New Laser MPE Limits published

[The_Doctor]

In the last post here I spelled out a problem, so I might as well have a shot at a solution too...

Expensive metering turns a scientist into a priest if it's so costly that only what amounts to a priesthood can afford it, but there is a neat way to avoid that risk. Make some tiny widgets with a fast photodiode, a differentiator built with the same low-loss parts you might use for a UHF tuner. Calibrate them with that expensive gear so that they drive a comparator that beeps through a piezo disk and flashes an LED if a light pulse hits it capable of exceeding MPE. Do it in serious bulk. That way the cost will be dirt cheap per unit, and several people will likely have one in any venue with a laser in it. It should be possible to make a simple unit cover a wide variation of power and duration for a given energy reading. The power of a firm like Casio might be well advised to make them anyway, just to let people feel safer about havign these new high power light sources in their homes.

[JStewart]

Clive,

HSE’s HS(G)95 guidance for laser show applications have underpinned what venues, show providers, and (knowledgeable) local authorities expect to see as acceptable laser safety management, since its publication in the mid nineties. In 2010 government legislation mandated the laser exposure limits, through implementing its obligations as an EC member state in relation to the Artificial Optical Radiation Directive. The law that was passed through parliament now sets the legally binding limits through regulation. HS(G)95 in that regard takes a back step, as the regulations that enact law take priority over any guidance.

The reason the HS(G)95 is up for review is that it’s a little dated with the law having moved on since its introduction. In days gone by, HSE would have updated the document themselves. But with a political drive to ‘appear to be seen’ as government reducing the grip it has on business, with H&S, (and the need to cut costs!), when this document came up for review, it was seen as good candidate to be reviewed, maintained, and published by industry. Which is what is starting to happen with PLASA’s panel having been assembled.

The important thing to remember is that HS(G)95 is not a ‘crowd scanning debate’ document. It is supposed to give a broad overview to laser safety management for laser show applications; ‘crowd scanning’ is only one small part the range of issues that need to be considered on a display installation.

I haven’t come across any issues with UV or IR being a greater problem than the primary concern of visible radiation exposure from laser projectors?

I don’t really understand why Casio would want anything to do with discussing ‘crowd scanning’ issues or want to get involved with a laser show safety management guidance document, when for their normal intended use, Casio products do not expose persons to light that would be considered a laser product.

I appreciate that there is an ongoing disagreement between some show providers and what the scientists have agreed are the appropriate exposure limits. But that aside, I’m not sure that even if the existing limits were relaxed, even by a 100 or 500x, if it would change the behaviour of many laser show users that make absolutely no attempt to quantify what they are putting into the faces of others.

James

[mixedgas]

Jeeze,

Where do I begin, Fast photodiodes and UHF tuners? Do you even own scanners and software?
That is non-sensible and shows no grasp of the physics problem.

The reason is not the "Big Boys", the reason for the UKs Functioning laser safety system is well, safety... HSE may be overkill, but at least they get the local and regional folks to inspect. Unlike here in the states.

I know some of the people who set up a great part of the UK system rules, and they were NOT profit motivated fools. In fact they were
a mix of academics and industry types. The one I know the most, wrote his Doctoral thesis on the problem. He went out to clubs, made measurements, and made his peer reviewed results available to the public. He showed up at ILDA conferences and asked for his work to be challenged, some of it was challenged. This is not rocket science, its a well documented issue. The papers are on-line to be read.

Here is a the deal, the difference between a good exposure and a bad exposure is the initial laser power, divergence, pulse width, and distance to the audience. Exposures are cumulative. So a bad wavelength calibration on a PD is not a real issue. The real issue is a accurate measurement of the laser's CW power, a good ballpark measurement of the divergence and pulse width. Then tracking the cumulative number of exposures. The PD is solely used for pulse width, not power.

A 100$ PD is almost overkill, in fact. Its calibration vs wavelength does not really matter, in fact, as long as its not saturated during the measurement, and properly terminated at the scope, it really does not matter much unless its blind at the wavelength in use.


But guess what, a standard LPM with a 5% error is not that bad. A decent photodiode is less 100$. A decent used oscilloscope is 100$ on Ebay, or 300$-450$ for a New Chinese scope fast and accurate enough to solve the problem. If you can't afford that, you probably should not be exposing eyeballs in a public environment. An accurate enough divergence measurement can be done with a 20$ Ebay digital micrometer, a piece of paper, and a tape measure.

The LPM needs some measure of traceability and needs to be commercial product. But even a used one is usually close enough on calibration. Of the five I've bought, most were within 5% even after a decade of use. Its easy enough to bounce test a used one off a new one or recently calibrated one. There are decent laser power meters all over the globe.

