0%
0%

# Measuring DVD-RW laser

Scientific approach to figuring out max drive current

Similar projects worth following
5.5k views
Finally, I decided to try something many people have made already: get hold of a powerful laser by ripping one off a DVD-RW drive. I have a plenty of those, and despite all are working, I decided to sacrifice one.

Why do I publish another project devoted to that common thing? Because while searching for what max current can I pass to that diode, I couldn't find any info online that I would trust.

So I'm trying to figure out max drive current myself! And I don't want to kill the laser!

OK, I still have no idea, what is the actual max current in terms of not killing the laser diode. But a lot if interesting stuff was revealed while measuring light output vs drive current.

The curve of my diode looks like this:

The curve starts to be irregular at currents of more than 0.3 A. The irregularity is accompanied by spatial oscillation of the beam at frequencies of MHz range.

I don't know if this is damaging to the diode or not, but for now I conclude that the diode, at least for optical purposes, should not be driven higher than 300 mA.

So, if you want to measure the max drive current for your laser, try discharging a capacitor through a resistor and the diode, and capturing current-vs-intensity curve in the process. From that curve, choose the point where it begins to be irregular, derate by 10%, and this the max current. Hopefully =)

• ### Attempts at explaining oscillation

I had two guesses at the reasons that may cause the oscillation of light output of the laser diode, that happens at high currents.

1. excitation of next mode. At low power, the lasing will happen at the frequency where the conditions are the most favourable. However, as the power is increased, other modes of resonator may come into play. Sum of two sinusoids at close frequencies manifests as beating, which might be what I observe.

2. oscillation may be caused by the onset of catastrophic optical damage. Say, the thermal runaway starts. It causes increased absorption, which reduces light intensity in lasing cavity. Reduced intensity causes thermal runaway to abort, and then it happens over and over again.

I can relatively easily rule out possibility 1. I am observing "beating" at frequencies of a few MHz. I can easily calculate the first beat frequency.

Wavelength of emission is about 650 nm. The length of crystal is about 0.3 mm, so is the length of cavity. That means the optical path of one loop of light travel is 2*0.3mm*3.49 (3.49 is refraction index, and 2 is because a full loop is a travel back and forth). And I get that in the path length, 3200 full waves fit. So, the lowest beat frequency is the frequency of emission light divided by 3200. That yields 143 GHz. That's way higher than what I observe.

I haven't yet come up with a way to proof or disproof explanation 2....

• ### Catastrophic optical damage

There were a few people here bringing attention to a possibility to kill a laser even with a very short pulse. Suddenly, I found a Wikipedia page about it: Catastrophic optical damage. Enjoy.

The article still refers to heat as the final reason of mirror death, but the thing is about local thermal runaway. It somewhat reminded me of a secondary breakdown phenomenon in bipolar transistors (BJTs).

• ### consequences

DeepSOIC03/03/2016 at 17:31 1 comment

When I was playing with the laser, I aimed my phone camera at it. The laser was operating in pulsed mode, with low average power but high peak power. I didn't notice any damage back then.

A month later I went skiing. Was taking pictures with my phone. And found some UFOs on photos!! What tha?! Look at these orange strokes in the sky (not at Valluga bahn =)

On most photos, one has to look carefully to notice, but on blue sky the damage done to sensor is very visible. Oops!

This leads to a very surprising and totally unexpected conclusion. DON'T LOOK INTO THE LASER WITH THE REMAINING EYE!! No, really, don't.

• ### temperature

I became curious, if temperature of the diode affects the onset of the wiggles in the L/I curve. Sure it does.

I can't say it's totally unexpected. Upon heating, the wiggles start at higher currents. But the output power at the onset drops. So the wiggles are not tied to specific output power, and to specific current.

• ### oscillation

The laser doesn't stop bringing surprises.

Now I've repeated the slow measurement with diverging lens, in hope of getting a video of the spatial redistribution.. And guess what happened? The laser is oscillating!!!

On the top is the overall trace, slow scale. On the bottom, I've zoomed into the starting piece of that widening. There is oscillation. The frequency of this oscillation is 950 KHz.

Previously, I wasn't seeing this oscillation, because my phototransducer was too slow to pick it up.

Go figure...

EDIT: the frequencies of oscillation are different in the bursts, with the bursts themselves not being simple. I see frequencies up to 8 MHz, which is probably the point where I hit the bandwidth limit again.

• ### wiggles are spatial

Breaking news! Those wiggles on L/I curves I got previously were tiny, because I've set up the measurement in such a way that the total intensity is registered.

But now I switched to higher-speed measurements, and I had to upgrade my photodiode transducer to become faster. Before, its settling time was about 10 us. Now I wanted to measure at microsecond scales, so I lowered the resistor from 100k to 1k, boosting the bandwidth by two orders of magnitude. This is at a cost of losing sensitivity by same two orders of magnitude, but I wasn't concerned about it, because the laser outputs quite insane amount of light.

Because the sensitivity dropped, I could no longer measure light by just aiming the photodiode at the spot on white surface. So I've put it directly into the beam.

