Transcript A few more details of the after pulsing using the above slides.

The next slide shows a simplified diagram of the pulser. A
capacitor is charged up to perhaps 10V. Switches T1 and T2
close. Some of the capacitor current runs through the diode.
Some current runs through the a resistor to ground.
To monitor the process we probed both sides of the diode and
sometimes compute the difference.
Third slide is one simple model. I let both switches close
simultaneously. The diode draws 100 mA for a few ns and then
the voltage is below threshold. The capacitor discharges the
rest of the way (pink) through the resistor. After 15 ns the
voltage across the diode is zero.
Simplified Version of the Pulser Boards (Based
on KAMLAND pulsers)
100 K
B+
Adjustable
Voltage
22 pF
100
probe
t1
pmt
probe
t2
FETs are switched on (SW1,SW2) to control the
firing of the diode. 2 P5050 probes measure
Voltage on diode.
Volts
Voltage at Capacitor
12
10
8
6
4
2
0
-2 0
Total Volts
sw1 on 5 ns
sw1 off 55 ns
sw2 on 5 ns
Diode Threshold 3 V
Diode draw 100 mA
22 pF 100Ohms
Sw1On
SW2On
a
5
10
15
20
25
30
time ns
MODEL: Switch 1 starts discharge through 100 Ohms R
Switch 2 starts discharge through LED (constant current).
The following slide just compares the measured voltage
between the capacitor and the diode with the simple model.
The two are consistent.
Never tried to measure the actual time sequence of the
switches t1 and t2. So I am not sure when the switches
actually close.
Volts
Voltage at Capacitor
12
10
8
6
4
2
0
-2 0
Total Volts
sw1 on 1 ns
sw1 off 50 ns
sw2 on 1 ns
Diode Threshold 5 V
Diode draw 300 mA
90 pF 100Ohms
Sw1On
SW2On
voltage 1GHz probe
a
10
20
volt 500MHz probe
30
40
50
time ns
Shows the simple model and measurements using 1GHz and a
500 MHz probe.
The slide that shows the afterpulsing shows
only the two voltages (not the difference).
One sees that initially the voltage across the
diode (pink to yellow difference). Both
voltages and there difference drop to zero
quickly. Then the capacitor recharges. As the
capacitor charges up a voltage develops
across the diode.
I assume that when the voltage difference
exceeds approaches threshold the diode fires
but in a noisy not consistent way. Evidence
for this is shown with low rate results (see
ahead)
8
6
4
2
0
-1.0E-05
0.0E+00
-2
-4
1.0E-05
2.0E-05
3.0E-05
Afterpulse Averaged
many WFms
-6
After pulse: Average over many WFMs
4.0E-05
The next slide shows the prompt pulse and the after pulse at a
lower rate. As we lowered the rate we found the after pulse was
smaller and also wider. I don’t know why but it did seem that at
low pulse rates we saw a good deal of single photon noise rather
than a definite pulse. Thus I characterize it as an enhancement
where the diode may be unstable in some way and tend to
generate light for awhile during the charge up cycle. There was
never any clear measured quantity that we could connect with
the after pulsing.
Voltage(V)
1.54E-05
1.43E-05
1.32E-05
1.21E-05
1.10E-05
9.84E-06
8.72E-06
7.60E-06
6.48E-06
5.36E-06
4.24E-06
3.12E-06
2.00E-06
8.80E-07
-2.40E-07
-1.36E-06
-2.48E-06
-3.60E-06
Singles trigger
Time(s)
.05
.00
-.05
-.10
-.15
-.20
-.25
Shows that the after-pulsing is an “enhancement rather
than a single pulse for some pulser rates.
We decided to put a resistor across the LED so that the voltage
difference during the recharge would be zero. The circuit below
shows the changes we felt gave a better performance.
The 2nd slide below shows the voltage at each probe (red/blue)
and their difference in light blue. This a short time scale so you
can see the voltage pulse that generates the light and pmt
response in green. Notice now due to the resistor the voltage
drop across the diode is normally zero whereas in previous
slides it was typically a bit more than 2Volts.
The 3rd slide below is an longer time scale with no after pulsing
evident. Both slides show the same signals but the time scale
has been changed.
100 K
B+
Adjustable
Voltage
22 pF
100
t1
39
1 K
Modified
Pulser
t2
probe
pmt
probe