EMF Exposure Calculator: Your Cumulative Dose Estimate
Electric car, public transport, mobile phone, household appliances, smart meter, high-voltage lines, cell towers… everyday electromagnetic fields (EMF) come from many different sources, at very different intensities. This calculator adds up your exposures to estimate a cumulative dose — and places it on time scales ranging from a single day to a whole lifetime.
Enter your habits for a reference period of your choice (day, week, month, year, decade or lifetime): the calculator automatically extrapolates the result to every scale, and compares it with recognised regulatory thresholds.
An awareness tool, not a measurement. The field intensities used here are illustrative orders of magnitude, not values measured in your own home: a real car model, appliance or installation can differ substantially.
Your exposure, source by source
Result
For your reference period (Week), the estimated cumulative dose is 202.35 µT·h. Converted to an annual average, that comes to 1.204 µT, or 0.602 % of the ICNIRP general-public threshold and 301 % of the Dutch precautionary threshold.
Time-scale grid
| Period | Cumulative dose | Average | % of ICNIRP threshold (200 µT) | Equivalent time at 200 µT | % of NL precautionary threshold (0.4 µT) |
|---|---|---|---|---|---|
| Day | 28.91 µT·h | 1.2045 µT | 0.6022 % | 8.7 minutes | |
| Week | 202.35 µT·h | 1.2045 µT | 0.6022 % | 60.7 minutes | |
| Month | 879.93 µT·h | 1.2045 µT | 0.6022 % | 4.4 hours | |
| Year | 10,558.33 µT·h | 1.2045 µT | 0.6022 % | 2.2 days | |
| Decade | 105,583.34 µT·h | 1.2045 µT | 0.6022 % | 22 days | |
| Lifetime (80 years) | 844,666.71 µT·h | 1.2045 µT | 0.6022 % | 5.8 months |
Breakdown by source (reference period)
| Source | Dose | Share of total |
|---|---|---|
| High-voltage lines/transformers | 125 µT·h | 61.8 % |
| 4G/5G cell towers | 44.8 µT·h | 22.1 % |
| Household appliances | 15.6 µT·h | 7.7 % |
| Home environment | 8.4 µT·h | 4.2 % |
| Mobile phone | 4.2 µT·h | 2.1 % |
| Electric car | 2.7 µT·h | 1.3 % |
| Train / subway / tram | 1.6 µT·h | 0.8 % |
| Smart meter | 0.05 µT·h | 0 % |
Comparison with reference thresholds
| Reference | Threshold | Your annual average | Position |
|---|---|---|---|
| ICNIRP — general public | 200 µT (instantaneous) | 1.2045 µT | 0.6022 % of threshold |
| Precautionary policy — Netherlands | 0.4 µT (annual average) | 1.2045 µT | 301.1 % of threshold |
| Swedish recommendation | ALARA principle ("As Low As Reasonably Achievable"): reduce exposure as much as reasonably possible, with no numeric threshold. | ||
Point source (smart meter, cell tower): field(d) = reference field × (reference distance ÷ d)²
Line source (EV cables, HV lines): field(d) = reference field × reference distance ÷ d
Dose(period) = dose(reference) × days(period) ÷ days(reference)
Important notes
This calculator is an awareness tool, not a medical diagnosis.
The effects of EMF vary between individuals. Some people may be more sensitive than others.
The figures are estimates based on studies whose results can vary. Real vehicle models, appliances and installations may give different results.
ICNIRP thresholds are based on established thermal effects. Non-thermal effects (headaches, sleep disturbance) are not accounted for by these thresholds.
Extrapolation over long periods (decade, lifetime) is a linear estimate. The reality of biological effects may be more complex.
What is a cumulative dose?
X-rays and radon come with a legal idea of dose: absorbed energy, weighting factors, a risk scale. Non-ionising electromagnetic fields have no such currency. There is no sievert for EMF. Field times time, here in µT·h (microtesla-hours), is a teaching proxy — how much field, for how long.
