The Galileo Energetic Particles Detector

Galileo EPD Handbook

Chapter 1. Instrument Summary

Original, Pre-Challenger Time-of-Flight Subsystem (continued)

The K element is superfluous since Tm will be exceeded if there is no K (or stop) pulse. The J component of the J · (Tm) pulse is formed from the pulse which is used to create the pile-up rejection window.  The Tm level is set for a 115 ns threshold.  The J · (Tm) pulse is approximately 4.5 microseconds wide.

The example at the bottom of Table 19 shows how the J · (Tm) pulse appears when an invalid event (where T>Tm) is detected.  A spike appears on the J · (Tm) output, and then the output goes high after approximately 4 microseconds as shown in the illustration.  Since the strobe pulse from the rate logic appears well within 4 microseconds of the event, an invalid event will always be indicated, even if the last part of the J · (Tm) pulse moves slightly because of temperature variations.

Table 20 summarizes the Time of Flight system performance.  With a preliminary set of delay lines the asymptotic FWHM response of the J and K channels at high energies is 250 pS, which is close to the target value.

TABLE 20.  Summary TOF Performance, E.M. with Preliminary Delay Lines

A linearity check was performed by fitting a curve to the 100 ns point and the extrapolated 0 ns point. The greatest deviation from the straight line was 1.5 ns. The best fit would have been plus or minus 750 pS.

If a pileup is detected, a 4.5 microsecond timing pulse, sent to the rate logic, blanks out the processing of any events during that timing pulse. If a pileup reject signal was generated in the 5 microseconds to 4.5 microseconds window, the circuitry would be ready for a new event at the end of the pileup reject window.

The TOF delay lines must have transmission and amplitude characteristics that will give the pulse shaping circuit an optimum signal to noise ratio. The delay lines must also be closely matched, so that the delay characteristics of the J channels are the same. One of the engineering model delay lines has been adopted as a reference. Figures 1-30 and 1-31 show tests conducted on the delay lines. A test setup consisting of a tunnel diode pulser with a 100 pS rise time was used to drive the delay lines through resistive matching networks. The delay lines have a 5 ns tap and there is an additional 18 ns of delay from the tap to the end of the delay line.

Figure 1-30. Comparison of delay line #1 with ref. delay line (J).

Reference delay line (first APL board): 5 ns: 3.9 Ω 18 ns: 14.9 Ω   Delay line #1: 5 ns: 4.2Ω 10 ns: 9.1Ω

Figure 1-31. Delay lines #2 and #3.

Tektronix FET probes (900 MHz response, 2 pF capacitance) were used to obtain the photographs in the figures. The photographs show that the delay lines match within about 100 pS. The amplitude of line no. 1 is off by about 5% due to a difference in the amount of copper in the printed circuit tracks used to make the delay lines. However, the amplitudes of the delay lines shown are close enough for the engineering model. Figure 1-31 shows the test results for delay lines no. 2 and no. 3. The amplitude variations are smaller, since the resistances are almost identical. There is about a 300 pS mismatch in delay for lines no. 2 and 3. Dr. McEntire commented that the delay variations shown were acceptable, but that they should not be any greater, because of other TOF error sources in the system.

Next: Calibrations 

QUICK FACTS

RELATED LINKS