Validation

Technology Validation

A working technology must be measurable.

VENDOR.Max is a non-linear electrodynamic installation operating in a controlled discharge-resonant regime. This page neither presents nor defends the physics of that regime.

We do not ask an independent laboratory to agree with our explanation of what happens inside the enclosure. We design the test so that agreement is unnecessary.

Measure the product. Inspect the available internal store. Choose the site. Measure the energy delivered to the load.

The previous build established the engineering viability of the architecture and produced the internal experimental material of the project. What is presented for independent validation is a prototype rebuilt on a new component base. Independent validation does not make the result exist — it moves it into a different status: a result recorded by the measuring party’s own instruments.

What validation is

Validation is measurement. A claim about a physical result, stated in advance, is checked by an independent experiment with a defined boundary, a defined load, measured inputs and outputs, and an acceptance criterion fixed before the first measurement is taken.

Independent validation here means measuring a physical result. Regulatory assessment of the product is not the subject of this page.

What is validated

Not a theory. Not a patent. Not our account of the mechanism. Not service life.

What is validated is a measurable result produced by a working installation under fixed boundary conditions. The measuring party is given an experimental question that requires no trust in the contents of the enclosure:

What measurable energy does the installation deliver to an external load under fixed boundary conditions, and for how long?

Answered by the measuring party’s own instruments

The answer is established by instruments belonging to the measuring party. Internal circuit design is neither required for it nor disclosed.

Structure of the test

Two experimental questions

The test answers two questions. They do not substitute for one another, and both are executed without disclosing the internal circuit.

01 — Output

What energy did the external load receive, and over what period?

Established by a traceable electrical measuring chain and independently cross-checked by calorimetry.

02 — Boundary and stores

What external energy connections exist, and can an internal store account for the observed result?

Established by an inventory of physical penetrations of the enclosure and by a conservative upper bound on the energy of the stores found inside it.

There is no third question

The test does not attempt to establish the source of the energy, and disclosure of the internal mechanism is outside its scope. The public account of the physical mechanism and the patent position of the project exist separately; the closed engineering implementation is not the subject of this test.

Method

The calorimetric load test

Electrical measurement of a complex electrical process is sensitive to phase, waveform and the mixing of peak, RMS and mean quantities. A thermal measurement is not computed from those quantities at all.

The kettle explains the method
The calorimeter executes it
The reference heater calibrates it

A calorimetric result is determined by the thermal effect in the load and does not require the delivered energy to be reconstructed from the shape of voltage and current. It does, however, require a closed thermal account of its own — which is why the calorimeter is calibrated by electrical substitution before the run and again after it.

The kettle explains the method

An ordinary domestic kettle with a known mass of water is understood by any reader: mass, initial temperature, final temperature, elapsed time. The change of state of the water is a physical fact that does not depend on how the source is built.

\[Q = m\,c_p\,\Delta T\]

Why a domestic kettle is not an instrument

Water-only accounting systematically returns less than the energy supplied: part of it goes into the vessel walls, the heating element and the lid, part is lost to convection, radiation and evaporation, the specific heat capacity of water depends weakly on temperature, and the automatic cut-off shifts the end point. The result of a domestic experiment is a lower estimate, not a measurement.

The calorimeter executes the method

The formal test uses a calibrated resistive heater, a known thermal mass, forced stirring, controlled losses, traceable temperature sensors and a measured mass of working fluid. The working window is chosen without a phase transition — approximately 25 to 75 °C, with the lower point above ambient so that heat exchange with the surroundings keeps the same direction throughout the run.

The reference heater calibrates the method

Before the VENDOR.Max run, the same calorimeter, the same heating element, the same mass of working fluid, the same stirring and the same temperature window are driven from a traceable source of measured energy. The calibration is repeated after the run. The substitution topology is fixed by the protocol: during calibration and during the run, one and the same heating element is driven in one and the same thermal system.

