Experimental Procedure to Discriminate Between Pull Gravity and EMRP Push Gravity Theory
| Scientific Paper | |
|---|---|
| Title | Experimental Procedure to Discriminate Between Pull Gravity and EMRP Push Gravity Theory |
| Read in full | Link to paper |
| Author(s) | Xavier Borg |
| Keywords | push gravity, EMRP, Xavier Borg, newton |
| Published | 2009 |
| Journal | General Science Journal |
| No. of pages | 11 |
Read the full paper here
Abstract
The aim of this experiment is to create a direct conflict between EMRP push gravity, and Newton's pull gravity. What is gravity is the question that has intrigued many scientists and philosophers alike, but for lack of any experimental evidence has never been satisfactorily answered. All we presently find in mainstream physics is no more than an absurd hypothesis of a pull type gravity innate within all matter, which surprisingly enough, is a concept specifically denied by Newton himself to whom the concept is most often erroneously ascribed. The main problem with past experiments is that both simple push and simple pull theories give the same results. The paper shows the theory behind the original way in which an innate pulling force and an external electromagnetic pushing force may finally be discriminated. Full experimental setup details and experimental results are given for replication purposes.
Overview
Xavier Borg's paper is unusual among push-gravity documents in that it proposes, builds and reports a bench experiment rather than arguing from principle. Its target is the observation that most tests cannot separate push from pull, "since both simple push and simple pull theories give the same results". Borg's claim is that the sign of one small effect differs between the two accounts, and that the difference is large enough to measure with a lead rotor on good bearings.
The theoretical setting is Borg's EMRP model (electromagnetic radiation pressure), in which gravity is not innate to matter but is the momentum transfer of extremely energetic incoming radiation — its frequency proposed to be "close to Planck's frequency" — that is slightly attenuated in passing through matter, so that bodies are pushed into each other's shadows. He opens by quoting at length Newton's letters to Bentley ("Pray do not ascribe that notion to me") to argue that innate attraction was never Newton's own position, and that the causal question was left open by the Principia itself.
The argument
The discriminating prediction
Take an elongated rigid body free to rotate about a horizontal axis through its centre of mass, carefully balanced. In a perfectly uniform field it would stay at any angle; but no such field exists near the Earth.
Under Newtonian attraction the field falls off as 1/r2, so the lower half of a vertical body is very slightly heavier than the upper half. The centre of gravity sits below the centre of mass, and the vertical orientation is the minimum-energy one: released at any intermediate angle, the body oscillates and settles vertical.
Under EMRP the situation reverses. Incoming radiation from below has already been attenuated by the Earth, so the net push is downward; but as the wave passes through the test body its intensity — and hence its push — falls, so the upper half of a vertical body is pushed harder than the lower half. The centre of gravity sits above the centre of mass, and the horizontal orientation is now the minimum-energy one. Borg's summary: "if the body settles in a position close to vertical, then, gravity is a pull innate within matter, but if it settles close to horizontal, then the gravity is an external push."
The EMRP force law
Borg writes the two-body force with attenuation along each body as
- F = k {M1e−(µ1x1) M2e−(µ2x2)} / r2
reducing to Newton's law when µ = 0. He draws from this the corollary that the "constant" G should in practice vary between k and k·e−µ1x1·e−µ2x2, which he offers as an explanation of why G is "presently the worst defined constant in physics". For simulation he recasts it in terms of the mass attenuation coefficient µm = µ/ρ. The key point of the geometry is that Newton makes the body effectively bottom-heavy as an inverse square of distance from the Earth's centre, while EMRP makes it top-heavy as an inverse exponential of distance from its own upper end — so for very small attenuation coefficients EMRP reduces to Newton.
Throwing in a few numbers
For a body of 6 kg and 20 cm long, Borg computes the Newtonian centre-of-gravity offset as 0.1 − (6.4×106(√(1+0.2/6.4×106) − 1)) = 7.8×10−10 m below the centre of mass. At 45° this halves, giving a torque of 6000 gF × 3.9×10−8 cm ≈ 0.2 mgF·cm toward the vertical. Against this, an unloaded good bearing has a static torque of about 0.11 gF·cm, and a loaded one about 6000 × 0.002 × 0.4 = 4.8 gF·cm — some 18,000 times the Newtonian torque. Borg's point is that the Newtonian effect is unmeasurable with such bearings, but that this does not matter, because EMRP's torque acts in the opposite direction: any torque toward the horizontal exceeding the friction floor is by itself non-Newtonian. Simulation of the attenuation iteration suggested a limiting case near µ ≈ 1.6×10−6/cm and a clear positive result near 3.4×10−6/cm.
Apparatus and result
The rotor is a pair of flat lead disks, 200 mm diameter and 8.4 mm thick, clamped rigidly by hardened steel rods to an aluminium sleeve on an 8 mm shaft, turning in 608-2RS hybrid ceramic bearings held in CNC-milled aluminium housings on a rotatable wheeled stand — the rotation of the stand being a control against alignment with the Earth's rotation or magnetic field. Bearings are pre-screened by sticking 0.2 gF of Blu-Tack to the outer ring: acceptable ones turn with as little as 0.1 g. A balancing disk of 10 cm radius serves both to balance the rotor at horizontal and vertical and to read out the torque, since the extra rim weight w needed to arrest motion gives 10w gF·cm.
