Differences between revisions 8 and 9
Revision 8 as of 2010-08-11 23:35:09
Size: 1082
Comment:
Revision 9 as of 2010-08-12 00:04:42
Size: 640
Comment:
Deletions are marked like this. Additions are marked like this.
Line 1: Line 1:
 . Subject: [LCGT-isc 00258] Re: [LCGT-isc 00254] 論点整理
 . From: Hiroaki Yamamoto
 . To: LCGT-isc at icrr.u-tokyo.ac.jp
 . Cc: Norikatsu Mio
 . Date: Wed, 11 Aug 2010 00:25:20 -0700
----
== Suspect on mirror optical losses by H.Yamamoto ==
Line 8: Line 3:
 . 私としては不可能と信じていますが、一応どの様な表面が必要か。 散乱によるロスは
  * 鏡の長波長(>1 mm) RMS = 1 nm で鏡あたりの散乱損失が 45ppm. '''0.75 nm で 25ppm'''
  * 鏡の短波長(<1 mm) RMS = 0.2 nm で '''6ppm'''
  * 訳の分からない点状散乱 '''10ppm''' (神様、仏様に祈ってやっと願いがかなうかどうか、15~30ppm が現実的な値)
 . この散乱だけで 61ppm (RMS(長波長)=1nm), '''41ppm (RMS=0.75nm)'''
 * Scattering loss of 45ppm by low-freq (>1mm) surface error of 1nm RMS (25ppm with 0.75nm RMS).
 * Scattering loss of 6ppm by high-freq (<1mm) surface error of 0.2nm RMS.
 * Unexplainable punctate scattering loss of 10ppm (realistic estimate would be 15-30ppm).
Line 14: Line 7:
These alone would make the total scattering loss already 61ppm (with 1nm LF RMS) or 41ppm (with 0.75nm LF RMS).
Line 15: Line 9:
 . シリカの場合、コーティングによる吸収が 0.5ppm 以下が可能と言われているが、表面が汚れると数 ppm 簡単に悪くなる。


 . この中で、金と時間で解決できるのが長波長領域での散乱ロス。 LIGO は RMS<0.5 nm を要求し、 Tinsley がそれより良い研磨を達成している。
 * Besides, absorption at the coatings, assumed to be 0.5ppm, can go worse by a few ppm easily.
 * LF RMS could be improved if we spend a lot of money and time. LIGO asked for RMS<0.5nm and Tinsley.

Suspect on mirror optical losses by H.Yamamoto

  • Scattering loss of 45ppm by low-freq (>1mm) surface error of 1nm RMS (25ppm with 0.75nm RMS).

  • Scattering loss of 6ppm by high-freq (<1mm) surface error of 0.2nm RMS.

  • Unexplainable punctate scattering loss of 10ppm (realistic estimate would be 15-30ppm).

These alone would make the total scattering loss already 61ppm (with 1nm LF RMS) or 41ppm (with 0.75nm LF RMS).

  • Besides, absorption at the coatings, assumed to be 0.5ppm, can go worse by a few ppm easily.
  • LF RMS could be improved if we spend a lot of money and time. LIGO asked for RMS<0.5nm and Tinsley.

Hiro_100811 (last edited 2010-08-12 08:09:05 by HiroYamamoto)