Pre-Construction Sound Study Substation DRAFT January 2011 D A T A AN AL Y S IS S OL U T I ON S
TABLE OF CONTENTS 1.0 INTRODUCTION...1 2.0 SOUND LEVEL MONITORING...1 3.0 SOUND MODELING...4 3.1 Modeling Software... 4 3.2 Modeling Results... 5 4.0 CONCLUSIONS...7 7 January 2011 Page i
LIST OF FIGURES Figure 1: Site Map with Sound Monitoring Locations and Proposed Substation Layout... 1 Figure 2: Hourly Sound Pressure Levels at Monitor A (Powerline ROW)... 2 Figure 3: Hourly Sound Pressure Levels at Monitor B (Woods)... 2 Figure 4: Modeled Post-Construction Levels, Transformer Fans Off... 6 Figure 5: Modeled Post-Construction Levels, Transformer Fans On... 7 7 January 2011 Page ii
1.0 INTRODUCTION Velco proposes to construct an electrical substation at a greenfield site in Jay, VT. The substation will include two 115/46 kv transformers as well as ancillary equipment with negligible noise emissions. In October 2010, RSG conducted preconstruction sound monitoring at two locations near the proposed location. After analyzing this data, RSG modeled post-construction sound levels for the substation. This report includes the results of pre-construction noise measurements and the predicted noise levels. 2.0 SOUND LEVEL MONITORING To determine ambient sound levels in the area, RSG staff visited the site in October, 2010, to set up two long-term sound monitors, which recorded data from 5pm on 21 October until 8:30am on 27 October. A site map provided in Figure 1 shows the locations of the monitors. Monitor A was set up in a powercut to the southeast of the proposed substation, about 300 feet from the nearest transformer location and 425 feet from Route 105. Monitor B was set up in a wooded area to the northeast of the proposed substation, about 0 feet from the nearest transformer location and 100 feet from Route 105. Figure 1: Site Map with Sound Monitoring Locations and Proposed Substation Layout 7 January 2011 Page 1
Hourly Sound Pressure Level (dba) Hourly Sound Pressure Level (dba) Equipment used included two ANSI Type 2 Rion NL-22 sound level meters set to log equivalent sound levels every one second and to record audio files every 30 minutes and during periods when sound levels exceeded 60 dba. Results are shown in Figure 2 and Figure 3 and summarized in Table 1. Figure 2: Hourly Sound Pressure Levels at Monitor A (Powerline ROW) 70 Leq L90 65 60 55 45 35 30 Figure 3: Hourly Sound Pressure Levels at Monitor B (Woods) 70 Leq L90 65 60 55 45 35 30 7 January 2011 Page 2
20 25 31.5 63 80 100 125 160 200 2 315 0 0 630 800 1000 12 1600 2000 20 31 00 00 Sound Pressure Level (db) Table 1: Sound Monitoring Summary (dba) 1 Monitor Daytime Nighttime Leq L90 L L10 Leq L90 L L10 A 54 38 45 57 49 39 41 52 B 55 34 46 59 36 39 53 Over the monitoring period, temperatures ranged from 30 to 69 degrees Fahrenheit, with an average of 46 degrees 2. Light rain occurred on 24 and 27 October. In addition, the weather leading up to the monitoring period was quite wet and caused a nearby stream to swell. The stream was located 0 feet southwest of Monitor A and could be heard at this site. The main sources of noise in this area were the stream, vehicular traffic, aircraft, biogenic sources (birds, etc), and weather-related sources (wind and rain). Over the monitoring period, the two monitors exhibit similar trends in L90 sound levels. They also exhibit similar peaks in Leq sound levels, though Monitor B generally has higher Leq levels because of its proximity to the traffic noise from Route 105. This is especially true during the daytime. The first three days of monitoring showed higher levels due to wind. In addition, 1/3 octave band levels were analyzed from the audio recordings. The results of this analysis are shown in Figure 1Figure 4 and Figure 5. Figure 4: 1/3 Octave Band Spectrum for Monitor A (Powerline) at 12:42am on 26 October 70 60 30 20 10 1/3 Octave Band Center Frequency (Hz) 1 Daytime is 7am 10pm. Nighttime is 10pm 7am. 2 Met data was retrieved from WeatherUnderground.com on 6 January, 2011 7 January 2011 Page 3
