The characterization facility, showing the 100-fs laser (upper left), vacuum flange for testing 33 mm MCPs (upper right), large vacuum chamber for testing 8 in. plates (lower left), and an O-ring-sealed 8 in.-square detector mounted on a large area 90 × 20 cm microstripline anode (lower right).
Schematic of the optics used to generate ultraviolet (266 nm) light from the pulsed infrared laser.
Schematic of the imaging and beam steering optics for the pulsed UV light coming from the laser optics (Fig. 2 ).
An example oscilloscope trace including both the trigger diode and the UV photodiode.
Raw, integrated UVPD signal for each of 1000 laser pulses, showing variations in laser power as a function of time. Occasional downward fluctuations 20% are due to instabilities in the regenerative amplifier.
A histogram of integrated UVPD signal for the same 1000 pulses, with corrections to the true UV pulse energy applied. The small number of non-empty bins in the low tail correspond to the downward fluctuations seen in Fig. 5 .
An example trace from the UV power detector. The output of the detector is a DC voltage proportional to the power by a multiplicative constant of 42 nW per measured volt. The ringing of the signal eventually stabilizes several nanoseconds after the initial pulse. We average this later part of the signal to determine the laser power.
The relationship between the integrated UV photodiode (UVPD) signal and the laser pulse energy in joules, which serves as a useful proxy for the energy of each laser pulse. However, it is convenient to translate these integrated signals (in units of volt seconds) into physically meaningful units. Each time we acquire 1000 laser pulses in the scope, we use the UV power detector. Comparing the average signal from this UV power detector with the average integral of the UV photodiode, allows us to relate the UV photodiode signal with pulse energy.
Probability of a laser pulse generating a MCP signal plotted as a function of pulse energy. At sufficiently high laser intensities, this probability approaches unity; as the laser is attenuated below roughly 10 × 106 UV photons per pulse, the fraction of laser pulses producing a MCP signal drops off and eventually approaches 0.
A schematic of the 33 mm MCP test chamber.
Schematic of a typical two-MCP stack mounted on the 33 mm test flange. Photons striking the photocathode produce electrons by the photoelectric effect. These electrons are accelerated across a potential gap towards the gain stage, which consists of two porous plates, optimized for secondary electron emission. These plates are typically held at field strengths of 1 kV/mm. Electrons accelerate down the pores and collide with the walls, producing an avalanche of secondary electrons. The amplified signal drifts across a final accelerating potential to the anode plane, where the charges form a signal. Insulating spacers between the two MCPs and between the top MCP and photocathode are represented by pale red bands. In our 33 mm test setup, we use Kapton rings.
Four configurations of the pulser calibration system.
Three configurations of the four-channel readout. The laser spot (shown as a blue dot) is focused on the central strip of a three-stripline cluster. The two sides of all three striplines and the trigger photodiode add up to 7 possible readout channels. Since the oscilloscope can only read 4 of these channels, we must select a subset of the channels for any given measurement.
A schematic of the 8 in. MCP test chamber. The laser beam enters from the right and is reflected onto the MCP via a fixed mirror mounted on a breadboard in the chamber.
The full 8 in. MCP detector stack.
A pair of demountable tile bases from Joe Gregar (ANL Glass Shop). These consist of a glass side wall fritted onto a 30-strip “frugal” glass anode and are used with an O-ring sealed window to test the MCP-spacer stack of the frugal tile at the APS laser facility. The pump ports are reinforced for mechanical strength.
The fully instrumented Demountable test setup at the APS at Argonne. A 4-tile tile-row consisting of a 90-cm anode, of which 1 tile is the Demountable test module, is read out with an analog card at each end. Each analog card carries 5 PSEC4 ASICS that digitize the 30 anode strips at 10 GS/s. The digital values are read out by an FPGA on a digital card on each end (this is via the blue cables); these in turn pass the data to a central card that combines the 2 ends and communicates with the PC. Three such tile rows would share a support “Tray” and a single central card to make super-modules. Each of the three empty anodes in this figure could be replaced with an active Demountable detector.
A histogram of our heuristic measure of pulse “significance” for a pair of MCPs operating in single photoelectron mode at modal gains of 107. The first large peak, near zero consists primarily of events with no pulse, while broader second peak corresponds to events with pulses.
The pulse height distribution for a pair of MCPs in single PE mode, overlaid with a polynomial fit to the shape.
Location of the signal centroid for several thousand laser pulses. The position of the signal is defined with respect to the stripline number of a three strip cluster, where the laser was directed over strip-2. This distribution was reconstructed from data taken in the 33 mm chamber with 1.1 mm striplines at 1.6 mm center-to-center spacing. The mean of the distribution is 2.114 and the RMS is 0.24 strip widths. There is a slight asymmetry, due to possibly the bias angle of the MCP pores with respect to the anode direction.
Fraction of integrated three-strip signal observed on the central strip, for pulses with centroid location within 0.4 units of stripline number of the central strip. This distribution was fit to a Gaussian with mean = 0.647, σ = 0.083, and χ2/ndf = 12.85/15.
Pulses of known charge are sent through the readout system with the pulse signal split into three components of varying size (50%, 41%, and 9% of the total signal). The pulse generator is triggered by the laser clock in order to accurately reproduce overlap between the signal and RF noise from the Pockels cell drivers. Each of the 3 signals is recorded by the oscilloscope. The signals are integrated and summed to reconstruct the charge of the original pulse. This figure shows reconstructed charge versus true charge for known pulses of various sizes, in units of elementary charge. Points are fitted with a line (slope = 0.9931, offset = −4.07 × 104).
An example of a MCP pulse showing the peak signal and the crossing times for a 50% and 25% constant fraction threshold. The baseline is determined by fitting the region before the pulse and subtracted out.
The measured TTS for a stack of 2 MgO coated MCPs, operating 1.2 mm thick with 20 μm pores at 8° bias angles. The stack was operated in saturation mode, with 1.2 kV across each plate, 300 V across the photocathode gap, 200 V between the MCPs, and 800 V across the anode. The signal was read from one side of a single stripline, fitted with a Gaussian (μ = 5.74 ns, σ = 17.4 ps).
Table summarizing the capabilities of the MCP characterization facility.
Article metrics loading...
Full text loading...