Laser power control - help me design a servo attenuator / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Laser power control - help me design a servo attenuator

1 2 3 next

Move Thread

LAN_403

hboy007

Sat Feb 18 2012, 10:39PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

I'm currently working with Nd:YAG and Nd:YAG pumped Ti:Sapphire lasers. I use them to optically excite quantum dot samples. My current goal is to perform pulsed and CW power dependent measurements. Unfortunately, the 532nm pump power is only stable to within a few percent and fluctuates on a timescale of 1-10 seconds and the CW output of the Ti:Sa is even worse (make it 20-95%).

Therefore, I'd like to build a motorized attenuator that autonomously controls the output power to within <1% around a safely reachable level. Perhaps someone can look over the parts selection I've already made? That would really save me some time. There's nothing like getting it right the first time.

The idea is to insert a linear graded attenuator and two glass plates (single side wideband AR coated) into the beam. All elements will be tilted at a small angle to compensate for beam shifts (the reflection of the attenuator won't be usable for monitoring so both glass plates are tilted in the opposite direction to the attenuator).

Here are my ideas:
* use a reflective graded ND filter 10x50mm

* I've obtained a Minebea 14PM-M201

stepping motor (35x35x20mm, rated 12V, 1.8Â°/step)
* L6219

will be used in conjuntion with the motor, allowing easy microstepping
* an ATMega16

will do for a main controller
* an FT232

will provide easy connectivity for configuration via UART

As for the intensity measurement before and after the attenuator,
* two BPW34

photodiodes with frosted glass attenuators (frosted microscope slides!

) in front of them will be used.

The PDs will be used in low-noise, low speed photovoltaic mode, so the opamp choice for the PD front end won't be critical.
* MCP6S22

seems to be a good choice to implement gain control for the ADC since the microcontroller ADC also multiplexed. Wouldn't it be good to sample both PDs at once with a sample&hold circuit, then digitize both channels sequentially? I would end up with two MCP6S21 and one HEF4066 quad bilateral switch. Add some jFET opamps like TLC084

for buffering.

The whole system will be designed to fit the Linos microbank system (or Thorlabs cage system, nevermind).

Linear guide rails for the filter are a thing I'll have to figure out how to implement in due time. I like the way the engineers solved this problem with chrome steel rods and brass bushings in optical drives (just google KEM-400AAA for a neat example) ...

Sorry if the above was a little incoherent, I just wrote it down in one go. All comments are welcome!

radiotech

Sat Feb 18 2012, 10:50PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Would a Pokels cell be of use ?

hboy007

Sat Feb 18 2012, 11:00PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

Pockels cell intensity modulators like

do exist and we also have Pockels cells. Since funds are limited and the power control feature is regarded as a nice-to-have, I'm not sure if this nice solution is within reach. The pockels cells we have will most likely be used in other parts of the setup (e.g. for time slicing for time-resolved measurements,

or in other experiments.

Steve Conner

Sun Feb 19 2012, 08:12AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

In situations like this I prefer to take a step back and look at the fundamental problem. Why is the pump power unstable in the first place? Does the instability have higher frequency components that a motor would be too slow to correct?

To take an example, the pump laser could possibly be stabilised with feedback from a photodiode into the driver. Maybe it already has that feature but it's broken :) Or maybe not so easy if it's an ancient Nd:Yag pumped by filament lamps.

And to take another example, some lasers naturally emit in the form of chaotic bursts of light, maybe the Ti:Sapphire is one of them when you try to run it CW?

Shrad

Sun Feb 19 2012, 09:02AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

this should be stable

the problem seems to be your pump laser

what is the type, brand and age?

you should stabilize the pump power and quality

do you have a vibration free beamtable? a Ti:Sapphire laser is very sensitive to vibrations if it is in a Z-fold cavity

also, what are power levels in your setup? 532nm pump power should exceed laser trashold of the Ti:Sapphire by some extend to get decent stability

it is also extremely sensible to temperature drifts, even the smallest, so you should use a liquid cooled mount for the Ti:Sapphire along with a nice chiller (needs a decent regulation) and you need to put your sapphire cavity inside a boxed beamtable with no or near no water vapor (an aluminium box with windows and a plexiglass cover, filled with CO2, will do)

hboy007

Sun Feb 19 2012, 12:15PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

Allow me to give you a little more information about the setup:

The main laser is a Coherent Verdi V10

"Verdi lasers are â€œsmartâ€ in that they have feedback
loops which can automatically adjust the diode drive
current to maintain the power at a constant value.
Monitoring the drive current is one way to track the
overall health and efficiency of the laser. "

There are indeed some power levels where the output is smooth over time. I haven't characterized the behaviour of the internal feedback loop at all power levels but the laser should run stable at as low as 100mW.
I also operated it at 1.5W with roughly proportional power fluctuations.

When pumping the Ti:Sa

, we use around 8W @ 532nm, producing 1.5-2W QCW in the 800-850nm region.

The whole setup is bolted down onto an actively damped optical table, both lasers are cooled in series by a Thermo Scientific Accel 250 chiller. There used to be another chiller that was installed for the last couple of years with some crumbs and blue-greenish residue in it. Clogging of the lasers did occur and I must assume that my colleagues did a proper cleaning in the course of this.

There won't be a way to purge the Ti:Sa housing with CO2 since the lab does not provide proper ventilation to ensure safety in this case.

Carbon_Rod

Sun Feb 19 2012, 11:26PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

The TEC units can still cause stability issues depending on the age of the laser.
The excitation cavity monitor diode method may not completely stabilize the DPSS output.

DVDR drives typically use a miniature Liquid Crystal Variable Retarder with a closed-loop optical power sensor on the output (or a splitter):

Cheers,

hboy007

Sun Feb 19 2012, 11:59PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

LC retarders would be a fine option and I recall we also have one or two of those somewhere. Beam splitter and polarization optics won't be a problem - but can the LC units handle 200-500 W / cmÂ² ?
Something for the "ideas thread": make a variable laser attenuator out of active shutter glasses with the filter removed

Carbon_Rod

Mon Feb 20 2012, 08:12AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Higher power setups usually already have one optical isolator to protect the driver.
I'm not sure if these would even put up with the output levels if you are still exceeding 500 W/cm^2 over a 20cm^2 approximate area ( 2" LC unit). Most LC units have a lower damage threshold of 500 W/cm^2 in CW mode.

After filtering the DPSS output for the wavelength you are after, the power level will usually be around 60% - 80% of the pump laser under ideal circumstances.

At $2k each, it would be advisable to double-check your exact part number for safe operating area ratings.

Proud Mary

Mon Feb 20 2012, 12:23PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Could the exposure control system of a good DSLR be of any use?

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob