Laser Power Meters

[White-Light]

Depends on what you call affordable.

At about £800 for sensor and USB meter, I'd recommend the Thor Labs PM100USB with a suitable sensor (sensitivity is more important than power handling when you're measuring a few milliwatts as a sensitivity in the milliwatt range can throw you over MPE on error tolerance) - I'd suggest a 302C as it has 1 micro watt of resolution. You also need a tripod attachment for the sensor.

Several big names I know use them although one uses the 302c thermal sensor and another uses a photo sensor I believe.

They're the dogs in my opinion. The software logs power trends and can be set into MPE mode after which you enter the sensor diameter, and it then automatically calculates and displays the output in MPE for you (mw per cm2), so there's no need to measure power and make separate calculations provided you use the same sensor and same mask (and ensure you overfill the sensor).

If you're worried about the max handling of 2W, don't be. If you're using it for MPE, simply start with the lowest possible BAM value (brightness) ie 10% or whatever the threshold is for lasing, and bring the laser up in power to MPE instead of starting at 100% and bringing it down. Do it that way and you'll never overload the sensor unless you're using a ridiculously powered laser. Only thing you can't do is measure total power output if your lasers over 2W, but then again that's irrelevant with MPE as all you need to know is the power at the audience when BAM'ed not the total full power of your laser.

BTW, there's a lot of software to download initially to update the software and get the meter to work.

[JStewart]

The Coherent Lasercheck, (or one of its rebranded counterparts such as that available from Edmund Optics), is probably about the lowest cost meter suitable for this type of work, that comes with a calibration. It’s a photodiode-based meter, so reacts quickly, giving an instantaneous output, and has the low sensitivity required for audience scanning assessments.

Although its quick, it, (as with most power meters), are not fast enough to measure an actual scanning effect). You can only use it to obtain a reading from a stationary beam. But if you can override your scan-fail system temporarily, this is not a huge problem. Be sure though that the beam being emitted from the laser system really is a CW beam, as software has a habit of inserting blanking points into the beam between refreshes, that affects the reading, making it seem safer than it is.

As others have pointed out, photodiode detectors don’t allow you to measure total system power from high power laser projectors. This is for two reasons. Firstly they are calibrated to measure specific wavelengths, because of the non-linearity of the silicon detector. So trying to measure a whitelight beam produced by 3 or more colours will not give a true reading. Secondly, the maximum they can measure without saturating is limited to the filter that they have on them, which usually works only to a few hundred mW or a W. Therefore if you want to measure ‘laser power’, you are better off getting a thermal type detector that can measure multiple wavelengths at once, and at higher powers.

Al (whitelight) mentioned the Thorlabs devices, which if you have a more money to spend may be a better bet, as you can buy different types of detectors for the USB device that would allow you to measure ‘total laser power’ and ‘low power’, if you bought both thermal and photodiode detectors. These devices can also be recalibrated by the manufacturer each year, (something that the Lasercheck can’t have done). Other manufactures sell separate detectors and meters for similar costs.

Just to clarify a point that Al made about the Thorlabs having an MPE mode, this doesn’t mean it provides an ‘MPE reading’. Instead the meter can report the measurement as a ‘power density’ (or irradiance to give it the proper name), that is ‘power per area’. So basically Thorlabs meter does the calculation internally, converting the power detected into an irradiance value (mW/cm^2). This is convenient (although not essential, as MPE is often expressed as an irradiance too. But if you know what the maximum power through an aperture is for a given MPE, you can take readings directly from the power meter in power meter mode, to do the comparison. – In fact 2014 update to the laser safety standard additionally gave the visible MPE in a form that is the power through a 7mm aperture.

(Al is right that the Thorlabs software does install a lot of underlying drivers the first time you use it, which are reliant on the .net framework also.)

The values you should be looking for are 0.4mW/7mm dia for intentional scanning/exposure, and 1mW/7mm dia for accidental exposure, if the people (e.g. performers), have been briefed to look away in the case of exposure. In recent years there have been changes to how the visible MPEs are treated. It can get quite complex very quickly unfortunately, and that would be a whole new discussion!

Hope this helps!
(BTW updates to the iOS software mentioned by Norty, (thanks!), coming very soon, as are Android versions…)

James

[White-Light]

Thanks James.

I thought it was self explanatory but perhaps you're right that I should have made it absolutely crystal clear, that when I said that it displays "MPE (mw per cm²)" it displays a mw cm² figure eg. 2.5mw cm², 1.33mw cm² etc not a figure expressed as a fractional actual MPE value eg 1.5 MPE.

For anyone familiar with the irradiance limits in mw cm² for both fast and non fast scan fail scenarios, it should make setting a laser within MPE limits very easy without the need for additional calculation provided you enter the sensor parameters correctly and overfill the sensor with the beam.

The only thing I can't recommend about this meter is the data logging which just takes the form of a notepad file with numbers in it. Somehow, I doubt an authority would accept that has evidence given that anyone could simply open notepad and type numbers. However, I think all is not lost as I would have thought a screen shot would suffice as it then shows everything - the software window, the measured value, the data trends, graph etc etc. leaving little doubt as to actual reading.

[jors]

Many thanks WL/James for your nice explanations.

I was also thinking on a meter that could be very very affordable for everybody. MPE metering stuff is not cheap, and it would be nice to find an affordable way (say... <250usd) to measure MPE.
Sure not that perfect that going with Coherent or Thorlabs PM100USB+302c, but better than nothing! ie a lot of people could use it...and this is just an idea... what about?:

LaserBee A
https://www.laserbeelpm.com/laserbee-a.html

Just setup BAM (+diverging lenses if necessary)...say.. at 10% POWER 1st (just above laser threshold), power off scan-fail and fire a static beam (not modulated) where audience should be (people near PJ). LaserBee on a tripod...and covering entire sensor with static beam.
As LaserBee sensor is 15x15mm (225 sq mm):
22,5mw reading on such sensor corresponds to 10mw/cm2 (10mW/100 sq mm).. so 22,5mW on this sensor should be ok MPE.
Do you agree?

Many thanks!

[norty303]

As we've said, cheap thermopiles are not known for their accuracy in the low mW ranges.

HOWEVER, somebody on this here forum actually went so far as to test their Laserbee (one of the very early ones) against a known calibrated meter and found that in the sub 100mW range they actually read high (and after that they showed to be particularly accurate. I've even gone so far as to dig it up for you...

http://www.photonlexicon.com/forums/...ee+calibration


See my post #7 above where I make cryptic reference to 'known characteristics'... If we 'know' it reads low, we 'know' its going to be less than whatever is on the display, so there's an added safety margin.


P.S. I don't agree on the figures you are using for the MPE as you haven't explained what your exposure timeframe is. You appear to be using 4x MPE which would be in a scenario using a very fast scanfail system (1ms or less).
If you are using no scanfail then you should be on 2.5mW/cm2, using a moderate one (like Guidos for example) you're just a bit over the 10ms response time which is 5mW/cm2.

P.P.S Guido provides a header which can be jumpered in order to disable the scanfail. I've got this routed to a jack socket on the rear panel so I can just plug in a blanking plug for exactly these sorts of things.