TY - JOUR
AU - Kavet,, Robert
AB - Abstract Measurements were conducted to investigate electric and magnetic fields (EMFs) from 120 Hz to 10 kHz and 1.2 to 100 kHz in 9 electric or hybrid vehicles and 4 gasoline vehicles, all while being driven. The range of fields in the electric vehicles enclosed the range observed in the gasoline vehicles. Mean magnetic fields ranged from nominally 0.6 to 3.5 µT for electric/hybrids depending on the measurement band compared with nominally 0.4 to 0.6 µT for gasoline vehicles. Mean values of electric fields ranged from nominally 2 to 3 V m−1 for electric/hybrid vehicles depending on the band, compared with 0.9 to 3 V m−1 for gasoline vehicles. In all cases, the fields were well within published exposure limits for the general population. The measurements were performed with Narda model EHP-50C/EHP-50D EMF analysers that revealed the presence of spurious signals in the EHP-50C unit, which were resolved with the EHP-50D model. INTRODUCTION A previous study reported on extremely low-frequency (ELF) magnetic fields in the 40–1000 Hz frequency range measured in a sample of electric and conventional gasoline vehicles(1). That study found that magnetic fields in both electric and gasoline-powered vehicles were similar in magnitude to those experienced routinely within residences in the USA, although the fields in the electric vehicles were moderately greater in magnitude compared with those in conventional gasoline vehicles. This article reports on the spectral content of both magnetic and electric fields of up to 100 kHz recorded in a small sample of vehicles that include all-electric, electric hybrid and conventional, gasoline-powered vehicles. The electric and magnetic field (EMF) environment of hybrid electric and all-electric vehicles is complex due to the variety of sources with different characteristics within the vehicles. The EMF environment of electric vehicles is likely to vary significantly as a function of spatial, temporal and frequency factors. Design parameters such as power cable routing and vehicle operational modes are important as well. EMF sources may include power cables, battery, drive motors, power-conditioning equipment, control circuitry, battery chargers and electrically driven accessories. The study's objective was to obtain insights into the peak magnitude of EMFs at frequencies of up to 100 kHz that may be present in electric and gasoline vehicles, and to compare these levels with EMF exposure limits published by the International Commission on Non-Ionizing Radiation (ICNIRP) and by the Institute of Electrical and Electronic Engineers (IEEE)(2–4). Because most of the vehicles included in this survey were rented at different locations in the USA, a test track or standard route was not available (the previous ELF study used a standardised route). Thus, each vehicle was driven on city streets and available highways during the measurement period. METHODS Instrumentation EMFs were measured with two different models of the Narda model EHP-50 EMF probe—analyser: an EHP-50C (SN 352WN01203) and an EHP-50D (SN 000WX10510). This device provides for isotropic measurements with a dynamic range of 140 dB (a 10-million-fold range of field magnitude) for EMFs (depending on the specific configuration of the device). A unique aspect of the EHP-50 instruments is their ability to simultaneously record each polarisation axis of the probe; switching time between the different axes is eliminated, removing the uncertainty associated with rapidly changing fields. Minimum detectable electric field strength is nominally 1 V m−1, and minimum magnetic field flux density is nominally 1 nT (0.01 mG). The EHP-50C and EHP-50D are battery powered and are connected to a personal computer (PC) via an optical fibre cable for spectral analysis. Built-in fast Fourier transform (FFT) spectrum analysis allows evaluation of the frequency content of the EMFs over the frequency range of 5 Hz to 100 kHz. Measured values of EMFs are displayed on the PC and saved to disc memory for subsequent analysis. Typical isotropy for magnetic fields is specified as 0.12 dB (variability of 1.4 % across all directions) and for electric fields as 0.54 dB (6.4 % variability). The associated software that operates with the analysers provides a ‘wideband’ value for the measured electric or magnetic field. This value is a representation of the overall field measured across the entire selected frequency range of the probe. Importantly, the wideband value does not include the contributions of amplitude