ASSAP: Paranormal Research
ASSAP: Paranormal Education
Privacy and cookie information ASSAP mailing list

Typical frequency response of EMF meter

Vigil equipment pages
Vigil equipment
Instrument baselines
Investigation techniques
Witnesses versus instruments
Paranormal equipment failures
EMF meters - what they do
What EMF meters measure
EMF meters - cause of readings
Analysing vigil data
Sound and radiation detectors
Negative ion detectors
Using still cameras on vigils
Static electricity and paranormal
Data loggers on vigils
Humidity and lighting
EVP infrasound IR thermometer

Frequency response

The graph above shows a typical frequency response of an EMF meter to magnetic fields. Along the bottom is frequency. The vertical axis shows sensitivity. At 50 Hz the meter has a sensitivity of 1. This means that a 50 Hz field of 1000 nT will show up, correctly, as 1000 nT on the meter. However, at a frequency of 1000 Hz, the meter is 10 times more sensitive! So a 1000 Hz field of 1000 nT will show up as 10,000 nT!

This frequency response is deliberate. It is weighted to mirror how magnetic fields are absorbed by the human body. That's because EMF meters were designed to monitor electromagnetic pollution not look for ghosts!

So when you see a reading of 1000 nT on your EMF meter, you've no idea what the real figure is because you don't know the frequency of the field. Most magnetic fields in a domestic environment will be mains frequency (50 Hz UK, 60 Hz US). But there are other frequencies possible and these will be either under- or over-represented. Even worse, there may be several different frequency mixed together to produce an overall figure that doesn't truly reflect any of them. So figures taken from an EMF meter are not particularly useful in terms of scientific measurement.

Different EMF meters will have different frequency responses making comparing readings problematic. If you're doing a positional baseline, it is better to use two meters of the same model.


What IS an EMF meter?

An EMF meter detects changing electromagnetic fields (hence E.M.F.). Most models only detect changing magnetic fields but some can measure varying electric fields too.

EMF meters were originally designed to look for electromagnetic pollution, though their main users seem to be ghost hunters these days (judging by adverts on the web). Most are therefore set to be highly sensitive to fields varying at 50 or 60 Hz, the mains frequency in the UK and the US respectively.

EMF meters can detect either magnetic, electric or both types of field together. Most, however, are designed to measure only varying magnetic fields. If it is not obvious what kind of field your meter is detecting, check the units it is using. Electric fields are measured in volts per metre (V/m) and magnetic fields usually in milliGauss (mG) or nanoTesla (nT).


As mentioned above, the units for electric fields are volts per metre (V/m). It is more complicated for magnetic fields because some EMF meter manufacturers insist on using obsolete units. Many use milliGauss (mG) even though scientific publications use nanoTesla (nT). To convert, 1mG is 100nT.

Computer sampling

Some meters come with ports to attach computers for automated recording. Also, some electronically-minded researchers have converted meters without ports so that they too can connect to laptops. This is a great idea as it allows much greater accuracy and frequency of readings as well as relieving observers of a tedious chore. You feed the output from the instrument into a data logger or specialist software.

There are, however, things to consider when doing computer sampling from EMF meters. For a start, if you take readings faster than the sampling frequency of the meter, you will not gain any higher frequency resolution even if it it looks like it!

Even more serious, you can only do frequency analysis if you have an accurate frequency response curve for the meter (see left) to apply corrections. The manufacturer may supply this, though it's unusual. Alternatively, you could get a linear sensor (like a fluxgate magnetometer) though these are usually rather expensive .

In all cases, you also need to consider the Nyquist criterion (where you need to sample at twice the rate of the top frequency you want to measure) for sampling and problems associated with aliasing (where higher frequencies affect the measurement of lower ones).You need to understand these before doing digital sampling.

Can EMF meters detect ghosts?

See here.



When choosing an EMF meter, there are several important specifications to consider.

Firstly, there are some meters out there that have no dials or displays for readings at all. They just buzz or show a light when a certain certain threshold is exceeded. Though it is true that the figures you get from EMF meters are not hugely helpful for paranormal research (see 'frequency response', left) they are certainly better than no figures at all. With a threshold detector, you've no idea if a reading was just fairly high or huge.

Perhaps the most important specification to consider is frequency response (see left). It is useful to get a meter sensitive to extremely low frequencies (under 10 Hz) as this region has been implicated in magnetically induced hallucinations. If you can get a model where the manufacturer supplies a frequency response curve (see left), that's far better than one without it.

Another important specification to consider is the number of measurement axes. Electromagnetic fields have direction as well as size (which is why compasses point north). If a meter doesn't mention axes it almost certainly has just one axis. The problem with a single axis meter is that if you rotate it, even slightly, during a vigil, subsequent readings will change. That's because it's at a different angle to the fields. This means you can't compare the earlier readings with the later ones. The solution is to fix the meter in place or use a tri-axial model. Single axis meters will underestimate every field they measure by differing amounts making comparisons between locations problematic.

Another important specification is sample rate (see 'computer sampling', left). Faster is definitely better.

The other important specification is scale or range. This specifies the maximum and minimum field the meter can detect. This is important because you may come across fields that are either too weak or too strong to measure. Try to go for meters with the biggest scale you can.
© Maurice Townsend 2007