Build a Low cost & simple Magnetic Pulser Categories Electromedicine Practical Health For those who are interested in building their own Magnetic pulser to use with the Beck protocol following is my updated post. "...The sicker a person is the more nutrients are needed in optimum doses to help the bodies reparative mechanisms..." "...eliminate all junk food i.e,. food containing any added simple sugars like table sugar or glucose as in corn syrup. This simple rule, comprehensible even to children, will eliminate nearly 90% of the additives commonly added to processed foods..."
Also note dehydrated patients may not respond well to any type of therapy! WARNING: This is an experimental device and uses 110Volt AC mains voltage, build at your own peril. If not comfortable, have a someone familiar with electronics like a TV repairman build it for you. Do not use 220 V AC without appropriate modifications! Well finally, I have got all the wrinkles out my prototype SCR Thumpy. And my new circuit has definitely got the power. You can actually feel an electric current pulse when used in the neck area - uncanny! This is subtle however. The only thing I have not got around to do is make it automatic - it still uses a press button to trigger the pulse. I don't like the auto types as the body gets habituated with non random pulses hence have to come up with a random pulser circuit one of these days. To calculate the output energy use the following: Formula:
W=(CE^2)/2
W=energy in joules: C = Capacitance in farads: V = Voltage across Capacitor in volts # capacitors 5* 6 7
#Joules 29 35 41
*Present circuit. Any SCR with PEAK current of at least 600 to 1000 amps should work. The one shown is 20 amp continuous with the appropriate peak rating. The lamps act as current limiters and protect the SCR against a short circuit. The circuit can be further simplified as discussed in point 3 below. I have build several of these and my experience has been:
1) The capacitors develop a memory and don't fully discharge its better to use a number of them in parallel. This reduces the internal resistance and provides a better result and less memory loss. The caps must be designed for flash applications. 2) The flash tube heats and develops some resistance so you need to have enough time between flashes for them to cool down. This has been eliminated in my circuit. 3) Using a high current SCR (this forces the caps to fully discharge by providing a longer connection than the strobe) and parallel caps from disposable cameras I can now consistently get 12 - 18 inch with #14 fender washers. And you can cycle them very fast (though not recommended). All for less than $30 Cdn. The most expensive part is the coil which can cost as much a $20 unless you build it yourself! One can further reduce the cost if at a latter date you don't want to upgrade to auto pulsing. This can be accomplished by removing the 10k resistor and the SCR and simply wiring the a push to close switch in line to the coil. Don't recommend this unless you just can't get an SCR or really need to reduce cost. MAKE SURE THE PUSH BUTTON SWITCH CAN HANDLE THE CURRENT AND IS MECHANICALLY ROBUST! More info regarding other coils options etc. is available at: http://www.keelynet.com/biology/thumind.htm If you do make the circuit I would love to hear how it turned out. Chris Gupta Coil winding instructions from Dr. Beck's paper are: "Junk VHS videocassette reels are cheap, plentiful and adequate for this application. Remove 5 screws from shell, remove reels and discard tape. Be SURE alternative spools (if used) are non-conductive or system will not work. Avoid shorter length VHS tape reels which may have center hubs larger than 1" dia. and won’t hold sufficient wire. Drill 3" holes through hub and through center of flange(s). Make two 4" discs from 3" thick plastic or fiberboard, drill 3" center holes and another 3" hole off-center so coil's inside lead wire can be pulled through. These 'stiffeners' will sandwich reel's flanges so they won’t warp or split as wire pressure builds up while winding