Formula for LD power output

[KreAture]

I've seen several claims that it's not possible to estimate the power outputs of LDs but I am starting to think it may be possible. If nothing else atleast we can ballpark it.

CW: 120 mW
Rohm RLD2WMGU1 (660nm): 0.9 mW/mA, threshold 55mA
Rohm RLD2WMGU1 (780nm): 0.85 mW/mA, threshold 55mA

PO: 1 W
Sony SLD332F 805nm: 1.1mW/mA, threshold 400mA
Sony SLD333V 790-840nm: 0.9 mW/mA, threshold 200mA

CW: 500mW / PO: 1 W
Mitsubishi SLD333V 638nm: 1 mW/mA, threshold 170mA

There are ofcource many other datasheets and if ppl check em and post the data I'd be grateful. Especially if you can dig up some blue LD specs. I am curious as to if they are in the same range.

The trend seems to be 0.8-1.1mW/mA making 0.9 a ok guess in my good eye.

There is also a general trend for about 0.5 % increase in total lasing eff for each °C drop in temperature down to about -10°C.
It must be pointed out however that this change in eff is not the result of the slope changing, but actually the threshold current dropping and as a result the LD is higher up on it's slope for a given drive-current.

The most important part of the estimate seems to be the key factor that relative eff is always measured. That is to say, they always subtract the threshold current. eff = Pout / (Id-Ithreshold)

If someone with a calibrated LPM could test a few of their diodes by finding their threshold currents and then doing a few readings of power vs current I'd be gratefull.

This estimate will ofcource be approx. The slope efficiency is usually less than 20% accurate and so is the threshold listed. Measuting threshold is not hard though so we can atleast eliminate that error. The slope efficiency is something we can't eliminate and it can vary between samples. We can however compare multiple identical diodes and pick the best one giving us output on the +side of our estimates.

For my CNC use I will definately be cooling my LD to around 8 below room temp. That will keep it as cold as I can without dew forming, and needing to insulate/moistureprotect it heavily.

 

[Things]

Getting a rough power estimate based on input power is easy, although not all diodes are ever the same, and minor defects in the optical cavity can affect their output power. This means some diodes are slightly more efficient than others.

However, since pretty much all of the laser diodes hobbyists use are ripped from peices of equipment, getting a chart specifying that diodes efficiency is pretty much impossible.

If we paid $2000 a piece for the diode from their manufacturers, they would include a chart specifying the exact efficiency of that diode.

 

[lasersbee]

I have a formula for you....

Take "A" (about the price of one Laser)
and invest it in "B" (a Laser Power Meter)...
then you will know for sure...:beer:

Jerry

 

[Wolfman29]

I like this idea, but it seems like it is essentially estimating, which is at least what I have been doing since I started here. For instance, it's a pretty good estimate of around 700-750mW from 500mA or so from a 12x blu-ray.

1.2A is enough to give 1W of power to a 445, generally, and 1.4A will give up to 1.3W.

Just these sorts of things, ya know?

 

[KreAture]

My point was that it can be ballparked based on general specs.
If someone would so some measurements we could see how well it adds up.

Just because someone claims they sold you a 500mW laser doesn't mean they did.
If you calibrated your LPM based on this laser then you may be calibrating to their estimate not the result of a calibrated instrument.

 

[lasersbee]

My point was that it can be ballparked based on general specs.
If someone would so some measurements we could see how well it adds up.

Just because someone claims they sold you a 500mW laser doesn't mean they did.
If you calibrated your LPM based on this laser then you may be calibrating to their estimate not the result of a calibrated instrument.

Exactly.... That is what LPMs are for...


Jerry

 

[KreAture]

Jerry. You yourself has said that you calibrate with a known output laserpointer.
How do you know the output?

Don't you find it just a bit nice to be able to use the datasheet or common estimates to see if the threshold current and drive current would at all add up right to the claimed wattage?

Some of the LD's have a very high threshold current. This makes it necessary to drive em quite high before getting good value for input. 200 mA threshold and 500 mA drive is really just 300mA into curve. A 50mA threshold LD with 350mA drive may put out just as much. (if it survives...)

 

[lasersbee]

Jerry. You yourself has said that you calibrate with a known output laserpointer.
How do you know the output?

Don't you find it just a bit nice to be able to use the datasheet or common estimates to see if the threshold current and drive current would at all add up right to the claimed wattage?

Some of the LD's have a very high threshold current. This makes it necessary to drive em quite high before getting good value for input. 200 mA threshold and 500 mA drive is really just 300mA into curve. A 50mA threshold LD with 350mA drive may put out just as much. (if it survives...)

I have never said that I use a known output Laserpointer to
calibrate the LaserBee products..

We us a Lab Quality Newport LPM with a Thermopile Head
and stable output Labby Lasers..

That's like saying it would be nice to calculate a car's speed by
the manufacturer's horsepower rating and the Octane value
of the gas used... A speedometer is much nicer and more
precise... IMO

I'm not dissuading you in your quest to calculate a Laser's
actual output power.... but there are easier and more
precise ways of doing it....:yh:


Jerry

 

[KreAture]

Glad to hear they're calibrated well then.

I guess I'm a spec-geek as I like to just look up the numbers and know what to expect.

I did find a few blue LD's with very high slope efficiency btw!
http://www.laser66.com/html/ML320G2-11.pdf

The projector diode ppl keep talking about seems to be a NDB7352 from Nichia but in 5.6mm package. http://redlum.xohp.pagesperso-orange.fr/electronics/data/NDB7352.pdf
Note the wide span of slope eff from 0.8 to 1.8 with a threshold from 150-300mA.
Theoretically the production variances in such diodes are such that higher thresholds usually deliver higher slop eff too, but without a sampling and tests one can't know if this is indeed happening with the projector diodes.
Still cool that they guarantee atleast 0.8 which is in itself quite good.