Background

Way back in around 2005/6 I remember waking up in the middle of the night and thought that I could hear a very low frequency humming sound. Well actually, I awoke and suddenly everything seemed to go dead silent, and I realised that the above sound had been present, if that makes sense. I remember telling a friend of mine that you can only hear it when it switches off! Initially I put the incident down to pure imagination and forgot about the matter.

The months went by and on occasion, in the dead of night, I would hear this very faint unusual humming sound. I thought initially that the sound might be originating from an appliance in the house and attempted to find the source by spectral analysis (both audio and electromagnetic) of various devices. No luck, nothing seemed to be emitting such low frequencies. From my experience of the humming sound I knew that it’s frequency was probably just below the threshold of human hearing which is usually between 15 to 20Hz at the lowest end to the spectrum.

Even more time passed and I just happened to stumble upon the phenomenon that is commonly referred to as “The Hum” via a random web search that I did in an attempt to find out what I had been hearing. It was of great interest to me that apparently “The Hum” is a world wide phenomenon that certain people are able to hear, although the actual cause still remains a mystery. See the Wikipedia article and other web references for background details.

Given what research has been undertaken previously into The Hum, explanations range from natural phenomenon to those of a man-made origin. As far as my Hum goes, I have always felt that the signal that I was hearing was of man-made origin, as it appeared to be composed of a single, narrow band signal, where noise from natural sources tend to be wider band with frequency and phase shifting present.

I tentatively asked one of my neighbours if they could hear anything, they said “no”, although much to my astonishment went on to tell me that people in Luton were also hearing it. Apparently some articles had been published in The Luton News at the time concerning this oddity. I recently (2025) found a website and a comment made about the original post by a person living at Shillington which is less than ten miles from Maulden where I live.

I started to hear The Hum again this year in February 2025 which finally prompted what is below in an attempt to understand what might be happening. So, that being said, and armed with my background in Signal Analysis, I decided to set-up a “Chris Lab” experiment (as my friend calls me) to see if any answers might be found. Certainly this “Hum” is not of any concern to me and more than likely anyone else either, and in saying that what is below represents more of a personal science experiment rather than anything else.

The first trial

The first trial that I ran was on Friday the 14th of March 2025 which started at just after 11am. I monitored and logged audio data from a bedroom in the house for a period of nearly five hours. The reason for the daytime trial was that I wanted to get an idea of background noise levels at that time of day.

The raw audio signal was sampled at a frequency of 44.1kHz and this was passed through a two stage chain of custom designed decimating anti-aliasing filters in order to produce an output signal having a sampling frequency of 200Hz, thus obtaining a signal I could then analyse in the low part of the audio spectrum, i.e. my interest was in frequencies below the human threshold of hearing at 15 to 20Hz.

The figure to the right are the results from the trial and show the power spectral density of the audio signal in the 0 to just over 100Hz range.

The two main bands of red at around 15 to just over 20Hz, and just under 40Hz, that run horizontally across the plot are probably due to traffic noise originating from local roads in the area. Certainly over the period of the trial there was an almost constant stream of traffic passing through the village on that particular day. Low frequency sound waves can travel long distances and it is quite possible that background noise from the Maulden by-pass and perhaps even the A6 trunk road and M1 Motorway that are only some miles away would contribute to the observed data.

If you look at the plot to the right, just before the one hour mark and extending to about 2.5 hours, there is a thin line corresponding to a frequency of around 7Hz. It appears to emerge of of nowhere and then disappears back.

I thought that it might be worth taking a closer look at this signal and designed a bandpass filter centering on 7Hz and using this to process my dataset to obtain a time domain representation of the signal. The results of which are to the side and show the the amplitude variations and burst like characteristics.

Closer inspection of one of the bursts lasting for over four minutes shows clearly that the 7Hz signal appears to be being modulated by another lower frequency signal. This suggests that two distinct causal processes might be involved in the generation of it. Even more puzzeling!.

I can’t be sure that this 7Hz signal is the source of the hum that I have been hearing, especially as the frequency is somewhat below the hearing threshold. Perhaps it might be possible. Whatever the case it was worth investigating a little further. Just a stab in the dark but it smacks of noise originating from a man-made source. One thing is for sure the signal is not a signal processing artifact generated either by my own algorithms or from that of the acquisition device, otherwise it would be present all the time and not fade in and out as it does.

The second trial

Trial two was a night time one. The results of which are below. The 20Hz and 40Hz noise bands that I mentioned above are fairly clearly the result of general traffic movements in the area as they gradually start to tail off at about 12am, and then start again around 6am. It was a Saturday night into Sunday morning, and hence still some traffic activity until the early hours. In general far less background noise at night than that obtained from the first daytime trial period.

The 7Hz signal puts in a second appearance at just after 1am, fading in and out until a little before 5am. The most notable difference is the appearance of a signal at just under 40Hz that spans the entire monitoring period. Again it looks to me like a man-made signal given it’s very narrow bandwidth. I have read various reports concerning frequency estimations of The Hum some of which include signals in and around the 40Hz region, perhaps this is the signal that I have been hearing.

Conclusions

The above results were encouraging to me, at least I now have two distinct frequencies that I can concentrate on in some upcoming trials, and a chance perhaps to dig into all of this a little deeper.

One of the hypotheses put forward by other researchers into The Hum is that the sound may be originating from underground pipelines that carry high pressure natural gas. A hypothesis that makes sense to me. Interestingly there is a natural gas terminal located just outside of Flitwick which is only around five miles from Maulden. The pipeline meanders across the countryside in a south easterly direction, crossing the A6 near Barton-le-Clay, onwards past Pegsdon, and over the hills towards Great Offley. Given the pipe length and pressures involved in pumping the gas, just maybe the pipeline is the cause.

Update on the 19/03/2025

On the 18/03/2025 I did a small survey consisting of 30 points scattered around the village in order to see if I could determine the spacial distribution of both the 7Hz and 37Hz signals.

Below are the results for the 7Hz signal. The lowest signal magnitudes are shown in blue, while the higher ones in red:

Here the bulk of the high magnitudes form a ring like pattern, with an area in the east focusing near The Knoll, moving southwards to the open farm land area in the south, and the moving northwards near the war memorial, and finally to the north around Kings Road area. The signal magnitude is these areas being around -32dB.

Given the results above, there are no apparent causes on the ground, to account for the spacial concentrations. The areas concerned are either houses or farm land. Perhaps the concentrations are due to standing wave patterns dependent upon ground elevation or similar.

While for the 37Hz signal:

The 37Hz signal has an epicentre that focuses in an area just south of the village to the left of the allotment grounds, which is open farm land. The 37Hz signal has a magnitude here of around -23dB.

The 37Hz signal has been very sharply filtered in order to eliminate any traffic noise present below this frequency. At the time that I did the survey traffic movements through the village were fairly quiet at just after midday. Much of the main traffic in Maulden is along Ampthill Road, and my results don’t show any real concentrations around this area.

I extracted the 37Hz signal sample that I took near the Maulden allotment grounds where the amplitude of this signal is at the greatest over the spacial survey, and is shown to the plot on the right.

Similar to the 7Hz signal above, the 37Hz signal is modulated. I extracted the modulation envelope using the Hilbert Transform in order that I could examine the frequencies of the envelope’s spectrum.

The principle frequencies present in the modulation envelope are 0.016Hz and 0.195Hz, with other minor peaks that tail off as they reach 1Hz.

All of this seems to get even more unusual. I think that my next steps with this investigation is to undertake a much broader spacial survey, i.e. one that would be made over some miles, in an attempt to try to get to the bottom of all this. Time and resources permitting.