March 5, 2008 online meeting minutes: Difference between revisions
No edit summary |
No edit summary |
||
| Line 1: | Line 1: | ||
present: Sergey Boyarinov, Sergey Pozdnyakov, Ben Raydo, Chris Cuevas | present: Sergey Boyarinov, Sergey Pozdnyakov, Ben Raydo, Chris Cuevas | ||
We discussed CLAS12 level2 trigger design. In recent weeks Sergey B. with Ben's help was working on level2 segment finder which suppose to search for the track segments in the single Drift Chamber superlayer. | We discussed CLAS12 level2 trigger design. In recent weeks Sergey B. with Ben's help was working on level2 segment finder which suppose to search for the track segments in the single Drift Chamber superlayer. First version of vhdl code was developed, it takes about 3.7 microsec and does not depend on the number of hits. Sergey will work on road finding part and then entire scheme will be reviewed again. | ||
Hardware design options were discussed. To implement CLAS12 level2 we have to modify current electronics installed in ADB crates. Three options were considered: | |||
* do not modify ADB crate backplanes; develop new boards instead of existing segment finders (called SFS) and connect them with new segment finder by 3 32-pair cables; existing MUX boards will be reused; main problem is the big number of cables (3 per MUX board) so new segment finder must have 21 high dense 32-pair connectors | |||
* new ADB crate backplane supporting 14 ADB boards (currently 12) to avoid splitting superlayers; still need new boards instead of old SFS modules | |||
* new ADB crate backplane and new MUX boards; redesigned MUX boards will support more efficient data transfer protocol to the new segment finder board over new backplane, and we can improve some other parameters as well (TDC signal's falling edge jitter etc); it maybe possible to implement programmable delay into TDC signals to minimize TDC window if existing FASTBUS 1877 TDCs are in use | |||
Chris estimated new backplane production cost as about $1800, and new MUX board production price will be studied. | |||
Revision as of 11:21, 10 March 2008
present: Sergey Boyarinov, Sergey Pozdnyakov, Ben Raydo, Chris Cuevas
We discussed CLAS12 level2 trigger design. In recent weeks Sergey B. with Ben's help was working on level2 segment finder which suppose to search for the track segments in the single Drift Chamber superlayer. First version of vhdl code was developed, it takes about 3.7 microsec and does not depend on the number of hits. Sergey will work on road finding part and then entire scheme will be reviewed again.
Hardware design options were discussed. To implement CLAS12 level2 we have to modify current electronics installed in ADB crates. Three options were considered:
- do not modify ADB crate backplanes; develop new boards instead of existing segment finders (called SFS) and connect them with new segment finder by 3 32-pair cables; existing MUX boards will be reused; main problem is the big number of cables (3 per MUX board) so new segment finder must have 21 high dense 32-pair connectors
- new ADB crate backplane supporting 14 ADB boards (currently 12) to avoid splitting superlayers; still need new boards instead of old SFS modules
- new ADB crate backplane and new MUX boards; redesigned MUX boards will support more efficient data transfer protocol to the new segment finder board over new backplane, and we can improve some other parameters as well (TDC signal's falling edge jitter etc); it maybe possible to implement programmable delay into TDC signals to minimize TDC window if existing FASTBUS 1877 TDCs are in use
Chris estimated new backplane production cost as about $1800, and new MUX board production price will be studied.