Power Output

The power output $P(t)$ is the quantity the entire Schubart Master Formula exists to compute. Where the other five factors describe the environment and the device, $P(t)$ is the result: the instantaneous usable power delivered by a converter operating in a continuous ambient field. Every other page in this section describes an input; this page describes the output they combine to produce. It is the left-hand side of the equation,

$$P(t) = \iint \Phi(E,\theta,t)\,\sigma(E)\,\Delta p\,v_{\text{ph}}\,\eta_{\text{conv}}\;dE\,d\theta .$$

Depth 1 — Intuition

Picture a waterwheel on a stream. The stream is the resource, the wheel’s design determines how much of the current it catches, and the axle finally delivers turning power to the mill. $P(t)$ is that delivered power — not the water in the stream, not the cleverness of the wheel, but the actual work handed off at the axle, measured moment by moment. The parenthetical $(t)$ is the whole point: the stream swells and slackens, and so the power at the axle rises and falls with it. $P(t)$ is not a single number stamped on a nameplate but a living quantity that tracks the changing environment.

This factor matters because it is the answer to the only question that ultimately counts: how much usable power is this converter producing right now? All the preceding factors are means; $P(t)$ is the end. It is also what makes the framework falsifiable — a predicted $P(t)$ can be compared against a measured one, which is how the abstraction earns its keep. Everything upstream is bookkeeping in service of getting this one function right.

Depth 2 — The physics

Physically, $P(t)$ is instantaneous usable electrical power, and its time argument distinguishes it sharply from a steady-state rating. Because the ambient resource $\Phi(E,\theta,t)$ fluctuates, the output tracks those fluctuations: $P(t)$ inherits its time dependence directly from the environment, filtered through the response of the device. A converter in a variable field therefore produces a variable output, and averaging $P(t)$ over a suitable interval yields the mean power one might quote for a datasheet, while the instantaneous form retains the dynamics that averaging discards.

$P(t)$ is always defined relative to a control surface — the boundary across which usable power is delivered — so it is meaningful only once the system boundary is specified; see system boundaries. It is also intrinsically a non-equilibrium quantity. A converter in equilibrium with its surroundings produces no net output, because the directed momentum transfer that a nonzero $P(t)$ requires vanishes there. A positive $P(t)$ is thus the macroscopic signature of a system exchanging with an environment that is itself out of balance, which places power output squarely within the framing of non-equilibrium energy conversion. The magnitude of $P(t)$ has no universal scale: it depends entirely on the resource and the device, and can range across many orders of magnitude between applications, from faint ambient fields to the more energetic environments considered in specialised cases such as neutrinovoltaics.

Depth 3 — The mathematics

Power output carries the SI dimension of power,

$$[P(t)] = \mathrm{W} = \mathrm{J\,s^{-1}} = \mathrm{kg\,m^{2}\,s^{-3}} ,$$

and it is obtained by integrating the full integrand over energy and angle:

$$P(t) = \iint \Phi(E,\theta,t)\,\sigma(E)\,\Delta p\,v_{\text{ph}}\,\eta_{\text{conv}}\;dE\,d\theta .$$

The dimensional bookkeeping is the essential check. The integrand carries the dimensions of the five factors combined — the flux density $\Phi$, the area $\sigma$, the momentum $\Delta p$, the velocity $v_{\text{ph}}$, and the dimensionless $\eta_{\text{conv}}$ — and the integration measure $dE\,d\theta$ supplies the energy and angular units. When these are multiplied and the two integrations performed, the energy, angular, and area dimensions cancel against those carried by $\Phi$‘s density structure, leaving power. The full dimensional reduction is worked through in mathematical architecture; the same accounting is summarised in variable definitions.

The double integral is not decorative. It expresses that $P(t)$ is an accumulation: contributions arrive from every part of the spectrum where $\Phi$ and $\sigma(E)$ overlap and from every direction the field spans, and the total power is their sum. This is why no single interaction determines the output and why the framework is written as an integral rather than a product. Two limiting checks confirm the structure. If the field is equilibrium-symmetric, the angular integral of the directed contribution vanishes and $P(t) \to 0$, as required. If the field is monochromatic and unidirectional, the integrals collapse to a single evaluation and $P(t)$ reduces to the product of the five factors at that energy and angle — the elementary case contained within the general one. Only the time argument survives the integration untouched, which is why the result is a function $P(t)$ rather than a constant.

In the balance

$P(t)$ is where the whole pipeline reports its result: the resource of the radiation flux density, gated by the scattering cross-section, directed by the momentum transfer, transported by the phonon velocity, and rectified by the conversion efficiency, all accumulated over energy and angle. It is the one factor that is computed rather than supplied, and its value can never exceed what the resource and the losses together allow — the framework’s guarantee of physical honesty, delivered as a number.

References

• H. B. Callen, Thermodynamics and an Introduction to Thermostatistics, 2nd ed. (Wiley, 1985). Foundational treatment of work, power, and the thermodynamic limits on energy conversion.
• S. R. de Groot and P. Mazur, Non-Equilibrium Thermodynamics (North-Holland, 1962; Dover reprint, 1984). Rigorous framing of power delivered by systems driven away from equilibrium.
• L. D. Landau and E. M. Lifshitz, Statistical Physics, Part 1, 3rd ed. (Pergamon Press, 1980). Standard reference for fluctuating, time-dependent macroscopic quantities of which $P(t)$ is an instance.