Granite is twenty to thirty five percent piezoelectric quartz, by volume. The footing of a stone building is a continuous low voltage source. It has never been logged on a Victorian one.

Quartz produces 2.3 picocoulombs per newton, or about fifty millivolts per megapascal of stress. Granite carries a reduced version of the same response.

Piezoelectricity was characterised in quartz by the Curie brothers in 1880. Quartz is the second most abundant mineral in Earth's crust and is strongly piezoelectric: the d11 coefficient (longitudinal stress producing axial voltage) is 2.3 pC/N, which corresponds to about 50 mV per megapascal of applied stress on a quartz crystal. The resonant quality factor is on the order of one million. The temperature stability is solid up to 573 degrees Celsius. None of this is in serious physics dispute; it is the basis of every quartz oscillator in every wristwatch and crystal radio of the twentieth century.

The reduction from pure quartz to granite output is geometric: in a single crystal of quartz, all the molecular dipoles align under the applied stress and contribute coherently. In granite, the quartz grains are randomly oriented, so the contributions from individual grains partly cancel. The empirical result is about a fifth of the pure quartz coefficient, which agrees with rock mechanics measurements published in the rock fracture and seismic precursor literature. A granite of higher quartz content (rose granite from the Aswan quarries, the Alpine granites, the Baltic granites) gives a stronger response. Limestone, sandstone, and brick are not piezoelectric in any significant way. The argument applies to a regional subset of pre 1914 ornamented architecture: New England, Scotland, the Alpine countries, the Nordic region, and the Canadian Shield. It is weak elsewhere.

A massive granite footing carries one to ten megapascals of compressive stress. Tens of millivolts of fluctuating potential follow every wind gust and every thermal cycle.

Static compressive load on a typical granite footing under a stone or masonry Victorian building is in the range of 1 to 10 MPa, depending on the building's mass and the footing's bearing area. At 10 mV per MPa granite output, that is 10 to 100 mV of static potential under the building. The static value is not interesting; charge bleeds off through the surrounding soil and the stone itself over seconds to minutes. What is interesting is the fluctuating component: every wind gust loads and unloads the building's frame by perhaps 1 to 10 percent of its dead load. Every thermal cycle through a daily or seasonal temperature change does the same. Foot traffic, vehicle traffic, distant railway vibration, and seismic microtremor are continuously modulating the load at a low level.

The fluctuating piezoelectric signature is therefore a low frequency noise source in the millivolt range, riding on top of the static potential. It is coupled directly to the iron framework of the building if the iron is bonded to the stone, which it usually is in pre 1914 construction. It is coupled to the earth through the wet soil under the footing. A modern continuous data logger could record this signature with a low noise voltmeter and a high impedance front end. The bandwidth is generous: the response runs from DC up to the megahertz range, so a logger with 10 Hz sampling captures the structural cycling and a higher rate captures any acoustic resonance.

At ten to one hundred millivolts, this is not a power source. The available current is microamperes; the available wattage is microwatts at best. The interest is not in the wattage but in the continuous low voltage modulation of the building's electrical state. If the integrated craft thesis is correct, then a Victorian ornamented building is not just a passive box. It is an electrically active structure: corona discharge at the top, weak vertical dipole field across the iron framework, capacitive coupling to atmospheric potential through the roof ornament, telluric current sampling through the earth electrode, and now a piezoelectric low frequency noise source at the foundation. None of this individually is a large effect. The integrated behaviour is a small electrical instrument that operates continuously, sources nothing meaningful in wattage terms, and has never been characterised on a surviving building.

Anomalous EM signals before earthquakes in granitic crust are the same phenomenon at six orders of magnitude larger.

The seismic precursor literature has a long running thread on anomalous electric and electromagnetic signals detected before major earthquakes. The peer reviewed candidate explanation is piezoelectric stress in granitic crust: as the rock accumulates stress in the build up to a slip event, the quartz content of the granite produces measurable surface voltages and weak radio frequency emissions. The exact source mechanisms and the reliability of the precursor signature for prediction are still debated; the underlying physics, that crustal granite under stress produces measurable EM signals, is in the published rock mechanics record.

The relevance to a single building footing is not that buildings predict earthquakes. The relevance is that the same physics is operating at both scales. The Earth's granitic regions are actively piezoelectric, not dormant. The continental crust under a Scottish granite town hall is producing tectonic scale signals continuously, and the footing under that town hall is in mechanical contact with that crust through the wet earth between them. The town hall's own granite footing produces a building scale version of the same effect under the building's own load cycle. Both signatures exist; only the geological scale one has been logged on instruments.

Low noise electrometer, exposed granite block, wind event, several hours of logging.

Experiment six in the ten experiments programme is the foundation piezo measurement: install a low noise high impedance voltmeter across two points on an exposed granite block of a historic building foundation, in a region with granite construction tradition. Log continuously across at least one wind event, ideally several wind events of varying strength, with simultaneous wind speed logging at the site. The signature should appear as a wind correlated voltage component in the tens of microvolts range.

The connection to the rest of the project. The roof end of an integrated Victorian building has been argued in detail through the theory page and is the subject of the RF audit at experiment one. The foundation end of the same building, in the granite construction subset, carries the piezoelectric signature this page describes and is the subject of experiment six. Together, the two measurements bound the building as a whole: an active vertical structure with a corona source at the top, a piezoelectric stress source at the bottom, and the iron framework connecting them. None of this is a power source. All of it is measurable. None of it has been measured on a surviving building. Nathan Stubblefield's 1898 earth battery (US Patent 600,457) is the closest historical precedent for the foundation end of this picture; the 1928 burning of his papers by his son after his impoverished death anchors the same documentary loss pattern this page treats as a regional construction feature.