3 The EXPLORER data

The Explorer detector, located at CERN, is a 2300 kg cylinder of aluminum alloy with primary quadrupole resonance close to 900 Hz equipped with a resonant capacitive transducer matched to a d.c. SQUID amplifier (Astone et al. 1993). The signal from the transducer is fed into a band-pass amplifier, sampled with a sampling time $\Delta t= 4.5$ ms, and analysed off-line with an adaptive Fast Matched (FM) filter (Astone et al. 1997). The squared output E of the filter is expressed in kelvin; its average value is called effective noise temperature $T_{\rm eff}$and represents the energy sensitivity of the detector for pulse detection, in the sense, that it gives the smallest energy variation that can be measured with SNR = 1. After the filtering, events were extracted with an adaptive threshold of $E_{\rm thr} \ge 19.5 \,T_{\rm eff}$ (Frasca 1997). The relation between the $T_{\rm eff}$ and the strain sensitivity $h_{\rm c}$ of the Explorer antenna, for a 1 ms conventional g.w. burst, is (Pizzella 1975; Bonifazi 1990): $h_{\rm c} = 8.1 \ 10^{-18} \cdot \sqrt{T_{\rm eff} \over \sin^4(\theta)}$where $\theta$ is the angle between the direction of the g.w. burst source and the antenna axis. The value of $\sin^4(\theta)$ was close to 1 over a period of 12 hours around the time of GRB 980425. The operation of the Explorer detector considering three days around the time of the burst, starting from day 114 UT, 1998 (April 24), was stationary. The average $T_{\rm eff}$ was of 10 mK, corresponding to an $h_{\rm c}$ (for best orientation) of 8 10^-19. The mean rate was of 5.7 events per hour and no events were observed, at this SNR level close, within a few minutes, to the GRB time. In a time interval of $\pm$ 5 min around the GRB 980425 trigger time, we found no events above threshold, with the largest peak (144 mK) occurring 33.31 s before (21.49.10.90847 UT) (see for example Kochanek & Piran 1993). In Fig. 1 we report the BeppoSAX GRBM data in the 40-700 keV band and the filtered Explorer data in $h_{\rm c}$ units. The difference between the initial time, obtained by fitting the light curve, and the trigger time, for this GRB, is of about 10 seconds.

  

Figure 1: BeppoSAX GRBM data in the 40-700 keV band and filtered Explorer data in $h_{\rm c}$ units, (local maxima in time bins of 0.29 s). The arrows indicate the initial time of the GRB, obtained fitting the light curve