Why this matters

The calorimeter is first calibrated by a source of known energy. VENDOR.Max is then asked to do the same thing. Two heatings are compared, not a calculation and a measurement. The effective heat capacity of the system and the loss coefficient are determined experimentally rather than modelled, and repeating the calibration afterwards makes any change in the measuring chain itself detectable.

Water energy only, idealised. Losses and the heat capacity of the vessel are not included; \(c_p \approx 4.184\) kJ/(kg·K). Illustrative figures — not a test result.
Water	Start	ΔT	m c_p ΔT	Equivalent	Ideal time at 2.4 kW
1.0 L	22 °C	78 K	326.4 kJ	90.7 Wh	136 s
1.0 L	24 °C	76 K	318.0 kJ	88.3 Wh	132 s
1.0 L	26 °C	74 K	309.6 kJ	86.0 Wh	129 s
1.7 L	22 °C	78 K	554.8 kJ	154.1 Wh	231 s
1.7 L	24 °C	76 K	540.6 kJ	150.2 Wh	225 s
1.7 L	26 °C	74 K	526.3 kJ	146.2 Wh	219 s

Two evidentiary methods and one observable indicator

The evidentiary result is formed by two measurement methods: a traceable electrical measuring chain and calorimetry. The visible energy counter does not enter the decision criterion.

Enclosure output port → visible energy counter → load / calorimeter

The traceable measuring chain operates on the same output port. The visible energy counter is an external element of the measuring assembly, located outside the enclosure boundary; its own consumption is not attributed to the calorimetric load. The measurement planes of the traceable electrical chain and of the calorimeter are fixed in the protocol before the run, and any difference between them is accounted for explicitly.

Visible energy counter

A compact plug-in energy meter is inserted between the output port and the load and continuously displays voltage, current, active power, power factor and accumulated energy. Its function is public observability of the experiment in real time rather than metrology: anyone present reads the accumulated kilowatt-hours directly from the display, with no data processing.

The rise of accumulated energy on the display makes the progress of the experiment directly observable. The moment at which the calculated bound on the internal store is formally crossed, however, is determined solely from the data of the traceable measuring chain and the adopted decision rule. Readings of the visible counter do not enter that calculation.

The counter is neither a deciding instrument nor an independent measurement channel: it sits in the same electrical circuit as the traceable chain and measures the same physical quantity by the same method. Independence is provided by calorimetry, not by a second electrical instrument. Instruments of this class have their accuracy specified for sinusoidal voltage and linear loads, and on a real output signal that condition holds only approximately. Counter readings are compared with the traceable chain on a known load before and after the run, and the deviation is entered in the report for reference.

Electrical energy recorder

A traceable instrument of known accuracy class keeps a continuous time-stamped record: voltage, current, active power, power factor, accumulated energy. Raw data are exported directly from the recorder and preserved in the primary data package without editing; the original exported files are attached to the report. This is the instrument that forms the electrical part of the result.

Calorimeter

An independent check of the same result by a different physical method, with substitution calibration before and after the run.

Agreement between two evidentiary methods of different physical nature, within the stated uncertainty, is considerably harder to explain by the error of a single instrument. All instruments used belong to the measuring party or are accepted and sealed by it before the run begins.

Continuous video recording keeps the display of the visible counter, the load and a clock in frame at the same time. The recording has the status of observability material and does not form part of the measurements.

Boundary

The test boundary and its passport

The boundary passport lists the physical penetrations of the enclosure relevant to external energy or material exchange. Its subject is power and service penetrations, not a complete energy inventory of the product: thermal and field accounting belongs to a later research level and is outside the scope of primary validation.

