Are there any rules of thumb for pulsed performance of CW diodes?

[Rocket]

I have a driver that can output 50A 5ns 20kHz pulses and want to use it to power a blue diode.

However, I have a few questions which are hopefully cheaper and faster to ask rather than determine experimentally!

1. Is there any danger of facet damage from these short pulses?

2. Can I extrapolate the peak power from the W/A charts in datasheets? Does it taper off between the ~1A region and the 50A region?

3. Is there any point getting a multi-Watt diode for this or will I be better off with a ~100mW diode? Maybe even a single-mode?

4. Are there any blue diodes particularly suited to ns pulsing?

5. What am I not considering that is really important...? (apart from a heat sink)

 

[diachi]

Just because you're pulsing it doesn't mean you can dump 50A into it. You can increase the current for sure, but you need to consider duty cycle and what the diode is rated for. Datasheets will typically have a pulsed rating in them. They'll also give you a slope efficiency graph/figure that you can use to figure out current vs power, which would tell you the peak power for a given pulse.

Pulsing a CW diode doesn't mean you'll get massive peak powers like you do with a Q-switched laser. If your laser does 1W at 1A CW and then you pulse it at 1A, your peak power will still be ~1W.

 

[Rocket]

Sure, thanks. That's understood but, for example, the datasheet for a PLT5 450B https://media.osram.info/media/resource/hires/osram-dam-2495842/PLT5+450B.pdf doesn't appear to have any information that tells me how big a 5ns pulse with a 0.1% duty cycle the diode could be expected to cope with. Is the information there but I'm missing it?

 

[diachi]

Sure, thanks. That's understood but, for example, the datasheet for a PLT5 450B https://media.osram.info/media/resource/hires/osram-dam-2495842/PLT5+450B.pdf doesn't appear to have any information that tells me how big a 5ns pulse with a 0.1% duty cycle the diode could be expected to cope with. Is the information there but I'm missing it?

It may be that they never intended it to be used for pulsed applications, thus didn't include specifications for that.

You'd either need to test it yourself or contact them to see if they have that data.

Why would you need a 5ns pulse at a 0.1% duty cycle...?

 

[Rocket]

Well the datasheet says "Efficient radiation source for cw and pulsed operation" so its clearly ok for pulsed operation

The laser is for a portable underwater lidar system. Hence the desired short pulses. The low duty cycle is due to that being the maximum duty cycle that the driver can maintain.

What will be the failure mechanism be for too-high-current pulse?

 

[diachi]

Well the datasheet says "Efficient radiation source for cw and pulsed operation" so its clearly ok for pulsed operation

The laser is for a portable underwater lidar system. Hence the desired short pulses. The low duty cycle is due to that being the maximum duty cycle that the driver can maintain.

What will be the failure mechanism be for too-high-current pulse?

Ooh, interesting application, some interesting challenges there I'd imagine.

There are various failure mechanisms that could occur, catastrophic optical damage, failure of the bond wires or failure of the semi-conductor itself are the ones I can think of.

I'd contact Osram or other suppliers and see what they offer that would be suitable for the task or if they have pulse operation data for other diodes you've looked at. They should have some experts that can help you.

There are pulsed diodes available that are designed to handle super short pulses at high currents, although I've only seen them at IR wavelengths.

 

[Alaskan]

All of the visible wavelength laser diodes I’ve seen are specified to do 2X the CW rating whether extremely short pulse or not. seen some IR laser diode arrays meant for LIDAR rated to as high as 50 kilowatt peak power in a short pulse but new they cost 2-3 thousand each when ordered in lots of 10.

 

[diachi]

All of the visible wavelength laser diodes I’ve seen are specified to do 2X the CW rating whether extremely short pulse or not. seen some IR laser diode arrays meant for LIDAR rated to as high as 50 kilowatt peak power in a short pulse but new they cost 2-3 thousand each when ordered in lots of 10.

Yup, that sounds about right as far as CW diodes go. That said, you can probably push them beyond that. We're getting over double the output power specified on some diodes we use. Of course, doing so sacrifices longevity.

 

[Rocket]

IR sadly no use unless you know of a magical doubling crystal

Would it be reasonable to try and push a nub to 15W then?

Do you think I could get much more if I was ok with a MTTF measured in the 100s of hours?

Never satisfied!

 

[Rocket]

Oops cross-posted.

It sounds like I just need to bite the bullet and break a few diodes in the name of research...

 

[Alaskan]

I've looked into this same thing awhile back, no solution I could find to get high pulsed power in the visible spectrum up into the kilowatt range, or higher, unless you use something like a 1064 nm ND:YAG and LBO crystal to double the wavelength to green. I want to do that myself, buying parts now. YouTube has a few videos posted from individuals who have done that.

 

[diachi]

Oops cross-posted.

It sounds like I just need to bite the bullet and break a few diodes in the name of research...

You could try the NUBM55 at 15W pulsed, I've seen them pushed to >9W CW. Might work long enough to be reasonable.

Keep in mind those put out a ton of waste heat, although being in water makes cooling them easier. <<< IGNORE THAT. Forgetting you're going for 0.1% duty cycle, should be no issue.

Beam specs are terrible on those too.

As always, use proper laser safety equipment (laser safety glasses, interlocks etc).

 

[Alaskan]

Finding a driver for extremely short pulses is the next challenge, if such a diode can produce them.

 

[Rocket]

Can definitely keep them cooled considering we are in the heat sink of the ocean

Is heat really going to be a problem on such a low duty cycle?


I have zero experience with high power diodes...(the current laser is a class 4 though and I’ve managed not to blind myself or my dog yet!) is it challenging to collimate the NUBM55 to a couple of mrad? Got an alternative diode to recommend with a better beam if so?

Thanks for the advice!

 

[Rocket]

Already have the driver specced - PCO7114-50-4 from directed energy.

 

[Alaskan]

Regarding your question about divergence, if you don't need a tiny thin beam, the divergence of any laser diode can be reduced by simply using a larger diameter longer focal length collimation lens, the divergence halved each time you double the beams diameter.

Sharing this last post for others who might be considering high power pulsed infrared diodes:

Directed Energy is one of the companies I was looking at for a high current short pulse driver too, found one for 900 dollars which would work for the kilowatt power output rated diode array I was looking into (not 50 KW as I wrote earlier, I guess I was thinking of a small YAG), found two of those high peak power IR diodes on ebay, bought them and then a couple of days later the seller said he could'nt find them to ship (right after I had told a well known surplus laser parts seller I bought them on ebay and asked a question about their specs, since i knew he wanted some of those too, may have been a bad move.... hummmm). Then I called LaserTel, the manufacturer of the diodes in Arizona and found they cost like 2700 dollars each in lots of 10, a bit more than the 65 dollars each I paid for the pair on ebay I never received.

Here is a link to the LaserTel web page showing the type of diodes I bought on ebay which got away: Conductively Cooled Laser Diode Arrays - S7


best image hosting site