How to Determine Duty Cycle

I am wondering how you would go about figuring out the duty cycle for a laser. I am posting in here because it is more of a general question but I do have a specific application for the question. I am building a fairly simple (in my eyes) laser pointer. Its a 1w 445nm pointer inside of a Cree led flashlight that has been stripped down as a host. How would I figure out the duty cycle for that application. It goes the same for other applications too thats why I posted it in here. I tried searching for a thread related to what I'm asking but gave up after going through a few pages with no results of what I wanted, so I apologize in advance if this topic has been beaten to death. So bottom line how to figure out the duty cycle! lol Thanks guys

IIUC.... you want to know how long you can run your
Laser before requiring it to cool down so that you get
the longest life out of your Laser....

Once the head of your laser starts heating up... shut it
off and let it cool down...
The Duty Cycle depends on many factors... including...

- current being run through LD
- thermal mass of heatsink
- thermal resistance between LD and outer radiating
surfaces
- external ambient temperature

just to name a few.....


Jerry

So it is more of a heat issue rather than an electrical issue. The duty cycle is simply referring to how long it takes it to get hot then cool off to turn it on again. I am used to 3 phase motors where it has a duty cycle referring to the run time for peak power until it drops rating...but on second thought it is probably including a heat factor where it no longer is as efficient due to swelling of bearing surfaces and adding friction and load. I know I am comparing apples to oranges but to me it is a good thing to know, so stuff doesnt sizzle on me and let the smoke out! lol Thanks again

As long as you respect the Max allowable current through
a specific Laser Diode and follow my above recomendations
you should be fine..

In your Motor scenario...with a duty cycle referring to the
run time for peak power... if you increase the peak power (or
input voltage) of a 220VAC motor by 50% you would need to
reduce the duty cycle by a very large magnitude...

Jerry

lasersbee said:

As long as you respect the Max allowable current through a specific Laser Diode

Click to expand...

I thought the duty cycle was only dependent of the temperature inside the diode/cristal. My question is: how to know what is the max temperature a laser can operate without damage? I've seen many comments about laser too hot or not, but if we put an thermocouple just near the center of the laser and measure its working temperature, we could know if its too hot or not. Just touching the host is not a good way, since if the heatsink works badly, you'll think it's cold but it's not. But what would be that temperature we could never reach for each laser?

Using Laserbee would help on this? I mean, looking the decrease of light power as the time increases? How much decrease is a signal we reach the laser duty cycle? 10%? 20%?

I think duty cycle is one of the most important parameter to evaluate a laser. What the value of a powerful laser if you have to turn it off after every 30s?

Vextor said:

I thought the duty cycle was only dependent of the temperature inside the diode/cristal. My question is: how to know what is the max temperature a laser can operate without damage? I've seen many comments about laser too hot or not, but if we put an thermocouple just near the center of the laser and measure its working temperature, we could know if its too hot or not. Just touching the host is not a good way, since if the heatsink works badly, you'll think it's cold but it's not. But what would be that temperature we could never reach for each laser?

Using Laserbee would help on this? I mean, looking the decrease of light power as the time increases? How much decrease is a signal we reach the laser duty cycle? 10%? 20%?

I think duty cycle is one of the most important parameter to evaluate a laser. What the value of a powerful laser if you have to turn it off after every 30s?

Click to expand...

This is true in diode science. However, most members here are only interested in how to run their hand-held laser safely. As Lasersbee said above, there are many factors that contribute and/or confound how to figure a "duty cycle."

For the majority of members here, getting the best heat transfer to a substantial heat sink, and then monitoring the temperature of the laser is about the best they can hope for.

Re: the 30 second on/off -- Many people are perfectly happy with short burns of bright light

Peace,
dave

Vextor said:

. What the value of a powerful laser if you have to turn it off after every 30s?

Click to expand...

My 3500w CO2 Mitsubishi Laser has a duty cycle of 100% because of a chiller tower. Without the tower you would have I would be willing to bet a zero duty cycle because of how powerful it is. I isnt just about keeping the laser cool in the 3500w laser too you have the bend mirrors as well as the final focusing lense to keep cool to maintain a 100% duty cycle. So keeping cool all the way around is the way to be hehe.

To answer your question though. maybe you only need 20 seconds of actual beam on time and dont need the full 30 seconds. If you need more duty cycle then provide better cooling to the LD and driver or go back to the drawing board to improve your design.

daguin said:

Re: the 30 second on/off -- Many people are perfectly happy with short burns of bright light

Click to expand...

Sure! No doubts! For any laser fanatic like us, it's always fun.

My point is: I buy a portable laser and the seller shows in his site a duty cycle 60s/30s on/off. How can I check that, without burning my toy?

Vextor said:

My point is: I buy a portable laser and the seller shows in his site a duty cycle 60s/30s on/off. How can I check that, without burning my toy?