Compiled before the run by the owner, checked by inspection of the measuring party, and extended by anything the inspection finds that was not entered.
Channel	How it is handled
External starting channel, where present	energy of the starting phase is fully recorded; after start-up the source is physically disconnected and the absence of further exchange is monitored
Internal battery of the regime branch	does not cross the external boundary; counted in the inventory of internal stores and in the analysis of their change
Built-in control and indication interface	located inside the enclosure boundary; not an energy input in itself
External service connections to the interface, where used	entered in the boundary passport; disconnected or measured during the main run
Output port	energy delivered to the load is measured
Protective conductor, where structurally present	retained for safety; verified not to be used as an unaccounted power input
Other external power connections	inventoried before the start; disconnected or explicitly accounted for during the main run
Material flow, where structurally possible	recorded as external exchange; where absent, the absence of the channel is stated

Start-up configuration

The start-up configuration is fixed before the test begins.

If the unit presented uses an external starting source, the energy of the starting phase is recorded in full; once start-up is complete the external source is physically disconnected and the state of the corresponding external channel is monitored by the measuring party throughout the main run.

If start-up is performed from the internal battery of the regime branch, there is no external starting energy channel: the unit is started from its own control element, and the battery remains inside the enclosure boundary and is counted as an internal energy store and regime buffer.

The control element and the indication panel are built into the product and sit inside the enclosure boundary; in themselves they are not energy inputs. If configuration or readout is performed over an external connection, that physical channel is entered in the boundary passport and is either disconnected or measured during the main run.

No term of the energy account is computed as the remainder of an equality. A quantity obtained by subtraction is not a measured quantity.

Discipline of scope

Controlling simple factors that could distort the result

Primary validation does not investigate every possible explanation of the observed result. It controls a limited set of simple factors capable of directly imitating or distorting the measured result.

Hidden internal store

We do not assert that none exists. The measuring party inspects the composition of the product, builds a conservative upper bound on the energy of every store found, and runs the test beyond what such a store could account for.

Hidden conductor

Every physical crossing of the boundary is inventoried before the run. Each channel is either measured or disconnected and sealed.

Site infrastructure

The measuring party may designate the test location. The unit is transported there switched off. No bench, table or mounting belonging to VENDOR is required unless it forms part of the object under test.

Dependence on the site

Checked by moving the test. The measuring party may select a different room or a different independent site and repeat the run. Retention of the result after a change of location removes the tie to the infrastructure of one particular laboratory as a simple alternative explanation.

Pre-heated enclosure

The thermal state of the installation is recorded before and after the run. This is not a full calorimetry of the product, but the explanation that the enclosure was warmed in advance ceases to be a free parameter.

Consumable substance

The mass of the installation is measured before and after the run. Weighing serves as a check on the state of the product and as an input to those conservative store estimates that explicitly depend on mass. The absence of a measurable change of mass does not by itself exclude a chemical or other internal store.

Measurement artefact

Two evidentiary methods of different physical nature address one result: the traceable electrical chain and calorimetry. The visible energy counter does not enter the decision criterion. A control run is performed in addition: the same protocol, the same assembly, the installation not operating. It establishes that the observed signal is not produced by the measuring chain itself.

Additional control check

Before the protocol is fixed, the measuring party may propose an additional control check that bears directly on the claimed result and requires neither an extension of the agreed scope of the test nor disclosure of the closed engineering implementation. Its inclusion is settled by the agreed protocol before measurements begin.

Internal stores

Inspection of composition and the upper bound on internal stores

An internal store is the one simple explanation that cannot be closed by inspecting the boundary. It is therefore closed quantitatively. The test admits two modes; the second is the primary one.

Mode A

Sealed enclosure, conservative mass bound

The enclosure is not opened. The upper bound is built from the mass of the product, as though its entire mass were given over to a store. This is the harshest mode and the longest in time. Its purpose is to show that the test remains feasible even if the enclosure is not opened for inspection.

Mode B

Inspection of composition

The enclosure is opened to inventory potential stores. The upper bound is built from the components actually found, which normally allows it to be reduced substantially against the assumption that the whole mass of the product is a store. The extent of that reduction is determined only after the unit presented has been inspected.

