I've suggested (& published in 21 journal papers) a new theory called quantised inertia (or MiHsC) that assumes that inertia is caused by relativistic horizons damping quantum fields. It predicts galaxy rotation, cosmic acceleration & the emdrive without any dark stuff or adjustment. My Plymouth University webpage is here, I've written a book called Physics from the Edge and I'm on twitter as @memcculloch

Monday, 16 October 2017

The Allais Effect

I'm getting slightly respectable in my old age, being invited to talk at St Andrews University, but I always found that un-respectable anomalies (no real anomaly is respectable) are the essential signposts to new physics and to show that I have lost none of my radicalism I'm going to talk about the Allais effect and a possible way that quantised inertia might apply. Please note that this proposal is not yet solid and is just an exploration at this stage.

The Allais effect was discovered by Maurice Allais, a Frenchman who won the Nobel prize for economics. He was using a Foucault pendulum during a Solar eclipse in 1954. Usually these pendulums swing to and fro in the same plane of space because of their inertia, so that, to us on the spinning Earth, their plane of oscillation appears to turn with a period of a day (at the poles, see comments).

The first component of the Allais effect is that during a Solar eclipse the plane of rotation of the pendulum rotates more rapidly than expected during the eclipse moving through about 10 degrees and at the end of it, it rotates back into the expected orientation.

The second effect was found in 1961 by Gheorghe Jeverdan who noticed that during the eclipse the period of the pendulum also decreases by one part in 2000 or dT/T=0.0005, where T is period.

The third effect was seen by Mishra and Rao (1995) and involves a reduction of apparent gravity, and then an increase, both of about 0.5 microgals or 0.5x10^-8 m/s^2.

If confirmed, then these observations would be a useful clue in the development from quantised inertia into horizon mechanics (a new complete dynamical model). I noticed a few months ago that quantised inertia agrees with the second effect but in a manner that I hesitate to mention, because it sounds a little wacky, even to me, but here's to bold suggestions and freedom of speech.

Consider Allais' pendulum. It sees huge accelerations within the hot Sun. As you'll see, the actual acceleration doesn't matter, which is lucky since I don't know it. Lets just assume it is a big number. So according to quantised inertia the acceleration is big, the Unruh waves seen are short and a large proportion of them are 'allowed' since they fit inside the Hubble volume. Suddenly, the Sun gets covered up by the Moon, and the main acceleration the pendulum sees now is the acceleration of the Moon around the Earth which is 0.0024 m/s^2. The Unruh waves it sees are now longer, fewer fit inside the Hubble volume, and a greater proportion are disallowed so the inertial mass of the pendulum drops. The change of inertial mass predicted by QI is

dm/m = (2c^2/Theta)((1/a1-1/a2)

where c is the speed of light, Theta is the co-moving distance to the cosmic horizon and a1 and a2 are the accelerations in the Sun, and of the Moon around the Earth respectively. So putting in values

dm/m = 2x10^-10 x ((1/0.0024)-(1/bignumber))
dm/m = 2x10^-10 x ((1/0.0024)-0)
dm/m = 8.3x10^-8

The period of a pendulum (T) is given by

T = 2pi.sqrt(lm/gM)

where l is its length, m is the inertial mass of the pendulum bob, g is the gravitational acceleration and M is the gravitational mass of the bob (M .ne. m in QI). So the variation of the period will be the square root of the variation of the inertial mass, in other words

dT/T = sqrt(dm/m) = sqrt(8.3x10^-8)

dT/T = 0.0003

The observed variation in the pendulum's period (Duif, 2004, data from Saxl and Allen) was

dT/T = 0.0005

So quantised inertia (summarised here) predicts in the right ballpark. I have to say that, even to me, the process of Moon-shielding of Unruh radiation sounds quite wild at this stage, and it predicts that there should also be a diurnal effect as the Sun sets and goes behind the Earth (see second reference, far from conclusive), but I think it is important to get these edgy ideas out there, just in case someone else can add a little to it, and to avoid the descent into safe irrelevance.


Duif, C., 2004. A review of conventional explanations of anomalous observations during Solar eclipses. https://arxiv.org/abs/gr-qc/0408023

Saxl E.J., M. Allen, J. Burns, 1980. Torsion pendulum: peculiar diurnal variations in period. Letter submitted to Nature. https://www.researchgate.net/publication/284186911_AllaisBook_SAB1980


Derek said...

I'd be careful about this, Mike. Any explanation of the Allais effect really has to explain the negative experiments as well as the positive ones. There have been many attempts to investigate the effect and some of them have seen it; some haven't. See Wikipedia for a pretty complete list.

