Loaded finite-window observation of a prepared collective reference¶

Declared question and candidate¶

This design admission follows the prepared matrix-response theorem. It asks whether a finite observation with an explicit passive sensor can preserve a predictive consequence and support an independent physical test. The analytical declaration and result below remain reusable under their stated premises; the research program alone selects further work. No apparatus, measured response, source acquisition or fine-node extension is supplied by this mathematical design.

Keep the complete C5 law, signed forms U, circular phases Phi, held unit capacities, fixed support, normalized structural clock tau and exact gamma=1/(1023*pi) of the owner. Write v=(1,-1,0,0,0), W_r=I+r*L^2, y=v^T*L*U. Add one sensor state z with

\[ W_r U'=(-LU+\gamma S(\Phi))/4+f-kv(y-z),\qquad \Phi'=\gamma LU/4,\qquad \tau_s z'=y-z. \tag{1} \]

The sensor loads the form row through its actual current; it is not an additive observation error. The selected observable y differs from the earlier v^T*U: it is proportional to a physical differential voltage in the declared electrical dictionary.

Primitive	Fixed analytical declaration
Response family and control	Fixed `r in [1/100,1/10]`, compared separately with `r=0`
Full preparation	The earlier centered winding `w=+1` or `w=-1` reference, `\|\|U(0)\|\|_2+\|\|theta(0)\|\|_2<=delta=2^-26`; the same continuous perturbation lift and closing-edge offset
Sensor preparation	Independently bounded `\|z(0)\|<=delta_s=2^-26`
Input	`f=a*v` on `[0,T]`, zero afterwards; `a=1/16`, `T=1/1024`
Sensor law	`k=1/64`, `tau_s=T`; no resets, phase forcing or support events
Observation	`J=(1/T)*integral_0^T z(tau) d tau`, with separate additive error `\|e\|<=eta=2^-26`
Identity criterion	Every oriented phase gap remains within `rho=1/10` of its prepared `2piw/5` during the pulse and later unforced evolution
Matrix-response criterion	Whole-preparation observed intervals for positive r and r=0 are disjoint with separation `>2^-21`
Phase-sensitive control	Compare with a passive linear reference having the same r, form/sensor initialization, input, clock and sensor but omitting the sine-current branch; determine whether permitted records must be disjoint or can overlap
Physical admission	Independent complete preparation, loaded transfer and matrices, delivery, clock and observation error/coverage under the same laboratory protocol
Calculation	Exact rational inequalities and certified exponential/trigonometric bounds; no trajectory, parameter scan, fitted response or physical dataset

The phase-blind reference tests what this observable resolves after linear calibration. It does not override independently established junction data. An independently justified sine-current law would be additional evidence, not something inferred from an unresolved pulse observation.

Source review concerns terrestrial methods and specifications only. Earlier frozen protocols, source archives, responses and the separate endpoint-voltage feasibility review remain unchanged. A conditional sensor equation is not a manufacturer performance claim or a hardware implementation.

Analytical result and admission boundary¶

Revision 34baddca freezes the declaration above. The following calculations use its complete loaded law, without a trajectory, measured response or parameter scan. The candidate retains the prepared identity and separates the positive-r matrices from r=0. However, its permitted records overlap those of the declared phase-blind linear reference. It therefore does not admit an organization-sensitive physical test. Independent full preparation, component/input/clock calibration and observation coverage also remain open.

Own storage, identity and resources¶

Write H_r=U^T*L*W_r*U/2+V(Phi), V=sum_edges[1-cos(Phi_j-Phi_i)] and V_*=5[1-cos(2*pi/5)], as in the matrix-response owner. For the loaded law, the correct storage and its derivative are

\[ H_{r,s}=H_r+\frac{k\tau_s}{2}z^2,\qquad H_{r,s}'=-\frac14\|LU\|^2-k(y-z)^2+(LU)^Tf. \tag{2} \]

Indeed, the form row adds -k*y*(y-z) to the original balance, while the sensor contributes k*z*(y-z). Their sum is the displayed loss. Treating the sensor as a detached reading would omit both terms. The form mean and common mean of the declared phase lift remain fixed: v and f are neutral and the phase row is unchanged. The solved input and loading channels are M_r*f and -k*M_r*v*(y-z).

