Author Topic: Laser diode efficiency  (Read 11203 times)

0 Members and 1 Guest are viewing this topic.

Offline npelovTopic starter

  • Frequent Contributor
  • **
  • Posts: 331
  • Country: bg
    • Microlab.info
Laser diode efficiency
« on: February 11, 2019, 06:17:28 pm »
I recently bought a 15W blue laser diode. I was wondering if I could roughly calculate if it's close to the rated output. Does someone know how efficient are these laser diodes? The input from the power supply is 23W. I can assume driver efficiency >=90%. I know it's not the way to measure laser output, but well... I just want to know if it's close to it's rated value.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15148
  • Country: de
Re: Laser diode efficiency
« Reply #1 on: February 11, 2019, 06:23:33 pm »
The efficiency of laser diodes can be pretty high. I remember around 50% efficiency for a 15 mW red one - though this was slightly better than genral data-sheet specs.

Still even a crude measurement is probably better than just looking at the power consumption. A poor quality diode tends to be lower output, not so much lower electrical input.
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7043
  • Country: nl
Re: Laser diode efficiency
« Reply #2 on: February 11, 2019, 06:37:31 pm »
Laser power meters all seem rather expensive.

How about taking a small plate of aluminium, painting it black, putting a resistor and temperature sensor and some insulation behind it? Then you can see how much power you have to put into the resistor to get the same steady state temperature as the laser (widen the beam a bit obviously, so you don't burn the paint off).
 

Online ejeffrey

  • Super Contributor
  • ***
  • Posts: 4031
  • Country: us
Re: Laser diode efficiency
« Reply #3 on: February 11, 2019, 06:47:12 pm »
That's basically how power meters work.

Agree that you are not going to be getting 65% wall to beam efficiency of of an ebay blue laser.  Red or IR lasers can have near 50% conversion efficiency from the DC power input to the output power, but blue diodes are considerably worse.  I wouldn't be surprised if your driver was only 65% efficient for diode input power of 15 watt.  The output power would be a fraction of that.
 
The following users thanked this post: grouchobyte

Offline npelovTopic starter

  • Frequent Contributor
  • **
  • Posts: 331
  • Country: bg
    • Microlab.info
Re: Laser diode efficiency
« Reply #4 on: February 11, 2019, 06:55:54 pm »
How about taking a small plate of aluminium, painting it black, putting a resistor and temperature sensor and some insulation behind it? Then you can see how much power you have to put into the resistor to get the same steady state temperature as the laser (widen the beam a bit obviously, so you don't burn the paint off).

I thought about aluminium, but since I don't have laser diode with known input I didn't know what to compare it to. It didn't occur to me that a resistor would transfer 100% of the energy to heat. Now I have the problem that I need a resistor with good heat transfer. I have 50W suitable for mounting on a heat sink, but they contain a lot of aluminium themselves, so it'll take more time to transfer the heat to the plate. The only way is to keep it running while the temperature rising stops. But the laser does not really have enough cooling to run for a long time at 100% power.
I need a small resistor with good mounting point for fast heat transfer, so I can measure temperature rise above ambient for given time.
I have 5W cement resistors, but I'm not sure they have that good thermal conductivity.
« Last Edit: February 11, 2019, 06:57:47 pm by npelov »
 

Offline chris_leyson

  • Super Contributor
  • ***
  • Posts: 1549
  • Country: wales
Re: Laser diode efficiency
« Reply #5 on: February 11, 2019, 07:03:10 pm »
Quote
The efficiency of laser diodes can be pretty high. I remember around 50% efficiency for a 15 mW red one
I also remember getting about 50% efficiency for red laser diodes in the 35 to 40 mW range, in good agreement with the data sheet. Blue laser diodes on the other hand were less efficient. We used a Coherent Laser Check for most of the measurements but it reflects maybe 50% of the light back off the sensor so you have to careful with your measurement technique. At 15W thermopile and bolometer techniques work well but not so sure how accurate a homebrewed power meter would be.
 

