down a suitable bulb. I'm
not finding much information by which to choose, so I'll
probably just end up buying a flashlight which has Halogen
bulbs as a starting point.

Doug, I'm likely to try one of Willie Hunt's LVRs as time
is getting short, but if you have a recipe for a circuit
that will work I'm all ears. I'm afraid I'm an electronics
tyro--I've soldered together kits in my time, but I have no
foundations in theory.