values occurring below 1.2 % of the full-frequency scan. In the case of a 10-kHz scan, the first 120 Hz of amplitude data are omitted; for a 100-kHz scan, the first 1.2 kHz of data are omitted. This is to avoid the issue referred to as ‘local oscillator zero feedthrough’ common to all spectrum analysers. For all of the vehicle measurements, the laptop computer was operated in battery mode without any connection to the AC mains, and the laptop computer was kept as reasonably far from the sensor as possible. No inverters were used during the measurements for recharging of either the computer or the EHP-50C or EHP-50D. All data are presented as indicated by the output of the device with no further corrections for frequency response. Procedure EMFs were measured by placing the EHP instrument within each vehicle while driving on city streets and highways. During each set of measurements, the vehicle radio system was turned off to avoid artefact from the vehicle's speakers. During the driving sessions (or runs) when EMF data were acquired, the driver employed normal manoeuvres such as accelerating, braking and using directional signals. During the sampling period, each vehicle was driven at speeds of up to 60 mph. In view of the fact that the exposure limits for fields <100 kHz are based on virtually instantaneous values, all measurements reported are reported in terms of instantaneous wideband peak values as opposed to time-averages. In most cases, measurements of the ambient ‘background’ EMFs were performed prior to and immediately following each series of measurements on most of the vehicles. In some cases, only a single background measurement was performed. The background measurement was taken without the vehicle running and consisted of placing the sensor, for magnetic fields, typically on the floor of the vehicle, most commonly in the rear-seat area. For electric fields, the sensor was held with a dielectric rod such that the sensor was suspended between the roof of the vehicle and the seat. During all of the electric field measurements, the sensor was never allowed to come into contact with any surface of the vehicle. During each vehicle's measurement period, the sensor was connected to a small laptop computer for executing the software application that interacts with the field sensor. The laptop computer was held by the operator while sitting in the rear seat of the vehicle, and the sensor was moved about the vehicle cabin in an effort to sweep out as much of the vehicle cabin space as possible while the software program was configured in the maximum hold mode of operation. In this manner, as each vehicle was driven, the electric or magnetic field environment was sampled such that the greatest field would likely be sensed during the measurement process. In most cases, at least two separate measurements of both EMFs were conducted so that a measure of the repeatability of the measurement could be deduced. In this process, separate measurements of the EMFs were alternated in time for most but not all vehicles. Each measurement spanned ∼5–7 min during which the maximum detected fields were sensed, across the frequency spectrum of either 10 or 100 kHz and recorded for subsequent analysis and display. Data were collected over a period of a few months from the spring of 2011 through the fall of 2012. During the sampling period, each vehicle was driven at speeds of up to 96.6 km h−1 (60 mph). RESULTS The sample was comprised of four gasoline and nine electric or hybrid vehicles (Table 1). For magnetic fields, there were a total of 38 driving runs in 11 vehicles measured with a 10-kHz bandwidth [3.5 ± 1.1 (mean) ± 1.2 (standard deviation) driving runs per vehicle, dr v−1], and 35 runs in 13 vehicles measured with a 100-kHz bandwidth (2.7 ± 1.1 dr v−1). For electric fields, there were 30 runs in 11 vehicles measured with a 10-kHz bandwidth (2.7 ± 0.8 dr v−1), and 30 runs in 13 vehicles measured with a 100-kHz bandwidth (2.3 ± 0.9 dr v−1). For both background magnetic and electric fields, there were 20 measurements related to 11 vehicles (1.8 ± 0.4 background measurements per vehicle) at 10 kHz and 21 measurements related to 12 vehicles (1.8 ± 0.5 background measurements per vehicle) at 100 kHz. Table 1. List of vehicles included in EMF measurements. Vehicle code . Description . Date of manufacture . A Compact gasoline station wagon 1993 B Midsize gasoline SUV 2004 C Compact gasoline hatchback sedan 2011 D Midsize hybrid sedan 2010 E Compact hybrid sedan A 2010 F Compact hybrid sedan B 2010 G Compact all-electric