Safety of Diverged and Heavily Diffracted Audience Projections

Hi there, last night I went to a party, which had a cheap $30 red/green laser projector that displayed various patterns across and into our crowd last night. The projector was located in the corner of the room, and has a 50mW green and 100mW red that are heavily diffracted. There was no actual "scanning" going on, that is, rapid XY shifting of a single beam. The lasers themselves were shown through two grid-matrix diffraction gratings that split the two continuous beams into ~400 beams.

That laser projector can be seen at this link:
Amazon.com: Green Red laser Stage Lighting Light DJ Christmas Party: Musical Instruments

Now, obviously, 50mW and 100mW are very damaging to the human eye. However, these beams have been so split up, that the power per beam should be much, much lower. On top of that, passing through the two gratings caused a little divergence for each beam. The projector was at least 15 feet from any persons eyes, and persons rarely stare directly into the projector.

Every post I read titled, "Is this safe", the answers are always the same. You need very advanced calculations, only professional installers should be audience scanning, legaleze (I'm not getting a permit for my house), a lot of knowledge about a very specific setup is required, it's only 'safer', and horror stories of such scanning going wrong. This does not deal with very powerful (200mW< or multi-Watt) lasers, this is for a common house party. There is no way you can just tell me, "yes, this is safe" or "no, it is not safe", and I'm not looking for that, here is my question.




The setup I want to be using.
The laser projector located at:
Double Laser DMX Projector (Sound Activated, Cloud Background)
100mW Red and 50mW Green laser [CVLC-G198] - US$101.55 : Ankaka.com has two continuous lasers, a 100mW red and 50mW green, containing two diffraction gratings to disperse the patterns. The split up beams (this has more splitting-up than the first link) are continuously moving. I will myself place an infrared blocking plate over the projector, as well as a diverging lens (basically, a double concave lens) to spread out each projected beam so it not only looks better, but is more safe. I can't get hard J^cm2 numbers since I don't own expensive measuring equipment, but each dot wouldn't be a very tight beam.

So now, how "safe" or "stupidly-confident" does this setup sound?

QuackMasterDan said:

Every post I read titled, "Is this safe", the answers are always the same. You need very advanced calculations, only professional installers should be audience scanning, legaleze (I'm not getting a permit for my house), a lot of knowledge about a very specific setup is required, it's only 'safer', and horror stories of such scanning going wrong.


The setup I want to be using.
The laser projector located at:
Double Laser DMX Projector (Sound Activated, Cloud Background)
100mW Red and 50mW Green laser [CVLC-G198] - US$101.55 : Ankaka.com has two continuous lasers, a 100mW red and 50mW green, containing two diffraction gratings to disperse the patterns. The split up beams (this has more splitting-up than the first link) are continuously moving. I will myself place an infrared blocking plate over the projector, as well as a diverging lens (basically, a double concave lens) to spread out each projected beam so it not only looks better, but is more safe. I can't get hard J^cm2 numbers since I don't own expensive measuring equipment, but each dot wouldn't be a very tight beam.

So now, how "safe" or "stupidly-confident" does this setup sound?

Click to expand...

Your setup should be fine, just put it back away from people a bit. The only thing about your post that attracted my attention was the highlited bit above. I was wondering if you could provide links to these 'horror' stories as I only know of one crowd scanning event that went wrong and that was at an event in Russia in a tent, which was stupid to start with, but that's the only thing I remember reading about. Have fun and stay safe.

Edit: my bad, edited bad info

By horror stories, I'm not explicit speaking of events where the entire crowd is exposed, and you're right, where multi-Watt, pulsed lasers were used on the crowd. The two well-publicized events are at:

Powerful lasers damage eyesight of some Russian ravers -- Engadget
and
Tomorrowland Festival laser incident, July 2009 in Belgium - International Laser Display Association

I was more speaking of users here discussing their close-calls or accidents. Two example links:
http://laserpointerforums.com/f53/how-i-almost-lost-my-vision-47478.html
and
http://laserpointerforums.com/f53/1st-laser-safety-concern-52621.html#post735154

There are lots of stories here of users catching a reflection of a 20mW or higher, and discussing how they have a grey spot in one eye or were almost hurt.

Check these links out. This should be everything you ever wanted to know. These have both been updated recently.

Great Powerpoint that goes over the basics.
http://dl.dropbox.com/u/1493830/ScanningAudiencesAtLaserShows_2009-09a.ppt

Report on Audience Scanning
http://dl.dropbox.com/u/1493830/audience-scanning_overview_latest.pdf

QuackMasterDan said:

Tomorrowland Festival laser incident, July 2009 in Belgium - International Laser Display Association

Click to expand...

That incident wasn't caused by the lasers on stage but by some dumb ass with a pointer in the crowd, shining in everyones faces.

