Electromagnetic noise level measurements on 3.5 MHz HF band during car drive

Published on: 2022/01/01 Updated on: 2022/02/05 Rev.1.2

Author: Chavdar Levkov LZ1AQ

I have made some measurements of the band noise level during car drive. On the top of the car was placed an active E-field antenna - a whip 1.2 m long with magnetic holder. The antenna base was connected to a ground square plate of copper foil with area of several square dm placed on the car roof in order to increase the capacitance between antenna base and the car compartment (approx 150 pF). AAA-1 amplifier in dipole mode was used (JFET High Z amplifier) https://active-antenna.eu/. The output internal noise floor of this amplifier is -116 dBm @ 1KHz BW. The band noise level was measured with Perseus SDR at 1 KHz bandwidth, RMS amplitude. The notebook, SDR, and AAA-1 amplifier were powered by a separate battery. The car was a diesel engine with no ignition noise. The noise difference when the engine was on and off was negligible. Records were performed with Perseus SDR with 1MHz bandwidth at 3.5 MHz center frequency during the drive. Later on, replaying the records, the noise level was measured at clean frequencies where there are no stations. A software was written which samples every 2 seconds the noise level from Perseus S-meter and produces xls file. Each sample consists of 3 noise level measurements and a their median is taken as a final value. The Perseus S-meter data are with 0.7 dBm resolution.

Fig.1 Ambient noise level at the active antenna amplifier output in dBm during car drive. The road passes through rural and urban areas. The mean car velocity was 70 km/h but in urban areas it was reduced to 40-50 Km/h so the X scale in distance(m) is variable. -73 dBm is equal to S9 S-meter units. -109 dBm= S3 and the peaks are S9+10 dB. This record was made on 14 Feb. 2020.

The first record Fig. was made at 14:30 local time. The length of the record was 45 minutes. Usually the atmospheric noise at this time is low, so the main contribution is from man made noise. The peaks in the chart are when the car passes urbanized areas – small villages and separate houses. A large and broad peak is observed when the car is near the 220 KV power line. Minimal detected noise level was -109 dBm at rural places far from any infrastructure. At my house I have 8 dB excessive noise compared to the lowest rural noise. The sharp peaks are near the air cables fixed on the poles on the streets. These are usually 220V mains, UTP (for I-net) and coaxial TV cables. I noticed that 10 to 20 m away from these cables the noise level drops sharply.

Fig.2 Ambient noise level at the active antenna amplifier output in dBm during car drive in relatively rural area. The mean car velocity was 70 km/h (+- 10 Km/h) so the X scale in meters is approximately linear. This record was made on 13 Feb. 2021.

The second record (Fig.2) was made at 12:00 local time. The length of the record is 13 minutes. Minimal noise is -112 dBm. The X scale is the recording time in seconds. The peak level of -65 dBm is at the time when the car passes under 220 KV power line. There is a power plant approx. 3 Km away at this point.

It is interesting to compare the level of the minimal rural noise obtained in our experiments to that given by ITU [1,9]. The very approximate antenna factor AF of AAA-1C (E-dipole mode) with 1m monople (or 2x1 m dipole) is 2 dB[m-1] https://active-antenna.eu/tech-docs/aaa-1c-addendum.pdf. The basic computation equation for 1 KHz bandwidth is:

(1) E [ dB uV/m ] = Po [dBm] + 107 + AF[dB m-1]
(explanation: in logarithmic values U [dBuV] = P [dBm] + 107; for Z=50 ohms)

If we assume that the minimal rural noise level is -110 dBm and AF= 2 then the conversion equation becomes

(2) E [ dB uV/m ] = Po [dBm] + 109 and for -110 dBm E = -1 [dB uV/m] at 1 KHz bandwidth

On Fig.3 is given a chart from ITU reports for 1 KHz bandwidth. At 3.5 MHz it is somewhere between -2 to 15 dBuV/m. Our results are somewhat close bearing in mind that the real AF of my setup is not known just calculated with idealized model presumptions, but in any case some rough comparisons can be made with published results.

Fig.3 Electric field strength versus frequency for atmospheric and cosmic noise (Daytime). Taken from [9]. The curves are for different noise parameter, which depend on geographic locations [1]. Europe is somewhere between 60 and 80. Bear in mind that these measurements are performed before "switched power supplies and digital transmissions" era

A demonstration of the near field noise differences are shown on Fig.4. Placing the antenna at different locations in my small yard give noise level differences up to 5 dBm. I am surrounded with other houses.

Fig.4 Relative noise level in dBm at 6 different points in the yard of my house on 80m band. The noise level was measured with the same setup as above but the whip was placed on the grass on a metal plate 40x 40 cm.