The transducer was overwhelmed by looking directly into the beam, so I decided to diverge the beam with a lens. And then! Woooow!

And then I saw that my tiny wiggles have become insane! Look!

And that insanity is changing, as I move the detector inside the beam:

On this pic, I've overlaid screenshots from three different points of the beam. It can be clearly seen, that wiggles are almost exactly inverted on top and bottom plots, which are on the opposite sides of the beam.

So, the wiggles are mostly redistribution of light in the beam, the while the total light output is relatively unchanged. Fascinating, isn't it!

• ### some info

Thanks to comment by @Brian, who mentioned a very googlable term, "kink current", I was able to find some info on laser diodes in general, that sheds some light on what is going on.

http://assets.newport.com/webDocuments-EN/images/AN05_Laser_Diode_Overview_IX.PDF

If the link goes dead, try googling for "laser diode kink current", or for the title: #5 An Overview of Laser Diode Characteristics - ILX Lightwave.

It does not include Brian's definition of kink current, but rather refers to kinks (that is, wiggles, noise, irregularities) on L/I curve.

And below is another graph from the internet, which shows more obvious wiggles on L/I curve.

This came from Keithley Application Note "High-Throughput DC Production Testing of Telecommunications Laser Diode Modules".

There, it's also noted that devices with kinks in their working current range should be rejected from production line, which gives me a good reason to state: nominal current of my laser diode is very likely less than 300 mA.

• ### now getting data from oscilloscope

For the first measurement, I read out the data from the captured screenshot of the oscillogram. It's good enough, but after repeating the process for a slower decay, I felt like it's too much trouble, so I started experimenting on extracting the data directly off oscilloscope. So I captured a waveform, and exported it into each and every format the oscilloscope can write.

Some of the files were binary, with sizes ranging from large (1MB) to very large(10MB). Export of full data into a csv made me play with a whopping 111 MB csv file. And it looks like it contains only one channel, I want two.

The best way so far is to export displayed data. It puts out reasonable amount of data, but with lots of noise. So the only thing I have to do to plot them is to boxcar-average it. The results follow.

This time, the L/I curve was measured much slower, with total length of the light pulse of about half a second. As you can see, the behavior isn't much different form the previous one, acquired in about 10 ms.

• ### photodiode transducer

As simple as it possibly can be. A big 4700uF 25V cap (instead of a battery - lasts days!), a photodiode, a 100k resistor, and that's it! Held together by a piece of scrap PLA 3d-print.

• ### Weird noisy intensity-vs-current curve

I've set up a rig to measure intensity vs current relationship for the diode.

Why? Because I'm expecting to see a saturating curve similar to those of LEDs. From that, I can guess max drive current - the one of LEDs is often around the point where the drop becomes apparent, but not seriously close to saturation.

I assume, that if I do pulse testing, I can probably seriously overdrive the laser diode, without causing serious damage. So off we go.

I've set up a simple circuit, where a capacitor (47uF) is discharged starting from 12 V through a 20-150 ohm resistor (adjustable) in series with the laser diode. This is done periodically, with an adjustable frequency of about 5 Hz.

The current and brightness waveforms are captured by 2-ch oscilloscope, and from these captured curves I can plot intensity vs current characteristic of the diode.

Once this was set up, it didn't take long to measure the curve.

This is what I get on the scope. Note that noisy part in the intensity plot (yellow) at highest currents! That is not noise! It perfectly reproduces at each pulse, and it is tied to diode current (if I adjust it, the noisy patch moves, and completely disappears if current does not exceed 300 mA). Interesing.

What's going on? I have no idea, but I'm sure it is the laser! Well, at least, I certainly excluded photodiode transducer, and the current waveform is also clean, so it's not something in the low-ohmic trimpot...

If you have any ideas, on why laser diode intensity vs current curve can have irregularities, or if anything else can be happening due to my setup... let me know in the comments!

Another thing to note is that the intensity drops to almost zero abruptly at about 50 mA. This, I think, is due to lasing threshold. It still glows at lower currents, but that can only be seen in the dark.

Finally, here's the intensity-vs-current plot!

Share

## Discussions

lantramnamcodon wrote 03/05/2018 at 17:25 point

i thank you for the knowledge,i feel angry because i brought a 200mw laser diode in a electronic component store,when the tester worker tested my laser,it was very hot,when i bring it home,the laser just can burn slightly,i accidentally broke the diode with 12v power supply.However i still feel that they did "destroy" my laser without responsibility.

Are you sure? yes | no

Keegan Reilly wrote 05/05/2016 at 15:53 point

This is late but congrats on the blog writeup!

Are you sure? yes | no

MECHANICUS wrote 01/05/2016 at 23:03 point

Wow you are such a freaking brainiac! Looks like you know what you are doing anyway.

Are you sure? yes | no

esot.eric wrote 01/05/2016 at 14:13 point

Awesome! Finally some difinitive way to test!

Are you sure? yes | no

esot.eric wrote 04/20/2016 at 08:00 point

define: difinitive... haha

Congrats on the blog-writeup, yo!