progresses. A 2" (or longer) 3-20 machine nut and bolt with washers through centers will clamp flange stiffeners and reel and also provide a shaft to hold in a variable speed drill motor or similar winding device if used. Then remove bolt and stiffeners. Specifications: Completely fill tape spool with #14 or 16 enameled copper magnet wire (130 to 160 turns) wound onto the 1" dia. hub and 3-2" OD spool with a gap width for wire of e". Scrape enamel insulation 2" from ends and tin. Pull inside end of magnet wire through hub and stiffener and to outside. ~130 turns (about About 1-2# should fill spool. Remove bolt, stiffeners, and finished coil. Now solder ends of 3 ft of heavy two-wire extension cord to each side of coil. Finished coil weighs ~1 LB 3 oz, has ~0.935 millihenry inductance, 0.34 W resistance, and takes ~20 minutes to hand
wind or ~3 minutes with drill motor. An excellent alternative is an AMS brand aircore crossover inductor for home audio, MCM Electronics, Centerville, OH 45459. (800) 543-4330 catalog # 50-940, #16 gauge, 0.58W, 2.5mH, 2-f" dia., TYPICAL LOW COST EMF PULSER CIRCUITS
I have also attached a file for the free CircuitMaker Student Version electronics software for those interested in modifying the simple circuit that I have developed. pulserworking.ckt Please share your experience so all can benefit. Thanks Chris Gupta
Improved Circuit of EMF Pulser By PJ van Coller
Building this circuit will eliminate the need for the light bulbs in the AC side of the circuit. This circuit is charged when you switch the switch on the “on” position and return it to it’s centre “off” position. The switch is a three position DPDT switch with a spring load to both on positions alternatively. This switch is used to prevent the input Triac and the output SCR to be “on” on the same time. This allows you to charge the Capacitor switching to the one position and to fire the magnetic pulse if switched to the opposite position. If the switch is in the middle, both the Triac and the SCR are off and the circuit is in the Off state. Leads of about 0.5m can be used to connect the coil to the device. Pleas note that you need at least 1.5mm PVC insulated wire to connect the coil because the current can spike up to 300A. The BT 139 is a AC switch that can switch 8A and the BT152 is a 13A SCR DC switch that can switch up to 300A spikes. The reason for the 4k7 resistor on the gate of the SCR is to eliminate false switching when the triac switches the AC voltage on. Without this resistor you may blow up the SCR and Triac. This circuit shoots a 8mm metal washer up to 3m in the air, which indicates n extreme magnetic pulse that will penetrate the body and even will go right through for optimum results.
The inductor is a 2.7mH coil made with 1.3mm insulated copper wire. NOTE: The inductor needs to be an air core. The filter coil is wound on a ferrite core with 1.3mm insulated copper wire. You can find these filter coils in old computer power supplies and can reuse them.
Designing a Magnetic Pulser How do you design an MPG, and calculate Peak Gauss output in the pulsed DC Magnetic Coil? Revised: August, 2001 The term Gauss is defined as "The electromagnetic unit of magnetic induction, equal to 1 Maxwell per square centimeter. [After K.F. Gauss]". The coil used in our Magnetic Pulser (MPG) provides the magnetic charge that is ultimately responsible for inducing the micro-currents (50-100 uA) of electricity into the conducting medium. It is very important to generate the necessary Gauss output from the coil in order to effectively induce this level of current at distances of up to 9" penetration. There are specific electrical and magnetic parameters on MPG design to ensure the minimum Gauss output is achieved. It takes a great deal of electrical energy, and the use of specific (usually expensive) electronic components, and good design practices in order accomplish this reliably and consistently. Critical components are: 1. 2. 3. 4.
The Power Supply. The Main charging Capacitor. The Switch ("charge dumping mechanism"). The Output Coil.
