• Laser
• Amplifier
• PowerCalc

Power calculation

P_arm = Plaser * T_PMC * T_IMC * T_FI * alpha_MM * G_PR * T_BS * G_arm

Current best guesses (as of 30 Apr 2019):

• T_PMC = 85%
• T_IMC * T_FI = 50% (it should be possible to make this ~90 % if we assume 50 ppm RT loss in the IMC and ~95% throughput on the IFI)
• alpha_MM = 98 %
• G_PR = 40-45 (once we put new folding mirrors in)
• T_BS = 50%
• G_arm = 270-280 (20-30 ppm round trip loss)

Giving P_arm = Plaser *2400. So if Plaser = 10W, Parm = 24kW. If we improve T_IMC * T_FI to 90%, then the overall gain becomes x4000, so 10 W from the laser becomes 40 kW in the arms.

As to the question "How much power do we need?"

I think this depends on the specific configuration we operate in, i.e. SRC detuning and homodyne phase. The general trend is that the optomechanical coupling crosses unity at higher frequencies with higher power. So the frequency at which the IFO quantum noise is at its lowest relative to unsqueezed vacuum is directly proportional to power. We hypothesize that it is easier to measure squeezing at higher frequencies due to lower technical noise. A set of example curves for just the quantum noises is (not optimized in any systematic way, just for illustration, but taking into account the non-unity transmission of the PMC and IMC and intervening optics):

In the lower plot, the dashed colored lines are the unsqueezed vacuum noise levels. So the squeezing is where the solid lines are below the dashed.

1. What should our PMC Finesse be so that its barely enough to satisfy filtering reqs and transmission efficiency?
   • Suppress angular modes by XX (note that the IMC also filters the beam jitter)

   • Suppress laser noise at the 11 & 55 MHz readout frequencies by XX (this is primarily a PMC function as the IMC transmits the sidebands.). For a finesse 700 cavity, the RIN attenuation at 11 MHz is ~x10, so we would need to detect 300mA of photocurrent to see the RIN of the laser assuming the RIN numbers quoted in Fig 4 of the amplifier paper).

   • Maintain a T_PMC > 0.9X (if the existing spec is accurate, the best possible T is ~88%). Potentially, we could use the metal PMCs, whose curved folding mirrors are spec-ed to be 200ppm, so we can achieve slightly higher transmission for the same finesse.

2. What's up with our T_IMC ?
   • let's clean the IMC mirrors

   • what's the MM frac between PMC and IMC? Judging by the IMC visibility, this is ~98%.

3. Is our T_FI too low?
   • measure this during the vent when we clean the IMC
4. What is the loss in the PRC?
   • clipping in the PRC
   • new PR/SR mirrors