User:Andy Maloney/Notebook/Lab Notebook of Andy Maloney/2009/06/08/Laser diode

Through many trials and errors, I believe I have finally gotten the laser to do what I want it to. Below are what I did both today and some on Friday. The goal was to get the laser through the fiber so that we can put a telescope at the end and run it directly to the microscope for tweezing. Or course, I'm using an ND 3.0 to do all this. That's why the beam "looks" dim in the images below.

First try
I tried to get the laser to go through a fiber using just the aspheric lens pair and the fiber. This did what I wanted it to, however, the power through the laser was not adequate. I used the 400 µm fiber to do this.

Second try
I thought that something fishy was going on so I tried using the 200 µm fiber, a 10x objective and the aspheric pair. Again, the power out was not adequate.

Third try
So I started to do things a bit more systematically with this try. I used: I first made a real image of the diode with the lenses in the far field. I then aligned the fiber by eye balling it and adjusted the fiber in x, y, and z until I maximized the power out. Again, this was not adequate. I also noted that there were lots of fluctuations with the power output.
 * 200 µm fiber
 * aspheric pair

Fourth try
I placed the lenses at their specified focal length from the LD and aligned the fiber as above. I got like 400 mW of power out but then something happened. The power dropped dramatically and fluctuated wildly. I took out the fiber and looked at the LD in the far field. I noted that there were crazy intensity fluctuations parallel to the LD. I decided to shut the whole thing off and let it cool/calm down. At this point in time I still had no clue what was going on.
 * 400 µm fiber
 * aspheric pair

While the LD was resting, I figured that it must be that I was creating an external cavity to the LD and it was messing with the LD power. This of course means that I must use the isolator before the fiber or else this problem will persist.

While we were in our meeting, I made the CCD camera take images of the LD every 60s. While away, the temperature increased from 18˚C to 20.7˚C. This irked me as I thought I was giving the TEC enough current to handle 1.5 hours worth of usage. Alas, I didn't so I bumped up the current to the TEC to 2.30A.

I noted that as the time progressed while the LD was on, the far field pattern shifted. Not sure if this has to do with the temp increasing or what but, it warrants further investigation.

The reason I wanted to take images of the LD over a time span is because I noticed that there were lots of fluctuations in its pattern. I have the data but I will wait till tomorrow to look at it. Oddly enough they are cause by vibrations.

Fifth try
Well, I just have to use the isolator. This means not only do I have to collimate the light out of the diode, I also have to send it through the isolator. Well, I got the LD to collimate after many different lens configurations. Most of which were my attempt to only use the aspheres we have. This proved to be fruitless due to the fact that the stages I am using do not have the resolution needed to put the aspheres in the proper spots to collimate the beam. So, the setup that did finally end up working was the aspheric lens pair and a 100 mm positive lens. I used the asphere pair to convert the NA from the LD to something small, in this case 0.16. This then allowed me to use a long focal length lens to collimate. Which, the translation stages I am using can handle.





After collimation, I had to reduce the size of the beam in order to get it through the isolator. To do this I used a 100 mm positive lens and a 35 mm negative lens. This reduced the collimated beam by 65% and again produced a (not so collimated but it doesn't matter) collimated beam. I say "not so collimated" because it isn't but, it is collimated pretty well through the isolator and that's all that matters.

The next stage was to get rid of the astigmatism in the beam (2 cylindrical lenses) and launch it into the fiber. As you can see, it works!



The fiber is nice because it gives me a circular beam and cleans up the modes somewhat because it acts as a spatial filter.

To do

 * I need to analyze the data I got from the CCD camera.
 * I need to optimize the placements of the cylindrical lenses so that I get rid of most of the astigmatism.
 * Optimize the setup to get the most power out.
 * Ensure there are no fluctuations coming out of the fiber, thanks to the isolator.