Lasers drift in power, so the other thing is doing the measurements pre-show, and not assuming it performs the same, event after event.



The Industry, has a moral duty to not go Cheap on safety. You, "Doctor", have a moral duty to look at the math and physics before you go spouting off about the problem's solution and the state of an industry you do not even participate in.

Steve Roberts

[The_Doctor]

So find a way to get that calibrated measure cheaply into the hands of many people. Whether I know the physics or not is irrelevant. You're going to have to get some gadget into the hands of thousands of people who don't. Instead of berating me, fix that problem, if you know how.

Is there really no type of photodiode that puts out a signal proportional to power received at those speeds?

(My post was late last thing yesterday. Probably shouldn't have been... The 'UHF' comment was based on one thing alone, an assumption of handling pulses about a nanosecond in duration. Basic electronics suggests that standard low-loss parts for UHF, i.e. 800 MHz or more, would be needed to handle the small signals. My point being that there are lots of those from the TV industry so they will be cheap.)

[White-Light]

James
The current implemented HSG guidelines for laser displays within the UK are HSG95, and as such exposure levels or safe practice hasn't been proven or disassociated.

The levels are based on more than 30 years of research and statistical data regarding eye injuries across the world.

That said, the levels are now known to be conservative and there's a lot of support for the approach to audeince scanning suggested in Greg Markov's paper which involves a proposed level of 10x the current MPE.

Would it not be prudent to ask Casio or one of the other commercial suppliers to get involved on the revision of the crowd scanning debate.
Casio produce a laser based video projector that people harvest components from, Mitsubishi, Coherent are also providers of such components.
Let's get some big boys on board and the HSE.

Personally I'd rather see this responsibility handed over to ILDA.

With most of the World's laser experts at its disposal - people such as Markov, Benner, Murphy, O'Hagan surely there's no organisation better qualified to set the level than the ILDA board.

[mixedgas]

OK, how I am I going to make this point to you? The MPE meter is but a somewhat useful toy. If you can't do the math, which is simple. le, you wont get the gig past a safety case in the US or UK


Here is your homework:

Two amusement parks are on opposite sides of a lake. The lake is 1.234 Kilometers wide. A 25 watt argon laser based show is ran by a bunch of 19 to 24 year old boys. The laser has a divergence of 1.65 milli-radian and a initial beam diameter of .80 mm.

Every night a 19 year old Buxom Blond who is majoring in Dance Theatre takes the stage at the other amusement park. She's really hot... The boys at the laser show use binoculars to watch the stage show across the lake. Amusement park B cannot afford lasers, so when they get to the 60's music part of the stage show, a bunch of fans blow the girls skirt up, like Marilyn Monroe in the famous picture. The boys at park A decide the cheap PAR can lighting is unworthy of lovely Marilyn, and illuminate her back side with the argon to boost her already heavenly glow.


Show your math and equations:

Part one

"Marilyn" turns around during the illumination. She's not wearing glasses, so assume 7 mm pupils, fully open. Is she Eyesafe?
What is the class of exposure at her face?

Part two:

Marilyn performs on a tower on a barge. US NFPA 115 code specifies 0.5 watts cm^2 visible light as the point at which a show laser constitutes a possible fire hazard. If Marilyn is wearing black tap pants under the white dress, how close to the laser booth does the barge have to drift before Marilyn wonders what is burning?

Part three:

Marilyn steps off the fake 20 foot oil derrick she performs on, so the boys aim the beam 25 degrees up in the air. What is the Nominal Ocular Hazard distance from the beam, assuming the laser is emitting 100% 514 nm light? What is the Critical Zone Exposure Distance assuming the laser is 100% 488 nm?

Part 4

After holding Marilyn when she nearly fell off the tower, her boyfriend the stage lighting guy at park B decides he needs a laser in the indoor show. He purchases a 70 mW ILT krypton with 4.5 milli-radian divergence and a initial beam diameter of 2 mm from Ebay. The laser is performing perfectly and the specs given are exact. He can't afford a galvo pair, so he aims the beam at a tilted mirror on a motor shaft. The cone described is 15 degrees full angle and the motor is spinning at 250 RPM. To a first approximation, what is the static beam diameter at 75 feet where the audience is? If the eye is 7 mm across, what is the approximate pulse duration at the eye assuming the beam is a true square wave with straight edges and not Gaussian. The motor slows down to 10 RPM, what is the pulse duration now? Bonus points, is the audience eyesafe?

Show your work and equations used. Do not use the on-line calculators, you don't need them.

Answers Friday if I can find time to do the math on the road. Nobody except the "doctor" post their answers till Friday.

I assure you everything you need to solve this can be done with the static beam equation except for part 4. Its all on line.