Simon Derricutt said...

Mike - given that QI predicts a loss of inertial mass then this should also apply to Quartz oscillators, balance-wheels and indeed most of the easy ways we use to time things precisely. In order to have a reliable time-reference, we'd need to design something that doesn't depend on a mass under the influence of a force. Since that also knocks out atomic clocks (vibrations of Caesium atoms) then this looks a little tricky. It's thus necessary to make a better clock before we start to place too much weight on the measured variations. Using current clocks, it seems to me that the pendulum and the clock should co-vary and thus give a null result. The only oscillation I can think of that doesn't depend on inertia is a constant voltage charging a capacitor through a resistor, with discharge when the capacitor reaches a set voltage - this isn't generally very precise but could possibly be designed to be so (precise temperature control and shielding, with precise trip-points). An LC oscillator, on the other hand, likely does depend on inertial mass (of the electrons) to some extent.

Given a precise time-reference that does not depend on inertial mass, you wouldn't need the complexities of the pendulum measurement and could compare against a range of inertial-mass-dependent time standards. Barometric pressure variations could be eliminated by running these smaller devices in vacuum, and temperature variations can be likewise controlled.

One other thing that ought to be measurable is the difference in speed of rotation of the Earth itself, though of course the tides and leaf-fall will impose some variations. These variations are currently measured, AFAIK (why we get "leap seconds" inserted into the world clock now and again) since telescopes give a very accurate measurement of angular rotation of the Earth. A plot of such speed variations against the relative position of the Moon might be interesting. If the Moon does shield Unruh waves, then if there's an eclipse anywhere on the surface of the Earth the rate of rotation should speed up (since the volume in the shadow has less inertia). This correlation should be distinguishable from the usual suspects. This data is probably already available somewhere, since sidereal time is closely watched. Maybe a question to the Time Lords is in order....

Laurence Cox said...

I would echo Derek and Simon's concerns on this and mention one small error in your posting. You said:

"Usually these pendulums swing to and fro in the same plane of space because of their inertia, so that, to us on the spinning Earth, their plane of oscillation appears to turn with a period of a day."

Actually, it is only a day if you are at the North or South pole. At any intermediate latitude the plane rotates more slowly:


Mike McCulloch said...

Derek, Simon & Lawrence: Caution noted, and Laurence I will correct my mistake regarding periods. Embarrassing. I hesitated before writing this, and it may be a bridge too far, but it is important to be able to discuss this sort of thing. It could be a clue to a more horizon-based dynamics and I need to imagine the mechanics of it. Simon: very good point about timing devices, for which I have no good answer yet.

Michael xChaos Polák said...

I would really like to code simple javascript based QI simulator... but in order to do this, I would really need to better understand the notion of "object seeing acceleration".

Mike McCulloch said...

Michael: A javascript QI simulator would be extremely useful, for me too, to visualise things. Does this blog entry help?:


and this one, that shows a way to visualise simple cases:


Zephir said...

the Unruh waves seen are short and a large proportion of them are 'allowed' since they fit inside the Hubble volume. Suddenly, the Sun gets covered up by the Moon, and the main acceleration the pendulum sees now is the acceleration of the Moon around the Earth which is 0.0024

Why the acceleration around Moon shielding Sun should get smaller instead of higher? IMO the Allais effect could confirm MOND/MiHSC theory quantitatively but falsify them qualitatively just due to its geometric shielding nature.

Zephir said...

BTW There are way more phenomena connected with Allais effects and eclipse, for example the temporary increase the inertia (with compare to gravity, which decreases). Also the gravity curve during eclipse gets more complex, than just "reduction of apparent gravity".

Mike McCulloch said...

Zephir: To answer your 1st question, the idea is that the Moon acts to block any knowledge of the accelerations within the Sun from the pendulum, but, as I said on the blog, this is just a hypothesis (the numbers agree but it could just be a coincidence). I like looking at anomalies and thinking about how to explain them, but of course I can't be sure that all the anomalies are real. To model the Allais effect properly in QI I'd need a model that can cope with the effects of horizons..

Zephir said...

MOND and MiHsC model (despite all efforts to explain gravity and dark matter by logic and geometry) is still based on formal regression of magnitude of quantum effects (fitted to diameter of observable Universe horizon) in similar way, like the epicycle model of Galileo era. These problems would arise, once you'll deal with geometry of Allais effect.

IMO you're still working with model which is dual to actual explanation of gravity and dark mater, which leads into similar paradox like the motion of Venus in geocentric system: despite it fitted the epicycle model quantitatively, it violated it qualitatively (by order of Venus phases).