20 25 31.5 63 80 100 125 160 200 2 315 0 0 630 800 1000 12 1600 2000 20 31 00 00 Sound Pressure Level (db) Figure 5: 1/3 Octave Band Spectrum from Monitor B (Woods) at 12:35am on 26 October 70 60 30 20 10 1/3 Octave Band Center Frequency (Hz) These levels reflect the presence of wind- and water-related noise. 3.0 SOUND MODELING 3.1 Modeling Software Modeling was completed for the project using Cadna A acoustical modeling software. Made by Datakustik GmbH, Cadna A is an internationally accepted acoustical model, used by many other noise control professionals in the United States and abroad. The software has a high level of reliability and follows methods specified by the International Standards Organization in their ISO 9613-2 standard, Acoustics Attenuation of sound during propagation outdoors, Part 2: General Method of Calculation. This is also the modeling methodology specified in the Michigan Siting guidelines. The ISO standard states, This part of ISO 9613 specifies an engineering method for calculating the attenuation of sound during propagation outdoors in order to predict the levels of environmental noise at a distance from a variety of sources. The method predicts the equivalent continuous A-weighted sound pressure level under meteorological conditions favorable to propagation from sources of known sound emissions. These conditions are for downwind propagation or, equivalently, propagation under a well-developed moderate ground-based temperature inversion, such as commonly occurs at night. The model takes into account source sound power levels, surface reflection and absorption, atmospheric absorption, geometric divergence, meteorological conditions, walls, barriers, berms, and terrain. A 10-meter by 10-meter grid of 4.0 meter high receivers was set up in the model covering 4.6 square miles around the site. This accounts for a total of about 1,198,270 modeled receivers. A receiver is a point above the ground at which the computer model calculates a sound level. In addition, discrete receivers were placed at 85 residences around the substation. The equipment to be used at Jay are 115/46 kv, 56 MVA autotransformers with a BIL rating of 5 kv. This a similar transformer as T1 at Velco s Blissville substation. The transformers are equipped with fans and can be operated with these fans off (ONAN mode) or on (OFAF mode). We modeled the sound pressure levels for each mode. The transformers were modeled as point sources with a height of 2.5 meters. The transformers are modeled as 15 db below the NEMA TR1 standard. This equates to a sound pressure level of 59 dba with 7 January 2011 Page 4
fans off and 62 dba with fans on. Spectral characteristics were derived from previous sound pressure and power measurements taken at the Blissville substation. Table 2 shows the transformer sound power levels by 1/1 octave band. Table 2: Modeled Sound Power Spectrum (dba) Unit Octave Band Frequency (Hz) 31.5 63 125 2 0 1000 2000 00 8000 T1 and T2, ONAN 68 65 72 73 66 56 49 46 77 T1 and T2, OFAF 62 68 80 72 68 65 60 55 45 81 The model used spectral ground attenuation. Ground outside of the substation was considered soft (G=1.0). Atmospheric absorption was based on meteorological conditions of 10 degrees Celsius and 70% relative humidity. 3.2 Modeling Results We ran the model for the proposed substation configuration with the transformer fans turned off and turned on. Results are displayed in Figure 6 and Figure 7 as noise contour maps. Within each figure, the lines emanating from the transformers are color-coded noise isolines, where dark red represents the highest sound level and blue represents the lowest. With transformer fans turned off, the highest sound pressure level at a residence was 20 dba. With transformer fans turned on, the highest level was 24 dba. These levels are at least 10 dba below the nighttime L90 sound levels recorded at Monitors A and B. dba 7 January 2011 Page 5
Figure 6: Modeled Post-Construction Levels, Transformer Fans Off 7 January 2011 Page 6
Figure 7: Modeled Post-Construction Levels, Transformer Fans On 4.0 CONCLUSIONS RSG conducted pre-construction sound monitoring at the substation and modeled sound levels for a proposed substation with two transformers. We modeled sound levels when the transformer fans are turned off and on. Our results indicate that sound levels at the nearest residence are well below the nighttime L90 measured during the monitoring period. Despite the increased background sound levels due to the wet season, the project sound levels would still be relatively low at other times. The project will therefore not have an undue adverse impact on aesthetics with regard to noise. 7 January 2011 Page 7