MILOM
Optical module timing precision

Next : Acoustic positionning system resolution

Last update : 01/09/2009

 

The excellent angular resolution of the ANTARES neutrino telescope (<0.3 for E > 10 TeV) relies on good timing resolution of the light signals recorded in the optical modules. The specification for the detector is that the timing resolution is limited by the transit time spread of the photomultipliers which have s ~1.3 ns. To achieve this all electronics and calibration systems are each required to contribute < 0.5 ns to the overall timing resolution.

Reference Clock Calibration

To obtain this necessary resolution in the electronics, a very precise time reference clock distribution system has been implemented. Figure 7 shows the variation in time delay measured with the clock calibration over the distance of 45 kilometres between the shore station and the SCM (String control module) at the bottom of the MILOM. Up until the present time, the standard mode of operation has been to turn off the power during the nights and weekends. This explains the gaps in the measurements and also the low points at the beginning of every operation period when the electronics temperature stabilises. The full span of the spread of the data in the plot over the two month periods of operations is seen to be ~1 ns.

Figure 8 shows the variation in the measured delay in the clock system within the MILOM between the SCM and the top storey.

(Figure 9 is not available at the moment)

Figure 7. Measurement of variation of delay in the clock system between the shore station and the SCM.

Figure 8. Measurement of the variation of the time delay in the clock system between SCM and TOP LCM, as a function of time during the operation of the MILOM line.

LED Beacon measurements of OM timing resolution

The complete timing resolution of the optical modules has been measured with the MILOM using the optical beacon system. As indicated in figures 3 and 10, the LED beacon is located in the first storey and the three optical modules in the second storey at a distance of about 15 metres. The LED beacon contains 60 individual LEDs synchronised in time and arranged to give a reasonably isotropic light emission. A small PMT internal to the LED beacon monitors the output light pulse time spread and amplitude.

Figure 11 illustrates the rise time of the emitted light pulse as recorded in this PMT and the typical risetime can be seen to be ~2.8ns.

Figure 10. Arrangement in sea of the timing calibration system with the optical beacon and the three optical modules

Figure 11. Data from the internal reference PMT in the LED beacon. The insert shows a typical signal waveform and the histogram shows a distribution of the risetime of the leading edge of the light pulse.

The test on the timing resolution in the optical modules is performed by pulsing the LED beacon with a frequency of a few Hz. Figure 12 shows the time distribution of signals in the three optical modules relative to the reference PMT in the LED Beacon. For events where two or more optical modules in the same storey record signals within a certain time window, figure 13 shows the difference between the signals of two adjacent OMs.

The timing resolution of one optical module can be estimated to be ~0.5 ns from figures 12 and 13; for figure 12 the contribution to the width is small from the reference PMT and so the width of the distribution is dominated from one OM whereas for figure 13, two OMs contribute such that the width is v2 0.5 = 0.75ns. This resolution is for large light pulses and so is not dominated by the PMT transit time spread but by the intrinsic electronics resolution. A detailed analysis of the timing resolution with varying light intensity is in progress to separate the various contributions, but this result already shows that the complete electronics contribution is <0.5ns.

Figure 12. Examples of the distribution of time of a signal in optical modules relative to the reference signal of the PMT in the LED Beacon after subtraction of an offset for the light propagation.

Figure 13. An example of the distribution of difference in time between signals between adjacent optical modules in the storey.

Figure 14 displays the data taken on different days with the LED Beacon system in the sea and shows on the same plot the analogous data taken in the laboratory dark room with laser light fed to the OMs by optical fibres tuned to have the same length. These data indicate that the relative time offsets in the optical modules are stable to ~0.5ns both in the sea and between the dark room calibration and the sea.

 The full optical beacon calibration system in the detector will contain both the beacons using LEDs, as in the data shown above, and beacons using lasers. The MILOM contains both types. The laser beacon on the BSS (Bottom string socket, i.e. the part near  the anchor) has so far not been used for a time calibration of the OMs but figure 15 shows the quality of the emitted light pulse as recorded in the internal reference light diode.

Figure 14. Data on the stability of the timing calibration from the LED Beacon system in the sea and the optical fibre/laser system in the dark room before deployment. The points are the means of distributions of the difference in time of hits in two adjacent OM. For this plot OM1 is the zero reference.

Figure 15. Data from the internal reference diode in the Laser beacon on the BSS of the MILOM. The insert shows a typical signal waveform and the histogram shows a distribution of the risetime of the leading edge of the light pulse.

Timing stability monitoring with OM internal LED

Each optical module contains an internal LED in order to monitor the stability of the photomultiplier. This LED is mounted on the back of the photomultiplier and illuminates the photocathode from behind as shown in figure 16.

In figure 17 a sample timing LED distribution from the OM internal LED in one optical module is shown. Figure 18 shows the mean of the peaks in such distributions for two optical modules on various days, both before deployment in the laboratory and in situ in the sea. It can be seen that a shift of ~3ns takes place between the shore and sea data, most likely due to temperature effects in the LED pulsing circuit.

Figure 16. Optical Module showing location of internal LED and the illumination of the photocathode.

Figure 17. For the OM internal LED, the distribution of the arrival time of the light pulse recorded in the PMT, relative to the time at which the LED was pulsed.

Figure 18. Data from the OM internal LED system. The points are the means of distributions such as that in figure 17, on different days both before and after the line was deployed in the sea.

Next : Acoustic positionning system resolution

Author : Thierry Stolarczyk