Here is a general description of GravityMagnet experiments, including what, why, and how we do them.
Universal Magnetech Co.
Click on the links below to find out more about the current projects going on in the GravityMagnet Invention Factory:
One of the first projects typically tackled by a high voltage scientist is the Jacob's Ladder. While the Jacob's Ladder is simple to construct, the operation is quite interesting. The spark starts at the bottom where the gap between the wires is close and then heat makes the plasma rise and stretch between the electrodes until there is no longer enough current to continue the arc. The maximum width of an arc depends on the current (or the volume of electrons flowing in the electricity). The initial break distance is dependent upon the voltage (or the potential energy of the electricity). Once the arc reaches maximum width, it breaks. Then, the whole operation starts over again.
The making of Jacob's Ladder has evolved into a type of underground art form. We have observed Jacob's Ladders powered by distribution transformers that run at over 50amps at 220VAC which emit strange magnetic fields. In this particular experiment, the metal garage doors would 'breath' in and out in synchrony with the arcs of the ladder.
The photo below is of a simple Jacob's Ladder. In particular, this photo used an exposure time delay to show the trail and evolution of the arc itself. You can see the sequence of stable arcs which formed and jumped like quanta in the process of climbing the ladder.
In the Jacob's Ladder shown in the image below, all that can be seen is a single arc as it makes its way upward. To the naked eye, a Jacob's Ladder appears to be a repeating ball of flame suspended between the two electrodes.
The photo below shows a sample of a more creative Jacob's Ladder setup. The acrylic tube enclosure is for safety and protection, as well as for esthetics.
The point of the Lightning Dipole Project is to be able to register lightning strikes simply from the difference in capacitance between a dipole and conductors in the wire. Ever time lightning strikes, a powerful EMF gradient is produced that expands outwards from the specific location of the strike. By aligning a simple dipole antenna perpendicular to the field gradient, a spike of HV can be measured.
Below is an inexpensive way to make a dipole antenna using standard RG58 cable.
We use Maxwell and Atomic pulse capacitors for bulk units like these. We often use TDK laser pulse capacitors, polypropylene capacitor networks in array formations, parallel and series. Bleeder resistors can be added for regulation and safety, and AC from DC networks are arranged in protective patterns within custom safety enclosures. Each individual string is arranged in series according to the plates. Each string is upside-down from the previous one so that manufacturing errors are distributed more evenly.
The best capacitors are rated in AC volts, but when building AC ratings out of DC capacitors, the general rule of thumb is to multiply by three. Basically, in order to manage at tank with 30KV AC waveform, you will need 90KV DC rated capacitors to safely handle it all.
Below is the layout that led us to design for the multipliers describes and constructed in the following images.
Below is the first multiplier that we attempted to make.
This small, singe-stage voltage multipliers can take a 12 KV NST and will drive a 38KV DC output. However, it arcs in air between the legs of the capacitors. We understand that hydraulic fluid is a very good (and inexpensive) replacement for the expensive Dahlia transformer oil.
The small copper rings are absolutely imperative to this design. They make it extremely easy to connect and disconnect parts, and they also eliminate the solder arc problem. This problem is very serious because the solder bits tend to arc badly in the air. With the rings, much higher voltages can be reached without the usual arcs.
Above are the diode strings that we originally used for the multiplier design. We later fazed these out for high voltage diodes that make it possible to make the entire multiplier much small in size. These diode strings worked well in the beginning, but it was extremely frustrating to replace them when some of the dioded became non-functional (ex. when they exploded).
This small sized, single-stage voltage multiplier takes a 7.5KV NST and runs up to 21KV DC. This is about as small as the multiplier design can get before it begins to arc into itself. This specific designs every so often because of the specific capacitors used; however, that problem can easily be fixed by submerging the multiplier in oil.
This double stack takes a 7.5KV NST and runs too high for my probe to measure. Some of the soldering is an issue here; however, more recent designs include the copper wire rings at the solder connection sites.
