User:Anthony Salvagno/Notebook/Research/2010/08/09/More work with the QPD and the AOM

We are going to try and do several experiments today and hopefully I will get more than one on here.

AOM Deflection vs qpd position
We want to check out the amount of AOM deflection when voltage is increased vs qpd position. We will take data sets for voltage measured on qpd vs voltage of AOM (QPD voltage relates to a position), and at each position we will calibrate the detector to be at 0V.

Comments
I think in the future I will do quick comments in a google doc, since making new feeds and stuff in FF is way more annoying. Anyways here are my thoughts on this experiment.
 * We are noticing a lot of drift in the laser in the y-direction (which we have never experienced before) as well as drift in the x-direction (due to the AOM power ramping issues). We watched the laser for a while and noticed lots of little blurps in the position so we are going to remove the AOM completely and just look at the laser directly on the QPD. Then we will go back to this experiment.

Experiment 1
We have setup the laser to shine directly on the qpd at low power (100mW).
 * 0 min: x=0.05V; y=-0.05V
 * After a little while we are still seeing the blurps in the position that we have been noticing for a while. The y-position is definitely drifting up. The plus side is that without the AOM, the laser seems to be a little less noisy. The position band is about 0.05V from min to max compared to almost double that with the AOM in there.
 * 15 min: x=0.05V; y=0.1V
 * 30 min: x=0V; y=0.175V

Experiment 2
We put the laser power up to 1W to see what would happen:
 * as power increased, the position of the laser changed dramatically. As power increased the sensitivity to noise on the detector increased.
 * at 1 W the drift is more noticeable and is occurring at a faster rate.
 * We put another filter on the QPD and the detector seemed to handle the noise fluctuations a bit more. That makes no sense to me. By this I mean that the width of the band of each position was like 0.01V and was very pixelated, but after the filter the width of the band changed to 0.04V and looked more like what we are accustomed to.

Experiment 3
We are taking time data points to map the drift at different powers (.5W, 1W and 1.5W) for 30 min at each power

Polarizations
We have completely devolved. We started trying to solve question 1 of the task list from Thursday. We now have come across so many problems with the laser we think that we can't even use it to solve some problems so we are going to try and figure out how to fix the basic problems to get us through the other questions.

We started thinking about polarization to see if we could reduce/elliminate the drift from the laser or to transform it all into one direction. We then discovered that the light coming from the laser is NOT linearly polarized. According to this page it should be vertically polarized.

I had to learn how a circular polarizer works so we could determine if it is circularly polarized. I found this page. We managed to find a 1064nm quarter wave plate and we are going to try and set that up.

Andy Maloney 01:06, 10 August 2010 (EDT): I'm not entirely sure why you would care about the polarization of the laser but, Pranav should know how to setup something to find Brewster's angle. Once you have found that, you can then use the two linear polarizers to find out how the laser is polarized. When you see a polarization ratio in spec sheets such as the one quoted, (linear 100:1, vertical) this means that the laser is 100 times more intense in the p-polarization (this case p means linear) than it is in the s-polarizaion (vertical).


 * Steve Koch 20:33, 10 August 2010 (EDT): I don't know what the problem could be. We definitely have 1064nm 1/2 lambda waveplates, which should rotate linearly polarized laser.  We also have polarizing beam splitting cubes for 1064 nm, which pass p-polarization and reflect s-polarization.  Without a doubt, the laser was highly linearly polarized last summer -- I checked this when using the other beam power stabilizer, which is polarization dependent.  I used a single 1/2 lambda plate and it worked just fine as far as being very linearly polarized.  My best guess is that the optics you're using are not for 1064 nm and were mis-labeled.  Otherwise, maybe the laser degraded more seriously?

Final Thoughts

 * There is definitely drift from the laser and the AOM (separately)
 * Can we investigate the effect of the drift from the AOM despite the drift from the laser?
 * With regards to polarization we believe it is not linearly polarized. We then stuck two linear polarizers in the path and tried to block all the light, but nothing happened. I don't know what that means.