From Curvity
Jump to: navigation, search

This is a reproduction of the Stavros experiment and closely follows the description as done by Jean-Louise Naudin of JLN labs.


Using a quarter wave antenna on the end of a pendulum, this experiment will test if the local gravity field around the antenna can be modified using a resonating EM field in the antenna. All pendulums, if long enough and not pulled too far from the center, obey a simple motion equation dependent on the length and gravity to produce a period but independent of the weight and initial starting point. The experiment will test, that for a constant length, if the period will change between a powered test and a non-powered control test. One factor highlighted by JLN testing is that the Q or the resonant quality of the circuit has a linear effect on results. For example, the higher the Q, the greater the difference that should exist between the control period and the test.

Curvity suggests a relationship exists between high power resonating electromagnetic waves and gravity in that energy from the former can affect the later. However, one problem with the experiment is that the power level, only 3W, may not be enough to see a distinct effect. Curvity suggests that a particle will encounter negative gravity in a positive gravity field if it's energy is raised far enough along the curve. The maximum extent of the curve is dictated by the equation, e=mc2, which is a huge wack of energy. For example, if a particle has the equivalent of mc2 in energy relative to the observer, in this case the earth, it will encounter the complete opposite maximum of positive gravity, in this case -9.8m/s2 away from the earth.

Although this experiment will operate at a very small percentage of the mc2 amount of energy, it might be possible to still see some affect. It is encouraging that both Stavros and Jean Louise Naudin have reported positive results that seem to correlate with a higher Q.


The Stavros experiment consists a pendulum with a quarter wave antenna attached to the end. Hanging inside the antenna is the signal source which consists of a common-emitter type "A" amplifier. Power is supplied down a pair of twisted leads in Cat-5 cable hung from the roof. The entire experiment was built and tested in a detached and insulated garage.

Below are pictures of the first of three antennas and the first of four circuits that were built.

First Build Attempt

The antenna and circuit are as shown above and produced the following waveform:

First measurement of waveform

Two things are notable, the amplitude is 66mV and the frequency is ~93Mhz. The expected result below is taken from JLN's website:

Expected waveform from JNL website

The experiment wasn't conducted at this point because the circuit wasn't producing the expected waveform, 60mV versus 800mV.

By dialing the variable resistor back and forth, which controlled the bias on the transistor, the circuit consumed anywhere from 1.3W (68mA at 30v) to 11W (390mA at 30v). Even when turned off, however, the circuit would produce the 66mV sine wave. The biasing resistor didn't change the amplitude as expected.

Because the circuit didn't produce the expected waveform, a part-for-part comparison was done.

Note: The part #s are from JLN's circuit diagram and "Theirs" means JLN's and Stavro's component specs.

Q1 - NPN transistor:

Mine - 2N49230S-ND (Digikey part #) in a T0225AA package, Vcb = 80v max, Ic max = 1A, HFE min = 30 (link to PDF datasheet)

Theirs - VCB max=60V, VCEmax=60V, VEB=5V, IC max 1A, P.Tot=8WC, hf=40/400 ( Ic=150mA)

C3 - 10nF cap:

Mine - B81141C1103M (manufacturer's part #), 10nF, 440VAC (link to PDF datasheet)

Theirs - 10nF cap

D1 - Diode:

Mine - 1N4148FS-ND, 100v, 4.0ns, DO-35 package (link to PDF datasheet)

Theirs - D1N4148

C1 - 150pF cap:

Mine - P4851-ND (Digikey part #), 150pF, 100V (link to PDF datasheet)

Theirs - 150pF to 1nF capacitor.

L1 - 1mH inductor:

Mine - M9838-ND (digikey part #), 1mH, (link to PDF datasheet)

Theirs - 1mH "VK-200" bead and twisted aluminum wire.


It is pretty hard to mess these up and I double checked them.

C2 was the same, but not important because it was used to reduce noise.

The inductor was then modified to include only 5 turns of copper wire:

Rewound inductor

The waveform now showed a 100mV peak to peak amplitude and the frequency was much closer to the expected frequency of 83Mhz, however it still wasn't strong enough.

The circuit was then modified with an actual "VK-200" which is a ferrite bead inductor with six aluminum conductors through it. Measurements showed the new inductor had the same performance as the modified one previously. The antenna was also modified with twisted ends, but that modification made no difference to the waveform.

Second Build Attempt

One thing that effects circuits operating in the Mhz range is stray capacitances due to copper traces and long leads on components and an second circuit was built with much shorter leads. A second antenna with copper caps was also constructed and both are shown below still sitting in the jig used to create the antenna:

Copper caps added to antenna

The results were a significant improvement with 300mV ptp waveforms, although there was modulation of the signal.

300mV peak to peak!

Higher capacitor values for C1 including 200pF and 470pF were tried in an attempt to reduce the modulations but didn't seem to make any difference.

Third Attempt

One problem with the previous two antennas was that they were only 68cm in diameter, not the 73 to 75cm as Stavros had suggested. Again, using copper caps for the ends, a third antenna was built that was the right diameter and it was tested with the second circuit. The amplitude, 300mV ptp, was the same as previous.

Different capacitors for C1 were also tried and the results were negligible.

Fourth Attempt

The last attempt at getting a circuit to work properly was to build using the "dead-bug" method of circuit construction and on a larger ground plane as shown.

"Dead-Bug" style with grounded transistor, oops.

Note: A heat sink from a previous circuit was added after this shot was taken. Fastening the transistor to the grounded ground plane as shown resulted in a short.

All the previous attempts used an eight element antenna and for this attempt a 16 element antenna was used (sorry no picture) with the same copper caps.

The results were worse.

Worse results with "dead-bug" style

The result was a wave form with about a 150mV ptp and frequency of 90Mhz.


  • All the graphs shown in each attempt above are the best case waveform for any position of the variable biasing resistor. IOW, before any measurement was recorded, the resistor was dialed back and forth until the best waveform was found, which was typically nearer the low end of the power, from 65mA to 75mA.
  • The amplitude was also affected by the proximity of the human body and the recorded waveforms above were recorded while standing as far back as possible.


Ultimately the experiment was abandoned without any period measurements because the Institute for Gravity Research (IGF) based out of Germany failed to reproduce this experiment. Although they used an external waveform generator and only allowed the pendulum to swing in one plane, their results were enough to satisfy that the negative results were valid.

Because three attempts were made and ultimately failed to reproduce the waveform, much less the difference in periods of the pendulum, the experiment was abandoned.