Let theta be the declared centered perturbation lift. In the relative cell max|D theta|<=rho=1/10, the earlier cosine bound gives V-V_*>=||D theta||^2/10. Initial admission follows from 2*delta<rho. A first boundary contact would require excess storage E_s=H_{r,s}-V_*>=1/1000. Using ||L||<=4 and completing the square in (2), including the sensor's nonnegative energy, gives

\[ E_s(0)\le(4+32r_{\max})\delta^2+\frac{kT}{2}\delta_s^2, \qquad E_s'\le\|f\|^2,\qquad E_s(\tau)\le\overline E_s=E_s(0)_{\rm upper}+2a^2T <2^{-16}<\frac1{1000}. \tag{3} \]

This excludes boundary contact throughout the pulse. After T, (2) is nonincreasing, so the identity remains inside the collar for all subsequent unforced time. The globally Lipschitz sine source and constant invertible matrices ensure continuation. This is retention of the supplied identity, not acquisition. The full preparation cost includes V_*; the excess in (3) is neither total storage nor a source for the external input.

For finite response bounds put X=||U||+||theta||. The sine source obeys ||S(Phi_w^*+theta)||<=4||theta|| globally; ||M_r||<=1, ||v||<2, ||Lv||=sqrt(20)<5, and gamma<1. The stable sensor convolution bounds sup|z|<=delta_s+5*sup X. Integrating both live evolution rows gives, when (2+20k)T<1,

\[ \sup_{[0,T]}X\le Z= \frac{\delta+2aT+2kT\delta_s}{1-(2+20k)T},\qquad B=(1+20k)Z+2k\delta_s, \tag{4} \]

and ||U(tau)-a*tau*M_r*v||<=delta+tau*B. These inequalities hold without freezing phase or relying on the identity flag. In particular, they retain unknown sensor initialization and back-action. The separate resource bounds are

\[ \int_0^T|(LU)^Tf|\,d\tau\le8aTZ,\quad \int_0^T\frac14\|LU\|^2\,d\tau\le4TZ^2,\quad \int_0^Tk(y-z)^2\,d\tau\le kT(\delta_s+10Z)^2. \tag{5} \]

Equation (2) retains the signed supplied work and both continuous losses. The growth allowance 2a^2T in (3) is not an equality for delivered work.

Sensor memory and finite matrix separation¶

The exact sensor solution is

\[ z(\tau)=e^{-\tau/\tau_s}z(0)+\frac1{\tau_s} \int_0^\tau e^{-(\tau-u)/\tau_s}y(u)\,du. \tag{6} \]

Eliminating z from the form row leaves the term -k*v*y(tau)+k*v*exp(-tau/tau_s)*z(0) plus the convolution in (6) multiplied by k v. Thus a reduced form law retains both sensor memory and initialization. Replacing z by y instantaneously would change the model. Integrating (6) over the declared window yields

\[ J=\frac{\tau_s}{T}(1-e^{-T/\tau_s})z(0) +\frac1T\int_0^T[1-e^{-(T-u)/\tau_s}]y(u)\,du. \tag{7} \]

For tau_s=T, let alpha=1/2-exp(-1)>1/8. Equation (4) then implies

\[ J=aT\alpha\,g(r)+R_J^{\rm actual},\qquad |R_J^{\rm actual}|\le R_J=\delta_s+5\delta+\frac{5T}{2}B, \qquad g(r)=v^TLM_rv=\frac{6+20r}{1+15r+25r^2}. \tag{8} \]

The positive kernel in (7), with total weight at most one, bounds the initialization and constant port remainder; its first moment is at most T/2. The neutral C5 identity L^2-5L+5I=0 supplies the exact gain. Its derivative has numerator -70-300r-500r^2, so it decreases strictly. For the fixed family, g in [32/11,2480/461], whereas g(0)=6. Writing b_J=R_J+eta, the entire permitted record sets lie within

\[ \mathcal J_+= [aT\alpha(32/11)-b_J,\ aT\alpha(2480/461)+b_J],\qquad \mathcal J_0=[6aT\alpha-b_J,\ 6aT\alpha+b_J]. \tag{9} \]

