Hi Phil, I was originally posting late at night, and didn't want to bother looking up the standards, so just posted a general warning. I have seen folks that just assumed that, since they weren't lasers, they must be safe. For those IR illuminators, we had to be sure that the maintance procedures took eye safety into consideration. I would get a headache just working with them in the lab!
Charlie
...........[snip]............
Maybe I'm being dense, but I'm still not convinced as far as it applies to my project. But I'm always ready to listen, so consider this as a request for further clarification rather than as a counter argument:
I explained in reply to another poster that the IR emitters will be placed 12' above ground, radiating forward parallel to the ground, and that the LED type I intend to use has a narrow beam angle, making it highly unlikely that a significant amount of radiation will fall on the eyes of anyone close by on the ground. Farther away - tens of metres - the intensity will be too weak to be hazardous. I also said that it will be a simple matter to place the LEDs in a tube, further increasing the minimum distance from which the rays can fall on anyone's eyes. Is there a fallacy somewhere in my reasoning?
..............[snip]..................
This is essentially what I said earlier and what I've just posted again in reply to Charlie E. before I read your post.
No it doesn't. The irradiance (power per unit area) on the retina is proportional to the radiance (power per unit area per unit solid angle) of the source. Radiance is independent of the number of similar radiating sources.
Yep, we had the same situation...
You have a maintainer that has to go up and adjust the aim. He stands on top of a ladder, and looks into the 'business' end at four inches. He then falls off ladder from pain...
You always have to consider that someone must install and maintain the system, and their safety counts too! ;-)
You don't even know the trouble I got into removing the safety cages in front of some laser rangefinders...
Charlie
I roughly worked out the radiance from a good IR LED (0.5A into a 0.6mm^2 die) is about the same as a black body at 900C.
Bit higher maybe because I am not sure what proportion of the die's area emits.
Do blacksmiths get horseshoe images burnt into their retina?
No, but welders need very dense goggles. Don't forget that the peak wavelength shifts towards the blue as the temperature goes up.
Eyes are essentially opaque to everything longer than (iirc) about 1440 nm.
Don't forget the spatial coherence issue--laser beams can be focused down to diffraction limited spots, but LEDs can't, at least not unless they're far enough away to be perfectly safe.
None of the above should be taken to suggest that you _can't_ hurt your eyes with LEDs, but for any given case, that's a matter of calculation rather than speculation. I'm going to take my own advice here, and not make stuff up.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal ElectroOptical Innovations 55 Orchard Rd Briarcliff Manor NY 10510 845-480-2058 email: hobbs (atsign) electrooptical (period) net http://electrooptical.net
I'd have to look at a datasheet for the receiver to see if they use what I am thinking of. However I have read the LIRC documents, and they show simple on/off pulses for the LED.
I don't know how I can make my explanation any simpler. What I am proposing is send a modulated light pulse and have a bandpass filter at the receiver. The modulation is the "unnatural" signal, that is the signal of which you have apriori knowledge. It is unnatural as in it doesn't match the signals that are ambient. The matched filter will accept only this signal. This is fundamental communications theory.
Look at this analogy. You have a twisted pair that you want to send a signal down. Say the environment has lots of 60Hz noise. Rather than just send DC down the twisted pair, why not send FSK with tones well above the line frequency noise. You could have mark and space filters, which is the most primitive FSK demod. The mark and space filters reject the ambient noise. I proposed an even simpler scheme where the tone burst is converted to OOK.
I don't know how I can explain my position any more simply either. Designing such a basic system is within my capability. But I don't really need to as all those features are already incorporated in the off-the-shelf IR receivers I used in the past. I send pulses at a frequency matching the receiver - 38kHz,
56kHz, etc. The receiver has a built-in pass band. Noise rejection is further enhanced by using burst and gap lengths (mark to space) as required by the receiver. The receiver rejects signals not conforming to those specs. The receiver decodes the bursts and outputs one pulse for one burst. For simple block/unblock applications, detecting the presence or absence of the output pulses is sufficient. For others, modulating the bursts by FSK, PWM or PSM.
What can I say. More power to the led and hope for the best.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.