How it all goes together: The power supply is used to charge the main capacitor. The main capacitor has the ability to store an electrical charge (it can be said that the capacitor gets "filled up" with electrons). The capacity of the capacitor is expressed in Farads. Because Farads is such a big value, most capacitors are rated in micro-Farads (uF). The larger the capacitor, the more electrons it can "hold". Our MPG has a capacitor rated at 600uF. NOTE: Capacitors connected in parallel add together their respective capacitance algebraically. Example, 2 x 300uF capacitors in parallel have an equivalent capacitance of 600uF. Capacitors connected in series add together their respective voltages algebraically. Example, 2 x 250 Volt rated capacitors in series have an equivalent voltage rating of 500 Volts. This means these two capacitors when connected in series can safely operate with a 500 Volt potential across them. It does NOT mean that they have increased their output, it means they are capable of handling a 500 Volt potential, should one arise! The main capacitor has a working Voltage Rating. This is the working potential it is designed to operate in. The higher the working voltage, the more electrical energy the capacitor can store as work without being destroyed. Work is expressed in Joules or Watt*Seconds. The formula for calculating the amount of energy a capacitor has is: E=1/2*C*V*V Where E is Joules or Watt*Seconds, C is the capacitance in Farads, and V is the Voltage as measured across the fully charged capacitor. Our MPG's capacitor is
rated at 350 Volts DC (highest voltage it is allowed to charge to) and has a working voltage (typical charging voltage) of 330 Volts DC. Applying the Energy formula we get: E=1/2*[600 x10 ^-6 Farads]*330 Volts *330 Volts = 32.67 Joules or Watt*Seconds. Note the Voltage term is a squared function? Energy goes up as a square of the voltage. It also goes DOWN as a square of the voltage. Let's say our 600uF capacitor is only charged to 1/2 of the 330 Volts: E = 1/2*[600 x10 ^-6 Farads]*165 Volts *165 Volts = 8.17 Joules. The voltage went down by 1/2, yet the Energy dropped by 4. This is one reason why it is so important to have a minimum charge voltage on the capacitor of at least 300 Volts DC. At 300 Volts, E = 27 Joules. Still quite respectable. Joules or Watt*Seconds is only one piece of the puzzle. This gives us potential to do work. But the coil needs Amps to make Gauss. Amps (or current) is generated when a Voltage is applied across an Impedance (resistor in DC terms). With all else being equal, the higher the Volts, the higher the Current and the higher the Gauss. The duration of the magnetic pulse (how long it lasts) is largely in direct proportion to the Capacitance in uF of the main capacitor. If the pulse duration is too short, it is not available long enough to do much work. SOTA's MPG has a pulse duration of ~2.5 mS (milli-seconds). To re-cap, we use a large capacitor to store adequate energy for the pulse, and we also need a high enough voltage potential across the capacitor to drive the right amount of current through the coil. The main capacitor must reach a minimum DC Voltage in order to have enough potential to generate the current (Amps) necessary through the coil, which in turn produces the intense magnetic field (Gauss). After being fully charged, the main capacitor's store of energy must be "dumped" through some sort of switching device to the coil. Immediately after discharge, the power supply must then re-fill the capacitor with electrical energy (Joules). When the capacitor gets fully charged, we must dump this high energy into the coil somehow. Originally a Xenon photoflash tube was used as a thyratron switch (the Xenon gas is ignited to a plasma which provides a low impedance path for the electron flow) for the do-it-yourself'ers. The Xenon tube presents about 1-3 ohms of resistance when ionized. It makes a good switch, but it does restrict peak current flow. Ringing can and does occur with the capacitor-coil combination because the current can back-feed into the capacitor. (This ringing affect can allow reverse-biasing of the main capacitor; degrading it's life-span very quickly or destroying the capacitor under extreme conditions.) When a capacitor is in series with an inductor in this manner, it is known as an L-C (Inductive from the coil and Capacitive from the capacitor) circuit. Xenon tubes get hot, and they waste energy in the form of heat and (of course) light. A better switch is an SCR (Silicon Controlled Rectifier) of appropriate voltage and current rating (800-1000 Volts @ 25 Amps is a good start.). An SCR provides a one-way path of current flow from capacitor to coil, inherently preventing ringing and therefore saving the capacitor from reverse-biasing. This one-way path ensures the output magnetic field is DC based or uni-polar. This means North magnetic pole will always be on one side of the coil, and will not change to South pole at any time.
