|
Size: 1483
Comment:
|
Size: 3245
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 13: | Line 13: |
| * Mirror loss = 45ppm per reflection. Is it possible ? * Actually, the current PRM design allows up to 65ppm without significant power loss (by Kentaro). Reference needed. |
* Mirror optical loss = 45ppm per reflection. Is it possible ? . --> '''[[Hiro_100811|Impossible]]''' by Hiro Yamamoto * With the current PRM (R=80%), the acceptable loss to achieve PRG=8.25 (100 W input power is assumed) is 58ppm. |
| Line 18: | Line 19: |
| For L = 3000.0m, it gives ROC = 7098.08m. | For L = 3000.0 m, it gives ROC = 7098.08 m. |
| Line 22: | Line 23: |
| * g-factor determines the beam spot size on the test masses. Thermal noise (especially coating thermal noise) is smaller when the spot size is larger. * So called Sidles-Sigg instability of the arm cavities by the radiation pressure is affected by the choice of g-factor. * The parametric instability is also dependent upon g-factor. |
* g-factor determines the beam spot size on the test masses. Some kind of thermal noise is smaller when the spot size is larger. * So called Sidles-Sigg instability of the arm cavities by the radiation pressure is affected by the choice of g-factor. * The parametric instability is also dependent upon g-factor. ==== Thermal noise and beam radius ==== Coating Brownian and thermoelastic noise are the largest noise sources; coating is the biggest at f<500 Hz and TE is the biggest at 500 Hz<f. Coating TN is inversely proportional to the beam radius while TE noise is almost independent from the beam radius at low temperature. |
| Line 27: | Line 32: |
| With the current design of the arm cavity power (450kW), the two eigen-frequencies of the radiation-pressure-induced angular springs are 1.7Hz and 0.88Hz for the Sapphire mirror of 30kg (moment of inertia = 0.173 [kg*m^2]). For positive g-factor, 1.66Hz becomes unstable whereas it is 0.86 for negative g-factor. | With the current design of the arm cavity power (420 kW), the two eigen-frequencies of the radiation-pressure-induced angular springs are 1.7 Hz and 0.88 Hz for the Sapphire mirror of 30 kg (moment of inertia = 0.173 [kg*m^2]). For positive g-factor, 1.7 Hz becomes unstable whereas it is 0.88 for negative g-factor. . Note: [[attachment:radiation_spring_freq.pdf]] == Mirror Size == Mirror size requirements are first set by the beam spot size on each mirror. The mirror radius should be larger than 2.7*(beam radius of 1/e^2) so that the diffraction loss is less than 1ppm. For LCGT, we do not want to have too many variations of suspensions. This is another factor to be considered when choosing mirror sizes. Different mirror size require different suspension design. So we don't want to have many different sizes of mirrors. Mirror size issues are discussed here. [[LCGT/subgroup/ifo/ISC/MirrorSize|Mirror Size]]. == MC length == * Currently the MC length is set 13.32 m. The frequency of the RF sidebands that can transmit the MC is an integer multiple of 11.25 MHz. Providing two sideband fields at 11.25 MHz and 45 MHz and using the beats of these frequencies, we do not have much room for one more SB, i.e. non-resonant SB for ASC, at a reasonable frequency. Tatsumi-san suggested that we double the MC length and make a room for a NRSB (e.g. 16.875 MHz). * Another issue on the SB frequency is that there may be a "bad number" frequency that overlaps with a radio frequency or something. == Original Slides == http://gw.icrr.u-tokyo.ac.jp/cgi-bin/DocDB/ShowDocument?docid=148 == Technical issues on the control scheme == Disscussion is [[LCGT/Technical/ISC|here]]. |
Current Issues of the IFO design
This page summarizes the current issues of the LCGT IFO design and provides links to pages discussing the details of each problem.
Contents
Arm Cavity Parameters
Finesse
1550 is the default value decided by the IFOBW working group.
Issues
- 1550 could be too high. Need more investigations on what could go wrong.
- Mirror optical loss = 45ppm per reflection. Is it possible ?
--> Impossible by Hiro Yamamoto
- With the current PRM (R=80%), the acceptable loss to achieve PRG=8.25 (100 W input power is assumed) is 58ppm.
g-factor (mirror ROC)
g1 = g2 = sqrt(1/3) = 0.57735 is the conventional number to avoid HOM resonances. For L = 3000.0 m, it gives ROC = 7098.08 m.
There are several factors to determine the g-factor.
- g-factor determines the beam spot size on the test masses. Some kind of thermal noise is smaller when the spot size is larger.
- So called Sidles-Sigg instability of the arm cavities by the radiation pressure is affected by the choice of g-factor.
- The parametric instability is also dependent upon g-factor.
Thermal noise and beam radius
Coating Brownian and thermoelastic noise are the largest noise sources; coating is the biggest at f<500 Hz and TE is the biggest at 500 Hz<f. Coating TN is inversely proportional to the beam radius while TE noise is almost independent from the beam radius at low temperature.
Sidles-Sigg instability
With the current design of the arm cavity power (420 kW), the two eigen-frequencies of the radiation-pressure-induced angular springs are 1.7 Hz and 0.88 Hz for the Sapphire mirror of 30 kg (moment of inertia = 0.173 [kg*m^2]). For positive g-factor, 1.7 Hz becomes unstable whereas it is 0.88 for negative g-factor.
Mirror Size
Mirror size requirements are first set by the beam spot size on each mirror. The mirror radius should be larger than 2.7*(beam radius of 1/e^2) so that the diffraction loss is less than 1ppm.
For LCGT, we do not want to have too many variations of suspensions. This is another factor to be considered when choosing mirror sizes. Different mirror size require different suspension design. So we don't want to have many different sizes of mirrors.
Mirror size issues are discussed here. Mirror Size.
MC length
- Currently the MC length is set 13.32 m. The frequency of the RF sidebands that can transmit the MC is an integer multiple of 11.25 MHz. Providing two sideband fields at 11.25 MHz and 45 MHz and using the beats of these frequencies, we do not have much room for one more SB, i.e. non-resonant SB for ASC, at a reasonable frequency. Tatsumi-san suggested that we double the MC length and make a room for a NRSB (e.g. 16.875 MHz).
- Another issue on the SB frequency is that there may be a "bad number" frequency that overlaps with a radio frequency or something.
Original Slides
http://gw.icrr.u-tokyo.ac.jp/cgi-bin/DocDB/ShowDocument?docid=148
Technical issues on the control scheme
Disscussion is here.