User:Andy Maloney/Notebook/Lab Notebook of Andy Maloney/2009/03/27/Optical tweezers

PID control

So I finished the new 5A power supply build yesterday. It's in its own little box with a fan and a super awesome fan guard that Larry thinks we don't need.

At any rate, we purchased the new power supply because the old one did not have enough juice to cool the 1W laser diode we have. So, after completing the build I increased the amperage supplied to the TEC to 2.1A. This seems to do the trick even when I have the diode running at full blast (1.8A).

To setup the PID control, I basically followed what Wikipedia suggested for manual operation of PID. Below are the steps I took:

1. Turn of switch 4.
2. Turn P share, I share, and D share completely counter clockwise.
3. Turn on the TEC.
4. Set the temperature to 18˚C.
5. Watch the temperature drop to 18˚C. My first attempt at this put P share to its max without the TEC reaching 18˚C. So, thinking this was a bad thing, I increased the current going to the TEC. After this (now at 2.1A from 2.0A) I returned P share to its off position and increased it till the TEC reached 18˚C.
6. Set the temperature to 22˚C and watch the temperature settle. If P share is set correctly, then the TEC will again reach 22˚C and oscillate around it for the exact number of times it took to oscillate around 18˚C.
7. I turned on the LD and noted that after about 5 minutes, the temperature began to increase. I then turned switch 4 on and increase I share just a little bit to make the temperature stabilize. I didn't do anything with D share.

I should note that it took me 3 hours to completely give up trying to make the PID work with I share on. I then said to myself, I wonder if using only P will work and yes, it did for a little bit with the LD at 1.8A. To keep the temperature stable, I then had to increase I share (and turn on switch 4) and it has been that way for 50 minutes now without a temperature change.

So far I have only been able to visually note changes in the LD's power output because I do not know anything about LabVIEW and how to read data into a VI. This is a deficiency that I need to correct at some point in time. Maybe I should just buy the damn student version for my Mac. At any rate, the LD power output is stable and has fluctuated around 783 mW ± 3mW. From my understanding, once we get an EOM, the stability of the laser will increase a lot more.

Update

Larry's program was able to show that the laser is stable both for power and temperature for extended periods of time. The program ran for a little over an hour with no problems and I had the laser running for around 2.5 hours with no hiccups. One thing to note is that the laser control box should not be directly next to the power supply because the PS gets pretty hot.

Note

Lihn stated that the power meter will not give the correct power measurement because of the amount of air flow we have in the lab. The power meter measures laser power by a thermal sensor and I guess if there is a lot of air flow around it, then it's measurements are off. I need to check this by making an enclosure around the power meter so no air flows around it.

Anamorphic prisms

Anamorphic prisms pairs do the exact same thing a set of cylindrical lenses does except that the prisms come in a nice little package. They expand one axis of the elliptical beam without disturbing the other. Our set gives a 4x magnification. This means that when we go to use it, the input beam should be 4x longer on one axis. That way the prisms will magnify the output beam so that the axis will be similar in size.