sedan 2011 H Midsize hybrid sedan 2010 I Compact all-electric sedan 2010 J Compact hybrid sedan 2009 K Midsize hybrid sedan 2011 L Compact all-electric sedan 2010 M Compact gasoline sedan A 2012 Vehicle code . Description . Date of manufacture . A Compact gasoline station wagon 1993 B Midsize gasoline SUV 2004 C Compact gasoline hatchback sedan 2011 D Midsize hybrid sedan 2010 E Compact hybrid sedan A 2010 F Compact hybrid sedan B 2010 G Compact all-electric sedan 2011 H Midsize hybrid sedan 2010 I Compact all-electric sedan 2010 J Compact hybrid sedan 2009 K Midsize hybrid sedan 2011 L Compact all-electric sedan 2010 M Compact gasoline sedan A 2012 Table 1. List of vehicles included in EMF measurements. Vehicle code . Description . Date of manufacture . A Compact gasoline station wagon 1993 B Midsize gasoline SUV 2004 C Compact gasoline hatchback sedan 2011 D Midsize hybrid sedan 2010 E Compact hybrid sedan A 2010 F Compact hybrid sedan B 2010 G Compact all-electric sedan 2011 H Midsize hybrid sedan 2010 I Compact all-electric sedan 2010 J Compact hybrid sedan 2009 K Midsize hybrid sedan 2011 L Compact all-electric sedan 2010 M Compact gasoline sedan A 2012 Vehicle code . Description . Date of manufacture . A Compact gasoline station wagon 1993 B Midsize gasoline SUV 2004 C Compact gasoline hatchback sedan 2011 D Midsize hybrid sedan 2010 E Compact hybrid sedan A 2010 F Compact hybrid sedan B 2010 G Compact all-electric sedan 2011 H Midsize hybrid sedan 2010 I Compact all-electric sedan 2010 J Compact hybrid sedan 2009 K Midsize hybrid sedan 2011 L Compact all-electric sedan 2010 M Compact gasoline sedan A 2012 As the measurements proceeded, it became apparent that the EHP-50C was recording signals above background (with vehicle off) at ∼19 and 38 kHz; this same characteristic signal occurred during measurements of different automobiles and in different geographic locations. As an example, the magnetic field traces from Vehicle B (gasoline) indicate the spikes associated with these frequencies blended to some degree with the spectra during vehicle operation, but were clearly evident as anomalous with the vehicle off (Figure 1A). This vehicle when operated produced negligible electric fields across the spectrum, and the anomaly's fingerprint is obvious for all modes of operation (Figure 1B). The manufacturer confirmed that a spurious signal at these frequencies originated in the sensor, and the investigators were provided with an EHP-50D, which did not experience the problem. Figure 1. Open in new tabDownload slide Magnetic (A) and electric (B) fields measured in a gasoline vehicle with the Narda EHP-50C low-frequency analyser showing anomalous signals at ∼19 and 38 kHz. Figure 1. Open in new tabDownload slide Magnetic (A) and electric (B) fields measured in a gasoline vehicle with the Narda EHP-50C low-frequency analyser showing anomalous signals at ∼19 and 38 kHz. The noise floors of the two EHP-50 models were compared in a relatively ‘quiet’ office environment (Figure 2A–D). The two instruments clearly exhibit different noise floors, and the EHP-50C's spectra show the presence of spurious signals near 19 and 38 kHz. Also, except for frequencies <10 kHz for electric fields, the EHP-50D exhibits a lower noise floor. Of significance, however, is that (a) spectra acquired across the 0–10 kHz span with the EHP-50C, generally, appear essentially similar between the two instruments and (b) the strengths of the spurious signals are relatively near the noise floor. Thus, the investigators do not believe that the artifactual peaks at 19 and 38 kHz significantly biased the wideband values reported in this study, which is relevant here because data for 10 of the 13 vehicles in the study were acquired with the EHP-50C. Figure 2. Open in new tabDownload slide Background EMFs in an office measured with the Narda EHP-50C and EHP-50D low-frequency analysers across the 10- (A and B) and 100-kHz (C and D) bands. Figure 2. Open in new tabDownload slide Background EMFs in an office measured with the Narda EHP-50C and EHP-50D low-frequency analysers across the 10- (A and B) and 100-kHz (C and D) bands. There were no magnetic field wideband measurements at either 10 or 100 kHz considered as outliers. There were 5 electric field wideband readings during vehicle operation considered outliers: 16.5 V m−1 (wideband) at 10 kHz and 14.4 V m−1 at 100 kHz both in the same run of Vehicle G (of 3 runs total); 29.6 V m−1 at 10 kHz in 1 of 3 runs for Vehicle D, and 14.1 V m−1 at 100 kHz in a different run for Vehicle D; and 