MCP-based detector: some results and perspectives

The timing resolution of photomultiplier tubes (PMT) based on shevron-type microchannel plates (MCP) has been studied inmagnetic fields. The same timingresolution with and without a longitudinal magnetic field up to 2.0 kGwas obtained as <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 85 &plusmn; 2 ps. It is shown that an increase of timing resolution in this magnetic field does not exceed25 ps (upper limit). The timing resolution of <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 31 &plusmn; 2 pswas obtained for narrow (10resolution) amplitude spectrumfrom Corone discharge. The counting rate of MCP-based detector was studied in function of the direction of the magnetic field.<P>The spatial and timing resolution for the MCP-based PMT were obtained using laser pulses as well. With laser pulses of 0.3 ns a timing resolution of <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> &#8773; 450 ps was obtained. Taking into account the amplitude correction narrows <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> to 140 ps. Using 100 fs-laser with the standard constant fractiondiscriminator gives a timing resolution from 20 to 40 ps depending on the read-out MCP region.<P>The perspectives of using an MCP-based detector for high-energy physics experiments are discussed.Abstract:<P><B>Summary:</B><P>We present the results of the systematic in-lab tests of the MCP-based detectors characteristics.Abstract:<P>In the first two tests two conventional PMT 165-2 tubes (produced by St.Petersburg) were used. These photomultipliers contain a photocathode, two microchannel plates in a shevron-type setup and anodes. A special stand was designed and performed for these tests. It consists of a magnet up to 2.0kG, CAMAC electronics and a PC with a set of programmes. The two detectors were placed face-to-face. The Conore discharge placed between the two PMT-MCP detectors was used as a light spark source. The light froma spark comes to both detectors. If a sheet of black paper had been placed between the source and detectors, signals did not exist. Signals from both detectors went to the amplifier-discriminator. Then both of the signals passed to the TDC through the special time expansion block. One signal had been submitted to "start" of TDC, the second one through a delay line had been given to "stop" of TDC. A TDC channel width was 78 &plusmn; 2 ps. A special block for an expansion of the timing signal was used. It gives a signal between "stop" and "start" to 100 times longer than in real time. After differentiation, the signals were given to TDC. A precise calibration of the timing track was doen. A 5 ns signal from the generator was divided by two. One had been given to "start" TDC, the second had gone through a 100 ps step delay line and given to "stop". This was a special delay line with an accuracy of 1 ps. The analysis showed that the TDC channel width with using of expansion block was <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif">= 0.7 ps and the jitter of all track was <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 10.7 ps.<P>As a first step, we wanted to obtain a good timing resolution for the checking of electronics. In this case a narrow (10resolution) amplitude spectrum was used. The timing resolution for one MCP was obtained as <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 31 &plusmn; 2 ps. For the next step, n amplitude spectrum was used that was closer to a real experimental situation (50 mV - 700 mV). The same timing resolution <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 85 &plusmn; 2 ps was obtained for both zero magnetic field and up to 2.0 kG magnetic field. We used a longitudinal magnetic field, and the angle between the direction of the magnetic field and the MCP channels was approximately 12 degrees. An estimation of the possible influence of the magnetic fiels was made. It was shown that an increasing of the timing resolution in the magnetic fiels up to 2.0 kG does not exceed 25 ps (upper limit).<P>The next tests were done using a laser. In this case, a one-photon regime was used. A new variant of the read-out was applied. A timing signal was taken from the "input" of the second MCP surface and an amplitude signal was taken from the anode. This makes it possible to combine an adequate amplitude analysis and a high timing resolution in one detector. The timing resolution for the laser impulse of 0.3 ns was obtained as <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 450 ps. Taking into account the amplitude correction, the resolution improves to <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 140 ps. Using a 100-fs laser with a standard constant fraction discriminator gave a timing resolution better than 40 ps. It is shown that the timing resolution has a dependence on the read-out region. Using a small (3 mm diameter) region gives <IMG WIDTH=10 HEIGHT=8 ALIGN=BOTTOM SRC="img5.gif"> = 20 ps.<P>The perspectives of using MCP-based detector for high-energy physics experiments are discussed.