Design figures, published so that the reader can follow the scale of the test. The measuring party determines mass and dimensions of the unit presented itself, with its own instruments, before the run.
Parameter	Value	Status
External dimensions	600 × 350 × 250 mm	design
Bounding volume of the enclosure	52.5 L	calculated
Mass of the assembled installation	11.5 kg	design
Mean volumetric mass of the product	0.219 kg/L	calculated
Load power of the worked example	2.4 kW	illustrative

The bounding volume is the volume of the external cuboid, not the volume available to house a store. Overall dimensions are given as a public passport of the product and do not enter the calculation of the upper bound: the Mode A bound is built from mass.

Mode A: the mass bound

Suppose the entire mass of the installation — enclosure, magnetic assemblies, electronics, heat sinks, switchgear — were given over to a hidden energy store. For a mass-based estimate this is a deliberately conservative allocation of mass: a real installation inevitably contains non-energetic structural elements.

Illustrative screening under the assumption of 11.5 kg of store. The figures show single coverage of the corresponding internal store and are not the required duration of a validation run.
Assumed available specific energy	Maximum store at 11.5 kg	Coverage time at 2.4 kW
300 Wh/kg	3.45 kWh	1 h 26 min
500 Wh/kg	5.75 kWh	2 h 24 min
700 Wh/kg	8.05 kWh	3 h 21 min
1000 Wh/kg	11.50 kWh	4 h 48 min

How to read this table

The times in the table show single calculated coverage of the corresponding internal store and are not the required duration of a validation run. Under the adopted twofold criterion, confirmed delivered energy must reach at least twice the value of the corresponding store; actual duration is determined from \(E_{\text{delivered}}^{LB}\), not from rated load power.

The illustrative public estimate uses an assumption of 1000 Wh/kg. It is not declared to be the physical or technological maximum of any class of store: it is a deliberately harsh assumption for the sealed-enclosure mode.

Mode B: the bound from components found

Before the run the measuring party weighs and measures the product, opens the enclosure, establishes its composition visually and identifies the components capable of holding a significant energy store. The enclosure is then closed, the load is selected and the required duration is calculated.

The upper bound is built from every class found, not from a single component:

\[E_{\text{stores}}^{UB} = E_{\text{battery}}^{UB} + E_{\text{capacitive}}^{UB} + E_{\text{other}}^{UB}\]

If the principal store is the battery of the regime branch, its rated energy is taken with a pre-established experimental margin; other significant stores are added separately. After a visual inventory, individual terms of the sum may prove negligible — but each is considered explicitly.

Scope of inspection, fixed by agreement before the test begins.
Permitted	Not permitted
visual inspection of composition	circuit measurements on internal assemblies
identification of components capable of storing energy	measurement of transformer and operating-regime parameters
reading markings and rated parameters of such components	oscillography of internal points
weighing and dimensional measurement	disclosure of topology, ratings and phase relationships
recording the state of seals	disclosure of control algorithms

Inspection of composition does not require disclosure of circuit design or of closed operating parameters. Its scope is limited to what is needed to identify and quantify internal energy stores.

What this check closes and what it does not

The inventory closes autonomous classes of store — those that exchange no substance with the surroundings. Classes requiring a supply of oxidiser and removal of products are closed by the boundary passport instead: by the list of physical penetrations of the enclosure, the absence of material flow and the control of mass before and after the run. An autonomous variant that holds both the fuel and the reagents needed to use it inside the enclosure becomes an internal store in its entirety and is included in \(E_{\text{stores}}^{UB}\).

Mandatory qualification

Excluding a bounded internal store excludes that alternative explanation. It does not by itself establish the source of the energy — and establishing the source is outside the scope of this test.

Decision rule

The decision criterion

The subject of the criterion is energy, not duration. The run continues until confirmed energy delivered to the load exceeds the upper bound on the internal store by a factor established in advance.