Steve

[The_Doctor]

Do your own homework. Here's mine:

A flat wall at 2500mm from the projector mirrors is used for this estimate. A line is scanned at 500Hz, 1100mm wide, about the best you'll get for a clean fast scan out of WideMoves for the angle capable of that scanned line. As a slow scan is more dangerous this is a good base to start with. It's the large number of people with cheap slow beamshow scanners that need helping here, so work with the conditions expected.

Inverse sine for half scan angle (To get right angled triangle to work with) of 550/2500 is 12.709 degrees, so total scan angle is just over 25 degrees.

Circumference being PI*2*Radius gets you 1090mm of arc length for Radius=2500mm. (Close to my flat-wall measure but now properly extensible for any radius once scan angle is known).

The beam is about 3mm wide at any distance from aperture to about 5 metres. (Already diverging from beam waist at lesser distance so it gets safer fast beyond that 5metres). At 500Hz, a full pass of the beam across and back takes 2ms. 3mm/2180mm is 0.001376, so the duration for a microscopic point at the centre of the scanned line is about 2.752µs. Faster scanners and greater distances will make that a lot shorter, but at this point it is mostly within range of a photodiode like BPW21, as used in a fast luxmeter circuit here: http://meettechniek.info/concepts/lux-meter.html

While a proportional photodiode will not go much faster if that one is anything to go by, could it at least give a proportional output with different power and duration at much higher speeds? If so, and it is repeatable for a wide enough variety of durations and powers, you can calibrate it for power. To get energy you need time, so what's wrong with getting that from a fast PIN diode sitting right beside it? Combine the two outputs and use UHF capable parts to prepare a signal a PIC chip can use, then calculate some worst-case figure for energy from that. Never mind spectral sensitivity, all you need to do is alert a user to a worst case state to avoid.

====

Some words about the differentiator you dismissed so carelessly. Tell me if you think this idea is stupid after you consider this carefully. A slow photodiode will not be able to respond fast enough to give you a stable power reading, so you would likely assert that it is useless for power at these short durations. But it will at least try to. A differentiator will tell you how fast its rising output is going. and if correctly mixed with a copy of the original signal, you get power. As this is a standard method to accelerate the response of a (very) slow thermopile power meter to give a good early reading, the same method can be applied to accelerate the response for a (much less) slow photodiode if you use low loss high speed electronics to do it.

[mixedgas]

You just gave me zilch in your problem to calculate if your eye safe. Nor enough to even start the problem.

You are ignoring a lot of factors relevant to the real issue of using a microprocessor for this. When we get done seeing your answers and math to the previous questions I'll talk to you. One of my problems is taken almost directly from a FDA publication on eye safety. So I can document my work.
The other is a bastardized version of a situation I know was once real.

I also have a sneaky punch I can hit you with if you don't work the problems, so if I were you, I'd get digging.
Consider yourself called out on this one.

Steve

[mixedgas]

. Quote:

"Would it not be prudent to ask Casio or one of the other commercial suppliers to get involved on the revision of the crowd scanning debate.
Casio produce a laser based video projector that people harvest components from, Mitsubishi, Coherent are also providers of such components.
Let's get some big boys on board and the HSE."
End Quote

The big laser based Cinema companies hired a few lawyers and engineers and formed their own group. They basically stated in their opening letter to worldwide regulators that they were professionals, the rules were designed for laser shows which are different and demonstrably hazardous. They further implied that laser shows are installed/operated by unlicensed, untrained professionals, unlike their systems. Based on this, they asked for their own set of rules, implied we are the bad guys, and divorced themselves from us by policy.

Sorry, nice idea but they saw the laser show industry as a threat, small potatoes financially, and unworthy/unable to help their cause. They had a point.

http://lipainfo.org/wp-content/uploa...-Pete-Lude.pdf


Steve

[The_Doctor]

Not doing it unless I feel like it. I'm not doing shows. I looked at the single notion of duration and power of a moderate scanner in a moderately close range. Either what I said is useful, or it's crap, but it does at least try to come up with something. You put your demands on people doing shows, not me. As to eye-safe, at some point you're looking at a measure of power and duration. You take only the worst case into account, and specify what that will be. As to accumulation, you make that clear to a user of the device. They can usually tell if a beam is going to repeatedly hit the same spot or not.

Instead of pointing your demands at me, ask yourself this: Why, in thirty-plus years at the top of the industry, did you not see this coming and have a better answer in advance? The electrical industry has put cheap household safety testers in the hardware shops for years. This problem is tougher, but I'm suggesting that it is solvable, and I'm trying to come up with ideas about that. All you're doing is suggesting the effort is useless, that the result will be 'a toy'. What you should be doing is coming up with one that isn't.