Are you sure? yes | no

Brian wrote 01/05/2016 at 00:56 point

In the laser diode business, we call the maximum current the "Kink current". It's the point where increasing the current through the diode does not increase it's optical output, as long as you hold the temperature constant. It will show up on the I-P(optical) curve as a roll off in output power. The optical power will start to decrease if you increase the current further.
For maximum diode life, always maintain the diode temperature below 30 degrees C, and do not operate within 10% of the kink current for long periods.

Are you sure? yes | no

DeepSOIC wrote 01/05/2016 at 01:27 point

Thanks for interesting comment. But that leaves a few questions.

"It's the point where increasing the current through the diode does not increase it's optical output, as long as you hold the temperature constant" At DC, or pulsed? (Basically the same question put in different terms:) keeping temperature of what? of junction, or of case?

"and do not operate within 10% of the kink current for long periods." does that mean I can fire the diode to up to 10% of kink current, or up to 90% of kink current? And, how long is long? Does it mean I can safely pulse the diode beyond kink current, provided the duration of pulse is small?

Are you sure? yes | no

Brian wrote 01/05/2016 at 01:37 point

At DC. Pulsed operation (with ns long pulses) will often allow you to run up to 3 times the current through the diode, with a corresponding increase in output power.

Temperature of the case. You have no direct control of junction temperature other than controlling the temperature of the case.

Within 10% of the kink current (more than 90% of the rated maximum current). So if the diode has a kink rating of 1A, don't run higher than 900mA for more than 1 second.

Pulsing the current allows the junction to cool between pulses and allows higher peak powers.

Poor control of the current pulses will result in ringing or current peaks that are too high and destroy the laser diode. Negative going currents due to ringing will destroy the laser diode.

I highly recommend these - http://www.lasorb.com/lasorb-overview/

The wiggly area in your graph above 0.3A is common as you are approaching the kink current, but can also be caused by changes in junction temperature, poor current regulation, or an issue with the measurement setup.

Are you sure? yes | no

flashcactus wrote 04/20/2016 at 12:56 point

hmmm. I think I've read somewhere that what really kills the diodes is not the current or temperature, but breakdown of optical surfaces due to too high radiation levels and that it also happens instantly and permanently, so the critical breakdown current is not to be exceeded ever, for any duration, however short. Moreover, that current varies between diodes and in different operating conditions; as such, the only drivers that would not damage the diode are those which have their feedback hooked to the laser's optical output.

Given all that, I don't understand how you could consistently drive the laser at several times its maximum current without any problems. What have I gotten wrong here?

Are you sure? yes | no

DeepSOIC wrote 04/22/2016 at 22:24 point

You are not the first one who says that here either. That there are two modes of failure: heat failure, and mirror failure.

I can guess, that for mirror, it is similar to a breakdown of a good insulator. Insulators can sustain very high voltage (E-field), but at the point it breaks down, the damage is quick and permanent.

I don't know, what the mirrors are made or in this laser, but I think they might be the faces of semiconductor crystal the lasing medium is made of. Semiconductors are much less prone to instantaneous breakage upon breakdown. For example, you can easily cause breakdown of a silicon diode, and if the current isn't allowed to grow too much, the diode isn't damaged. Same may apply here: the semiconductor breaks down, that causes light absorption, which stops lasing positive feedback. This can actually explain the oscillation of light output that I see!

Are you sure? yes | no

RoGeorge wrote 01/05/2016 at 00:47 point

Does your DVD-RW LASER diode have 2 or 3 pins?

Usually, for high power LASER diodes a monitor photodiode is integrated into the LASER diode package.

Driving with constant current and ignoring the optical feedback might either destroy the LASER or drive it at sub-optimal power. Junction temperature may vary widely during use, so optimal current also vary. It is almost impossible to drive reliably a LASER diode at nominal power without optical feedback. For more details, you may want to look for dedicated ICs LASER diode drivers or other pages like this one: http://www.teamwavelength.com/info/laserdiodedrivers.php

Are you sure? yes | no

DeepSOIC wrote 01/05/2016 at 01:07 point

It has three pins, but only two of them were soldered. I diode-probed the third connection to see if there is a photodiode, but it looks like it's open.

There is a special chip on this assembly, that looks like an optical feedback thingy. It is under a diffuse semitransparent coating, directly in line with the laser diode, behind the mirror.

Are you sure? yes | no

RandyKC wrote 03/03/2016 at 18:51 point

Shine some light on it and see if the resistance changes would be my suggestion. (Lens could also be a filter)

Im wanting to do some laser etching using an existing cnc.

Are you sure? yes | no

## Similar Projects

Project Owner Contributor

### Open Source Laser Power Meter

B. A. Bryce

Project Owner Contributor

### CD/DVD mechanisms and cartesian thinggie[s?]

esot.eric

Project Owner Contributor

### BluBEAM - a scanning laser microscope

andreas.betz

Project Owner Contributor

Ted Yapo

# Does this project spark your interest?

Become a member to follow this project and never miss any updates