Results 1 to 7 of 7

Thread: building a simple laser power meter for high peak power

  1. #1

    Default building a simple laser power meter for high peak power

    I want to be able to test the output from a IPG fiber laser, 1064nm output 10 watts average power, 5kW peak pulse power (0.5mJ in 100nS pulse @ 20khz rep rate). I just need to determine if it is close to spec.

    I have a Coherent lasercheck (only good to 1 watt I think and not really meant for this) and a Synrad powerwizard 250 watt which is meant for CO2 laser thermal measurments. The latter might somewhat work if I could know the reflectivitiy of the black annodized target for 1064nm vs 10,600nm

    I looked through my junkbox and found some large area photodiode power detectors from IIRC a 1990 vintage medical yag laser with fiber delivery. The two smaller mounts in the photo contain one of these diodes with some sort of white diffuser which is pointed at a optical element at 45 degrees...I assume some sort of power takeoff beamsplitter arrangement. The larger mount seems to have the same photodiode mounted behind a diffuser and was intended to take the direct output from the 100 watt fiber (CW though).

    I have an Agilent MSOX3024A DSO which I could use to measure pulse length, rep rate, etc.

    How to calibrate one of these sensors when I don't have a known 1064nm source? I think I have an old dpss laser pointer....could remove the filter, run it through a prism to separate out the 532 and 808nm and hope that some 1064nm is leaking through? Then I could measure this 1064nm with the Coherent lasercheck and get a rough baseline for power when I shine the same beam onto my photodiode?

    Ideas? Thanks. Note in the following photos of the sensors, the little blob seems to be a thermister, as I measure 10.9k at room temp across it and 6k when I wave a heat gun near it.

    Click image for larger version. 

Name:	photodiode1.jpg 
Views:	10 
Size:	257.0 KB 
ID:	39162Click image for larger version. 

Name:	photodiodecloseup.jpg 
Views:	8 
Size:	161.4 KB 
ID:	39163Click image for larger version. 

Name:	photodiodefront.jpg 
Views:	8 
Size:	190.7 KB 
ID:	39164

  2. #2

    Default

    Ok, I decided to read the data sheet on the 250 Synrad Powerwizard and it claims better than 96% absorption rate at 1064nm. It also says it can handle 1J pulses of 50uS length, which is 20kW, over an area of 1cm^2. The IPG is supposed to put out 0.5mJ pulses 100nS length which would be 5kW but the unfocused beam is 6mm which is an area of 0.28cm^2. Should still be ok since this would be 18kW over a square cm, which is below the rating. The average power being only 10 watts is also well below the 250 watt rating of the device.

    So I will try that, and use my oscilloscope and the photodiode to measure the pulse duration from some reflected photons. I should verify with an led and my function generator that the photodiode can respond well to a 100nS length pulse.

  3. #3
    Join Date
    May 2009
    Location
    UCSB
    Posts
    715

    Default

    It sounds like you more or less have it figured out, what you will want to do is measure the average power using the thermal power meter, and check the pulse width using the fast photodiode/scope. (If these two measurements are within spec, you are more or less guaranteed that the peak power is also within spec). Worth mentioning, those large area photodiodes are almost certainly too slow to accurately measure the pulse width from your laser, however you should be able to see the pulse train and confirm that the laser is at least operating in the correct q-switched regime.

  4. #4
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    A back biased photodiode such as Thorlabs DET-10A works wonders in this time regime. You'll need a 50 ohm terminator on your scope if it does not have a internal 50 Ohm setting. 1 ns is good, as it lets you see any ringing/missing pulse issues with the Q-switch etc.

    The Thorlabs FDS010 low area photodiode (1 ns rise time) would be fine if you reverse bias it, and its 1/2 the cost of a DET-10.

    You simply need a silicon photodiode with a very, very, small area and about 9-12 volts of reverse bias.

    You don't need to put the PD in the beam, you have lots of light, the scatter from a beam terminator should be more then enough.

    I could dig and find something cheap at Digikey, however it is best that you get a screened device with a known risetime.

