NEXT
User name johna
Log entry time 04:17:06 on March 5,2006
Entry number 165260
This entry is a followup to: 165249
keyword=More detail on raster-size scan
Mark Jones suggested looking at the scan in more detail: singles rates, and random-subtracted coincidence rates. Here are the singles rates, along with the randoms and real coincidences from the end-of-run scaler files.
run 6258 6256 6257
Raster 3mm 4.5mm 6mm
T1(kHz) 101.0 104.1 105.7
T2(Hz) 279.8 304.0 312.7
T3(Hz) 13.1 13.8 13.8
I(uA) 1.192 1.211 1.192
Rand(Hz) 3.4 3.8 4.0 (=T1*T2*120ns)
Real(Hz) 9.7 10.0 9.8 (T3 - random)
The uncertainties on these are slightly below 1% for coincidences, and much less than 1% for the others. The following are normalized rates, e.g. (T1 rate)/current, normalized to the small raster size:
raster 3mm 4.5mm 6mm
T1/I 1 1.014 1.047
T2/I 1 1.069 1.118
T3/I 1 1.037 1.053
Real/I 1 1.014 1.013
So the singles rates and total (real+random) coincidence rates both show a small but steady increase with raster size, while the estimated real rate (which will have about a 1% uncertainty) is pretty consistent. Of course, this depends on the assumption of a 120ns coincidence window - a larger window makes the raster-size dependence even smaller (e.g. a 150ns window gives 1, 1.006, 0.998).