14.4 V m−1 for 1 of 5 runs in Vehicle J. There were 2 electric field wideband background readings considered outliers: 14.0 V m−1 in 1 of 2 readings in connection with Vehicle F (100 kHz), and 14.4 V m−1 in 1 of 2 readings in connection with Vehicle G (100 kHz). Thus, three of seven outliers were related to measurements of Vehicle G, and two related to Vehicle D. Within each frequency band, the spectra from vehicle operations with the maximum electric (including outliers) and magnetic field wideband readings were plotted against the general public EMF limits for frequencies <100 kHz published by the ICNIRP and the IEEE (Figure 3). The maximum 10 kHz electric field was for Vehicle D, Run 1 consisting of four sweeps, whose spectra (Figure 3A) had a composite wideband field of 29.6 V m−1. For the 100 kHz band, the maximum was recorded for Vehicle G, Run 2, with a composite of 14.4 V m−1 from four sweeps (Figure 3B). In both cases, two of four sweeps displayed a ‘double hump’ in the lower segment of their respective spectra. Figure 3. Open in new tabDownload slide Maximum measured electric (A and C) and magnetic (B and D) fields across the 10- and 100-kHz bands. The IEEE and ICNIRP general public exposure limits are shown for comparison. Figure 3. Open in new tabDownload slide Maximum measured electric (A and C) and magnetic (B and D) fields across the 10- and 100-kHz bands. The IEEE and ICNIRP general public exposure limits are shown for comparison. Evidence of a double hump was also evident in background electric field 100 kHz readings of Vehicles F and G—that is, with the vehicle off—which argued against its presence as due to vehicle operation. Figure 4A illustrates that for Vehicle G, the 0–100 kHz Background B (after the vehicle was operated) and Run 2 magnetic fields were essentially identical. The readings for Background A, prior to vehicle operation, were at a level one would expect in the absence of any obvious sources. Vehicle F had similar results with respect to the before-and-after background readings (Figure 4B). The two runs for Vehicle F were not considered of the magnitude to qualify as outliers, nor were their spectra consistent with the same double hump. Magnetic field traces in four vehicles during vehicle operation also exhibited double humps (1 at 10 kHz, 3 at 100 kHz, see Figure 3D as an example), but were not considered of a magnitude to be classified as outliers; although not classified as outliers, it is possible that the magnetic field double humps were unrelated to vehicle operation. Figure 4. Open in new tabDownload slide The presence of a double hump in the 100 kHz electric field frequency spectra for Vehicles G (A) and F (B). See text for discussion. Figure 4. Open in new tabDownload slide The presence of a double hump in the 100 kHz electric field frequency spectra for Vehicles G (A) and F (B). See text for discussion. For the analysis comparing electric with gasoline-powered vehicles, the field value assigned for each vehicle was estimated as the average of its wideband readings across all runs after excluding outliers, as described above. Background fields were pooled from all vehicles. Because of limited sample size, it was opted to plot the minima, mean and maxima for electric fields and magnetic fields during vehicle operation, as well as EMF background fields, all with both the 10- and 100-kHz bandwidths. The results did not discriminate between fields measured with the EHP-50C and EHP-50D. The plots (Figure 5) show that the mean background fields were all lower than the mean EMFs measured during vehicle operation, but with some degree of overlap of the background with vehicle operation ranges. The mean values of EMFs for the electric vehicles tended to be greater than for gasoline vehicles, with clear overlap of the ranges. As indicated earlier, none of the measurements exceeded either ICNIRP or IEEE EMF exposure limits for the general public. Figure 5. Open in new tabDownload slide Minimum, mean and maximum EMFs measured in electric and gasoline vehicles across the 10- (A and B) and 100-kHz (C and D) bands; also shown is the range of background fields measured. Figure 5. Open in new tabDownload slide Minimum, mean and maximum EMFs measured in electric and gasoline vehicles across the 10- (A and B) and 100-kHz (C and D) bands; also shown is the range of background fields measured. DISCUSSION This study, focussed on measuring EMFs up to 100 kHz in electric and gasoline-powered vehicles, was conducted as a follow-up to a previous study that reported ELF magnetic