Offline ajb

  • Super Contributor
  • ***
  • Posts: 2785
  • Country: us
Re: Laser diode efficiency
« Reply #6 on: February 11, 2019, 07:49:47 pm »
That's not a 15W module.  I don't think anybody even makes a laser diode known to do more than 7W in a single can format (Nichia NUBM44, which doesn't even officially exist, but you can buy them as pulls from some sort of diode array). 

Anyway you really can't establish the module output power from the input power, ESPECIALLY for a random eBay module.  It depends on the quality of the diode and how carefully it's being driven--as you increase the drive current, the efficiency will crest before the output power does (which in turn happens just before the diode dies, typically).  It also depends on the quality of the collimating optics (basic transmission and how well matched they are to the diode).  Diode age and mistreatment will diminish optical output and therefore efficiency further.

But as a very rough ballpark, I would expect a high-power collimated blue diode to do something like 20-50%.  In reality, I suspect hat your estimate of driver efficiency is optimistic, and just based on the diodes that are known to be available and the basic construction of that module, I would guess that you're probably getting something like 3.5W at most.  It's POSSIBLE that you're getting as much as 7W, but I wouldn't bet on it. 
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Laser diode efficiency
« Reply #7 on: February 11, 2019, 08:11:03 pm »
Laser diodes are notoriously inefficient. It really depends on the laser and it’s structure. Blue lasers are probably of the less efficient kind and I expect you may get less than 10% efficiency. Typically lots of power is wasted in heat in a laser so if your input power consumption is 23 watts and you are expecting 15 watts of optical power, you might  be sorely dissapointed. The best way to measure optical power is with a properly calibrated power meter. Trust me on this. I have spent more than half of my 45 year engineering career in fiber optics and test and measurement with a stint at Coherent.

@grouchobyte

Huh? Laser diodes are incredibly efficient, especially compared to other types of lasers, for example a typical Ar-ion laser is doing well to reach 0.01% efficiency, the common HeNe lasers are not as bad but still quite low while 50% is not uncommon for a diode laser. Even LEDs made for lighting are only around 50%.

The commonly available 445nm blue diodes can deliver 1W output after the collimator for around 5W input so ~20% efficiency.
 
The following users thanked this post: grouchobyte

Online ebastler

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: Laser diode efficiency
« Reply #8 on: February 11, 2019, 08:30:46 pm »


'"15W" 450nm blue laser engraver module dies after 3 minutes'
Not worth buying, I guess.  :--
 

Online ebastler

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: Laser diode efficiency
« Reply #9 on: February 11, 2019, 10:05:19 pm »
Grouchobyte, we are looking at laser diodes with direct 445/450 nm emission here; not diode-pumped crystals. Progress never stops! ;-)
 

Online ejeffrey

  • Super Contributor
  • ***
  • Posts: 4031
  • Country: us
Re: Laser diode efficiency
« Reply #10 on: February 11, 2019, 10:10:29 pm »
Laser diodes are notoriously inefficient. It really depends on the laser and it’s structure. Blue lasers are probably of the less efficient kind and I expect you may get less than 10% efficiency. Typically lots of power is wasted in heat in a laser so if your input power consumption is 23 watts and you are expecting 15 watts of optical power, you might  be sorely dissapointed. The best way to measure optical power is with a properly calibrated power meter. Trust me on this. I have spent more than half of my 45 year engineering career in fiber optics and test and measurement with a stint at Coherent.

@grouchobyte

Huh? Laser diodes are incredibly efficient, especially compared to other types of lasers, for example a typical Ar-ion laser is doing well to reach 0.01% efficiency, the common HeNe lasers are not as bad but still quite low while 50% is not uncommon for a diode laser. Even LEDs made for lighting are only around 50%.

The commonly available 445nm blue diodes can deliver 1W output after the collimator for around 5W input so ~20% efficiency.

Yes, compared to an argon ion laser -- I used to use a large frame Ar+ laser operating at 351 nm.  Something like 30 kW in, 1-2 watt out.  Flashlamp pumped Nd:YAG lasers are similarly bad.  However blue laser diodes are considerably less efficient than red lasers and considerably less efficient than blue LEDs.  They are definitely not in the 80-90% that would be required for the OPs laser to have 15 watt output with 23 watts to the controller.
 

