No promise we cannot demonstrate. Here are our field observations and our exploratory research — freely available, methodology and limits included. Every claim is labeled by its level of evidence.
20 studies · methodology and limits included · updated 2026
01 · Circuit physics
The physical principle of the circuit
Our circuits invent no physics. They are tuned conductive loops: three classical notions fully describe what they can do — and what they cannot.
Established · classical physics
Faraday's law
A varying magnetic field through a conductive loop induces a voltage in it. This is the principle of the transformer and of the contactless badge. Our circuit is a loop; the electromagnetic environment varies constantly.
Established · classical physics
The natural frequency
f₀ = 1 / (2π√LC)
An LC circuit has exactly one natural frequency, set by its geometry: the inductance L of the loop and the capacitance C. It is not programmed — it is drawn.
Established · classical physics
The quality factor
The Q factor measures selectivity: the higher it is, the more the response concentrates in a narrow band around f₀. That is what separates a tuned circuit from a plain piece of metal.
The right analogy: the crystal radio
A radio with no battery. It produces no energy: it draws on what already crosses the air, and its only intelligence is to select a single station from it. Our circuits do nothing stranger than that — and nothing more than a crystal radio could do.
Physical distinction · established
Passive does not mean linear
These are two independent properties: passive means no energy source, linear means a component returns only frequencies already present at its input. A component can be passive AND non-linear — a diode is exactly that: no power supply, and yet it rectifies the signal.
The crystal radio illustrates it precisely: the tuned circuit selects the station, and it is the detector — passive, battery-free — that extracts the sound. A fully passive device can therefore produce a low-frequency component, provided it contains a non-linear element.
We do not claim this is the case for our circuits: the question is being verified. We write it because the distinction is real and is often stated wrongly.
Physical precision
Two physical objections — and our answers
①
“At 7.83 Hz the wavelength runs to thousands of kilometres — an object a few centimetres across cannot resonate at it.”
The objection would be right for an antenna. It is not right for a tuned circuit: the resonant frequency of an L-C-R circuit is given by f = 1/(2π√LC) and does not depend on the size of the object.
But the practical conclusion goes the objection’s way: the patent’s own worked example — 0.5 µH and 12 nF — gives a resonance of about 2 MHz. The resonance of our circuits is in the radiofrequency domain.
②
“A passive, linear circuit cannot create a frequency that does not exist at its input.”
This objection is correct, and we adopt it as our own. Such a circuit cannot impose a low-frequency envelope on a carrier. A crystal radio does not create the modulation: it demodulates one that already exists, and it is the detector — not the tuned circuit — that fills that role.
What a passive resonant circuit can do is more modest: it selects a narrow band of the ambient radiation and returns it to its immediate surroundings.
The consequence, stated plainly
We do not claim that our circuits produce a signal at 7.83 Hz or at 0.1 Hz.
These values are design references — they describe how the profiles were conceived and named, not a frequency emitted. The only physical quantity we can claim today is the resonant frequency of the circuit, in the radiofrequency domain.
What we name · what we measure
Design references
0,1 Hz
7,83 Hz
1 kHz
≈ 2 MHzMeasured resonance
Design references — the names of the profiles, not emitted frequencies.The only physical quantity claimed: the circuit’s resonance, in the radiofrequency domain.
What they are
Four distinct tunings, verified with instruments.
What they are not
Four emitters of biological frequencies.
The question that matters
Does a different tuning produce a different effect? That is exactly what our pre-registered specificity study has to settle.
Quality control · exact scope
What instrumental verification establishes — and what it does not
Every circuit is individually verified before shipping by measuring its resonance, and our profiles are mutually discriminable with instruments.
What this control covers
The conformity of the circuit — its resonance, and its distinction from the other profiles.
What it does not cover
No physiological effect. No instrument applied to a circuit will demonstrate an effect on a person.
Proof of effect is established on the body — not on the test bench.