The reported outcome is that the cores turn toward the horizontal from both 45° and 225°. The arresting weight was 0.16 g, giving 1.6 gF·cm; adding the bearing friction Tf = µfWr = 6000 × 0.001 × 0.4 = 2.4 gF·cm gives a total of 4 gF·cm in the EMRP direction. Fitting the iteration to this torque yields an upper limit of µ = 2.6764×10−6/cm for lead at the gravitational wavelength, making lead about 0.6 million times more transparent to EMRP than to 500 keV gamma rays (µ = 1.64/cm), and about 42 times more transparent than thin air is to the same gammas (µ = 112×10−6/cm).
Assessment
The experimental design is the best thing in the paper, and it is genuinely clever. Borg has identified a sign asymmetry rather than a magnitude, which is exactly the right strategy when the effect is far below the noise floor of the apparatus: he does not need to beat the friction to measure the Newtonian torque, only to see rotation the wrong way. Testing from both 45° and 225° is a real control, because a simple centre-of-mass offset would give a torque that is 360°-periodic while a genuine field-gradient effect is 180°-periodic — and the reported behaviour is 180°-periodic. The write-up is honest about the friction floor, gives full replication details, and states an upper bound rather than a value.
Most of the arithmetic is correct. 6000 × 0.002 × 0.4 = 4.8 and 6000 × 0.001 × 0.4 = 2.4 gF·cm both check; 10 × 0.16 = 1.6 checks; the transparency ratios check exactly (1.64 / 2.6764×10−6 = 6.1×105 and 1.12×10−4 / 2.6764×10−6 = 41.8); the quoted air attenuation is right for 500 keV; and the stated geometry is self-consistent — two lead disks of 20 cm diameter and 8.4 mm thickness at 11.34 g/cm3 come to 5.99 kg, matching the 6 kg used throughout. The factor of one half at 45° is also right, since the gravity-gradient torque goes as sinθcosθ.
Two things do not check. First, the Newtonian baseline is computed with the wrong average. For a uniform rod of length L at radius R the centre of gravity lies L2/6R below the centre of mass; Borg's expression, which takes the geometric mean √(R(R+L))), gives L2/8R. The correct offset is 1.04×10−9 m rather than 7.8×10−10 m and the correct torque about 0.31 mgF·cm rather than 0.2. This is 25% and changes nothing, since the term is negligible either way — but it means the "Newtonian" comparison figure quoted is not the Newtonian one.
Second, and seriously, the friction coefficient is quoted inconsistently in the same paper. On page 4 the loaded bearing coefficient is 0.002, giving 4.8 gF·cm; in the results on page 10 it is 0.001, giving 2.4 gF·cm. The friction term is 60% of the headline 4 gF·cm, so this single unexplained factor of two moves the result between 4 and 6.4 gF·cm. Worse, the arresting-weight method does not measure the driving torque at all: the rotor stops when the driving torque no longer exceeds counterweight plus static friction, so 4 gF·cm is an upper bound, and the only firm statement the data support is that the anomalous torque lies somewhere between the friction floor (2.4 gF·cm) and that bound. There is one measurement, no repetitions, no error bars, and no null control — the obvious one being an aluminium rotor of the same mass and dimensions, which on EMRP should give a very different torque and on any mundane explanation the same.
The deepest problem is internal to the theory and can be checked from Borg's own equation with his own number. His derived µ = 2.6764×10−6/cm for lead corresponds to µm = µ/ρ = 2.36×10−7 cm2/g. The Earth's radial column density is roughly 5.5 g/cm3 × 6.37×108 cm ≈ 3.5×109 g/cm2. Substituting into his own exponent gives µmρ1x1 ≈ 8×102, so the Earth's factor e−µx is of order 10−360: the Earth would be totally opaque to the gravitational radiation, its interior would contribute nothing, and terrestrial gravity would not be proportional to the Earth's mass at all. For the shadowing mechanism to remain unsaturated — which it must, since surface gravity does track GM to high precision, and the Cavendish and lunar-laser-ranging determinations of GM⊕ agree — one needs µmρx ≪ 1 for the Earth, i.e. µm smaller than about 3×10−10 cm2/g, at least 800 times below the value the experiment requires. Since the rotor torque scales linearly with µm in this regime, the largest torque compatible with ordinary gravity is of order 0.005 gF·cm — a thousand times below the bearing friction. Borg's measured torque and Borg's own force law cannot both be right.
The same tension shows up in the G corollary. If G really varied by the factor e−µmρx for the attenuating test body, gravity would be composition- and thickness-dependent, and that is precisely what Eötvös-type torsion balance experiments measure: the Eöt-Wash bounds on differential acceleration of unlike materials, at the level of a few parts in 1013, constrain any such shielding term by many orders of magnitude more tightly than a bearing experiment can. Nor is G the worst-defined constant in the sense implied; its uncertainty (a few parts in 105) reflects the difficulty of laboratory torsion measurement rather than a genuine variability, and the scatter between experiments is not correlated with the density or size of the attracting bodies as EMRP would require. The experiment is worth replicating — the rotation toward horizontal, if real, wants explaining — but the paper's own arithmetic argues that whatever produced it is not a gravitational shadowing effect of the magnitude claimed.