Both use the same alpha. Exact rational substitution already gives

\[ \inf\mathcal J_0-\sup\mathcal J_+ \ge\frac{aT}{8}\frac{286}{461}-2b_J>2^{-21}. \tag{10} \]

The implementation also encloses alpha tightly through the shared rational exponential bounds. Independently widening alpha to unrelated coarse values in the two records would discard their correlation and needlessly weaken the separation. This is a finite loaded-response distinction between specified matrices, conditional on the complete preparation and error bounds.

What the same observation cannot distinguish¶

The declared reference keeps the same r, U(0), z(0), input, clock and sensor:

\[ W_r\widehat U'=-L\widehat U/4+f-kv(\widehat y-\widehat z), \qquad \widehat y=v^TL\widehat U,\qquad T\widehat z'=\widehat y-\widehat z. \tag{11} \]

It consumes no phase coordinate. Its own passive storage is Uhat^T*L*W_r*Uhat/2+k*T*zhat^2/2; its derivative is (2) with hats. The comparison matches calibrated linear components and omits the sine-current branch. Independently established junction data could exclude this reference; the present calculation tests what the finite pulse adds by itself.

For the actual complete law, ||theta(tau)||<=delta+gamma*tau*Z. Write E=sup_[0,T]||U-Uhat||. Equal sensor initialization and (6) give sup|z-zhat|<=5E. The form difference therefore satisfies

\[ E\le(1+20k)TE+\gamma\delta T+\frac{\gamma^2ZT^2}{2},\qquad |J-\widehat J|\le5E\le \frac{5\bar\gamma(\delta T+\bar\gamma ZT^2/2)} {1-(1+20k)T},\quad \bar\gamma=\frac1{3000}>\gamma. \tag{12} \]

The first term bounds both diffusion and actual sensor back-action; the remaining terms bound the full sine source about the twist. These are global Lipschitz estimates, not a tangent reduction. The fixed declaration gives Z<1/8000. Even the looser substitution Z<=1/4000 proves

\[ |J-\widehat J|\le \frac{47003}{1930258022400000000}<2^{-44}<\frac\eta{100}. \tag{13} \]

This establishes overlap of actual allowed noisy records, not just overlap of two outer enclosures: the common record (J+Jhat)/2 requires opposite errors of magnitude |J-Jhat|/2<=eta in the two models. Hence no uniformly error-robust discriminator exists for this pulse, scalar observation and declared error class. It does not assert equality of the noiseless responses, equality for every error realization, or impossibility for another experiment. The matrix separation in (10) remains valid. It cannot be attributed to a resolved phase-organization effect by this observation.

Independent preparation cannot be inferred from energy and winding¶

Within an independently admitted centered collar, L>=I on neutral forms and phases gives

\[ E_r\ge\tfrac12\|U\|^2+\tfrac1{10}\|\theta\|^2, \qquad (\|U\|+\|\theta\|)^2\le12E_r. \tag{14} \]

Thus an independent E_r<=delta^2/12 certificate is sufficient there, with separate sensor admission. It does not establish the collar itself or make phase-dependent energy an independently available reading.

There is a global counterexample even with exact target energy. Set winding w=1, kappa=2*pi/5, U=z=0 and choose oriented gaps

\[ d(q)=\left(q,\frac{2\pi-q}{4},\frac{2\pi-q}{4}, \frac{2\pi-q}{4},\frac{2\pi-q}{4}\right), \quad V(q)=5-\cos q-4\cos((2\pi-q)/4). \tag{15} \]

At q=kappa+rho, the local bound used in (3) gives V(q)-V_*>=rho^2/8=1/800>0. At the limiting endpoint q=pi, V(pi)=6-2*sqrt(2)<16/5, whereas V_*=(25-5*sqrt(5))/4>55/16; hence V(pi)-V_*<-19/80. Continuity supplies a q0 in (kappa+rho,pi) with exactly V(q0)=V_*. All five gaps are in (0,pi) and sum to 2*pi, so winding remains one. Reconstruct a continuous lift by cumulative gaps and subtract its perturbation mean. This preserves the gaps and places both states on the same mean leaf.