  1. The minimal noise level can be measured and compared with the noise level at your house to estimate how noisy is your QTH. I am " loosing" 8 dB S/N ratio at my house.
  2. Even several Km away from the high voltage (220KV) power line there is a noise contribution.
  3. The hills have screening effects to high voltage power lines noise. Note that at the top of the hill (where there is a vision to the power plant) the noise levele is 4 dBm higher
  4. The noise level in certain points at urban areas can be up to 50 dB higher than the rural noise.
  5. Even at "quiet" urban locations the expected noise level will be 5 to 10 dB higher than the minimal rural noise.
  6. The noise emitted from the air street cables (which is near field) drops sharply 10 to 20 m away from them. The noise origin is probably from switched power supplies or what else might come in our mind. PLT systems are very rare here. At distances above 50 m away this noise is not detected.
  7. One of the way to reduce this (near field) type of noise is to use compact receiving antennas. Then every meter away from the noise source matters!
  8. The chart on Fig.1 is an example of the electromagnetic pollution which we suffer.
  9. The antenna and the amplifier are similar to those recommended by ITU for measurements of the ambient noise. The setup is not calibrated but some rough comparisons can be made with published documents.
  10. It must be pointed out that these minimal rural noise levels are measured during daytime period. At night times when this band (80m) is opened the atmospheric noise is higher (with 10 dB or more) and in some cases the man made noise in urban locations might become negligible.

Here are some links for those interested more deeply in these problems. The main resource is [1]. On this site (ITU) there are other documents on this topic and the access is free. Owen Duffy, https://owenduffy.net had made measurements with similar setup in residential area on 40m band. His measurements are given for 9 KHz noise bandwidth. In order to compare the results with 1 KHz bandwidth we must subtract 9.5 dB. On Fig.2 in [13] the median noise level is between -1 to +8.5 dBuV/m which is lower than the residential levels obtained in my measurements on 80m. My other measurements also show that the ambient noise (both man made and atmospheric) on 40 m band is lower than that measured on 80m band. On his site there are also several useful calculators on these topics and in the blog section there are discussions about short whip on a car vessel.


[1] Recommendation ITU-R P.372-15 (09/2021) Radio noise. Int.Telecommun. Union, Geneva, Switzerland, https://www.itu.int/dms_pubrec/itu-r/rec/p/R-REC-P.372-15-202109-I!!PDF-E.pdf
[2] HF Interference, Procedures and Tools, 2007, NATO RTO TECHNICAL REPORT TR-IST-050, https://apps.dtic.mil/sti/pdfs/ADA473220.pdf
[3] VERON Manmade Measurement campaign, 2019, http://hf.r-e-f.org/c4_iaru_r1/16_Vienne/VIE16_C4_15_VERON_Provisional%20Results%20of%20Measurement%20Campaign.pdf
[4] Measurement Methodology and Results of Measurements of the Man-made Noise Floor on HF in The Netherlands, 2019, Koos (T. W. H.) Fockens, Peter (A. P. M.) Zwamborn, Frank Leferink, https://vienna.iaru-r1.org/wp-content/uploads/2019/03/VIE19-C7-007-VERON-Noise-Floor-Measurements.pdf
[5] Issues Concerning Radio Noise Floor Measurements using a Portable Measurement Set-up, 2019, Koos (T. W. H.) Fockens, Frank Leferink, https://vienna.iaru-r1.org/wp-content/uploads/2019/03/VIE19-C7-008-VERON-Noise-Floor-Measurements-Issues.pdf
[6] Man-Made Noise in Our Living Environments, 2010, Frank Leferink, F. Silva, J. Catrysse, S. Batterman, V. Beauvois, A. Roc’h, https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7911080
[7] The Background Noise on the HF Amateur Bands, 2017, RSGB, http://rsgb.org/main/files/2017/12/221216-Noise-leaflet-issue-2.pdf
[8] Noise Level Estimation in the Shortwave Frequency Range, 2011, E. Lossmann, M. A. Meister, U. Madar, https://eejournal.ktu.lt/index.php/elt/article/view/454
[9] Natural and man-made terrestrial electromagnetic noise: an outlook, 2007, C. Bianchi, A. Meloni, ANNALS OF GEOPHYSICS, VOL. 50, N.3, https://www.annalsofgeophysics.eu/index.php/annals/article/view/4425/4508
[10] HF radio reception compatibility test of an in-house PLC system using two brands of modems. 2003, Koos Fockens, PA0KDF, http://www.arrl.org/files/file/Technology/ModemRPRTVeron11-04-03.pdf
[11] Koos Fockens, PA0KDF personal home page at QRZ.com
[12] Martin G8JNJ, VDSL interference, https://www.g8jnj.net/vdsl
[13] Owen Duffy, Ambient noise survey, Hackett ACT on 40m, 06/03/2009, https://owenduffy.net/software/fsm/cs/03/index.htm
[14] Owen Duffy, Ambient noise measurement using whip on vehicle – #1 – estimate Antenna Factor, https://owenduffy.net/blog/?p=23386

Revision 1.2: There was an error in Eq.1 (thanks to Mr. Owen Duffy for spotting it) which leads to 4 dB lower level of the E. I have added additional explanations and links.