how to make high illumination light source

Just had a quick look at Luxeon AB12 note, basically if the max continuous current is say 350ma, you can only have a maximum pulsed current of about 500ma, so this will not quite double the light output.

It really comes down to integration time, what is it? and what leds are you using?

Can you get a pre integration time trigger signal from your camera?

Martin

On a sunny day (Thu, 15 Nov 2007 10:38:00 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

Yes, you mean high current pulse I think? Look up the LED specs for Imax, and possibly for light output versus I. Whatever yo udo, the dissipation in the LED is also limited, so if you have 2 mS on @ 2xI, and 2mS off, you could perhaps gain

I have no answers but this post did get me curious enough to read a little about xenon flash tubes on

D from BC

So...you have a clue towards a method that will work nicely.

Brendan Gillatt wrote:

A strobe light, which is the same basic technology, can easily be found to shoot that fast.

(Just found in a web search, not associated in any way with either:)

DT-311A - flash rate up to 35,000 flash per minute (583 per second)

Lists a 1000 FPS "special application" system.

"Huge charge currents" are relative - a tiny camera flash operating from a battery supply might think "huge" at a current that is easily supported by a unit plugged into a wall outlet. Several units, flashing in sequence, can be used to obtain even higher effective rates, if charging time for a single unit is too long.

--
Cats, coffee, chocolate...vices to live by

I found the max current spec of high power Luxeon LEDs not to be much higher than the average spec. Don't know why. You might say I am working on the same problem using LEDs. The big problem is getting an even concentration of light using multiple LEDs. I have allready used 45 watts continuous LED illumination using 9 LEDs. There is one LED strobe LED device from Scimedia. 3 or 5 watt.

greg

Thought of using an HID globe such as are used in video projectors? The one I've been playing with is 100W and produces a ridiculous amount of light. A bit pricey, but then so are white LEDs.

On a sunny day (Thu, 15 Nov 2007 21:30:42 GMT) it happened snipped-for-privacy@comcast.net (G) wrote in :

illumination

Yes, I did some white LED pulse experiment some weeks ago, I wqas mainly interested how much brighter it *looked*. It did not look brighter then normal on, with ratio 1:4 and 4x current in the peak. Some publications say that the eye should hold on to the max, but I think this may not be true. Maybe you will have to go with a triggered stroboscope. There is no real limit to what those can do.... Else some spots?

I'm just playing with one of these for a strobing idea ...

They say it's as bright as you can get and can be used for flash units. Certainly it is too bright to look at in continuous operation and still looks good in the mS, (high current) pulse area.

15x15cm is a lot of surface though.

If you can set the integration time of the sensor, why do you need an intermittent light source?

--
John

2 big ideas can be used to increase illumination for cmos image capture......Focus & Impulse

Large area to "matched area focus is obvious concept, as you want to collect and apply all (many) photons at the filed of view... Genrally you can apply a high peak energy, stress, electric field, illumination for a relatively short time nanos to mili seconds, while thermal limits occur in the 1 second time frame, so impulsed light can create a relatively;y high short term value, enabling image capture via integration on CMOS array.

Best regards

Marc Popek

PS.. Have significant experience in realizing these and related circuits and system. Perhaps your orga\\nization can benefit from this expertise?

Martin Griffith mart_in_medina@ya___.es posted to sci.electronics.design:

At 500 fps, 1 ms for exposure and 1 ms for sensor readout sounds reasonable.

OP also seems to be looking at very long burst or continuous operation, so sync may be a "different" solution.

A bit more info about the camera/sensor is needed, isn't it?

Martin

yes, im wondering how this can be achieved. (for example a circuit or technique commonly used)

yes i can, what i need is to make sure is the area is sufficienly illuminated during that period. continuous illumination is also fine, but im not sure such a illumination source is available that will operate on car battery.

the sensor is LM3730, now KAC 3730. it is 580 fps, 101x128 gray scale CMOS sensor from kodak. the lens is taken from an old web camera.

On a sunny day (Mon, 19 Nov 2007 03:30:39 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

Ah! You can get 2x more light by getting a better lens (1 f stop). LOL

A thing about most monochrome CMOS and CCD sensors: They respond well to infrared out to about 1,000 nm, in fact better to most infrared wavelengths (up to 1,000 nm) than to visible wavelengths. Halogen lamps produce a lot of such IR in addition to visible light, even more such IR than visible. 12 volt halogen lamps are cheap, easy to get, and easy to get working.

- Don Klipstein ( snipped-for-privacy@misty.com)