From: W.J. Llope (llope_at_physics.rice.edu)
Date: Fri Jan 24 2003 - 12:03:48 EST
i'm already typing these replies - and i'll send drafts
to this list.... i'll include some of your text below....
At 10:16 AM -0600 1/24/03, Frank Geurts wrote:
>tof folks ... please read the answers to Hank and Tonko's questions
>below and send me any comments before I send them/the list the
>replies...
>
>thanks,
>frank
>
>
>---
>
>>>Hello -- Fred has just given me this nice document describing the TCD. I
>>
>> have a
>>
>>>question for each subsystem - If we issue a Level0 trigger, but for some
>>
>> reason
>>
>>>do NOT issue either a Level2 accept or abort, what does your system do?
>
>the system will time-out (and thus will be busy). Our understanding of
>last year's operation was that not sending data after an L2 accept is a
>capital crime. If we received an L0 but do not receive an L2 accept
>or abort within some time period the system will asume it actually
>missed the L2 altogether. Still having a valid L0, it decides to
>send the event to STAR as a good event.
>
>>
>> --
>>
>> I would like to expand this question:
>>
>> a) At which point in time (relative to L0 or L2 or ...) does your system
>> ship data to DAQ?
>
>relative to L2-accept or in case an L2 gets lost, right after the
>time-out (which is now set at 100ms).
>
>>
>> b) What are the points in time where a L2_ACCEPT or a L2_ABORT would
>> cause problems - either total lunacy or ignored/corrupted events?
>
>- L2 Aborts:
>L2-Aborts are ignored if they are not related to the L0 token in the
>system. If we do have an associated L0 token then that event is
>discarded, the DAQ busy is cleared and the local trigger reset.
>
>- L2 Accepts
>L2accepts are always served:
>* if there is an associated L0 event, then that data will be shipped to
>STAR-DAQ (the busy is cleared and local logic is reset)
>* if there is no L0 event (which means that it got lost otherwise we
>wouldn't be getting an L2-accept in the first place): an empty
>(filled w/ 0s, but similar sized) event is shipped to STAR-DAQ
>
>
>
>>
>> c) At which points in time does your system need to be protected
>> by the internal TCD BUSY?
>
>tofp combines two busy signals for the TCD busy: one busy signal is
>set by the local trigger logic upon receiving a pvpd coincedence. It
>triggers digitization and protects the tofp system against new
>coincidences before it received the L0 which is associated with this
>particular coincidence. For that the logic opens a window in which
>it expects the L0 command to arrive.
>A second busy signal is generated by the local DAQ system. It is
>raised in case a L0 token arrives. It will be lowered only in three
>cases: after the L2 accept, after the L2 abort and after an L2
>time-out.
>
>
>>
>> d) What about L1_ACCEPTS? Currently unused, but one never knows...
>
>is not implemented and therefore ignored.
>
>
>
>---
--W.J. Llope, Ph.D. Res. Assoc. Professor T.W. Bonner Nuclear Laboratory phone: 713-348-4741 Rice University, MS-315 fax: 713-348-5215 6100 S. Main St. Houston, TX 77005-1892 e-mail: llope_at_physics.rice.edu WWW home: http://wjllope.rice.edu/default.html
--
This archive was generated by hypermail 2.1.4 : Thu Jul 24 2003 - 00:39:33 EDT