Offline LaserSteve

  • Super Contributor
  • ***
  • Posts: 1354
  • Country: us
Re: Laser diode efficiency
« Reply #11 on: February 12, 2019, 04:24:33 am »
Scroll down to the I/V/P pictures at the bottom of this page.Crude but effective.

https://sites.google.com/site/dtrslasershop/home/diodes/445-m140-didoes

Steve
"When in doubt, check the Byte order of the Communications Protocol, By Hand, On an Oscilloscope"

Quote from a co-inventor of the PLC, whom i had the honor of working with recently.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Laser diode efficiency
« Reply #12 on: February 12, 2019, 05:15:00 am »
I'm sure the 15W claim is bogus, the highest power single blue laser diodes I'm aware of can produce about 7W which is an absolutely crazy amount of power from a visible laser diode but it's not 15W. It's possible something new has come out since then but I haven't really kept up. I have one of the old 1W 445 diodes and it's scary enough, it has a tendency to set things on fire across the room. I played with gas lasers for years and previously the most powerful I had was a 50mW multiline air cooled argon that draws over 1kW. 1W or more meant a large frame ion laser needing 3 phase power and water cooling and then suddenly you could carry a 1W laser in your pocket that runs off a battery. I'm actually surprised that in ~6 years of these things being available we haven't been hearing about large numbers of eye injuries.
 

Offline Wolfram

  • Frequent Contributor
  • **
  • Posts: 414
  • Country: no
Re: Laser diode efficiency
« Reply #13 on: February 12, 2019, 10:22:01 am »
I checked a few datasheets for high power multimode laser diodes in the 450 nm range, and it seems like the typical efficiency of the raw diode is around 30 %, give or take. Here's an example: http://download.insaneware.de/0019.pdf
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20356
  • Country: gb
  • 0999
Re: Laser diode efficiency
« Reply #14 on: February 12, 2019, 10:50:39 am »
Lasers, unlike LEDs, rate optical output power instead of electrical input power.
Normally both are given for LEDs. Visible LEDs are normally specified in lumen output and invisible/far ends of the visible spectrum are typically specified in radiometric power output. Yes 50% efficiency is typical for royal blue and far red LEDs. It wouldn't surprise me if laser diodes can approach this efficiency.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 10384
  • Country: nz
Re: Laser diode efficiency
« Reply #15 on: February 12, 2019, 11:52:34 am »
15W  :o, and to think i'm scared of my 300mW
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15148
  • Country: de
Re: Laser diode efficiency
« Reply #16 on: February 12, 2019, 01:08:41 pm »
15W  :o, and to think i'm scared of my 300mW

A high power laser is scary and may need a special permit.

I remember getting scared from the first blue laser pointer made from a diode laser.  Not that bright (probably due to 408 nm wavelength), but the presentation showed a curve for the laser diode to be able to go up to 120 mW.  With a fluorescent paper it got quite bright - so it was probably above todays legal limits for this use - though maybe not in 1999.

So for some reason even the first blue laser diodes where quite powerful.
 

Offline npelovTopic starter

  • Frequent Contributor
  • **
  • Posts: 331
  • Country: bg
    • Microlab.info
Re: Laser diode efficiency
« Reply #17 on: February 12, 2019, 01:34:49 pm »
The state of the art 808nm pump laser has around 40% efficiency. This should say something.

Lasers, unlike LEDs, rate optical output power instead of electrical input power. If it rates 15W, it draws at least 37.5W. With a 90% efficiency driver, it should suck in 41.7W.

Your 23W looks more like an 6W NUBM144 over driven to 7W~8W.

Some laser modules rate peak power, that is the peak power when operated in QCW mode. Many so called high power Chinese laser engravers do so.

That was the kind of answer I was looking for. I don't care if I assume it's 6W while it's actually anywhere between 4 and 8. It's roughly 3 times less so I can submit my return request.
Also actual power doesn't matter to me. All that matters is that the laser should be able to mark steel as I saw in a video of AppliedScience.
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Laser diode efficiency
« Reply #18 on: February 12, 2019, 06:25:17 pm »
In the 90's I worked with systems containing 830 nm IR laser diodes ranging in power from 15 mW to 20 W.  Here's a few points I picked up over the years.  It should be applicable to blue lasers but as I haven't worked with them I'd like others to chime in.

