A SAR rating — Specific Absorption Rate — is the maximum rate at which your body absorbs radiofrequency (RF) energy from a phone, measured in watts per kilogram of tissue. In the US, the FCC requires every phone sold to stay under 1.6 W/kg averaged over 1 gram of tissue, tested under worst-case conditions in a lab.
That’s the short answer. The longer answer is more interesting: a SAR rating is a pass/fail compliance ceiling, not a safety score. It doesn’t measure what your phone actually delivers to your head during a normal day, and even the FCC says you can’t use it to compare one phone against another. Here’s what the number really means — and what it quietly assumes.
SAR is a compliance test designed in the 1990s — a ceiling every phone must stay under, not a graded safety score.
What a SAR Rating Actually Measures
SAR quantifies energy absorption: how many watts of RF power end up deposited in each kilogram of your tissue. When your phone transmits — during a call, uploading photos, hunting for signal — some of that RF energy is absorbed by whatever is closest to the antenna, usually your hand and head.
The US limit for phones is 1.6 W/kg averaged over 1 gram of tissue, which applies to devices transmitting at or below 6 GHz. That partial-body limit sits on top of a whole-body limit of 0.08 W/kg for the general public.
The Number on the Box Is a Worst Case
Here’s the detail most SAR explainers skip: certification testing runs the phone at its maximum transmit power, on its worst-performing frequency band, in the worst-case position. The FCC is explicit that reported SAR values reflect the device’s highest possible energy absorption — not its typical output.
Your phone almost never operates that way. Modern phones use adaptive power control, transmitting with just enough power to reach the tower. With strong signal, transmit power can drop to a small fraction of the tested maximum. With one bar in an elevator, it climbs toward that maximum. Two identical phones — one in a strong-coverage suburb, one in a rural dead zone — can deliver very different real-world exposure despite the same SAR sticker.
How SAR Testing Works
Since you can’t put probes inside a living human head, compliance testing uses a stand-in: the SAM phantom (Specific Anthropomorphic Mannequin).
A Mannequin From a Military Survey
SAM is a fiberglass-style shell shaped from 90th-percentile adult male head dimensions drawn from US Army anthropometric data — a large man’s head, by design. The shell is filled with a liquid whose electrical properties mimic averaged human head tissue at the test frequency. A robotic probe sweeps through the liquid while the phone transmits at full power against the phantom’s “ear,” mapping where energy absorption peaks. If the hottest 1-gram averaging volume stays under 1.6 W/kg in every tested position and band, the phone passes.
The design goal was a conservative test: a big head absorbing a worst-case signal should over-predict exposure for most users. Critics have long countered that one adult male mannequin can’t represent everyone — children’s heads are smaller, their skulls thinner, and their tissue properties different — a concern a federal appeals court ordered the FCC to address, as we cover in our review of why the FCC’s 1996 exposure limits are considered outdated.
The Separation Distance Fine Print
Head testing presses the phone against the phantom. Body-worn testing doesn’t: manufacturers test at a small air gap — typically 5 to 15 millimeters — on the assumption you’ll carry the phone in a holster or bag, not against skin.
That gap matters because absorption falls off steeply with distance. In 2019, the Chicago Tribune commissioned independent lab tests of popular smartphones at 2 mm — a phone in a pants pocket — and reported that several exceeded the FCC limit at that distance. The FCC ran follow-up testing and reported the phones compliant under its standard procedures. Both sides were arguably right, because they were testing different assumptions: the certification test was never designed to represent a phone riding in your pocket against your body all day.
Check your EMF exposure
See cell towers, power lines, and substations near any US address.
Search Your AddressThe 1996 Assumptions Behind the Limit
The FCC adopted its current RF exposure limits in 1996, drawing on recommendations from the NCRP (1986) and ANSI/IEEE (1992). The chain of logic behind the number:
- Find the effect threshold. Animal studies from the 1980s identified behavioral disruption starting around a whole-body SAR of 4 W/kg — the point where absorbed RF energy heats tissue enough to change behavior.
- Divide by a safety factor. For the general public, that threshold was cut by a factor of 50, producing the 0.08 W/kg whole-body limit (occupational exposure uses a factor of 10, yielding 0.4 W/kg).
- Set a local hotspot cap. Because a phone concentrates energy in one spot rather than the whole body, the separate 1.6 W/kg partial-body limit caps localized absorption.
Notice what the entire framework is built on: preventing tissue heating. If RF energy doesn’t measurably warm you, the 1996 logic says you’re protected. Whether decades of research on non-thermal biological effects should change that framework is one of the most contested questions in EMF science — the research runs both directions, and we’ve covered it in depth, including a study spotlight on 5G phone SAR and brain temperature modeling. The short version: SAR limits were never designed to answer questions about chronic, low-level, non-thermal exposure. They answer one question — “will this device heat you?” — using assumptions about devices, usage, and bodies frozen in the mid-1990s, when a phone was something you used for minutes a day.
Why SAR Is a Ceiling, Not a Safety Score
The most common mistake people make with SAR ratings is treating them like crash-test stars: phone A scores 0.9, phone B scores 1.5, so phone A must be meaningfully safer. The FCC itself rejects this. In its consumer guidance, the agency warns that a single SAR value “does not provide sufficient information” about typical RF exposure to reliably compare phone models.