Click to expand...

I think that might go back to what Lasersbee first said in his first response to my original post. I would quote him but I'm not the greatest with multiple quotes haha :thinking:

If I actually needed more than 60 seconds on, on a regular basis, I would invest in/build a lab style laser with active cooling and full size (not smd) components.

I have a Lab style 445nm Laser in the shop that
we use for testing... It runs happily for 30 minutes
at ~1200mW before the passively cooled Heat Sink
begins to get warm.... then we shut it down for a
cooling off period...


Jerry

It will always get warm, there is no way around that, even if you build a heatsink the size of a shoebox around a laser diode.

The question is, is it getting 'too warm'? There are only 4 variables that determine if you can run a laser continously, and if not, at what duty cycle you can:

- maximum operating temperature of the laser diode (Tj)
- the heat produced by the diode (P)
- ambient temperature (Ta)
- thermal resistance all the way from the junction to ambient (Rja)

The first three should be easily answered using the datasheet for the diode and a thermometer, so its down to the latter. When building a labby, you probably have a heatsink of known performance (K/W), or you can compare it to a heatsink of similar size and shape and make an educated guess. The same goes for thermal resistance from junction to heatsink: junction to diode is in the datasheet, diode to aixiz, and aixiz to heatsink would have to be guesstimated. Taking a bit of safety margin its not hard to calculate how large a labby must be to run continously.

The temperature calculation is straightforward:

Tj = Ta + P / Rja

There is, however, a common misconception about heatsinks: Its NOT the mass of the thing that matters for duty cycle or performance, its the thermal resistance thats heavily influenced by shape, airflow etc. The only thing the mass (thermal capacity) does is buffer some heat. With a big hunk of metal inside a flashlight you can run the laser for some time heating up the metal, but once its hot you have got to wait for it to cool down too.

So a lot of metal on the inside of a host would make the difference between, say, 10 seconds on 20 seconds off and 1 minute on 2 minutes off, but it will remain 33% maximum on time if the outside is identical.

In some of our hand-held builds we also have the problem of OTHER components overheating (driver, switches, etc.)

Peace,
dave

Benm said:

The temperature calculation is straightforward: Tj = Ta + P / Rja.

Click to expand...

Yes, this is science! If you are building a new one or want to disassemble a laser, this is the way.

Excuse me if I insist, but my goal, and I think it's the same for Frazman187 at the very first post of this thread, is not to design a new laser, but once you have one in your hands, how to find out it's duty cycle measuring something and getting a quantitative value. In my case, I restrict the question to a laser pointer bought somewhere.

If there is some direct or indirect measurable indication (a symptom) that the cycle has been reached, what would it be? The outside temperature can't be, because if it has bad heat dissipation, you won't have a good indication of it's inner temperature. Would it be a power decrease? Any kind of fluctuations? Or there is no method to measure it?

Ofcourse there is a way to measure it: Dissipate a known amount of power in the laser diode, let it run for a long time (up to equilibrium) and measure the temperature of the laser diode and ambient. Knowing the power burned, you can easily calculate the K/W value for the entire host.

There is, ofcourse, a catch: With a host that is obviously too small for CW operation, this method of testing will damage it. The best way would be to temporarely remove the driver, replace it with a current source of your own, and measure the voltage too. This could even be below lasing threshold, as long as its know, you're good.

Measuring the case temperature of the diode could be done with a thermocouple inserted somewhere in there... or you could measure it using the laser chip itself at a specific current, since the forward voltage will drop by a few mV for every 10K temperature rise. You would need to set up a control for that laser diode, so its not as easy in practice.

There is no simple trick to asses the cooling performance of a host by just its dimensions and rough shape - the details really really matter. You could -calculate- the typical thermal resistance of a smooth cylinder of dimensions comparable to your host, but you will probably end up far too low if your laser has any kind of texture or surface features.

Vextor said:

Yes, this is science! If you are building a new one or want to disassemble a laser, this is the way.

Excuse me if I insist, but my goal, and I think it's the same for Frazman187 at the very first post of this thread, is not to design a new laser, but once you have one in your hands, how to find out it's duty cycle measuring something and getting a quantitative value. In my case, I restrict the question to a laser pointer bought somewhere.

If there is some direct or indirect measurable indication (a symptom) that the cycle has been reached, what would it be? The outside temperature can't be, because if it has bad heat dissipation, you won't have a good indication of it's inner temperature. Would it be a power decrease? Any kind of fluctuations? Or there is no method to measure it?

Click to expand...

If you understand science, then you must understand the concept of confounding variables. Getting a new laser in hand, without being able to control for all of the variables that are present, which will affect heat dissipation, and then trying to test it, will give you trash for results.

Peace,
dave