It lets you stack an hour at an induction hob against eight hours on the Tube, or see whether the meter even registers next to the phone. It does not mean “how much biological harm”. Two hours at 1 µT is not the same as a minute at 120 µT, even if the product looks close. If anything happens below thermal limits, it almost certainly does not add up the way ionising radiation does.
So the µT·h on this page ranks sources and stretches them over time. It is not dosimetry. It is not a survey of your flat.
Point source versus line source
From a few metres away a meter, a mast or a compact transformer behaves like a point source: the magnetic field falls roughly with the square of distance. Double the gap, quarter the field. A long cable, an overhead line, the high-voltage loom in a car is a line source. Decay is 1/d. Double the distance and you only halve the field.
Point: B(d) = Bref × (dref / d)²
Line: B(d) = Bref × (dref / d)
That is why three extra metres from a smart meter move the needle far more than three extra metres under a 400 kV span — the line comes with you. Distance is not one price. The calculator encodes it: meter and mast as 1/d², car cables and pylons as 1/d. Train, phone and appliances already have distance baked into the usage category.
Why limits differ from country to country
ICNIRP — the body WHO leans on — sets public reference levels. At 50/60 Hz this page uses 200 µT (2010 guideline): an instantaneous ceiling, built from established nerve and heating effects, plus a margin. Many states copy it. Others add a much lower precautionary planning value, without claiming a proven effect at that level.
The Netherlands aims near 0.4 µT as an annual average for new lines — five hundred times below 200 µT. Sweden talks ALARA (“as low as reasonably achievable”) without pinning a single number. UK public exposure has followed ICNIRP-aligned guidance; the US conversation often quotes IEEE and FCC figures that are not identical, especially for radiofrequency. A health ceiling (no heating, no nerve stimulation) and a planning goal (keep new classrooms off the right-of-way) are different jobs.
Putting your annual mean against both bars — 200 µT and 0.4 µT — shows the gap. A tiny ICNIRP percentage can still eat a chunk of the 0.4 µT band. Both sit in the results table on purpose.
What the research can settle, and what it cannot
On power frequencies, a few older epidemiological papers linked living near overhead lines with a small excess of childhood leukaemia around 0.3–0.4 µT. Association is not cause. Counts are small. Later work is mixed. WHO puts these fields in group 2B (“possibly carcinogenic”) — the same drawer coffee sat in for years, not tobacco’s. For radiofrequency, IARC also used 2B (phones, glioma). Later cohorts have not closed the file.
The lab is clearer on tissue heating and, at very high 50 Hz fields, nerve stimulation. Below that, “non-thermal” findings (sleep, semen quality, oxidative stress) do not replicate cleanly. UKHSA, like ANSES and others, usually lands on: no solid proof of harm at everyday levels, enough uncertainty to skip pointless exposure — especially children, especially a handset against the ear for hours.
This page does not umpire that fight. It parks orders of magnitude next to public limits, and it will not turn a µT·h into a diagnosis.
Low frequency versus radiofrequency
Mains electricity, trains, a traction motor, the fridge, the meter’s 50 Hz side: ELF, quoted in tesla (here µT). The phone, Wi-Fi, 4G/5G masts: radiofrequency, usually V/m or W/m². Different wavelength, different absorption, different legal units.
Folding them into one µT·h column is a convention of this calculator, so the handset and the fridge can share a ranking. That is not how an accredited lab writes a report. A µT stand-in for a mast does not replace SAR or power density. We say so out loud, so nobody thinks one number “measures the waves”.
Thermal and non-thermal effects
ICNIRP limits — and the national rules that shadow them — rest on what is established: tissue heating at RF, nerve and muscle stimulation at very high ELF. Take the effect, add a reduction factor, publish a reference level. Headaches, broken sleep, self-reported “electrosensitivity” — the cluster often labelled non-thermal — are not in that formula.