AFAIK, all laser diodes have a threshold current which has to be exceeded to produce coherent light ie., make them "lase".  Below the threshold, they work as LEDs, often with a different wavelength.  Above the threshold, the additional current goes into radiation and so the incremental efficiency is remarkably high.  Diode lasers emit from a very small area on the edge of the die called a facet.  Small diodes have a power limit because the power density at the facet can get so high that it damages the facet and the diode stops lasing.  This limit was usually just above the rated power and could be easily exceeded with a power supply surge.  We joked that the burn rate of laser diodes was measured in dollars per nanosecond because it happened so fast.  Larger bar lasers were a bunch of small lasers that had an overall power dissipation limit and were rarely damaged except by large, fast current pulses.

I prepared a few of the laser safety reports and applications to the CDRH to get approval to sell our equipment in the US.  There are a lot of misconceptions about what Brightness really means in physics.  (hopefully I have the correct terminology)   Brightness is the power intensity received divided by the solid angle subtended by the emitter.  In other words, how much power received per area per the apparent size of the source.  This is very important for laser safety.  For an unfocussed system, ie. no lenses, it doesn't matter too much if the power falling on a surface comes from a point source or from something much larger.  For a focussed system, such as the eye, the lens can concentrate the power in a much smaller area.  A laser has a very small exit aperture and the solid angle it subtends is very small.  This means that the power intensity where an image has been formed is very high and can cause damage.  Laser beams usually have small divergence and so any spot formed is also quite small and can also be a hazard.

We used laser power meters with black corrugated sensors to measure the laser power outputs.  They were accurate everywhere except for the film plane where the power intensity was too high.  We also had some integrating spheres and silicon sensors to more accurately measure lower intensities.  I don't know what sort of sensor could accurately measure laser radiation that can ablate metal.  Perhaps a defocussing lens to spread the beam out and a larger area sensor?  Large area silicon photodiodes have a fairly well defined response to radiation of various wavelengths and produce a photocurrent proportional to incident power so it should be possible to build a power meter with available parts.

Cheers,

 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Laser diode efficiency
« Reply #19 on: February 12, 2019, 06:30:56 pm »
I don't see how you're gonna mark steel with even a true 15W visible diode laser, most of the energy will just bounce off. 15W far IR from a CO2 laser might do it. In order to mark something, it has to absorb the energy, most metals make pretty good reflectors at visible wavelengths.

Regarding pointers, in the US I believe the legal limit has always been 5mW but nobody seems to enforce that. You can easily buy pointers from China that are well over 100mW.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20356
  • Country: gb
  • 0999
Re: Laser diode efficiency
« Reply #20 on: February 12, 2019, 06:56:58 pm »
I don't see how you're gonna mark steel with even a true 15W visible diode laser, most of the energy will just bounce off. 15W far IR from a CO2 laser might do it. In order to mark something, it has to absorb the energy, most metals make pretty good reflectors at visible wavelengths.
Metals are even more reflective at infra red, than visible. The reason metals reflect light so well is because they conduct electricity. As a general rule, at lower frequencies, there are less losses, so more reflectivity. To melt steel the frequency needs to so high, the free electrons in the metal can't move fast enough, so deep UV might do.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Laser diode efficiency
« Reply #21 on: February 12, 2019, 07:10:09 pm »
I've seen steel being cut by a CO2 laser but I don't know what the power was, quite a big one though. According to this https://fslaser.com/Applications/Metal a 90W CO2 laser can mark steel directly without using a special coating although they mention a fiber coupled laser is better I'm not sure what wavelength fiber laser they're referring to. CO2 is 10,600nm so very far IR.

Either way a 15W laser isn't gonna engrave steel unless maybe with a marking coating.
 

