Main INDEX
Monthly INDEX
PREV
NEXT
User name Michaels/Gayou
Log entry time 11:18:19 on January13,2005
Entry number 139172
This entry is a followup to: 139022
Followups:
keyword=Trigger restore, tests
We restored the trigger for the production running with protons on
R-HRS, i.e. the "R-slow" setup. We did the change procedure in halog
139022, modified slightly so that the true coinc time (TC) appears
in the middle of the dynamic range of the TDC in the pulser setup.
The slight modification affects the values of Ldelay1 and Rdelay2.
Next, we test this with pulser setup by putting N nsec on L-HRS and
subsequently N nsec on R-HRS in separate runs, to show that TC moves
the expected way and that there is no delta-function goofiness like
described in halog 139024 which affected the "Lslow" setup.
CODA Run .... Delays ....... Coince Time (TC) measured on L,R
............. on pulser ..... ctimel ..... ctimer (raw channels)
1777 ......... 0 relative. ... 1919 ..... 2068 (recall 1=0.05nsec)
1778 ......... +30 to LHRS ... 1322 ..... 2684 (change = 600 = 30nsec)
1779 ......... +30 to RHRS ... 2497 ..... 1468
1780 ......... +60 to RHRS ... 2894(!) .. 850 ... problem w/cimel
1781 ......... +60 to LHRS ... 707 ...... 3322
All these are ok except ctimel for run 1780. But this is understood
from my notes because the width of T3 was 74 nsec, and as T1 slides
late it can make T5 over this interval but the retiming on LHRS has
only a 47 nsec gap between L1A and RT. So, we reduce T3 width to
40 nsec (must be less than 47). Do another run with RHRS by 45 nsec
which is at the edge of the overlap (verified on scope):
1782 ....... +45 to RHRS .... 2795 ..... 1161 ... Good !
Final run to check the nominal setup
1783 ...... 0 relative .... 1918 ..... 2072
Finally, we remove the pulser. Test completed.