EXPERIMENT 1 (aborted).

The experiment is under preparation and the first tests will be done probably around 7 to 15 january 1999 .

If the theoretical interpretation of my experiment of 1991 is experimentally true, there are a lot of chances to obtain a positive result.

EXPERIMENTAL SETUP ( figure below ).

Remark: the experimental setup is a bit complicated, but it is because I use low cost components from the surplus of the NATO supply center in Luxembourg (close to Belgium).

A 10 dBm waveguide microwave oscillator at 9.192 Ghz is sent to a 3dB power divider (Magic Tee) to supply two special networks of coaxial cables made differently. One ouptut of the power divider pass through a phase modulator controlled by a low frequency generator at 1kHz before to be sent to a network. The LF amplitude is chosen in order to have about a Lambda/10 peak-to-peak phase modulation (0.0108 ns). The loss in the modulator is balanced by an attenuator in the other arm (output) of the power divider before to be sent to the other network.

The two outputs of the networks (at the same amplitude) as thus added by an other Magic Tee whose output power depends of the phase relationship between them (max= phase, min=180° out of phase, medium = phase-quadrature).

Thus, obviously, as the phase of one signal is slightly phase modulated, there is an amplitude modulation of the output signal with a maximum modulation at 1khz in phase quadrature. On the contrary, when the signals are in phase or out of phase, the output modulation is very weak at the frequency of 2kHz (not a sinus, but periodic).

The signal at 9.192 Ghz (together with the modulation) is amplified by a microwave amplifier (20dB) to balance a bit the loss in the cables before to be sent to a schottky diode biased by a DC current. The ac modulation at 1khz is thus extracted by a capacitor and filtered by an active (amplifier) bandpass filter at 1kHz before to be monitored by an oscilloscope.

As (see theory below) in one network a maximum phase shift of 0.28 ns is expected for a 180° rotation of the apparatus , and as the period of 9.192 Ghz is 0.109 ns, we may expect 2*0.28/0.109= 5.13 = 5 quadrature crossings for an ether-wind of 360km/s.

THEORY

Cables 1 have a velocity w1=0.61.c in the rest frame.

Cables 2 have a velocity w2=0.81.c in the rest frame.

I Measurement arm.

If the ether-wind (360km/s) comes from the top, the velocities in the cables 1 (D'=15 m each) are, according to GTWMC, [w1+v]/(1-bb) with b=v/c, v =360km/s=ether-wind velocity or velocity of the Earth in the rest frame. The time delay in these two cables is:

2.D'(1-bb)/[w1+v].

In the cables 2, the velocities are [w2-v]/(1-bb), and the time delay:

2.D'(1-bb)/[w2-v], with a total for the four cables of:

delta-t1= 2.D'(1-bb)[1/(w1+v)+1/(w2-v)].

Now if we perform a 180° rotation of the apparatus, we obtain by the same reasoning the following time delay:

delta-t2=2.D'(1-bb)[1/(w1-v)+1/(w2+v)].

Thus, in the measurment arm, a phase shift is expected of the following time amplitude:

delta-t12=delta-t1-delta-t2= 2.D'(1-bb)[-2v/(w1.w1-v.v)+2v/(w2.w2-v.v)], about:

4.D'v.[-1/(w1.w1)+1/(w2.w2)]= -0.28 ns.

II. Oscillator drift and temperature compensation arm.

This arm is made of the same lengths of the two sorts of cables, but arranged differently to cancel the ether-wind phase shifts.

If the ether-wind comes from the top, the delay in the first cable 1 is: [w1+v]/(1-bb), and

[w1-v]/(1-bb) in the second 1.

In the first cable 2, the velocity is [w2+v]/(1-bb) and [w2-v]/(1-bb) in the second.

Thus the total delay is: D'(1-bb)[1/(w1+v)+1/(w1-v)+1/(w2+v)+1/(w2-v)]

If we perform a 180° rotation of the apparatus, the total delay becomes:

D'(1-bb)[1/(w1-v)+1/(w1+v)+1/(w2-v)+1/(w2+v)].

Thus the difference is nil.

III. Conclusion.

A phase shift of 0.28 ns is expected between the two arms from the ether-wind and is insensisitive to the temperature of the cables if they are at the same temperature, and also insensitive to a (slight) change of frequency of the oscillator. Tests already performed prove that the oscillator is stable enough for the experiment.

[Return ]