Measuring longer objects

If the leading edge of the object reflections undergoes a further reflection from the loudspeaker and returns to the microphone before the original trailing edge arrives, then the measurement of the object reflections will be disturbed. For this reason the measurements presented so far have a time length of $2 l_1/c = 2 \times 7.37/343 = 0.043$secs where $l_1$ is the length of source tube between the microphone and the source (see figure 5.2). The length of instrument that can be measured must be less then $2 l_1$ since the primary reflections from the end of such an instrument would arrive at the end of our time window simultaneously with the source reflections. It is not realistic to expect that such a limit will be approached in practice, however, since we would lose the secondary reflections from the object which carry on after the last primary reflection has arrived. In practice the maximum length measurable depends both on $l_1$ and the radius of the object in question since long sections of large radius will mean low losses and many secondary reflections will return before all of the reflection sequence is finished.

Increasing $l_1$ will mean that the source reflections will arrive later and the measurement time window will be longer. The aim of this section is to measure long objects so we will therefore use a reflectometer with $l_1 = 9.68$m rather than the 7.37m used up to now. This option is not without its drawbacks. Firstly, since the losses in the source tube reduce all frequency components, a higher amplitude of sound must be input in order to preserve the signal to noise ratio on sampling the reflections. This is easily achieved by turning up the volume on the audio amplifier used in pulse production, although when taken to extremes this could damage the amplifier and speaker or cause non-linear sound propagation not accounted for in the analysis. Also, the high frequencies are attenuated proportionally more than the low frequencies meaning that the further a pulse travels, the wider it gets. The source tube section $l_2$ must therefore also be lengthened to allow the longer input pulse to completely pass the microphone before the object reflections arrive. For the reflectometer in this section $l_2=6.52$m was chosen (3.10m was used previously).