IMO you're quite close to the physically correct explanation of dark matter and gravity, but still not quite - which is the price for quantitative approach: the theories which work well quantitatively (being fitted to observations) don't work well at logical level (just because of caveats of formal regression used).

Zephir said...

The reversed causality of your model is in the fact, you're attributing the local effects of gravity to very distant areas of Unruh horizons, which introduces the similar inversion of coordinates, like the Ptolemists did in their description of Solar system. The local effects should always have local origin - or something gets rotten in the Kingdom of Denmark...

Mike McCulloch said...

Zephir: Pls have patience. I am progressing with a more coherent model in which I can derive everything in QI neatly from the uncertainty principle & relativity alone (see my paper last year in EPL for the start of this):


The whole structure will eventually come out. It is a bit like the debate between Newton (an empirical Brit) and others (continental rationalists) who said his 'action at a distance' was ridiculous. It is, but if that's the way nature works (non-locally) so be it. We already see it in EPR. The logic will become apparent when we have a deeper understanding.

Zephir said...

/* I can derive everything in QI neatly from the uncertainty principle & relativity alone */

After then you finally wouldn't need the Unruh radiation and horizon stuff..;-) Hypotheses non fingo. You're an oceanographer and the effects like gravity and dark matter all have their counterparts in behavior of waves around islands. As you probably know, no need of action at distance is required for their explanation there.

Zephir said...

BTW Is it really possible to derive dark matter effects and another violations of relativity with using of relativity in mathematically fully consistent way?

Michael xChaos Polák said...

There really seem to be 24h variations no one told me about!

But aren't there expected to be tidal forces anyway? Is this really new physics?

Simon Derricutt said...

Michael - very interesting paper, thanks. Tidal timings would be obvious over the 15 days, given that it takes on average 24 hours and 50.5 minutes for the Moon to be in the same place (or nearly, since the orbit isn't circular). The difference between a sidereal day and solar day (sidereal day is about 4 minutes shorter) can't be resolved over this experimental period. Still, given that the oscillations centred around noon it's a reasonable bet that it's solar position that's more important, unless by chance the experiment was started at exactly the right date.

That's not to say this is necessarily new physics. Could be the silk thread produces torsion based on temperature or humidity (easy to check for, though) or *something else* we haven't thought of. This is an incredibly sensitive device. With the device placed in a locked room 40ft underground, though, I'd expect most such mundane explanations can be dismissed fairly easily and demonstrated to not be systematic errors. It would be interesting to see what happens if the disc was in vacuum rather than in air, though of course the vacuum may modify the Silk thread.

Alternative methods for a torque-free suspension may also be interesting (magnetic levitation comes to mind, though there it's possible that magnetic changes could produce a small torque) but it does look like there may be some real but very small effect here.

Mike McCulloch said...

Zephir: Yes, it is possible to model galaxy rotation perfectly using just relativity and quantum mechanics in a self-consistent way (QI). See my paper here (eqns 11-20):


and note that after a minor amendment suggested by Jaume Gine, QI can now be derived exactly this way.. Paper to be submitted..

joesixpack said...

Unruh wave shielding?

So can we theoretically manipulate Unruh waves into vectored thrust, even like a laser beam and possibly as a source to create warp-bubbles (or simply stretch reference frames) or even "tractor beams"?

The shielding concept makes proposals like effective "anti-gravity" easier to understand.

I have seen sonic tractor beam research, very interesting (of course). These really work. How hard is it to create Unruh waves in a lab to create the same effect? Smolyaniyov's paper would imply all we really need is a laser and a source to diffract the photons. If Unruh waves can be concentrated and amplified, the precision of such a technique would be simpler. Obviously, at large distances, EM-based energy works better.

Mike you had an idea elsewhere about collecting Unruh waves as an energy source.

I find this perplexing with regard to the quantum sea and expanding the universe. Are the perturbations of the dirac sea Unruh waves? Are they caused by the expanding universe or are they simply waves caused by cosmic objects inertia being sent across the quantum foam in spacetime? Or both? Or do the quantum particle pairs create Unruh waves?

Perhaps we shouldn't think in terms of generation and drive, but of "redirectors" of Unruh waves. The goal perhaps ought to be a passive machine with no moving parts that can create directed Unruh waves without changing the energy into another kind of energy and back at all.

But how do we do this without creating very large machines? I mean as in keeping the total volume of any complex machine using this technology down to 1000 cubic metres or less, or possibly down to 1 million cubic metres - so for argument's sake, the size of a large multirole fighter plane or space shuttle, but no larger than the largest aircraft or ship used right now.