KGB_Productions said:

Check these links out. This should be everything you ever wanted to know. These have both been updated recently.

Great Powerpoint that goes over the basics.
http://dl.dropbox.com/u/1493830/ScanningAudiencesAtLaserShows_2009-09a.ppt

Report on Audience Scanning
http://dl.dropbox.com/u/1493830/audience-scanning_overview_latest.pdf





That incident wasn't caused by the lasers on stage but by some dumb ass with a pointer in the crowd, shining in everyones faces.

Click to expand...

Wow, thanks for the rapid feedback, those two links are great.
Yes, the first link is a legitimate horror story. The second one has listed in bold-yellow font that the show was performed safely, and user-held laser-pointers caused the damage. The point of the linked threads is to show how my searching through safety discussions often include some pretty intense warning. Media-hype and exaggeration of audience dangers don't make searching any easier. =\

Look, lets simplify this before everybody in the world gets the idea that grating based displays from China are intrinsically safe. They are not. A lot of people in this thread are constructing safety fallacies, without running any numbers.

The gratings are cheap, so what is the power in the zeroth order? My bet is, much greater then class IIIA. The zeroth order is the undiffracted part of the beam, and it goes straight thru. All grating based laser displays have zeroth order, and its usually pretty hot. So now you have at least one "hotter" then average beam in the mix,per emitter, so be default, under the rules, your calculation is based on that beam, not the average of the others.

The laser is cheap, we have all seen Chinese lasers ship at over 2X rated power, and since this is not a stabilized laser, it can vary all over hell's half acre with temperature and voltage changes. There is no scan fail in the unit, no sensor to terminate the beam if the motor fails, and no means of cutting the laser off if the laser power surges.

Next up, what is the definition of safe? Answer, there is not one in this case. If we go by the popular PL assumption that if it is below 5 mW Its ok, because that is class IIIA, You need to read up on what IIIA is. IIIA is a designation for a laser that is not intrinsically safe, it is a laser with a low probability of damage. Say 1 in 100,000 chance of observable damage.
That standard is based on animal tests in the 70s. You have 10 pairs of eyes in the room, does the umber become 100,000/10 = 1 in 10,000? No but the odds do change toward a greater chance of a incident?

What is the real odds, I cant do the math for that. Why? I don't have a statistical way of assuming how many times a person will look towards the laser and be in the path for a hit. But the odds do shift towards damage.
If I do the math, I have to thus assume the viewer never faces away from the laser, and that each scan hits them. Otherwise I need measurements. Why? If you dont have the data, the rules say assume worse case.

Next up, the scan speed of the laser is zero for the zeroth order, and not very high for the rest. So there is little or no gain in safety from speed.

Distance is your friend, however. Inverse square law somewhat applies to that diffracted, poor quality beam. So the further back you get, the less of the pupil is filled and the density is lower.

Next up, what about secondary reaction. A incident need not be a retinal burn. Just getting flash blinded, tripping over the dancer next to you, and spraining a ankle is a injury. So it is wise to get the laser farther away from the crowd.

Another variable is time. The longer a viewer is in the room, the more likely a event happens.

I have the formulas for slow sweeping beams. But I need solid measurements to give you the numbers.

So you have a fixture emitting laser light with no quality control and no reasonable means of insuring that people will be far enough back from the unit to be safe, and no means of scan failure or power level safeguards.

Thus you CANNOT assume this thing is intrinsically safe.

It has a perhaps low probability of a injury, but that is not a zero injury rate by a long shot.

Without a milliwatt meter making checks every night, you have a non zero risk.

Its a clone of a Chauvet "FAT BEAM" but without the intrinsic safety tests and quality control Chauvet did to get US approval.

What is stopping the maker from installing a 200,300, or 500 mW laser, NOTHING.

Therefore until you have real numbers, do not assume via a "Bandwagon" fallacy that you have a safe device.

Steve

I've spent many hours reading about laser safety, last night, I read the entirety of the International Laser Display Associations report on audience scanning here: http://dl.dropbox.com/u/1493830/audience-scanning_overview_latest.pdf and would recommend it to you, it was very thorough. I'm well aware of the classes of laser systems, and that diffraction gratings do have zeroth beams that never move (but are still split and weakened). Chinese based lasers are crap, aren't measured, leak IR like crazy which throws the numbers more, and in my guesstimating for this post I'm assuming that the numbers for those lasers are too high, maybe double.

It does not concern me if the display suddenly stops moving and only throws out a single beam which could be prevented with a system that rapidly kills the projector within 1ms to save the crowd. That is completely insane overkill, this laser display is used in my house, not a stadium, and simply shines downward over and onto the heads of my friends and I while we watch movies or play games. Usually, we're staring at the screen unless someone stands up to get snacks, and doesn't stare into the laser projector while doing so. Though he may get a glance or two, which if they are above 5mW, burns on the retina are still accumulating. I'm not just ignoring the dangers of glancing beams, I'd like to avoid those and those are the real danger.