The constructed state and the prepared twist have identical zero nodal and sensor voltages, winding, means and total own storage, yet the constructed state lies outside the required collar. Its existence follows without locating a root or evaluating a trajectory. Consequently these aggregate preparation quantities cannot globally certify the local phase prior.

Physical dictionary and method evidence¶

Under the separately supplied electrical reference, choose c,g,a0>0 and r=c_J/c. The laboratory clock is tau=b*t_lab, with both clocks zero at observation start. With P the neutral projector, retain

\[ C_{\rm phys}=cL^++c_JL,\quad G=gP,\quad b=4g/c,\quad v_0=a_0b\gamma/4,\quad V=v_0LU,\quad j=a_0\gamma^2g^2/c,\quad I_{\rm ext}=cv_0bf. \tag{16} \]

A series resistor-capacitor branch across the v port has capacitor voltage V_s=v0*z, conductance g_s and capacitance C_s. Its exact dictionary is

\[ k=\frac{g_s}{cb}=\frac{g_s}{4g},\qquad \tau_s=\frac{bC_s}{g_s},\qquad k\tau_s=\frac{C_s}{c}. \tag{17} \]

Thus the fixed candidate requires g_s=g/16 and C_s=c/65536. Kirchhoff's current equation and the capacitor equation give (1), including loading. The physical storage scale is c*v0^2=a0*j, so physical sensor energy is C_s*V_s^2/2 and (2) transforms with the same clock as every row. These equalities specify a conditional component model; they do not establish its fabrication or accuracy. Any subsequent voltmeter loading must also be admitted rather than silently removed.

The physical mean is J_lab=v0*J over t_window=T/b. If the stipulated structural error eta is to cover a uniform mean-voltage error e_V, a necessary compatibility condition is

\[ e_V\,t_{\rm window}\le\frac{a_0\gamma T\eta}{4}. \tag{18} \]

Slowing b alone leaves this product unchanged. For the conventional supplied Josephson conversion a0=h/(4*pi*e), it is approximately 3.7254e-31 V s at the declared T and eta. The exact SI constants h and e come from NIST SP 330, Section 2. They are not derived from TNFR. Equation (18) is a requirement of this protocol, not a noise floor, uncertainty principle or universal impossibility theorem. Independent timing, input, transfer and parameter errors would also need their own propagated allowances before using (9) as a physical prediction.

The Keithley measurement handbook describes resistance, bias and capacitive loading (Section 2.2.1), bandwidth and settling (2.6.4), finite integration (3.2.8), and distinct accuracy/uncertainty/noise considerations (1.4.2-1.4.4). These support modeling sensor state and averaging. They do not certify this exact one-pole law, the full C5 preparation or the joint deterministic error bound. The source receipt records reviewed sections and limits. It claims web-text review, not archived source bytes, measured arrays, admitted apparatus or an evaluated response.

Reusable consequence and remaining question¶

The detached exact bound uses shared loaded-pulse admission for the pulse and preparation primitives, sensor energy, forced growth and identity bound. It separately admits the observation error and short-window norm denominator, then rebuilds the response bounds in (3)-(13). Its independent controls check matrix gain, sensor storage, convolution coefficients, arithmetic domains, preparation obstruction and the physical scaling. This private proof consumer installs no runtime law or SDK pathway. Earlier frozen protocols and responses retain their original models and budgets.

The short-window mean supplies a scoped obstruction and reusable sensor accounting. The subsequent recovery-area result conditionally distinguishes restoring phase feedback from the same calibrated passive linear response under a separately declared observation and error contract. It does not identify a unique restoring law or supply independent physical preparation and measurement. Repeating the short-window mean with tighter arithmetic does not supply those missing premises. The sole active task and its acceptance conditions remain in the execution plan.