Offline ajb

  • Super Contributor
  • ***
  • Posts: 2785
  • Country: us
Re: Laser diode efficiency
« Reply #22 on: February 12, 2019, 07:33:46 pm »
I prepared a few of the laser safety reports and applications to the CDRH to get approval to sell our equipment in the US.  There are a lot of misconceptions about what Brightness really means in physics.  (hopefully I have the correct terminology)   Brightness is the power intensity received divided by the solid angle subtended by the emitter.  In other words, how much power received per area per the apparent size of the source.  This is very important for laser safety.  For an unfocussed system, ie. no lenses, it doesn't matter too much if the power falling on a surface comes from a point source or from something much larger.  For a focussed system, such as the eye, the lens can concentrate the power in a much smaller area.  A laser has a very small exit aperture and the solid angle it subtends is very small.  This means that the power intensity where an image has been formed is very high and can cause damage.  Laser beams usually have small divergence and so any spot formed is also quite small and can also be a hazard.

"Brightness" can mean a lot of things, but is typically a photometric description rather than a radiometric one, since the apparent brightness of a source depends on its spectral content in conjunction with the spectral sensitivity of the human eye.  In physical terms it refers to the total luminous flux as well as some combination of the emitting or receiving area and emitting or receiving solid angles.

Generally laser safety considerations are based on irradiance (incident *power* per unit area), without regard for the emission or reception angles (except insofar as the emission angle determines the irradiance as a function of distance).  The angles do have an impact on the risk of injury, but in most applications it's not worth splitting hairs about.  A multi-watt beam with an initial diameter of a couple of mm and only a few mrad of divergence obviously maintains a rather high irradiance for quite some distance.  (Pulsed laser safety measurements must consider pulse energy and repetition rate, which is not irradiance per se, but still the angle is not generally a significant factor.)  You could certainly produce a more accurate hazard model with a more sophisticated optical model, but no one wants to be responsible for the empirical aspects of that study. . .

Quote
I don't know what sort of sensor could accurately measure laser radiation that can ablate metal.  Perhaps a defocussing lens to spread the beam out and a larger area sensor?  Large area silicon photodiodes have a fairly well defined response to radiation of various wavelengths and produce a photocurrent proportional to incident power so it should be possible to build a power meter with available parts.

Power sensors for up to 120kW are available, but obviously not cheap!  A lot of the magic is in the absorber, with high power sensors using cone-shaped deflectors and annular absorbers to spread the incident power out over a large area.  A crude measurement can be made using an integrating mass and a temperature sensor, as long as you can calibrate the absorbance and thermal mass (here is a very DIY example).  The trouble with photodiodes is the damage threshold is usually low and poorly specified.  You can send only a fraction of the beam to the diode (reflective sampler or attenuator), but then you have to be able to figure out what that fraction is!

I don't see how you're gonna mark steel with even a true 15W visible diode laser, most of the energy will just bounce off. 15W far IR from a CO2 laser might do it. In order to mark something, it has to absorb the energy, most metals make pretty good reflectors at visible wavelengths.

CO2 won't mark plain metals well (at least not until the ~100W level), they're too reflective even at 10um.  I believe most marking lasers are UV fiber (~300nm?).

Quote
Regarding pointers, in the US I believe the legal limit has always been 5mW but nobody seems to enforce that. You can easily buy pointers from China that are well over 100mW.
Basically, but this covers products, not necessarily components. 
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Laser diode efficiency
« Reply #23 on: February 12, 2019, 11:35:02 pm »
You can buy fully assembled ready to use laser pointers in the hundreds of mW range all over ebay and various direct from China sellers. Totally illegal in the US but I've bought a few and never had any issues getting them. Fortunately I'm sensible enough to not go pointing them around carelessly but a lot of people aren't.
 

Offline jolshefsky

  • Regular Contributor
  • *
  • Posts: 227
  • Country: us
    • Jason DoesItAll
Re: Laser diode efficiency
« Reply #24 on: February 13, 2019, 12:51:41 am »
When you're talking efficiency, you're asking about power-in versus power-out, right? So for practical accuracy, P[in] = P[out] + P[loss], and P[loss] is going to be heat, so why not measure the electrical power input and the thermal power output (e.g. heat sink dunked in known quantity of water; measure temperature change over time; calculate power) and subtract?
May your deeds return to you tenfold.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf