A test procedure has been devised to increase the accuracy with which lightning currents on a protective wire can be determined from raw current measurements. The procedure was conceived specifically for determining lightning currents on a steel cable used to protect the space shuttle launch pad against direct lightning strikes. The cable is hung between (a) a mast on top of the launch pad and (b) two grounding points about 1,000 ft (≈300 m) away from the mast.
The measurements are made by use of current sensors at the grounded ends of the cable. The measured currents are distorted versions of the lightning currents in the following sense: Each section of the cable is, in effect, a lossy transmission line of characteristic impedanceZ0 terminated in impedances different from (and generally smaller than)Z0. The mismatches between impedances give rise to reflections at the grounded ends and at the mast, so that the measured currents are superpositions of incident and reflected currents. The measurements are further complicated by attenuation (both ohmic and radiation losses) of currents that have traveled along the cable to the measurement points. In general, the characteristic impedance, the ohmic and radiation losses, and the terminating impedances are unknown and are functions of signal frequency.
The problem thus becomes one of determining frequency-dependent reflection coefficients, then using these coefficients to construct a transfer function that expresses the relationship between the raw current measurements and the incident lightning current. The transfer function can then be used, in turn, to correct the measurements for reflections and attenuations to obtain more-accurate estimates of the incident lightning current. The present test procedure (see figure) yields the needed reflection-coefficient information, without need for explicit knowledge of the unknown impedances and losses. The steps of the procedure are the following:
- One end of the cable is disconnected from ground, creating an open-circuit (infinite-impedance) termination. The other end is connected to ground through a 50-Ω resistor.
- A pulse with a duration of about 250 ns is applied across the resistor. About 5 µs later, this pulse returns after reflection from the open-circuit end. Because of losses, the amplitude of the returned waveform is less than that of the applied waveform. An oscilloscope at the resistor is used to observe and record the applied and returned waveforms.
- The returned waveform is Fourier-transformed to obtain a complex spectral amplitude Sopen(ω), where ω ≡ 2π× frequency.
- The end opposite the resistor is connected to ground, creating a nearly short-circuit (low-impedance) termination.
- A measurement like that of step 2 is performed. Because the impedance of the termination is less than Z0, the polarity of the returned waveform is the reverse of that observed with the open-circuit termination.
- The returned waveform is Fourier-transformed to obtain second complex spectral amplitude, Sgrounded(ω).
- The ratio Sgrounded(ω)/Sopen(ω) is then calculated. This ratio is one of two desired frequency-dependent reflection coefficients.
- The terminations and test equipment at the two ends of the cable are interchanged and steps 1 through 7 are performed to obtain the other frequency-dependent reflection coefficient; that is, Sgrounded(ω)/Sopen(ω) for measurements from the opposite end.
Pulses Are Applied at One End of the cable and measured after reflection from the other end of the cable with grounded and open-circuit terminations.
This work was done by Pedro J. Medelius formerly of I-NET, Inc., for Kennedy Space Center.
Inquiries concerning rights for the commercial use of this invention should be addressed to
the Technology Programs and Commercialization Office
Kennedy Space Center
Refer to KSC-11952