    Steve
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  5. #5

    Default

    Thanks Steve. In a circuit board from a junk laser I found a OSD 15-0 silicon photodiode which at least gives me something with a data sheet. It is still fairly large but has a rise time of 9ns with a 10V reverse bias according to the data sheet. I tried reverse biasing it with a linear supply and a 50 ohm load and monitoring the voltage across the load with my 200mhz Agilent DSO but there is hella noise and ringing which masks the few mV of signal. I think I will scrounge a coax from my junk box and make a proper little shielded box for the photodiode, with a 9V battery for bias and set my scope input to 50 ohm termination. This should get rid of some of the noise and the ringing.

    data sheet for the photodiode is here: http://www.osioptoelectronics.com/Li...odes.sflb.ashx

    Nice little app. note here describing fast pulse laser measurements with these diodes here: http://www.osioptoelectronics.com/ap...cteristics.pdf

    It confirms the method you suggested and that I had tried with poor shielding/wiring.

  6. #6

    Default

    Ok, I constructed a much cleaner setup for the OSD15-0 photodiode. I used a small square of copper board and dead bug mounted the RC filter, photodiode, and a short length of coax. I used a E3615A linear supply to provide the 10V reverse bias. For the led, I chose a green because the response of the photodiode peaks at 900nm and falls off to about the same amount at 550nm and 1064nm.

    The measurement oscilloscope is an Agilent MSOX3024A 200mhz 4GS model with built in 20mhz function generator. The function generator was able to directly drive the led through a 10 ohm resistor with a 100nS 4V pulse repeated at a 20khz rate. The rise/fall time on the function generator isn't great (roughly 16ns) but it is convienent to use. Trace one (yellow) shows the output from the function generator and Trace three (blue) shows the photodiode current through the 50 ohm input termination of the oscilloscope. I am only seeing 7mV of signal, but the photodiode data sheet says about .25A/W for responsivity at 550nm. I measured the voltage across the 10 ohm current limiting resistor on the drive led and found it to be 100mV during the pulse. Thus the led has about 3.9V across it and 10mA running through it, or a power input of 39mW. This is a very old green led (>15 years)...maybe getting 5mW to 10mW of light output power? If the photodiode was capturing 50% of 5mW that would mean the reverse current should be about .25A x 0.0025 = 625uA. The voltage generated in the oscilloscope across the 50 ohm input impedance would then be around 30mV.

    Ok, well I am seeing 7mV so at least it lands within reason considering the efficiency assumptions I have made about the led output. There is also the chance that the led response isn't that super.

    Maybe I will try and find a spec led and a spec photodiode and repeat the experiment. Right now though I think this setup would tell me at least that I am getting near 100nS pulses and would tell me the repetition rate very exactly.


    Click image for larger version. 

Name:	photodiodesetup.jpg 
Views:	10 
Size:	207.7 KB 
ID:	39174Click image for larger version. 

Name:	pdiode1.jpg 
Views:	7 
Size:	84.7 KB 
ID:	39175Click image for larger version. 

Name:	pdiodeschem.jpg 
Views:	6 
Size:	69.9 KB 
ID:	39176

  7. #7

    Default

    Well I got the IPG fiber laser today and built an atmel microcontroller circuit to interface to the 25 pin connector, gave it 24V, donned my OD10 goggles, put up some laser mark paper in front of a beam stop and set it to fire it for 5 seconds. It immediately blasted a pencil size hole though the paper (guess I really didn't need that heh heh). I then measured the power with my Synrad 250 watt powerwizard thermal meter and took two readings of 8.9 and 9.0 watts. I would guestimate that the real average power is somewhere around 9.2 to 9.3 watts, but who knows how many variables there were in my two measurements. I have also not played with the rep rate...just using whatever rep rate is built into the unit. I think the spec is 9 to 11 watts average power at 20khz rep rate.

    Tommorow when I have more time I will try out the photodiode/oscilloscope and see if I can see the pulses. Then it is time to start thinking about focusing this little beauty. Can't believe how compact and low input power (24V at 5 or 6 amps). Wasn't it just a few years ago we were messing with dorm fridge sized arc lamp pumped yags requiring 3 phase and having horrible thermal lensing?

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •