Estimating adjustment factors to predict vibration at research facilities based on measurements in a subway tunnel

Abstract

The Sound Transit System in Seattle, Washington recently opened the University Link (U-Link) extension. U-Link extended the existing light rail transit (LRT) system to the southern tip of the University of Washington (UW) campus. The potential for the LRT system to generate vibration that interferes with vibration sensitive research at UW has been a topic of concern and debate for over 10 years. Sound Transit and UW negotiated a Master Implementation Agreement (MIA) that includes specific limits on the maximum vibration levels inside a number of University buildings that can be generated by LRT operations. The limits are in terms of 1/3 octave band vibration velocity levels (RMS) from 2 to 100 Hz. A unique feature of the MIA is that substantial financial penalties may be assessed if the MIA vibration limits are exceeded at any of the 24 buildings identified in the MIA. Part of the MIA agreement is that Sound Transit would install a vibration monitoring system in the subway that will identify all trains that generate vibration levels within 10 decibels of the limit (a warning) and will identify all trains that cause vibration levels that exceed the MIA limit (an alarm). The vibration monitoring system (VMS) consists of monitors at nine locations, three in the University of Washington Station, three in the crossover box immediately south of the UW Station, and three in the bored tunnel portion of the U-Link extension. This paper presents the approach taken to estimate the relationships between the vibration at the VMS sites and the vibration at the buildings identified by UW as vibration sensitive along with some of the unexpected results of the detailed vibration testing that was performed. The relationship between the vibration measured at a monitor and the vibration generated at a sensitive building site is referred to in this paper as a Vibration Adjustment Estimate (VAE). Some of the challenges to developing credible estimates of the VAEs were: (1) There were no 1/3 octave bands where train vibration at the measurement sites inside UW buildings had a clearly discernable effect on the ambient vibration; (2) For the 1/3 octave bands of 16 Hz and lower, the vibration at the VMS sites in the subway from train operations is often lower than the vibration from trucks and buses on nearby roadways; even though the roadway is 60 m from the subway and the VMS sites are at the subway wall about 3 m from the tracks; (3) In some cases, the agreed upon limits are lower than the existing ambient vibration; (4) Because of the long history of discussions about the potential for LRT vibration to interfere with vibration–sensitive research at UW, it was critical that our work be deemed satisfactory and credible by Sound Transit, the Sound Transit vibration consultants, UW, and the UW vibration consultants. The findings of this study are relevant to the next extension of the Sound Transit light rail system to the north (Northgate Link) because the subway for that extension passes substantially closer to many UW research facilities. Northgate Link is currently under construction and is scheduled to be completed in 2021. The findings also are relevant for other rail transit projects that would pass close to vibration sensitive research facilities.