December 16th 2008,
Argonne, EFI Chicago
Measurements of Micro-Channel Plate Devices
Three Micro-Channel Plate devices (MCPs) and associated hardware (fast transmission line card) are been tested, having in view:
-1 To cross-check the measurable electrical properties of the tubes and readout card with the expectations, such as the inter-electrode resistances and capacitances, the photo-cathode response to the daylight when biased under a few Volts,
-2 To validate these devices in terms of signal amplitude, rise-time, gain vs high-voltage dependence and light pulses rate, using a fast pulsed LED light source (4ns rise time).
-3 To compare the dynamic electrical properties of a transmission line readout card
with the simulations using a fast pulse simulating an MCP anode output, and validate its use for a two-dimension readout of the MCP devices with a one-dimension fast timing electronics.
-4 To test the fast transmission line card when connected to an MCP device illuminated in the conditions of 2- and validate the concept of 2-dimension readout using pico-second timing.
MCP 1 (25 mm pore device) has been connected to a transmission line card where 32 lines of 50 W impedance at 1.6mm pitch have been implemented on a Rogers 3014 support, using an Epotek 4110 silver epoxy glue.
MCP 2 and 3 (10 mm pore diameter devices) are connected to a card ganging three sets of 256 signals.
A first test of the resistances of the 1024 anode connections to the transmission line card to MCP1 using silver epoxy Epotek 4110 resulted in the histogram shown Figure 1
Figure 1. Histogram of the 1024 anode pads connections of MCP1 (precision 1 W).
MCP capacitances.
MCP / 1 / 2 / 3Cathode-MCP in (pF) / 46.4 / 45.7
Cathode-MCP out (pF) / 34.6 / 36.1
Difference (pF)_ / 11.8 / 9.6
Daylight induced voltage drop across the HV bridge (MCP in).
MCP / 1 / 2 / 3No light (V) / 2.40 / 2.76
Daylight (V) / 2.45 / 2.84
Difference (mV) / 50mV / 80
1- Tests at Chicago
Figure 2. MCP pulse amplitude as a function of the High Voltage. f=1kHz
Blue LED 160V
Figure 3. MCP pulse amplitude as a function of the rate. High Voltage: 1.5 and 2 kV. Important: pulse dynamics is not due to MCP response, but to the slow LED (4ns rise time) used at EFI Chicago.
Figure 4. 25 mm pores MCP on the 32-transmission lines card
Figure 5. MCP3 signal at 2kV, LED, 1 kHz.
Figure 6 a and b. MCP1 signals at the left and right ends. LED, 2kV, 1 kHz
The delay difference between the two ends corresponds to a propagation velocity of .48c, as simulated.
2- Tests at Argonne
2-1 MCP1 (35 mm pores) measurements
Tests have been done with a set-up comprising (fig 1.):
A pulse generator triggering the laser from 1 Hz to 100 MHz
A laser:
- l 635nm
- Light pulse duration 80ps
- Pulse tne with a set-up comprising:
o pulse jitter 10ps
Attenuators
- Optical density. See Table 1 below.
A lens
- f=5cm
A movable stage (X,Y,Z) with micrometer precision remote control
Two digitizing oscilloscopes: 2 GHz bandwidth, 40Gs/s, and 15 GHz, 40GS/s,
Figure 7. Laser test stand at Argonne.
Table 1. shows the number of photo-electrons (Npe) obtained for the attenuators used.
Filter / Density / Npe1/6 / 3.0 / 5
2/6 / 2.5 / 18
3/6 / 2.0 / 50
4/6 / 1.5 / 158
5/6 / 1.0 / 500
6/6 / 0.0 / 5000
MCP1 has been installed with the laser focused onto the front plate window, in front of the transmission line to be observed at the two ends.
Figure 8. MCP1 + PCB test. Pulse amplitude vs rate at 2kV high voltage, attenuation 10 (density 1). Laser focused on MCP1 front-plate. Spot size: 1mm (eye estimated). The amplitude was increased later by a factor of 3 improving the positioning of the MCP relative to the stage.
At very high laser light inputs and MCP gain (no attenuation giving 5000PEs, high voltage at 2.3 kV), two peaks are observed in the pulse waveform, the primary (first) and a secondary (two rise-times after the primary) At 10 kHz, the first peak decreases and the secondary peak starts growing, at 100kHz the secondary peak is two times larger than the primary. This should not be an issue at regular light levels (<500 PEs).
Figure 9. Pulse amplitude vs HV. No attenuation (5000 PEs). Laser focused.
Figure 10. Rise-time vs high voltage. No attenuation (5000PEs)
Figure 11. Amplitude vs attenuation. High Voltage is 2kV, rate is 2 kHz.
A better positioning of the MCP with respect to the stage resulted later in an increase of signal by a factor of 3.3. This was done moving the MCP in X and tuning the position to have the maximum signal, whatever the Y position. Then the MCP could be moved in Y to check the pulses at the ends of the transmission line
At 500 Hz rate, 2.2 kV high-voltage, a signal of 19 mV was obtained for 50PEs, 2.2kV high voltage.
2- Transmission line measurements
At 500 Hz rate, 2.2kV high voltage, using the ANL set up comprising a x 100 amplifier together with TAC and CAMAC ADCs, the time distribution on each signal showed the following spreads:
Npe / A / B / Difference158 / 4.4 / 4.4 / 6.2
50 / 5.1 / 5.2 / 7.4
Figure 12. Time difference histogram obtained with TAC + ADC for:
Left-hand: Single position
Right-hand: Same, with 10mm moved MCP merged
Filter 4: 158 PEs
The time difference yields a velocity of 64ps/cm against 68ps simulated.