Need a safe suggestion on laser class

Hello, all.
I recently have been poking around on this thread: led orb.
I want to make one myself, but with red green and blue laser diodes.
I'm having trouble deciding on a good class, and need input.
1) I want everything to appear at the same brightness. I want to do this to have the maximum effect for all colors.
2)i plan to have a safe operating distance of 2 feet. This is important in calculating maximum eye exposure time.
3) the inner ring alone will be spinning just under 1000rpm.
4) the estimated eye exposure time is .000383 seconds
5)the estimated exposure rate is 13 times a minute.

I was thinking class 2m plain, or class 3b with an optics lense

tuskiomi said:

Hello, all.
I recently have been poking around on this thread: led orb.
I want to make one myself, but with red green and blue laser diodes.

Click to expand...

First of all, are you trying to make a laser "ball" with the laser diodes like in the video? Doing it with lasers will result in something that acts more like a disco ball but with a much higher rate of rotation since the lasers emit light only in a narrow beam unlike LEDs.

Secondly, I wouldn't recommend making one just because of the safety issue. Having lasers that point at people and fly around randomly doesn't sound very safe to me.

If you really want to make one you need to make sure that the individual lasers are under 1mW or class 2. It doesn't matter how quickly your ball is spinning, it needs to be under that power. What if the ball's motor fails and the lasers become stationary?

You also need to check the individual laser outputs with a reliable laser power meter. Just because on eBay it says it is 1mW doesn't mean it is.

If the green laser is 532nm or DPSS and not a diode laser you also need to worry about IR leakage.

Have you thought about 3-5 watt LEDs instead of laser diodes? Unless your specific goal is to have the orb painting lines of color on the walls all around it then the high power LEDs might be a worthy choice. If you do build one with lasers I'd recommend incorporating circuitry to shut down the lasers if the orb slows down below a safety threshold speed to ensure no one's eyes get...slow scanned.

WizardG said:

Have you thought about 3-5 watt LEDs instead of laser diodes? Unless your specific goal is to have the orb painting lines of color on the walls all around it then the high power LEDs might be a worthy choice. If you do build one with lasers I'd recommend incorporating circuitry to shut down the lasers if the orb slows down below a safety threshold speed to ensure no one's eyes get...slow scanned.

Click to expand...

I have thought about the LEDs. It seems like they're take all the fun out of it, turning into a rave machine.
the lines on the walls are specifically what I want to happen. like I said, I don't think i'll be using anything more than class 2, without a consensus.
that pretty much limits my power to 1mw, and at worst case scenario, i'm guessing you get flashed 30 times a minute, for .001 seconds, is a total of .03 seconds of exposure a minute, and at a safe rating of .25 seconds of exposure, that's an estimated duty cycle of 8 minutes. certainly i'll have to do something about the RPM throttling back.
keep in mind that the .0003 seconds of exposure, is if you're staring at it from 2 feet away. I'm guessing people will be further than that if I hang it from the balcony in my house (22 feet from ground, 4 feet from overlook) it will be much safer.

just read up on the 532 IR. seems more serious than I once thought.can I measure IR with my Raspberry pi 2? I heard that the IR in cameras causes them to reset via a photo-electric field.

tuskiomi said:

just read up on the 532 IR. seems more serious than I once thought.can I measure IR with my Raspberry pi 2? I heard that the IR in cameras causes them to reset via a photo-electric field.

Click to expand...

Staying under 1mW should be fine. If you want to use a 532nm laser just get an IR filter. The IR element usually shows up as violet on a camera unless it has an IR filter of its own.

Hmmm..... I'm thinking that 22' off the ground you're going to need rather more than 1mW. More like 50mW of red, 15-20mW of green and 50mW of blue. My wife gave me one of those little pocket laser spirograph toys for xmas a few years back. It's 5mw red and actually pretty cool...from 10'...in a perfectly dark room. Worthless though at greater distances or with pretty much any other light present. One of these days I'm gonna stick a single mode green diode in the thing and that should be a definite improvement. You might want to take this over to the laser shows section. The folks there should be able to give you some more definite suggestions as to what sort of power you'll need but I'm thinking that you're going to want class III diodes. + rep for thinking this out carefully with safety first in mind.

The folks over in the light show area seem to be inlo laser projectors? Still cool, and i'd like to have their input, but i'm a skepticist in this case. Not to say, that they won't have valuable input. More in a sense that this is foreign to all of us.

any ideas on where i can buy diodes? I have been poking around, but no luck in finding raw diodes. I don't care much on what the diode is rated for; so long as it is above spec.

I've designed 3d parts for this and will print them off. This will be a nice project to post about

I want to run a diode rated for 1 watt at only 20mW. Is it dangerous to underload 445nm diodes? I'm guessing i'll only be turning out .3V or .4V to get that kind of current.

tuskiomi said:

I want to run a diode rated for 1 watt at only 20mW. Is it dangerous to underload 445nm diodes? I'm guessing i'll only be turning out .3V or .4V to get that kind of current.

Click to expand...

^There are blue diodes which are rated to output at much lower then 1W fortunately. Not sure this might be of help, but might be worth checking out

Link: OSRAM 445nm 450nm 50mW 3 8mm to 38 Blue Laser Diode LD Single Mode PL450B | eBay

-Alex

"any ideas on where i can buy diodes? I have been poking around, but no luck in finding raw diodes."

DTR's laser shop. He's got what you want.

And no need to use an expensive 1 watt blue diode. Or a fragile DPSS green module. Red, green and blue are all available as single mode diodes with really nice beam specs.

"I want to run a diode rated for 1 watt at only 20mW. Is it dangerous to underload 445nm diodes? I'm guessing i'll only be turning out .3V or .4V to get that kind of current."

This is making me think you ought to do a little reading. Adjusting the power output of laser diodes is a matter of adjusting the current through them. The voltage will change but not by very much. Take a look at one of the many diode tests on the forum.

If you want lower power i would seriously recomment diodes like the Osram part from the link. They have a much better beam profile compared to the multimode 445 nm diodes from projectors and such.

From a safety aspect you should really have some mechanism that shuts off the lasers if the scan speed lowers for whatever reason. This would apply to anything you use in public that's over 5 mW.

There are tools to calculate the MPE for a give scan pattern that can hit spectators, but those only apply if the device is working perfecty well. Looking at that LED orb i can see serveral failures that could cause it to slow down or even come to a full stop, pointing the laser in any random direction that could potentially hit someone in the eye. This could be due to some part flying off causing imbalance, a motor failure, or simple some (foreign) object getting lodged in the mechanism.

For 'normal' laser shows that scan the audience failsafes are usually in place. These systems are usually pretty well contained, but it's still detected if galvo's are moving properly and have not stopped for any reason. This could be the galvo burning out, but also some mechanical obstruction that prevents it from moving despite being electrically intact. Galvo's have buitin feedback that can be monitored, but this orb construction has none of that so would be fairly difficult to diagnose in realtime.