Falsifiable prediction
“Deprived of an ambient field, inside a Faraday cage, our circuit should fall silent — an active device would carry on.”
The test is simple and inexpensive. It can contradict us: that is precisely what makes it a test.
Ambient load — the electromagnetic environment as the only source. Official Q-Technology infographic.
02 · Public domain
The patented prior art
The principle is not ours, and we do not claim it. It was patented, and the patents have since expired: they are in the public domain, readable and verifiable by anyone.
The timeline
The idea that living systems respond to precise frequency registers was the subject of a patent filed on 14 December 1995 by Michel Marignan (FR 2 742 339), then of an international extension filed on 13 June 1997 by Alain Baudry and Michel Marignan (EP 0 988 085 · US 6 461 375 · CA 2 294 631). Alain Baudry designs our circuits today. This portfolio has expired: the principle is in the public domain.
The patent places the source of the energy in the organism. We place it in the electromagnetic environment. It is a divergence of interpretation — and it is testable, through the Faraday-cage prediction stated above.
Caveat. A patent establishes novelty, never efficacy. We therefore cite none of the results claims it contains.
Per the manufacturer’s technical documentation
What the manufacturer’s technical documentation confirms
It describes activation of the circuit by the high-frequency components of the ambient electromagnetic field; a circuit comprising a high-frequency oscillating circuit, a coupling self-inductance and a low internal resistance; no internal energy source; and a mode of operation belonging to the radio-frequency identification family.
20 × 20 × 1.3 mm
Dimensions
0.8 g
Mass
An object of that size cannot resonate at low frequency — which converges with everything above.
03 · Safety
Safety and compliance
What we can attest about exposure, and what we cannot.
Per the manufacturer’s technical documentation
Compliance and exposure
Per the manufacturer’s technical documentation, the device was tested by an independent test laboratory, verifying its conformity with the applicable European standard for the assessment of human exposure for radio-frequency identification devices.
What this establishes
Emission stays below the safety thresholds used.
What this does not establish
Neither the existence of a useful emission, nor any effect.
We cite it for what it is — a safety element verified by a third party — and we will not make it say anything else.
What we say about use
Passive, non-invasive, no battery, no electrode. No characterisation has been carried out in the presence of an active implant. If in doubt, seek medical advice. This does not constitute a medical claim.
04 · Peer-reviewed
The literature that bears on our class of object
An inductively coupled passive resonator placed near the body measurably alters the local electromagnetic field. The principle is used in medical imaging, where local enhancements of a factor of 3 to 10 have been demonstrated in vivo.
Alipour et al.Scientific Reports 11:23034 (2021)Peer-reviewed
Shchelokova et al.Nature Communications 11:3840 (2020)Peer-reviewed
Hai et al.Nature Biomedical Engineering 3:69–78 (2019)Peer-reviewed
Limit 1 — the exciting field comes from a machine
In imaging, the wave that excites the resonator is produced by a powerful, tuned device — not by the electromagnetic noise of a room.
Limit 2 — the demonstrated effect is an image
What these works establish is a gain in image quality. Not a physiological effect.
05 · The open question
Can a weak field act?
The literature shows that a very low sensitivity threshold is possible in living systems. It also shows that nobody knows by what mechanism.
Peer-reviewed
Engels et al., Nature 509:353–356 (2014)
Migratory birds lose their magnetic compass under ambient electromagnetic noise — double-blind, at levels below regulatory thresholds.
Peer-reviewed
Pakhomov et al., J R Soc Interface (2017)
A sensitivity threshold on the order of 2 to 3 nanotesla.
Limit 1 — an organ we do not have
These thresholds concern a dedicated sensory organ, magnetoreception, which humans do not possess.
Limit 2 — a degradation, not an improvement
The demonstrated effect is the loss of a function under noise. Nothing there establishes that a weak field can improve anything.
The objection, cited by us. Player & Hore, J Chem Phys 151:225101 (2019) conclude that the radical-pair mechanism must remain “highly speculative”.