The power is rated at 50mW at the incident point of that laser, let's just assume it's 100mW for the crappy quality of Chinese lasers. Thus, if I stuck my eye up to the glass from one inch away, I'm looking into 100mW. When I'm 20-40 feet away, like in my house, since the projector is attached to the roof, I'll very likely be catching a single beam, and of the ~400-500 beams being projected, I doubt they are above even 2mW (again, stupid guesstimation). I made it clear in my first post that I don't have testing equipment or professional advisers to perform advanced calculations and quality control for my living room, and don't think it's necessary. It's dangerous to assume, "I think that's safe enough from me looking at it", but let's do the stupid process of reasoning it out.

Let's be practical, it's 100mW, being split up 400 times, shown 20-40 feet away, and I'm sticking two double-concave (150mm focal and a 75mm focal - pretty strong) diverging lenses over the projector for each color to fan them out strongly. Even then, the zeroth beams that don't move have still been split into ~40-60 beams of the total ~400, on top of being diverged by the lenses. Being flash-blinded and mentioning the danger of tripping over a power cord or soda bottle in my house... that sounds like paranoia.

Can you please provide me the formulae for measuring slow-sweeping beams, though I doubt that applies well except to very fast 20k ppm lasers like this is not anywhere close to. I, being non-knowledgeable and inexperienced, will simply use measurements that are far too high to make myself feel more safe. I was hoping this thread would avoid the factors I mentioned in my first post, the extreme caution and fear. I'm perceiving my safety level on this is actually going too far already. Extreme, horrible, worst-case scenario discussions, where caution is necessary, for my little 100mW system (not a stadium level-system with 3W lasers where worst-case does matter) aren't necessary. Could this discussion be about something I will actually face with a low power home-based setting? The setup sounds very practical with low danger though there is and always must be a very low chance of everything going horribly wrong, let's be realistic.

I'm a former ILDA member, I'm well aware of ILDAs current position. What I have a problem with is the low cost hardware being distributed without quality control. That is what is happening, and it is only a matter of time before incidents start.

OK, I can give you a pdf of the math, but the only free source, I'm aware of, with the required exposure tables is down for a website remake. That book is US Army and the 300 page PDF on safety is down until they redo the web site.
Since this is now a ANSI standard, it can be hard to find the exposure tables without paying for them.

Here is the basic scheme of things:

OK, I tried to do the attachment but I cannot resize the PDF to match LPFs requirements. I will need to rescan the basic document some day.

For intrabeam viewing the standard was class I before audience scanning was restricted and re evaluated in the US. You need to measure the exposure, do the math and compare the derived numbers to the allowable exposure limit time for class I for a given wavelength in the AEL table.
This is the CW laser procedure, pulsed or qswitched is another ball game.

THE REQUIRED DISCLAIMER, this is a FRACTION of the whole procedure for determining if the audience scanning show is safe. Each individual effect in a show must be measured and calculated with the actual gear at hand, you cannot assume any two projection devices are the same, and you must calculate the cumulative exposure for the whole show. In other words, you cannot just stick this in a spreadsheet and assume your safe. You need calibrated gear to do this, and you cannot just make simple assumptions about your laser, YOU MUST MEASURE and YOU MUST PERFORM QUALITY CONTROL AT EACH USE OF THE SYSTEM. THIS IS BY NO MEANS A COMPLETE GUIDELINE TO AUDIENCE SAFETY, NOR IS IT INTENDED TO BE SUCH A DOCUMENT>

As this is NOT peer reviewed, you cannot assume I have not made mistakes.
Thus this is for personal amusement and education only.

End of what I stick into avoid the potential lawsuits or the possibility of seeing this reprinted in "Make" magazine as a "How to make a cool laser effect" expose.

The old rule sort of says you can expose until the effect becomes class II or hit the allowed cumulative exposure.


From a BRH document, which is what CDRH used to be called, roughly quoted, slightly modified, single beam example, you'd have to integrate over time for the multiple beams from a grating.
I've kept the math the same.

I. This is the case for the full open pupil which is 7 mm diameter, unaided viewing.

II. Set up a calibrated detector attached to a time measuring device (oscilloscope) behind a 7 mm mask at the viewing location.