\(E_{\text{delivered}}^{LB}(\Delta t) \ge 2\,E_{\text{stores}}^{UB}\)

(1)

Here \(E_{\text{delivered}}^{LB}\) is the lower bound on the energy confirmed as delivered to the external load over the run, allowing for the agreed uncertainty budget, and \(E_{\text{stores}}^{UB}\) is the upper bound on the energy of internal stores built according to the selected mode.

It reads simply: the possible store was established, and at least twice that amount of confirmed energy was delivered.

Estimating the required time

If the mean power over the window is steady, then \(E_{\text{delivered}}^{LB} = \overline{P}^{\,LB}_{\text{delivered}}\,\Delta t\), from which

\[\Delta t \ge 2\,\frac{E_{\text{stores}}^{UB}}{\overline{P}^{\,LB}_{\text{delivered}}}\]

This is a derived estimate for planning the run. The decision is made on integral energy, not on time: if power varies during the run, the integral remains the natural quantity of the criterion. Rated load power does not enter the calculation.

On the factor of two

The twofold factor is fixed by the protocol before the test begins. It is not a confidence interval, an uncertainty safety factor or a statistical quantity: the uncertainty already sits inside both bounds. It is a pre-announced additional margin over calculated coverage, and it requires no statistical justification.

On orders of magnitude

In the sealed-enclosure mode the bound is built from the whole mass of the product and the required duration is measured in hours. In the inspection mode the bound is built from the stores actually found and proves substantially smaller, and the required duration falls with it. Specific values for Mode B are determined after the unit presented has been inventoried and are not given here.

Conduct of the test

Site, quarantine and chain of custody

The test requires no VENDOR site. The place of performance is determined by the measuring party.

Choice of site

The protocol can be executed at a site selected by the independent party and may be repeated after a change of site.

Quarantine and chain of custody

The installation is handed over switched off, is sealed by the measuring party and remains at its disposal for an agreed period before the start, with no access by VENDOR personnel. Seals are photographed when applied and checked after the run is complete. The list of seals and of any incidents forms part of the report.

Evidence

Repeatability, reproducibility and the control run

A single successful run is an observation. Beyond that, two different statuses are distinguished.

Repeatability

Several runs of one unit presented under agreed conditions — the same protocol, the same load conditions, the same measurement window — show the stability of the result within a single test configuration.

Reproducibility

A stronger status arises when the result is retained under independent repetition with permitted factors varied: time of performance, composition of the instruments, operator, site, and later the unit itself.

Runs are performed several times, spaced in time; the spread between runs is entered in the uncertainty budget as a separate line.

The control run

The same protocol is executed with the installation not operating, using the same assembly and the same instruments. The control run establishes that the observed signal is not produced by the measuring chain itself.

Fixing the protocol in advance

The version and date of the protocol are fixed before the test begins. Acceptance criteria are determined before the first measurement is obtained and do not change once the run has started. Any amendment before the start is issued as a new version with a new date.

Symmetry

Possible outcomes

An independent experiment is not obliged to end in confirmation. The procedure is symmetrical: it serves confirmation, refutation and the localisation of a discrepancy equally well.

Claim confirmed: The claimed result was reproduced within the established criterion.
Additional boundary input identified: A previously unaccounted external energy channel was found.
Measurement artefact identified: The observed result is explained by the measurement method.
Result not reproduced: The installation did not reproduce the claimed regime under the agreed conditions.
Result not determined: The run did not reach the duration required to exclude a bounded internal store, or the evidential bound on stores proved insufficient. This is a regular outcome of the procedure, not a failure of it.

Right of publication

A requirement written into the agreement with the measuring party before measurements begin: the right to publish the report in full, whatever the outcome, belongs to the measuring party. VENDOR receives no right of veto, no right of editing and no right to defer publication.

Limits of the method

What this protocol does not establish

Acknowledging the limits of a method is part of the method.

