Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I How to estimate total energy when dealing with beamsplitters

  1. Nov 6, 2018 #1
    I'm having a brain freeze right now and I need some help. :frown:

    I have a 50/50 beamsplitter (non-polarizing) with 5% tolerance. Test measurement (with a laser at a certain wavelength) gave that I'm transmitting 5.36 units and reflecting 4.6 units (49.5 transmitted; 42.6% reflected). Units refer to a conversion unit of my detector that I can convert to nJ or W. Without the beamsplitter the total power is 10.82 units.

    If you add up the reflected and the transmitted powers, I get in total 9.97 units (there seem to be some losses in the the glass), using this total value I get the corresponding transmitted and reflected to 53% and 46% respectively.

    In another measurement (using another laser with the same wavelength as above) I can measure the reflected power P= 1,32 units. I need to know the total power coming into the beamsplitter. The transmitted is unknown in my setup since a camera is placed in this end. How can I determine the total energy from the 53:46 or 49.5:42.6 division by knowing the reflected percentage?
     
  2. jcsd
  3. Nov 6, 2018 #2

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Couldn't you take out the beamsplitter and empirically measure both the transmitted and reflected, the same way you did with the first light source? Unless you want to assume that the percentage of absorption is always the same for both light sources, then there is no direct way to determine the amount of transmission.

    Zz.
     
  4. Nov 6, 2018 #3
    The point of the beam splitter is that I'm only measuring the reflected beam in my actual setup (since the transmitted beam goes into a camera). The laser has the same wavelength (and is the same type of laser) so it should be the same splitting ratio from unit to unit.

    In the above I measured the reflected and the transmitted powers to determine the actual splitting ratio.

    Questions is how can I determine the total power entering the BS by measuring the reflected power (and without measuring the transmitted power every time). Can't I determine this from the pre-measured ratio? If so, how can I do this?

    Thanks
     
  5. Nov 6, 2018 #4

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    If it is the same wavelength (which it appears to be), then you should be able to "calibrate" it using your first measurement. So such calibration must be done first at this particular wavelength, i.e. before you start the experiment, you need to measure the input power, the reflected power, and the transmitted power. You have to do all 3 because real beamsplitter (or crystal) absorbs some of the power. Then you have to explicit assume that, for a given power at the same wavelength, the ratio of reflected to input power remains the same. In fact, you have to assume the ratio of reflected to transmitted power also remains the same.

    Zz.
     
  6. Nov 6, 2018 #5
    I agree. So my embarrasing question is if I get a reflected transmission of 1,32 nJ. How much is the input power if I have splitting ratio of 52:46 (T:R)?
     
  7. Nov 6, 2018 #6

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    This doesn't seem right: the beamsplitter is absorbing 10% of the incident power? That's very noticeable and would create observable effects (heating, damage, etc.).
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted