Post by mbowry on Nov 3, 2015 9:18:48 GMT -8
Proceeding with analysis of precision deadtime scaler runs that were performed either side of the 26Na beamtime.
I have a script that looks at all GRIFFIN channels with pulser inputs and calculates the deadtime (source plus pulser method of A.P.Baerg), based on some user inputs like the number and length of different patterns in the cycle.
Looking at the 10 channels with pulser inputs for run #5196 (0 us deadtime only), I get strange results (picture 2). I can't believe that we have a (negative) 90 us deadtime in these channels. I think there is a step that is missing. For this run we had 40 second cycles with two 'source only' and two 'source plus pulser' patterns each (picture 3). The pulser frequency was 2kHz.
I'd also like to discuss how we expect to calculate the uncertainty in these measurements. For the source only patterns, we can assume our normal gaussian statistics. For the source plus pulser runs this is no longer valid - for this I am using some metric similar to the maximum likelihood method.
I have a script that looks at all GRIFFIN channels with pulser inputs and calculates the deadtime (source plus pulser method of A.P.Baerg), based on some user inputs like the number and length of different patterns in the cycle.
Looking at the 10 channels with pulser inputs for run #5196 (0 us deadtime only), I get strange results (picture 2). I can't believe that we have a (negative) 90 us deadtime in these channels. I think there is a step that is missing. For this run we had 40 second cycles with two 'source only' and two 'source plus pulser' patterns each (picture 3). The pulser frequency was 2kHz.
I'd also like to discuss how we expect to calculate the uncertainty in these measurements. For the source only patterns, we can assume our normal gaussian statistics. For the source plus pulser runs this is no longer valid - for this I am using some metric similar to the maximum likelihood method.
Attachments:
