diy laserdiode measurements with solarcell

[Krutz]

hi everyone!

i am about to build more red ld stuff, i have quite a few diodes from Pioneer DVR 109 16x burners. they die at quite random currents, instantly from COD. like between 180 and 400mA, dont remember exactly. therefore my thread about "how to find out the maximum current by watching" some time ago.
will get a thermal powermeter soon (die4thing) hopefully, unfortunately they take quite long for individual measurements. its second problem is that its not here yet, anyway.
so i dug out my solarcell i bought long ago, figured what was wrong with it (foil over the contacts to first rip off), made it into a box, and tried how to get some readings out of it.

here it is, my uncalibrated laserpowermeter, christmas-edition:

 

[Krutz]

the box was a small box of tea. the solar cell has several "segments", i taped all but one with aluminum foil (perfect for 100% dark cover).
inside its dead black, just for the heck.
in the top is a small lens, to enlarge a collimated beam a bit. no, no special reason neither.

 

[Krutz]

you can obviously measure voltage and current across the cell. voltage would top out quickly, but current works great. between 0.09 and 5mA with the laserdiode i will show later. i guess that small cell could measure easily 500mw before topping out, estimated with hands full of pointers and my hqi desktop lamp.
with that box i have a "dark reading" of down to zero! would be 0.02v, but since i use the current reading anyway..
oh, if you build something like that, watch out for reflected beams. my cell is basically a thin film on a thick piece of glass.
built it in some hours, contacting the cell was the longest and most stupid part.
find some solar calculator, good enough. for relative readings, anyway, which was my goal.

here comes my graph.
X is mA, Y is reading in mA on the cell. so no way to calculate power, dont even know if the whole beam hit the cell. i didnt touch the setup while testing, put the diode in a heatsink with fan, and waited a moment between readings. even with that much passive cooling, the output drops slowly but steadily, and immediately.

 

[Krutz]

when i got those diodes, like a dozen from a multiauction, i sent one to someone to measure it. i never really believed his graph, couldnt accept such a low reading.. he did in 25mA steps, and used a coherent field mate with silicon head. without heatsink at all, bare diode, said it wouldnt be necessary with a reading-time less than 3 sec per step.
anyway, i attach it for reference.

 

[Chris.]

Nice photovoltiac....
My only worry would be that the sun's energy is about 1000W per square meter(at ground level, sunny day) = 1 000 000 square millimetres. assuming a laser dot occupies 3mm^2, that cell is designed to take 0.003W per 3mm^2. That's 3mW, I think a laser dot might damage it so that its output would change with respect to time.

But, that's just theory, can you explain your graph to me?

 

[Krutz]

oh, of course i will make *many* tests once i have my prrrecious powermeter. different wavelengths, responsetime, "calibration" of solarcells, silicon diodes, leds, you name it!

but for now:
thats an easy way of finding out where a particular LD will rather melt than produce more output.. look at those last few lines in my graph's numbers: same output, more current. that extra input is more heat, and less runtime. and less lifetime too of course!

anyone with me? i expect pics in, say, 48 hours!

manuel

 

[Zarniwoop]

You might want to bounce off a piece of glass to reduce what hits the cell.
Otherwise I'd expect a 100mW beam to fry your cell.

 

[Benm]

Interesting result for sure... looks quite good. I think the lens to spread out the beam over a larger area of panel is certainly a good idea. Also, this should not introduce much error as long as all incoming power your want to measure is a paralel beam (or close). Fairly typical for laser pointers and diodes in an aixiz with the lens in the proper position.

I guess the main disadvantage is that you need to make a calibration curve for each time you build one of these, and also for all wavelengths. This makes it all very difficult to get quantitative readings unless you already have a power meter, or a few lasers of known power.

Still, even taking relative measurements is probably useful - especially when for example tweaking something for maximum power.

 

[Hemlock_Mike]

Krutz --

This is the way most of us started. Anything to get a reading of some kind - often referenced to whatever looked good. Your use of the lens is a good idea to prevent solar cell damage.
Use this for learning - -eventually, you will get a meter !!!

Mike

 

[Rhith]

I like this simple "meter". Very festive for the holidays to boot! I'm interested in what you can pull off once you get a proper meter so you can compare.

 

[Krutz]

thanks for the notes!

yes, i never intended to do some "real" powermeasure with that, only relative measurements. but just for that alone its great, like finding the knee of a diode, or comparing uncomparable lightsources.. different wavelengths, lasers with lightbulbs, apples with bananas, you name it!
so, if i burn holes into the cell, it still reads accurate, since it reads relative anyway.. hehe
this here is a thin-film cell. take a regular multicrystalline blue cell, no way you will fry it (ok, perhaps with other stuff than diodes *smile*). its all metal, perhaps 1/4 or 1/2 mm, and somewhat reflective. cheap too, and should be somewhat linear too.

oh yeah, cant wait for my powermeter! i think that combination is just perfect.. a calibrated heat-meter (wavelength-independent), and a (hopefully linear-response) semiconductor sensor (solar cell, silicon cell or similar) for quick measurements. "calibrate" the silicon "meter" at every session, so you can compare the both readings, and wont even care about wavelength!

manuel