Transcript Document
Setting PMT#5 (Ch3) DTACQ 4.46 V 3.22 V #4 1 k #1 Intensity ratio versus frequency 1051222b series 19 shots (1051222906 ~ 1051222924) Frequency goes from 10 to 100 kHz 5 kHz increment 1051222c series 19 shots (1051222925 ~ 1051222943) Frequency goes from 100 to 10 kHz 5 kHz increment Intensity ratio = (fft amplitude at dtacq #4) / (fft amplitude at dtacq #1) Function generator DC offset not stable in time? So, I checked the stability of function generator 1051222d series 10 shots (1051222944 ~ 1051222954) Fixed frequency at 50.00 kHz Shot taken every 3 minutes So x axis is time (not frequency). Function generator DC offset slowly decreases in time. The rate is approximately (0.862 – 0.870) / 0.5 hour = - 0.016 /hour Several hours later, another series was taken 1051222b series 1051222c series 1051223a series 19 shots (1051223902 ~ 1051223920) This time, the order of frequency change is random 5 kHz increment Several hours later, the intensity ratios decreased maybe because only LED intensity itself decreased. (Raw signal plots imply this) But the slope (that is, the frequency response) seems to be the same. Absolute intensity change in frequency or time 1/2 1051222b series (10 100 kHz in order) 1 k resistor – dtacq #1 LED – PMT#5 – dtacq #4 1051222c series (100 10 kHz in order) Absolute intensity change in frequency or time 2/2 1051222d series (constant 50 kHz) 1 k resistor – dtacq #1 LED – PMT#5 – dtacq #4 1051223a series (random order in frequency) Plot files Location : /home/jinseok/mse/invessel/20051117PMT/pmt pmt_analysis_1051222b.ps (41 plots) pmt_analysis_1051222c.ps (41 plots) pmt_analysis_1051222d.ps (26 plots) pmt_analysis_1051223a.ps (41 plots) pmt_total_000.ps (2 plots) pmt_total_001.ps (2 plots) contain the raw data and fft plots for all frequencies including the plots on the previous pages. On the plots, Red : the signal from DTACQ #1 (Resistor) Blue : the signal from DTACQ #4 (LED – PMT) Tentative summary The typical measurement time (with the frequency increment of 5 kHz) from 10 to 100 kHz is about 30 minutes. The frequency response plots indicate the change of frequency results in about 0.08 decrease in the intensity ratio. Then, the transient DC offset contribution to this change can be 0.016/0.08 x 100 = 20% But, again, the frequency response test after several hours later indicates that this contribution is the same. If we do this test for, say, 5 different input intensities (using the neutral filters), the total test time will be 5 x 30 min = 150 min = 2.5 hours Then we expect the decrease of 0.016/hour x 2.5 hours = 0.04 in the intensity ratio, assuming the rate of change in the DC offset is constant during the test.