It does not establish the source of the energy. Excluding a bounded internal store closes an alternative explanation but does not answer the question of the source. That question is outside the scope of the test.
It does not yield an efficiency figure. End-to-end efficiency is defined for a single control volume across all ports with stores taken into account, and does not follow from port measurements of individual planes.
It does not close uninventoried classes of store. Only the classes entered in the inventory are treated as closed.
It is not a complete energy inventory of the enclosure. Scientific and metrological survey of thermal and field channels belongs to a later research level and is not a condition of the first independent record of the functional result.
It does not establish patent novelty and does not bear on the legal status of the solution.
It does not replace the developer’s own engineering tests and is not replaced by them.

The list of diagnostic classes of discrepancy — an incomplete boundary passport, an unaccounted input channel, a contribution from stores, measurement error, incorrect signal processing, incompatible quantities, non-stationarity of the regime — marks the start of an investigation, not a closed list of possible explanations.

Independent party

Who can be the independent measuring party

This page names no organisations. A named laboratory reads as an arrangement already reached, and none has been. Criteria are named instead.

accreditation for testing in the relevant field;
its own instruments with traceability and valid calibration or verification certificates, including an electrical energy recorder with continuous recording and exportable raw data;
no financial interest in the outcome;
the right to select the test site independently;
the right to publish the report independently;
the right to propose additional control checks before the protocol is fixed;
willingness to sign the protocol and the decision rule before measurements begin.

Disclosure

Both questions of the test are executed without disclosure of circuit design or closed operating parameters. Inspection of composition is limited to what is needed to identify and quantify internal energy stores.

Status

The unit presented and current status

What is offered for independent validation is a specific unit, and its status is stated without borrowing from its predecessor.

The previous build

It established the engineering viability of the architecture and produced the internal experimental material of the project. That is a developer’s result, not an independently confirmed metrological one.

The new build

Following the relocation of the laboratory to the European Union and its restoration, the prototype is being rebuilt on a new component base. It is this new unit that will be presented to the independent party.

Rule of inheritance

Characteristics of the unit presented are not inherited automatically from the previous build and are established afresh by measurement. Past internal tests transfer no metrological status to the new unit.

The open question is independence

The claimed class of observation must be obtained on the unit presented, at a site and under the control of an independent measuring party, according to a decision rule fixed before the test begins.

For an independent test VENDOR sets a contractual requirement that the report be published whatever the outcome. Once performed, published reports are placed on this page. The raw data package and the text of the protocol are provided under controlled access.

Objections

Questions

The questions a technical reader asks first.

Will the device have to be disclosed?

The enclosure is opened to inventory internal energy stores: the measuring party establishes which stores are present and how much energy they are capable of holding. The scope of inspection is limited to that task — circuit measurements, transformer and operating-regime parameters, oscillography of internal points and control algorithms are not part of it.

How is a hidden internal store excluded?

Not by assertion. The measuring party establishes which stores are physically present, builds a conservative upper bound on their combined energy and runs the test beyond what such a store could account for. If the enclosure is not opened for inspection, the bound is built conservatively from the full mass of the product and the required duration increases.

Why heat water instead of using an electrical instrument?

Both. The thermal method does not require the delivered energy to be computed from the phase relationship and the shape of the electrical signals; the electrical chain, in turn, gives time resolution. Agreement between two measurements of different physical nature is harder to explain by the error of a single instrument. The formal test uses a calibrated calorimeter, not a domestic kettle.

What happens if the result is not confirmed?

The report will be published. The right to publish it in full and independently of the outcome is secured to the measuring party by agreement before measurements begin.

Does successful validation mean the source of the energy has been established?

No. It means the external result has been recorded independently and that the internal store found does not account for it. The question of the source is outside the scope of the test.

Is a VENDOR site required for the test?

No. The place of performance is determined by the measuring party, and the run may be repeated after a change of site. Retention of the result across a change of location removes the tie to the infrastructure of one particular laboratory.

Continue

Where to go next

The engineering account of the operating regime, the metrology of active power and the record of the developer’s own testing are held by separate pages.