Very High Speed (500 MHz) Laser Diode Driver

[someone153]

Hey everyone!

I have been doing some research for the past few weeks but have come up empty. I am looking to either build or buy a high speed (500 MHz) laser diode driver for a 445 nm laser diode.

I have seen that using a bias tee is probably the easiest method, but I also want to take advantage of the power of these diodes. I can't seem to find one that will work around 500 MHz and allows for a modulation current of 500 mA.

If anyone could help point me in the right direction, I would greatly appreciate it.

Thank you!

 

[Cyparagon]

Why do you need that speed?

It's done in bluray drives, so it's possible, but probably not with a multimode diode like you want.

 

[someone153]

I am looking to test the limits of the diodes that I have. I intend to use them for data transmission so the faster the better. I guess I figured that 500 MHz would be more than enough so I just wanted to have enough room to work with.

 

[shokkunlasers]

why would you use multimode 445nm diodes for data transmission?

 

[someone153]

why would you use multimode 445nm diodes for data transmission?

I have my reasons...

I'll just say that I need powerful blue lasers and the 445's fit the bill.


EDIT: I'll link this paper too!
http://dspace.mit.edu/openaccess-disseminate/1721.1/59310

 

[rhd]

This is outside my area of expertise, but I would think a multimode 445 is exactly the wrong type of laser to use.

I would think higher (red) wavelengths would be better suited to data transmission in the air. They would presumably travel farther / be less impacted by the atmosphere.

Then I would also presume that, all else being equal, a single mode diode would be better than a multimode diode, because of the better divergence and the higher power density at a distance.

Putting that together, and assuming use of hobby-level (inexpensive) diodes, I'd think you'd want to use a $5 single mode 650nm LPC-8XX diode at 300 mW, before turning to a 1W+ 445 multimode.

But I may be wrong about some of the assumptions above.

 

[Sigurthr]

Could DIY it by rectifying the output from a 70cm (440MHz) transmitter and using that as the drive signal for a very high speed transistor. Layout would be a bitch. Would provide easy FM modulation since you can just use the audio input line on the Tx.

 

[someone153]

This is outside my area of expertise, but I would think a multimode 445 is exactly the wrong type of laser to use.

I would think higher (red) wavelengths would be better suited to data transmission in the air. They would presumably travel farther / be less impacted by the atmosphere.

Then I would also presume that, all else being equal, a single mode diode would be better than a multimode diode, because of the better divergence and the higher power density at a distance.

Putting that together, and assuming use of hobby-level (inexpensive) diodes, I'd think you'd want to use a $5 single mode 650nm LPC-8XX diode at 300 mW, before turning to a 1W+ 445 multimode.

But I may be wrong about some of the assumptions above.

Your thinking is absolutely correct, but I do want to use a blue source. Maybe I should look for a single mode blue diode as the beam quality will be higher... Thanks! That's something I will look in to.

Could DIY it by rectifying the output from a 70cm (440MHz) transmitter and using that as the drive signal for a very high speed transistor. Layout would be a bitch. Would provide easy FM modulation since you can just use the audio input line on the Tx.

A single transistor! That's so simple that it might just work! All I would need would be a high speed transistor (as I have an 500 MHz function generator). Any recommendations on a transistor? Why do you think layout will be difficult?

 

[Sigurthr]

Well, if you use a fixed series resistance from a CV supply you'll get a constant current out of it. By using FM instead of AM modulation the thermal drift effects on amplitude won't be of concern. You'd need a temperature controlled resistor and some kind of optical feedback network to use it on AM though to counter for drifts.

Layout will be difficult because even though you're talking DC you'll have induced voltage spikes from switching half an amp at 500MHz from parasitic inductances in the layout. Parasitic capacitances will round off corners (again, not much of an issue for FM). The net effect of the two can potentially result in unwanted oscillations. You'll also need to decouple the supply sufficiently. It is by no means simple, but it is potentially inexpensive.

At 500MHz you're well into the advanced RF techniques when it comes to circuit design.

No recommendation on discretes, I work at 2MHz and lower, sorry!

 

[phenol]

PD84002
you will need a transmission line to the diode like they do it in dvd/bluray sleds. good luck with that.

 

[Tomorrows System]

I have my reasons...

I'll just say that I need powerful blue lasers and the 445's fit the bill.


EDIT: I'll link this paper too!
http://dspace.mit.edu/openaccess-disseminate/1721.1/59310

I was doing underwater project and 445nm diodes are best for it because of the water transmittance. But I needed 100kHz only.

My fastest driver for now on is around 2MHz but after some test with 445nm diodes I got results that after 300kHz the power is dropping, it is too fast for them, even if current is flowing. I can try to reactivate the tests.