“We do not claim to know the mechanism — the best specialists do not settle it. That is why we base nothing on it.”
06 · The parameter
Why asymmetry
Before saying what we measure, we have to say why this parameter. Left-right strength asymmetry is not an indicator we invented: it is an established clinical object, with its thresholds, its tests and its decisions.
Established · clinical literature
In return to sport
Grindem H. et al. — Br J Sports Med 2016;50(13):804-808
Delaware-Oslo cohort: symmetry > 90 % → 84 % reduction in the re-injury rate.
Kyritsis P. et al. — Br J Sports Med 2016;50:946-951
Discharge criteria not met → 4× the risk of graft rupture.
Guan Y. et al. (2022) — systematic review of prospective cohorts
Asymmetry ≥ 15 % → increased injury risk.
Limb Symmetry Index — the return-to-play clearance threshold
LSI 90 %
0 %
100 %
The 90 % threshold (Limb Symmetry Index) is the reference standard for return-to-play clearance. Common professional language, not a theory.
In low back pain
Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH — Spine 1994;19:165-172
Multifidus asymmetry of 31 % ± 8 % on the symptomatic side, versus less than 5 % in healthy subjects; correspondence at the vertebral level in 24 of 26 patients.
Hides JA, Richardson CA, Jull GA — Spine 1996;21:2763-2769
Multifidus recovery is NOT automatic after resolution of a first acute episode.
Beneck GJ, Kulig K — Arch Phys Med Rehabil (2012)
Multifidus volume reduced by 18.1 % at L5-S1 in chronic unilateral low back pain.
Multifidus asymmetry — reported magnitudes
Symptomatic side — Hides 199431 %
Healthy subject — Hides 1994< 5 %
L5-S1, chronic unilateral low back pain — Beneck 201218.1 %
The pain goes away. The asymmetry stays.
It has been published since 1996, and it is the reason we measure this parameter rather than another: it persists after the symptom has gone, and it can be tracked over time.
PainMultifidus asymmetry
The limits — in the same module
⚠
Hides 1994 states it itself: no correlation between the degree of asymmetry and symptom severity. More asymmetry does not mean more pain.
⚠
The 10-15 % threshold used in sport is not always substantiated — a systematic review (J Sports Sci Med, 2021) notes that fifteen of eighteen articles use it without referencing its origin, and that reported effects are inconsistent.
⚠
Opar D. et al. (2015) find no increase in hamstring injury risk at 10, 15 or 20 % — a contradictory result.
⚠
The direction of causality is not established. Does asymmetry promote pain, or does pain inhibit the muscle? Pain-related reflex muscle inhibition is well documented, and argues rather for the second direction.
What this changes · what it does not
What these limits change: they forbid claiming that asymmetry causes pain. We do not claim it.
What they do not change: asymmetry remains measurable, it persists, and restoring it is a recognised rehabilitation objective. That is all we need to justify measuring it — and nothing more.
And one point in favour of our approach: several authors argue for individual assessment rather than a universal threshold. That is exactly what a before/after measurement in the same person, with the same operator, does.
07 · The studies
All the studies, freely available
Every document downloadable, methodology and limits included. These are pilot studies — exploratory, honest about their limits, and set to grow with more participants. Our multi-subject compilations sit alongside our longitudinal series and our n=1 observations; the conflict of interest is declared where it applies.
Asymmetry ÷ 2 to ÷ 5
The weaker side rebalances — observed across 6 independent subjects
Across a series of isometric strength tests on a dynamometer (Kinvent), run on 6 people and several muscle groups (shoulder, knee, psoas), it is the weaker side that gains most: left/right asymmetry drops markedly with the technology (e.g. 15.9% → 4.1%).
Uncontrolled series (fixed order, no blinding) — randomized blinded replication planned. Functional observations, internal data.