NOTE: Although a typical SCR is a fast operating device, there will always be a "dead-time" where the device is in a conductive state (known as tq which is typically = 35uS). This can allow reverse voltages to appear across the main capacitor which may eventually lead to a much shortened lifetime or the complete destruction of the capacitor. So, in order to prevent such an event we need to suppress this high-power content reverse voltage spike across the main capacitor. The simplest, effective and most economical way is to place a highcurrent diode in reverse, across the leads of the main capacitor. The CATHODE (-) lead of the diode connects to the POSITIVE (+) terminal of the capacitor. Remember, this reverse-voltage spike can contain many joules so you must use an adequately rated diode. We use an MR756 silicon rectifier. It is rated at 600 Volts DC, 6 Amps continuous and 400 Amps peak surge. I measured over 80 Amps of current in the reverse-voltage spike. WARNING: If you do not use a similar rated diode, it may very well blow up in your face! I know, because I had this happen many times while taking measurements. Scares the heck out of you! Once the charge is dumped to the coil, the current flowing through the coil generates a moving magnetic flux as governed by the following equation. We must look at the coil as a DC electromagnet, because in essence that is what we have created. Thus: Bpeak = Inductance (L in Henries) of the Coil*Peak Amps*10^8 / (N*A) Where Bpeak is Peak DC Flux Density in Gauss, L is the inductance of the coil in Henries, Peak Amps is the measured RMS Peak Amps flowing through the coil, N is the number of turns in the coil, and A is the cross-sectional area of the air-core section of the coil. SOTA's MPG coil is rated at 2.5 milli-Henries, has 270 turns, and has a crosssectional area of 3.22 cm^2. We measure 150 Peak DC Amperes on each pulse through our coil. Substituting we get: Bpeak=0.0025 Henries*150 Amps*10^8 / (270 Turns*3.22cm^2) =43,133 Gauss* [*This 43,133 Gauss measurement is at the intercoil winding flux. Magnetic flux intensity follows the inverse square rule - the further you are away from the magnetic core, the faster the Gauss intensity drops off. At face of Magnetic Coil, the Gauss drops to ~6,000 Gauss.] As you can see, all the components come together to generate the required Gauss output. In order to drive the necessary amps through the coil, the coil must have a low impedance or DC resistance. The use of 18 AWG or thicker wire ensures this. Our coil has a DC Resistance (DCR) of about 0.5 ohms. Theoretically, at 330 Volts across the coil, we could drive: Current = Volts/Resistance = 330 Volts/0.5 ohms = 600 Amps through the coil. So, the coil's DCR is not a factor for most MPG designs. The Xenon tube injects a 1-3 ohm DC resistance into the path. So, theoretically Current = Volts/Resistance = 330 Volts/1 ohm = 330 Amps best case, and 330 Volts/3 ohms = 110 Amps worst case. Therefore the Xenon tube's DC resistance can significantly impact on peak current. The SCR, being a PN junction, does not present a impedance per say just a voltage drop (1.5 Volts, which is insignificant). A typical 25 Amp rated SCR has a peak one-cycle surge current of 350 Amps. This is more than adequate to handle the peak currents we are likely to encounter. This is why an SCR makes an ideal electronic switch for MPG's. It's also FAST. And the speed of the switch governs the Rise Time of the pulsed wavefront. The rise
time is a result of the change in Current over a change in Time, or dI/dT. The higher the dI/dT, the more current you will induce into the conductive medium. SOTA's MPG has a rise time of <1.8 uS (micro-seconds). Remember, why we are making an MPG in the first place? To generate, through inductive coupling, the micro-currents (50-100 uA) of electricity in the conductive medium. In order to inductively couple 50-100uA of current, you need a fast moving magnetic field of high enough Gauss intensity. Patents on pulsed magnetic field therapy actually call for Gauss outputs of 2 Tesla (1 Tesla = 10,000 Gauss, so 2 Tesla = 20,000 Gauss) and even up to 20 Tesla (200,000 Gauss). So, I believe that unless your MPG outputs at least 2 Tesla (20,000 Gauss) it is not effective as a pulsed magnetic field therapy unit and should not be used as such. NOTE: In order to test the actual induced currents I performed a simple experiment: I placed a single wire across the MPG coil and connected a 2,000 ohm resistor in series with the wire. This then would mimic the 2,000 ohm impedance found in the human tissue. I used my 'scope to measure the maximum voltage impressed across the resistor from pulses delivered from the MPG coil. I got a result of about 0.5 Volts Peak, which mathematically calculates out to 250 uA of current. I can make a fairly conservative conclusion that the MPG coil is capable of inducing greater than the required 50-100 uA into a 2k ohm conductive medium. The main capacitor has definite ESR (equivalent series