That is not the same as calling the complaints imaginary. Expert panels still lack a reproducible mechanism and a number they can write into the limit. Sitting well below ICNIRP answers the heating question. It does not, by itself, end a precaution debate or a medical work-up if symptoms are real.
This tool is not a diagnosis and not a measurement. Someone who links symptoms to EMF needs a clinician, not a µT·h score. The reverse holds too: a low score does not rule out sleep, screens or stress.
Smart meters: claims versus measurements
Communicating meters send short bursts (power-line carrier and/or a radio hop). Independent surveys — Ofcom-style RF checks, utility ELF measurements a few metres into the room — usually sit far below the limits, and often below a hob or a microwave while those are on. Not zero. Not a mast in the cupboard either.
The field drops fast (point source, 1/d²). Three metres instead of half a metre shows up immediately in this model. Moving the bed away from the meter cupboard, when you can, is the lever that matters. Disabling the meter is not. Pinning a whole symptom list on it without a measurement or a diagnosis is not either. Here it is one source among others, with a distance and a duration — not the villain of the piece.
Cutting everyday exposure
The physics is dull and it works: distance almost always wins. Speakerphone or a wired headset instead of the handset on the ear. Router out of the bedhead. Induction hob: don’t park a mug against a live zone. In the car the HV cables are stronger at the feet and the boot than at the temple if you set ~50 cm — that is why the form asks “head to cables”.
- Move what will move (meter cupboard, router, laptop brick).
- Shorten what will shorten (phone-to-ear calls, standing in the microwave’s face).
- Leave on what has to stay on (fridge, lights): low rate, long hours, often a weak lever.
“Anti-wave” paint and pendants rarely earn a place in a serious budget. A badly designed Faraday nest can even make the phone wind its power up to hold the network.
Electric cars: what independent tests show
German and Swiss campaigns, plus a few magazine probes, see mainly 50–1000 Hz from the traction cables — higher under boost, higher at the feet and the boot than at the temple. Real-use figures are typically a small slice of the 200 µT ICNIRP level. Sometimes they sit in the same band as a train. Sometimes a little above a petrol car, depending on how the loom is routed.
Not “zero EMF, therefore magic”, and not “a microwave on wheels”. Model, probe point and driving mode (boost, regen) all move the number. Hence pedagogical orders of magnitude in the form, not the type plate of your car. If you want this vehicle’s figure, you need a probe, not a web scale.
Phones and masts, forced into µT
A handset’s SAR is in W/kg; a mast is in V/m. Turning those into a µT/h rate, as this page does, is a device so everything can be added. It ranks habits (phone to the ear ≫ data in a pocket ≫ Wi-Fi) better than a “5G yes/no” row. It does not let you say “my mast is X% of the RF ICNIRP limit”.
In this model a call without a headset often outweighs a mast a few hundred metres off. That matches drive-test style surveys: the transmitter nearest the brain is still the handset. The mast is treated as a 24-hour ambient, so the dose spreads out even at a low rate.
Reading the table from a day to a lifetime
Once the reference period is in, the table scales by a rule of three: week = 7 days, year = 365.25, “lifetime” = 80 years. Linear. It assumes today’s habits last 80 years, which they will not — cars, flats and jobs change.
dose(period) = dose(reference) × days(period) ÷ days(reference)
The lifetime column is for the eye, not an 80-year risk forecast. The useful column, against precautionary policy, is the annual mean in µT, set beside the Dutch 0.4 µT and, far above, the 200 µT ICNIRP figure.
If you need a real measurement
Ofcom maps RF in the UK. Network operators and accredited labs measure 50 Hz near lines. A consumer broadband meter gives a ballpark, rarely a figure you can take to a tribunal. If a pylon or an open-plan office is a genuine worry, a standards-based survey (height, averaging time, the right quantity) is the number that counts — and it will usually be in the unit of that band, not in µT·h.
A pacemaker or similar implant has its own distance notices. This calculator does not know them. That conversation belongs to the clinic and the manufacturer, not a web score.