At the moment, I am planning for a project involving lasers (yes, I have been lectured on laser safety). I am looking for a cheap way to build a pulsed laser in the near-infrared range. Since laser diodes are typically relatively cheap, I thought that it would do well in satisfying this requirement of my project. I managed to find a relatively cheap laser diode of appropriate wavelength, but it is not pulsed. Is it possible to just take a normal (continuous) laser diode and pulse it (by building an appropriate pulsed laser driver, or something)? Or does this not produce the same effect as pulsed laser diodes? If this is indeed practically feasible, then how can one go about doing this (links to resources/information are appreciated)?

  • \$\begingroup\$ Link to the data sheet of the device. \$\endgroup\$
    – Andy aka
    Apr 30, 2020 at 8:42
  • \$\begingroup\$ @Andyaka I added a hyperlink, but I'm really asking generally. \$\endgroup\$ Apr 30, 2020 at 8:45
  • 1
    \$\begingroup\$ Well, it's a thermal management problem; you can overdrive it and it will almost certainly work, but by how much can you do this before it destroys the die? You either buy a batch of them and characterise this yourself, or find one where the manufacturer has already done this. \$\endgroup\$
    – pjc50
    Apr 30, 2020 at 9:01
  • 1
    \$\begingroup\$ @Andyaka Oops, I confused pjc50's comment with yours. My apologies. I meant "short bursts of very high power", which is the purpose of pulsed lasers in a photonics context. \$\endgroup\$ Apr 30, 2020 at 9:02
  • 2
    \$\begingroup\$ I think it would be a lot more effective to give specs you are aiming for (repetition rate, pulse energy, duty cycle) and your application (ablation, time of flight, etc) and then ask about how to build a driver for that specification. Those specs are going to change the solution from being trivial to expensive to completely impossible. \$\endgroup\$ Apr 30, 2020 at 16:10

1 Answer 1


Often, a pulsed laser diode is just one that is not very efficient, so that if it were operated continuously (we say "continuous wave" or "CW") it would overheat and destroy itself. Most often the first lasers designed of any particular type (including the first laser ever, a pulsed ruby laser) are pulsed, and only later are more efficient ones capable of CW designed. This includes laser diodes.

So you can certainly take any laser diode and operate it pulsed.

Doing it practically will have a few difficulties.

  • If you want to achieve very short pulses, and the laser isn't designed for that (very low capacitance, mainly, when we talk about laser diodes), then you might not be able to get the pulse width you want with any particular laser.

  • If the laser isn't characterized for pulsed operation, you won't know how strong a pulse (what pulse energy) you can drive without damaging the device or reducing its lifetime.

  • If you want a very particular laser wavelength, you might find the laser produces a slightly different one when operated pulsed than it does when operated CW.

Also, if you want pulses in the 1-100 ns range, and not too much peak power, and are willing to do some experimentation (and destroy a few lasers) to find how hard you can drive your laser without damage, then you will probably be able to do it. If you want sub-nanosecond pulses (and particularly if you want 10 ps or less), you'll probably want to choose a different laser type with a gain-switching or mode-locking mechanism and a longer cavity to achieve that. Diode-pumped fiber ring lasers seem to be quite popular for producing sub-picosecond pulses, for example, depending on the desired wavelength.

  • \$\begingroup\$ The first part of your last paragraph describes my requirements precisely! I’m looking for 1-100 ns pulse width, low peak power (for biomedical applications, and require skin/eye safety), and wavelength will be in the NIR range. I’m thinking of getting some cheap 800nm diodes and using them for experimentation, and then moving on to the higher wavelength, more expensive diodes when I have a better idea of how to implement the idea. Since you seem to have a good idea of what I’m interested in, do you have any resources/suggestions/advice/ect.? \$\endgroup\$ Apr 30, 2020 at 15:20
  • \$\begingroup\$ (In particular, how to go about converting a continuous laser diode into a pulsed one, related mathematics for doing calculations of the characteristics of the new laser, etc.? Textbook references would be fine, if you know of any that have sections focusing on this.) \$\endgroup\$ Apr 30, 2020 at 15:25
  • \$\begingroup\$ It shouldn't be too hard to get a 10 ns or longer pulse out of just about any low-power NIR laser just using a pulsed current source to drive it. Getting shorter pulses will depend on the exact laser chosen. The only other thing is if this product needs to last more than maybe 100 hours, you need to take design for reliability seriously. In my work (telecommunications) we typically use over 100 lasers for reliability studies for any new product, for example leaving them running for months at elevated temperatures, to prove out the reliability. \$\endgroup\$
    – The Photon
    Apr 30, 2020 at 15:30
  • \$\begingroup\$ My original concern was that continuous laser diodes cannot be converted to pulsed lasers, since pulsed lasers require some mechanism of accumulating energy in the cavity and then dumping it out, whereas continuous-wave lasers do not possess this capability. Outside of laser diodes, active or passive q-switching is used for this. So, just to clarify, are you saying that continuous laser diodes actually possess this capability to accumulate energy and dump it in pulses? I just want to be sure, because obviously this is different from generating pulses by “switching it on and off”. \$\endgroup\$ Apr 30, 2020 at 15:43
  • 1
    \$\begingroup\$ @The Pointer They don't accumulate any significant energy like a mode locked laser does. You're just turning them on and off really quickly. \$\endgroup\$ Apr 30, 2020 at 15:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.