A Development of Front-end Electronics System Adopting Pipeline Transfer

Research Project

Project/Area Number	04640293
Research Category	Grant-in-Aid for General Scientific Research (C)
Allocation Type	Single-year Grants
Research Field	Economic statistics
Research Institution	KYOTO UNIVERSITY
Principal Investigator	KURASHIGE Hisaya Kyoto Univ., Fac.of Sci., Instructor, 理学部, 助手 (20205181)
Co-Investigator(Kenkyū-buntansha)	TANIGUCHI Takashi NLHEP, Instructor, 助手 (60163630) SASAO Noboru Kyoto Univ., Fac.of Sci., Professor, 理学部, 教授 (10115850)
Project Period (FY)	1992 – 1993
Project Status	Completed (Fiscal Year 1993)
Budget Amount *help	¥600,000 (Direct Cost: ¥600,000) Fiscal Year 1993: ¥600,000 (Direct Cost: ¥600,000)
Keywords	Data-acquisition / Pipeline / Trigger / パイプライン転送 / イヴェントトリガー / Tkoバス / VMEバス
Research Abstract	We have developed a new data acquisition system, which has a unique pipeline operation scheme for front-end modules. We have implemented a pipeline architecture for front-end signal processing on the TKO BOX system. TKO BOX system was originally designed for TRISTAN experiments at KEK.Operations in TKO modules can be divided into three stages : sample-and-hold, A/D conversion, and data read-out. Pipeline transfers between these three stages make it possible to process the next event during data read-out. As a result, dead-time due to read-out is significantly decreased. This scheme is applicable to conventional modules without flash ADC_S. We have adopted a minimal pipeline scheme for simplicity. In order to realize this scheme, we have modified TKO specification and developed new TKO front-end modules, a new gate controller, and a trigger supervisor. We have tested this new data acquisition scheme for calorimeter calibration at KEK PS.Three TKO BOXes were used for the test system, and it takes 512msec to scan all channels. At a trigger rate of 500 Hz, dead-time is found to be 7.2% including dead-time of 4.6% due to A/D conversion of 100msec, while dead-time comes up to 23.4% without pipeline operation. The system worked very well and no critical errors occurred in three months of calibration run.