fast analogue modulation

Hi, I saw that DPSS lasers are more difficult to drive, why?
I need to make a fast analog modulation (the best would be 65KHz, but even 30KHz would be ok). There is any way? I have no problem with electronics, so neither with analog amplifiers. I am able to produce a low output impedance amplifier capable of high current (class D).

Please help me if you can :thanks:

You might want to search the forum for a product called a FlexmodN2

Here is a pic of the product:


And some output/input graphs

High speed linear modulation:


High speed modulation with no overshoot:


They are available here
http://hacylon.case.edu/ebay/laser_diode/FlexMod01.php

PM drlava for more details on the product or download the user manual here
http://hacylon.case.edu/ebay/laser_diode/FlexMod2 Manual.pdf

EDIT: You may also want to look into the Laserboy correction amp... If you haven't already got a DAC for your setup.

Hope that helped! -Adrian

Bob, no matter how well you regulate the current of the pump diode, there are too many factors to accurately modulate the Nd laser at as high a rep rate.

Thank for this thread. I like it it is really helpful to me.

The issues preventing DPSS lasers from being modulated at very high frequencies have to do with the processes involved in DPSS. It takes a small amount of time (a few ms) for the output to stabilize after the initial burst of pump energy from the diode, so as the modulation rate increases, the output power decreases dramatically until there is no usable output. As far as determining just how fast your specific DPSS laser will modulate, experimentation is the only way to find that out.

Few milliseconds?
I have to get much more faster. I know there is a way as 1080p projectors exist.
Does someone know any source of fast green lasers? I need only 5mW, or even less...
Maybe such a low power laser can be fast regulated?

Why it's so slow? What if I keep temperature constant (e.g. with a peltier cell), I plot the output vs input chart and I digitally correct the output?
I would need something like a thermometer, a fast 8bit ADC/DAC, a photodiode, a microcontroller, and a peltier cell. But if it isn't temperature related I don't know what to do...

Bob, no matter how well you regulate the current of the pump diode, there are too many factors to accurately modulate the Nd laser at as high a rep rate.

Click to expand...

I see... The OP sounded like he wanted to run a laser on his set up, but hasn't made his mind up yet and on a side note was asking about the DPSS lasers.

Few milliseconds?
I have to get much more faster. I know there is a way as 1080p projectors exist.
Does someone know any source of fast green lasers? I need only 5mW, or even less...
Maybe such a low power laser can be fast regulated?

Click to expand...

May I ask what this is going to be used for? Would you be able to use a different colour instead? Correct me if I'm incorrect but you will not experience this much trouble if you stuck with a laser diode. It might be your only choice....

Hope that helped -Adrian

kingbowser said:

Few milliseconds?
I have to get much more faster. I know there is a way as 1080p projectors exist.
Does someone know any source of fast green lasers? I need only 5mW, or even less...
Maybe such a low power laser can be fast regulated?

Why it's so slow? What if I keep temperature constant (e.g. with a peltier cell), I plot the output vs input chart and I digitally correct the output?
I would need something like a thermometer, a fast 8bit ADC/DAC, a photodiode, a microcontroller, and a peltier cell. But if it isn't temperature related I don't know what to do...

Click to expand...

Laser diodes by themselves can be modulated into the MHz with no problem.

It's not that the laser doesn't emit any light during the first few ms, it's just that the output unstable in terms of power output and beam shape.

The issue is thermal lensing in the laser crystals. When the pump diode's energy is absorbed into the crystal, it causes the crystal to change shape minutely due to expansion from the heat. When you run a DPSS laser CW it overcomes this initial instability very quickly and the output becomes more stable and tends to stay that way, becoming more and more stable as the laser warms up. With rapid pulses of pump energy, this thermal lensing happens over and over, and as the pulses become shorter it gives the crystals less time to stabilize. To the best of my knowledge there is no fix for this. Even the most advanced, stable DPSS lasers in the world do not respond well to direct modulation.