Strength & motor control · multi-subject series (Kinvent)
Observed · multi-subject · uncontrolled
Shoulder abduction — Jon
Dynamometer: peak force up on both sides, asymmetry roughly halved.
Transparency note. These are pilot studies. The Strength series (Kinvent) covers several independent subjects on a dynamometer, but remains uncontrolled (fixed order, no blinding). The EEG, physiological and longitudinal observations are single-subject (N = 1); where the subject is also the inventor, the conflict of interest is declared. All are exploratory — they generate hypotheses to confirm, and the body of work will grow with more participants and randomized blinded protocols. No medical claim.
08 · Method
What measuring actually means
Before claiming a difference is real, you have to know what the protocol can distinguish. In dynamometry, that answer does not come from the device.
< 1% to > 50%
The range of minimal detectable change reported in dynamometry across studies — with the same instrument.
It is not the sensor that decides
It is the standardisation of the protocol: position, instruction, trial order, operator. Between-operator variance accounts for 70 to 80% of total error.
Our commitment
“We are establishing the minimal detectable change of our own protocol, and we will publish that figure. Until it is established, we describe what we observe without claiming that any given difference is significant.”
09 · Transparency
Three tiers, six open questions, three commitments
Every claim on this page carries its level of evidence. What remains open is written here, before anyone asks.
Proven
Peer-reviewed literature, cited with its full reference — limits included.
Observed
Our internal testing, with its sample size, protocol and limits declared.
Hypothesis
Exploratory. To be confirmed — or dropped if measurement contradicts us.
The six open questions
01The exact signature of the circuit in use.
02The detection threshold of our protocol.
03Whether the effect survives blinding.
04The stability of the profile over time.
05The mechanism.
06The real level of field returned to the surroundings.
The three commitments
Protocols written before the first measurement.
An independent third party holding the key to the blinding, alone.
Publication of the results, including negative ones.
The limit we accept. Because the exact calibration is not published, a third party cannot verify the specificity of our circuits. That is precisely why a blinded result counts for more, not less: if the effect exists, it will show up in a sham-controlled study, without any value needing to be disclosed.
The trap we have to name ourselves
The quality of this document is not a data point about the product. A page can be perfectly calibrated on its own weaknesses and still describe something that does not work. “They are so transparent that they must be right” is false reasoning — and transparency at this level is, mechanically, more persuasive than a dishonest pitch. We write this because it is true, and because we do not want the rigour of the text to stand in for proof. It is not proof. Only a result is.
10 · Sources
References
Every source cited on this page — including those that contradict us, flagged as such. That contrast is what separates this document from a commercial pitch.
Hides JA, Richardson CA, Jull GA — Spine 1996;21:2763-2769. DOI
Beneck GJ, Kulig K — Arch Phys Med Rehabil 2012. DOI
Guan Y. et al. (2022) — systematic review of prospective cohorts.
Bishop C. et al. — J Sports Sci Med 2021;20:594-617. DOI
Contrary source
Opar D. et al. (2015).
Contrary source
Passive resonators · weak fields
Alipour A. et al. — Scientific Reports 11, 23034 (2021).
Shchelokova A. et al. — Nature Communications 11, 3840 (2020).
Hai A. et al. — Nature Biomedical Engineering 3, 69–78 (2019).
Engels S. et al. — Nature 509, 353–356 (2014). DOI
Pakhomov A. et al. — J R Soc Interface 14, 20170364 (2017). DOI
Usselman RJ. et al. — Scientific Reports 6, 38543 (2016). DOI
Player TC & Hore PJ — J Chem Phys 151, 225101 (2019) — major objection to the radical-pair mechanism. DOI
Contrary source
Rubin GJ. et al. (2005/2010) — reviews on electro-hypersensitivity.
Contrary source
The invitation
Build one. Test it.
The principle of our circuits is public, the patent has expired, its description is complete and its example values are published. Any laboratory can build one and test it. We invite exactly that, and we will respond to any serious request for collaboration or independent replication.