III. Find the hottest and slowest moving beam in the system, measure its total power and full pulse width across the 7 mm target as well as its frequency. Do not use full half width maximum measurements, calculations are based on "leading edge edge to falling edge" measurements. The tables assume a Gaussian beam profile, so if you have black areas in the beam, the measurements become far more complex. Use this worse case data for EVERY beam in the system, and beware the zeroth order. (this line edited by Steve to get the "assumptions" into a short paragraph)

IV. Determine the linear velocity of the beam across the detector.

VI. Determine the approximate beam diameter Db at the location of the detector, this is defined as the 1/e^2 power points "a" plus the beam divergence "Theta" times the distance "d" between the laser source and the detector. Db = a + (theta times d) [Steve's note, many diodes and low cost DPSS do not have simple 1/E^2 points as they are not TEM00 mode]

VII. Determine the pulse width (t) by dividing the sum of the detector aperture diameter (Dd) and the beam diameter (Db) by the beam velocity (v) ie: T = (Dd + Db)/v


VIII. Determine the stationary power P collected by the 7 mm diameter detector at the target location, P = Pe (.7/Db)^2 where Pe is the power of the beam at the exit of the laser.

VIII. Determine the energy in the pulse, by multiplying the stationary power P by the pulse width. E/P = (p)(T)

IX. Determine the pulse rate or frequency (f) from the measurements of the scanning device.

X. Calculate the Energy rate or average power using the values obtained in steps V and and VI, E/Sec = Pave = (E/p)(T)(pulses per cycle)

XI. The above values are compared with the emission tables in the "Performance Standard for Laser Products" in order to determine if accessable radiation in the effect is class I for all emission durations. If class I, determine the time for the accumulated energy to exceed the class I limits. Thus the energy rate in step VII and divide it into the limit from the table to determine the allowable exposure time.....

Procedure ends.

It gets more complex to do the cumulative, the period of which is 10^4 seconds. Say you get 80 seconds for your effect to switch from class I to class II, thus you can only use it for a total of 80 seconds it is class I in the allowed exposure period of 10^4 seconds, ie you can use it for 4 periods of 20 seconds in the show or 1 long 80 second period. All effects add toward the cumulative limit.

The table is divided according to wavelength among other things.

------------

Fun, is it not? wait till you see the variations for lissajous patterns, flat lines, sinusoidal scans, ramp waveform scans, sawtooth scans, single pulses, mirror balls and cumulative. ILDA suggests a simplified technique, making a measurement and adjustments at the point in the audience that the beam fills the 7mm aperture, but that technique results in a minimum power show with weak effects. You still need a really good power meter to do the ILDA simplified technique.

Steve

Wow, I was not expecting that level of complexity for determining safety of a laser display. As far as equipment for measuring... I have none of that. If I actually did have the gear, the math might indeed be fun =)

I'm still on the fence about this laser-display going into my house. There are so many variables, I see there is no way to even guesstimate safety without at least some numbers (is the 100mW red really 100mW or 50mW or 500mW)? I'd like to have this laser display running once a week for Friday nights when everyone gets together for watching neo-noir films and playing video games. Aside from Prototype's initial comment, "Your setup should be fine", I haven't really gotten anywhere closer as to how safe my eyes and those of my guests will be with these two diffraction gratings and double diverging-lenses. For the time being, I'll make the potentially dangerous assumption that the power levels are safe enough due to each beam being split up hundreds of times on top of being diverged. That said, it would be a waste to have eye damage over relatively mild entertainment after a few months of using this thing.

A quick guesstimate says if you keep folks 10 feet back, use your diverging lenses in series and really back the current knobs off on the drivers, you will probably be more then OK. Ie take the green down to below lasing threshold and slowly bring it back up till its stable.


Get things so that the average power is about 100 microwatts across 7 mm, much more then that and your friends will be asking you to turn it off anyways, as the flashes will be very bright.

When in doubt reduce power and expand the beam. Increasing the scan rate, contrary to myth, does not help you much, it just increases the PRF.

If its not comfortable, its not safe.

You can also try proper chrome or metallic neutral density filters inside the laser, provided they are not at a crossover point in the beam.

These are Schott Glass, which does not bleach in the visible, but you might need a stack of them.

http://www.surplusshed.com/pages/item/m2535.html

http://www.surplusshed.com/pages/item/pm1024.html

or install with epoxy on a firm mounting some 50/50 or 70/30 beamsplitters in stack and route beam energy into the case of the unit. . You can get them in dielectric glass from One Stop laser Shop.

I'm not against having fun, but I am against setting a precident here for declaring a generic laser eye safe. Fair nough?

Steve

Oh, and expand the crap out of the beam, inverse square law is really your friend, and it makes beams in smoke/dust look so much better in short distance , low power shows.


BTW, If your serious, you want a water or glycol based cold haze, which is done with a little tiny air compressor and not heating elements.

If not that, try watching the sales after halloween, wall mart and the craft stores blow out tiny fog and haze machines for 19$ after halloween, witch is the only thing good about that pagan holiday now that I am too old to metabolize candy into brain/muscle and not fat. Just turning one on till it warms up will probably fog a video room to the proper level for a hour, even without hitting the button.

Steve