It is possible to modulate DPSS lasers relatively fast, so I wouldn't count them out until you've proven that yours won't do what you want it to by actual testing.

Thank you for your help

Are youo thinking about using a LUT for modulation? This was my idea long time ago, but never got the time to try that. This way is may be possible to get past the mode hops and output breakins a PID regulator would hang up on.

Or you could use a AOM...

It's gotta be less than a few milliseconds, because most (all?) analog DPSS can get at least 10Khz. I've seen many advertising 30Khz.

What is it for?

They can be modulated that fast, yes, but the power output never comes anywhere near what it should be. It's not that the laser produces no output for the first few ms, it's that the output is unstable during that time. The faster the modulation, the dimmer and more unstable the output. This can usually be demonstrated by connecting a DPSS laser (with a driver that will support high-frequency modulation) to a function generator and applying a signal with increasing frequency at a 50% duty cycle until the system reaches it's maximum rate. The output will drop steadily as the upper limits are approached.

I'd recommend using an external modulator (AOM or EOM) in order to maintain power, stability and beam shape while modulating at rates in excess of 20-30 kHz. Trying to do this internally will only cause problems.

ND:VANDATE has a upper laser level storage time of 800 or more microseconds.. So if you have a blanked DPSS laser, you need to slam in a lot of pump diode to get past the threshold of lasing. As you blank, things go the other way and the 800 uS storage time takes a while to bleed off. This is the upper limit on speed for a SMALL doubled yag laser. Larger lasers with bulk crystals are even slower in CW mode.

The lasing threshold varies all over the place based on the immediate past history of the crystal set and the thermal history of the heatsinking...

Take your left hand, spread your fingers apart and hold in it front of you. This is the output spectrum of a low cost 808 nm pump laser in the dpss, its about 2-3 nanometers wide and has several spikes across its wavelength spectrum. Now factor in that a .1' shift in diode core temp moves the wavelengths up or down about 1 nanometer. The adsorption band for Nd:YAG or ND:VANDATE is also a series of spikes of max adsorption next to gaps of much less adsorption. It is also temperature driven, but not as bad as the diode is. Spread your fingers on your right hand to represent your pump adsorption. Now slide your hands across each other, and see that as you move your hands, there are places that sometimes have a good pump and other times do not. Since the pump diode is so tiny, its very hard (well, pretty much impossible) to have the huge mass of the pump diode chiller react to the tiny cavity size changes from the pump current that shift the wavelength.

The change in cavity length changes the resonant frequency of the pump diodes cavity and moves it across its gain curve... The gain bandwidth of a diode is large, but follows a bell curve, so as you move towards its edges the power can really fall off..

If you want to see a example of this, watch a high power green dpss pointer "ramp up" to power, or "ramp down" right after you hit the button. Or place your pointer in ziplock bag to prevent condensation and toss it in a freezer for about 5 minutes, then hit the button. You'll see all sorts of effects as the pointer trys to make it back to the room temperature equilibrium it was tuned at. Pump diodes are graded by wavelength at the factory under steady state heatsinking.. Often times the diodes that are not near the +/- 5 nanometer wide adsorption band in yag show up on ebay, and are real lemons if you buy them to repair a laser

Professional high wattage DPSS display systems and laser TVs use a EO or AO modulator for this reason.

Modern Laser TVs use vertical cavity surface emitting lasers (vecsels) or similar tiny lasers, in arrays , the vecsels are small enough to quickly reach thermal equilibrium, but are very limited in power. A vecsel is so small that the doubling crystal and host laser crsytal can be vacuum deposited on the laser at the wafer level.. The down side of the is a green laser for a TV emits a 1 or 2 cm square beam with the quality of a flashlight and a huge divergence..

DPSS 473 nm blue is even worse, the temperature of the whole assembly often has to be held to .01 degree C or less for peak power...

Novasel laser arrays would be pretty much useless for a pointer or classical vector laser scanning.

See:

http://www.photonics.com/Article.aspx?AID=32111

Steve