fields measured in both classes of vehicles, also while driven(1). That study, which included eight electric and six gasoline-powered vehicles, employed six EMDEX Lite data loggers within each vehicle that record within a bandwidth of 40–1000 Hz. This study had a similar sample size—nine electric and four gasoline-powered vehicles—with measurements taken with a single Narda EHP-50 instrument used to scan the vehicle interior within bandwidths of 120 Hz to 10 kHz and 1.2–100 kHz. The spectra were highly variable and complex, with signatures for which background fields (i.e. unrelated to vehicle operation) were not readily discriminated from those associated with the vehicle itself. Nonetheless, even for the spectra with the highest fields regardless of source, there were no indications of fields approaching ICNIRP or IEEE EMF exposure limits for the general public. The data left an impression of marginally higher fields associated with the electric vehicles, but the maximum–minimum field ranges for the electric fields enclosed those shown for the gasoline-powered vehicles in both frequency bands (Figure 5). Vassilev et al.(5) reported on magnetic fields of up to 10 MHz in 8 electric and 3 gasoline-powered vehicles, with qualitatively similar results for the frequency bands in common with the study reported here. The study described in this article took the approach of an environmental characterisation, whereas Vassilev et al. included elements of source characterisation indicating the highest fields near the battery and close to the feet. Even though the magnetic fields near the battery and feet were ≤20 % of ICNIRP's general public limit, it is not clear under what scenario people would be exposed near the battery on any kind of regular basis. With respect to exposures near the feet, IEEE has far less stringent exposure limits for arms and legs, compared with organs and tissues in head and torso: 3.79 f−1 mT−1 for 10.7 Hz to 3 kHz (f in Hz), and 1.13 mT for 3 kHz to 5 MHz; ICNIRP limits have no such distinction of extremities from head and torso. Finally, and perhaps most importantly, this study provides a reminder of checking instrumentation for possible artefact. After spikes at ∼19 and 38 kHz were discovered, the manufacturer became involved in confirming the presence of artefact and providing a redesigned probe that eliminated the issue. Although the artefact did not affect the results or conclusions to any significant degree, investigators are advised to not automatically take instrumentation performance for granted. To conclude, though limited by a small sample size of vehicles (but similar to those of preceding studies(1, 5)), this study revealed the apparent high variability of EMFs associated with their operation, both in terms of magnitude and spectral distribution. If further investigation of electric vehicle fields is pursued, an emphasis on measurement protocol is suggested as a means for reducing potential variation in recorded field values during operation of the vehicles. FUNDING This study was supported by the Electric Power Research Institute. REFERENCES 1 Tell R. A. , Sias G., Smith J., Sahl J., Kavet R. ELF magnetic fields in electric and gasoline-powered vehicles . Bioelectromagnetics. 34 , 156 – 161 ( 2013 ). Google Scholar Crossref Search ADS PubMed WorldCat 2 ICNIRP . Guidelines for limiting exposure to time-varying electric and magnetic fields (1 Hz to 100 kHz) . Health Phys. 99 , 818 – 836 ( 2010 ). PubMed OpenURL Placeholder Text WorldCat 3 IEEE . IEEE standard for safety levels with respect to human exposure to electromagnetic fields, 0–3 kHz . Institute of Electrical and Electronic Engineers , Report No. IEEE Std. C95.6 ( 2002 ). 4 IEEE . IEEE Standard for safety levels with respect to human exposure to radio frequency electromagnetic fields, 3 kHz to 300 GHz . Institute of Electrical and Electronic Engineers , Report No. IEEE Std. C95.1 ( 2005 ). 5 Vassilev A. , Alain Ferber A., Wehrmann C., Pinaud O., Schilling M., Ruddle A. R. Magnetic field exposure assessment in electric vehicles . IEEE Trans. Electromag. Compat. 57 , 35 – 43 ( 2015 ). Google Scholar Crossref Search ADS WorldCat © The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
TI - ELECTRIC AND MAGNETIC FIELDS <100 KHZ IN ELECTRIC AND GASOLINE-POWERED VEHICLES
JF - Radiation Protection Dosimetry
DO - 10.1093/rpd/ncv533
DA - 2016-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/electric-and-magnetic-fields-100-khz-in-electric-and-gasoline-powered-v48OFGhJar
SP - 541
VL - 172
IS - 4
DP - DeepDyve
ER -