resistance), and can be a main contributor to the resistance to current flow through the coil. SOTA's custom professional strobe capacitor used in the MPG has an ESR of only 0.076 ohms. At 330 Volts, it has a short circuit current capacity of Current = Volts/Resistance = 330 Volts/0.076 ohms = 4,342 Amps! (This amazing current capacity, due to the extremely low ESR, is un-matched by standard camera photoflash capacitors.) It is important to note that the lower the ESR the less HEAT will be produced, and the longer the capacitor will last. This is the main reason why SOTA's MPG has such high performance. Capacitors having a high ESR (not good) get HOT during operation - very HOT! Heat destroys the dielectric, and burns the capacitor out. In addition, the build-up of heat further INCREASES the ESR. A higher ESR lowers the current on each pulse, which of course lowers the Gauss output on each pulse. As mentioned above, the lower the ESR the longer the capacitor will last, and generally the faster you can charge and discharge it. SOTA's custom professional strobe capacitor is rated at over 30 million pulses! This compares to 5-10,000 pulses from your typical photoflash capacitor found in most cameras. Remember, cameras were NEVER intended to be magnetic pulsing units. Charging a capacitor follows simple electrical rules. For example, our 600uF capacitor is to be charged to 330 Volts. We know that the energy required is: E=1/2*[600 x10 ^-6 Farads]*330 Volts *330 Volts = 32.67 Joules or Watt*Seconds (that's Watts TIMES Seconds) Now, power is Watts/Second (that's Watts PER Second, not Watts TIMES Seconds!). So, if you want to charge this big capacitor in say 5 seconds, you will require 32.67 Watt*Seconds / 5 Seconds (the Seconds term cancels out, leaving you with Watts) = 6.534 Watts input to the capacitor. Now, most power supplies have a great deal of trouble raising 12 Volts from your typical battery (or 6 Volts from 4 x AA cells) to the required 330 Volt charging potential. If you get 50% efficiency, you are doing extremely well! So, at 50% efficiency, the power supply requires: 6.534 Watts / 50% = 13.068 Watts. At 12 Volts input, this translates to 1.089 Amps (I=V/R). Your battery (or Wall Adapter) must be capable of supplying at least 1.089 Amps to get the job done without overheating (in this example, you would choose a Wall Adapter with a 1.2-1.5 Amp rating.)
Using 4 x AA cells, which equals 6 Volts (4 x 1.5 Volts each) would require 2.178 Amps draw from the batteries. Those 4 x AA cells won't like you very much! They will get real HOT, and won't last more than probably 2-5 pulses. So you see, a high powered MPG requires a substantial power supply and the correct selection of critical components. Period. (And this gets expensive.) You can't fool Mother Nature nor attempt to by-pass her fundamental electrical rules. Most photoflash units simply cannot, and do not, perform to the level of a professionally designed MPG like the units we at SOTA Instruments Inc. offer. I hope you enjoyed reading this dissertation on MPG's. Your feedback and comments are very much welcomed. Sincerely and Best Regards, Russell J. Torlage, CTech
posted by Chris Gupta on Tuesday August 19 2003 updated on Saturday September 24 2005 TrackBack URL of this article: http://www.newmediaexplorer.org/chris/2003/08/19/build_a_low_cost _simple_magnetic_pulser.htm
Related Articles Mobile Phones Increase Tumor Risk, Study Says Further to the post: Em Fields On Brain Tumor Incidence - Chemicals And Cell Phones Chris Gupta ---------------------- Mobile Phones Increase Tumor Risk, Study Says Thu Oct 14, 1:38 AM ET STOCKHOLM (Reuters) - Ten or more years of mobile phone use increases the risk of developing acoustic neuroma, a benign tumor on the auditory nerve, according to a study released on Wednesday by Sweden's Karolinska Institute. The risk was... [read more] November 03, 2004 - Chris Gupta Em Fields On Brain Tumor Incidence - Chemicals And Cell Phones ..."In a series of studies, Hardell and colleagues have examined the relationship between the side of the head habitually used in operation of cellular and cordless phones and a possible relationship to the site of brain tumors ([Hardell et al., 2003]). The risk for ipsilateral (same side) use significantly increased the risk for astrocytoma for all types of phones, but use of the phone on the opposite side of the... [read more] October 20, 2004 - Chris Gupta BIOELECTROMAGNETIC MEDICINE - THE BOOK ...""In the decade to come, it is safe to predict, bioelectromagnetics will assume a therapeutic importance equal to, or greater than, that of pharmacology and surgery today. With proper interdisciplinary effort, significant inroads can be made in controlling the ravages of cancer, some forms of heart disease, arthritis, hormonal disorders, and neurological scourges such as Alzheimer's disease, spinal cord injury, and multiple sclerosis. This prediction is not pie-in-the-sky. Pilot studies... [read
more] September 16, 2004 - Chris Gupta