DCC performed vibration analyses to determine potential vibration effects of a proposed rail line that would pass by NASA's Neutral Buoyancy Lab in Houston, Texas. The NBL (shown in Figure 1) is a large pool that contains full-scale models of space vehicles where astronauts train underwater to simulate the effects of weightlessness. NASA was concerned that ground-borne noise, caused by train vibration, might interfere with radio communication with astronauts and divers training in the NBL.
Figure 1. NASA Neutral Buoyancy Lab-Underwater Sound Measurements
DCC employed 'transfer mobility' techniques whereby vibration measurements of a real train at one site can be 'transferred' to another site where no train activity exists. Vibration propagation through the ground can differ substantially between two different locations depending on soil types, and differences in geology.
The differences in soil conditions between two sites can essentially be eliminated by imparting a known force into the ground and adjusting the train vibration measurements based on the measured differences in vibration propagation characteristics. Figure 2 shows a seismic shaker that DCC used at the NASA site to impart up to 50,000 lbs of force into the ground.
Figure 2. Seismic Shaker
Based on the measured vibration propagation data at the two sites, DCC developed one-third octave transfer functions, which coupled with the train vibration measurements result in a predicted train vibration spectrum at the NASA site. DCC also measured shaker-induced vibration on the pool structure and concurrently measured underwater sound using a calibrated hydrophone.
Figure 3 shows the results of the analysis, which indicate the following: 1) underwater sound induced by ground-borne vibration will likely be lower than ambient sound levels. Cranes, diver respirators, and other equipment generate significant underwater sound, which would mask train-induced sound. 2) Train-induced sound will be likely lower than the threshold of hearing underwater. 3) Normal conversation levels via radio communication with divers would be substantially greater than ambient levels in the pool. Figure 3 shows an example human speech spectrum adjusted upwards by 62 dB* to account for the difference in reference pressures and impedance between air and water.
*To convert sound pressure level in air to water, two correction factors are added- one for the different reference pressures, and one for the difference in impedance. Thus dB=20log(pwater/1ÁPa)=20log(20)= +26 dB, and dB = 10log(Zwater/Zair)=10log(3600)=36 dB. Correction = 36+26 =62 dB