Below is a more compact layout using the previously mentioned copper rings. Rachel made this one. It runs quite well in air. I have ordered the parts of a 240KV DC extension of this prototype.
Below is the best design we have found yet: it works in air, takes in a 7.5KV NST, and returns 21KV DC. Electrical stress is evenly distributed on each component. The full wave design is perfect for NST input and beats the regular CW multipliers that are typically seen instead.
Below is pictured an example of a quick swapt IGMT or power MOSFET heat sink section. I will need a huge heat sink for this project. I will also need to be able to swap out the silicon quickly and easily in order to compare and switch different parts because they blow up regularly. Get a white plastic barrier strip from Lowe's and dissect the little metal tubes inside. These tubes are useful for making a screw-down quick swap power lead attachment device. With this setup, I can swap out the transistor in about two minutes. This heat sink is from an ancient computer CPU. The addition of a blower fan would be good.
However, there is one problem:
The GATE on the IGBT can drift from "on" to "off" at random, especially in close physical proximity to a running Tesla Coil. This is a natural circumstance in the laboratory, so, a 10k resistor is used to tie the GATE to the SOURCE, right down near the IGBT. This technique is shown in the above photo. In order to slam the capacitor in the GATE of the transistor as quickly as possible, a UCC27321 is used to drive the N-channel MOSFET IRFP450A transistor. A 12 ohm resistor limits the current into the GATE, because experience shows that the UCC27321 will explode trying to source or sink current into the IGBT if it is hard wired. This makes the RC time constant into the gate nice and small. The pulse signal is nice and square up to around 200khz.
By analogy input to the UCC27321 is now susceptible to the same random drift as the GATE was. Since this IC is an inverting unit, it's input is tied through another 10k resistor to a nice regulated 12V supply.
By thinking about it, any algorithm, pulse generator, or feedback mechanism has the possibility of leaving the transistor in the on state. This quickly creates a huge DC current that explodes the transistor in under a second. During normal experiments, there is not time to realize this situation has happened, and you have bits of hissing plastic flying at you.
Above is an 8 pin IC socket free form soldered and then GOOP glued into a phenol board. This is the gate driver, and obtains its power from a well stabilized LM7812 connected into the bus bar. This regulated supply can be seen directly above the bus bar. This regulated supply is smoothed with a 35V 2700 capacitor. All input and transfer signal wires are shielded.
Below is a monster overkill 25 amp 21 Volt DC supply. This is the DC power that is switched through the fly back transformer that drives the plasma globe. The smoothing capacitors are 500,000uF at 32V electrolytic capacitors. These values are ridiculous overkill but allow the project to continue without concern for overloading the main DC supply. The driver power supply is regulated at 12V with its own smoothing capacitor, and I can see no transients at either supply output.
The picture above shows the primary of the fly back rewound with 21 turns #14 magnet wire. The acrylic stand holds the fly back upright. The winds of the new primary and small pieces of duct tape are dipped in cyanoacrylate (super glue), so the base of the fly back is encased in solid transparent plastic. The high side winds measure 480 ohms. This AC fly back is ancient, and was originally made for Zenith televisions.
For an AC Fly Back Driver With Adaptive Feedback:
One evening, audio vibrations from a telephone speaker suggested I look at the ZVS topology (Zero Voltage Switch) circuit. If the metric of "goodness" is maximum output power then this is the best fly back driver circuit that exists. Initial observation suggests an adaptive feedback that seeks resonance is happening. How cool is that? This circuit was designed by Vladmiro Mazilli.
Experiments show that the resonant frequency of a standard light bulb can be established.
As the circuit is engineered certain instabilities are observed.
The Fringe Science section includes science which is not considered to be 100% true, but is often full of points of interest and ideas that warrent further investigation. This section includes possibilities, mysteries, conspiracies, unexplaned phenomena, and much much more. Click on the links below to find out more of these fringe ideas:
SEG (Scalar Electromagnetic Gradiometer)
and the Rife Generator
T.T. Brown and Antigravity
Last Edit Date: 12/06/2010