The energy comes from the antenna
In 2002 the US Federal Trade Commission charged two sellers whose ads said a patch over a phone’s earpiece blocked up to 97 to 99 % of its radiation. We simulated a head beside a phone. A 1.5 cm metal patch over the earpiece cuts the power absorbed in the head by about 7 %. A closed metal ring around the antenna cuts it by about 50 dB, so the simulation can show a real block when there is one.
The verdict
Nothing there
In a 2-D wave simulation, a 1.5 cm metal patch over the earpiece cuts the power absorbed in the head by about 0.3 dB, or 7 %, at 900 and 1800 MHz, not 97 to 99 %, because the energy comes from the antenna. A closed metal ring around the antenna cuts it by about 50 dB, so the simulation can see a real block.
The patch over the earpiece cuts the power absorbed in the head by 0.33 dB at 900 MHz and 0.30 dB at 1800 MHz. The claim needs at least 15 dB at both.
The power the phone radiates changes by 0.04 dB at 900 MHz and 0.02 dB at 1800 MHz, so the patch does not mute the phone either.
At three points just inside the head, behind the patch, the field drops by 0.9 to 1.2 dB.
A closed metal ring around the antenna cuts head power by 49.5 dB. The same patch moved to the back of the phone changes it by 0.02 dB.
The claim, as presented. Sellers advertised that a patch over the earpiece blocked up to 97–99 % of a phone's radiation; the FTC alleged these claims were false and unsupported.
US Federal Trade Commission, “FTC charges sellers of cell phone radiation protection patches”, February 2002 · the episode
The registry’s note on sources. Read from the source named here by a research subagent (2026-10-09); paraphrase, not quotation. No video was watched.
Re-read before the run, 2026-10-12. The press release describes two sellers of patches placed over a phone’s earpiece. Their ads said the patches blocked up to 99 % (97 % in one summary) of the energy reaching the ear canal or brain. The FTC alleged the claims were false and unsubstantiated, and that most of the energy comes from the antenna and other parts of the phone. These are allegations, not findings. The test here is of the physical claim only.
Preregistered in PREREG.md (commit 74a5bb7) before any episode code; one addendum after run 1, declared as such. Self-graded, with no outside review. The run’s RESULTS proposed nothing there, and on 2026-10-18 the lab’s verdict engine set the same verdict. Lab files are listed at the foot of the page.
The setup
A 2-D wave cross-section, in which every object is an infinite cylinder. It gives ratios and shows where the energy goes; it does not give exact absorbed power for a real phone. The head is a lossy disk 9 cm in radius. The phone is a conducting chassis 1 cm from it, with an external stub antenna above the chassis, as on 2002 phones.
The patch is a 1.5 cm conductor on the phone’s face, over the earpiece. The frequencies are 900 and 1800 MHz, the GSM bands of 2002. The readout is the power absorbed in the head with and without the patch, from the same source. “Blocks 97 to 99 %” means a cut of at least 15 to 20 dB.
The checks
Every check below was written down before the episode code. Each planted fault had to turn its check red once.
| Check | Result | Numbers |
|---|---|---|
| The wave engine’s own tests and planted bugs | pass | 15 passed; all 8 mutations caught |
| Free-space decay against the 2-D Green’s function | pass, run 2 | 2.9 % (900 MHz), 0.1 % (1800 MHz); limit 5 % |
| … against a 3-D 1/r decay instead | red, as it should be | 43 % off |
| Source power equals head plus radiated power | pass | 1.8×10−15 |
| … with the head left out of the audit | red, as it should be | 0.14 |
| The claim: head power cut by the patch | fails the claim | +0.33 dB (900 MHz), +0.30 dB (1800 MHz); the claim needs at least 15 dB |
| Positive control: a closed ring around the antenna | pass | 49.5 dB |
| Null: the same patch on the back of the phone | pass | +0.02 dB |
| Grid: 6 mm against 5 mm cells | pass | 0.35 vs 0.33 dB; 0.32 vs 0.30 dB |
The miss. In run 1 the free-space check failed at 900 MHz, 10.9 % off against a 5 % limit, so by the rule written down first that run carried no verdict. The absorbing edge was made thicker and gentler, in an addendum declared after the fact, and every run was repeated. Run 1’s claim numbers, 0.32 and 0.30 dB, were not used.
Other readings
Reported alongside the gates, with no pass or fail attached.
| Variant | Cut in head power | Cut in radiated power |
|---|---|---|
| Patch over the earpiece (the claim), 900 / 1800 MHz | 0.33 / 0.30 dB | 0.04 / 0.02 dB |
| The same patch, with the antenna at the bottom of the chassis, as on modern phones | 0.00 dB | — |
| A 30 cm metal plate between the whole phone and the head | 39 dB | 11 dB |
In this geometry about 15 % of the phone’s power ends up in the head at 900 MHz, and about 9 % at 1800 MHz. The rest leaves as radiation.
Shielding the head is possible: it takes a plate as big as the phone’s near field, and that plate also cuts what the phone sends out.
- Real 3-D absorbed power (SAR), and the phone’s own transmit-power control.
- Tissue permittivity: the engine has loss but no refractive-index contrast.
- The products’ actual materials and sizes.
Read: the FTC press release (fetched 2026-10-12). Prior art: the FTC’s own technical point, and antenna and shielding textbooks. No novelty claimed.