Why the regulator undercuts its own metric:
| What SAR tells you | What SAR doesn’t tell you |
|---|---|
| The phone stays under 1.6 W/kg at full power in test positions | What the phone emits during your actual use |
| A worst-case peak in a standardized adult phantom | Absorption in your head, or a child’s |
| Compliance on the worst frequency band | Exposure across bands, WiFi, and Bluetooth combined |
| A lab ceiling at fixed distances | What happens in a pocket, or with weak signal |
The comparison problem is structural. A phone with a lower maximum SAR may reach that maximum frequently if its antenna performs poorly and it must transmit near full power to hold signal. A phone with a higher tested ceiling but a more efficient antenna may spend its life transmitting at a small fraction of it. The sticker ranks lab ceilings, not lived exposure.
That doesn’t make the number useless — a ceiling is still information, and all else equal there’s nothing wrong with preferring a lower one. If you want the actual numbers for current handsets, our lowest-SAR phones guide ranks them — with these caveats attached. Just don’t mistake a lower ceiling for a measured safety margin.
US vs European SAR Ratings
If you look up an iPhone’s SAR value, you’ll find two different numbers — and they aren’t comparable:
| Standard | Limit | Averaging mass | Where |
|---|---|---|---|
| FCC | 1.6 W/kg | 1 gram of tissue | United States |
| ICNIRP | 2.0 W/kg | 10 grams of tissue | Europe and much of the world |
The averaging mass is the real difference. Averaging over 10 grams smooths out sharp local hotspots, so the same phone in the same test typically reports a lower number under the European method than under the US 1-gram method — nothing about the phone changed, only the math. That’s also why “Europe allows more radiation than the US” is too simple; the two regimes measure differently, and neither is straightforwardly stricter across every scenario. For how exposure limits vary worldwide — including countries that set limits far below both — see our international EMF limits comparison.
What Actually Determines Your Exposure
Since the sticker can’t tell you your real exposure, what does? Three things, roughly in order:
1. Distance
RF power density falls off with the square of distance. Moving a phone from against your ear to arm’s length on speakerphone cuts absorption enormously — far more than any difference between two compliant phones’ SAR ratings. This is the same physics that makes your phone a bigger RF source than the cell tower down the road: proximity dominates.
2. Signal strength
Weak signal forces high transmit power. Calls in basements, elevators, moving cars, and fringe-coverage areas push your phone toward the maximum power it was SAR-tested at. Strong signal lets it whisper. If you want to reduce exposure, coverage quality matters more than the number on the box.
3. Habits
Speakerphone and wired earbuds, texting instead of long calls against your head, not sleeping with the phone under your pillow, carrying it in a bag rather than a pocket during heavy streaming — these swamp the differences between phone models. One thing that doesn’t help: stick-on shields and most EMF-blocking phone cases, which can backfire by weakening signal and driving transmit power up.
Your phone is also just one source in your environment. If you’re mapping your total picture, check your address on EMF Radar to see cell towers, power lines, and substations near you — the ambient side of the equation that no SAR sticker covers.
The Bottom Line
A SAR rating answers exactly one question: can this phone, at full power, in a standardized test against a large adult phantom, heat tissue beyond the limit set in 1996? Every phone legally sold in the US answers “no.”
What it doesn’t answer: how much RF energy you absorb in real life, whether one compliant phone is safer than another, how the test translates to a child’s head, or anything about non-thermal effects — questions a federal court has ordered the FCC to revisit with a reasoned explanation. Treat SAR as what it is: a coarse regulatory ceiling from another technological era. Then put your effort where the physics is — distance, signal, and habits.
Frequently Asked Questions
What is a good SAR rating for a phone?
There’s no graded scale — SAR is pass/fail. Any phone sold in the US must test under 1.6 W/kg, and the FCC cautions that comparing SAR values between compliant models doesn’t reliably tell you which exposes you to less RF in typical use. Lower is fine to prefer, but the number reflects a worst-case lab test rather than a promise about everyday exposure.
Is a lower SAR phone actually safer?
Not necessarily. SAR is measured at maximum transmit power, and real phones adjust power constantly based on signal strength — a low-SAR phone with a weak antenna can emit more in daily use than a higher-SAR phone with an efficient one. Distance and habits (speakerphone, texting, not pocket-carrying) change your exposure far more than the difference between two compliant phones.
How do I find my phone’s SAR rating?
Most manufacturers publish SAR values in the phone’s legal or RF exposure information — typically in Settings under legal/regulatory info — and on their websites. You can also look up any US-certified device in the FCC’s equipment authorization database using the FCC ID printed on the device or in its settings.
Why is the SAR limit different in the US and Europe?
The US tests at 1.6 W/kg averaged over 1 gram of tissue; Europe follows ICNIRP at 2.0 W/kg averaged over 10 grams. Because averaging over more tissue smooths out hotspots, the two numbers aren’t directly comparable — the same phone reports different values under each method without emitting anything different.
Does SAR apply to 5G?
Partly. SAR testing covers transmissions at or below 6 GHz, which includes the low- and mid-band spectrum carrying most 5G traffic today. Millimeter-wave 5G operates above that range, where energy is absorbed at the skin’s surface and compliance is evaluated with power density limits instead of SAR.