Return to the INE Main Page

Date: Tue, 16 Apr 1996 10:44:22 -0700
Mime-Version: 1.0
From: (Evan Soule)

From: (Ralph Hartwell)
Organization: The Energy Machine Information System 504-733-8380

Note: The views expressed herein may or may not represent the position of Joseph Newman and, as informational material, are provided here from submissions by other individuals interested in the technology


(C)opyright 1991 by -

Ralph M. Hartwell II
715 Jefferson Heights Avenue
Jefferson, Louisiana 70121-1110

The motors demonstrated by inventor Joseph Newman to date have been of two types. The rotating magnet armature version, similar in appearance to a conventional DC electric motor, and the reciprocating or "vertical" design, which resembles a giant solenoid magnet. This discussion will concern itself with the first type of motor, the rotary Newman machine.


The rotating magnet Newman motor is deceptivly simple, apparently consisting of nothing more than a large coil of wire, a rotating magnet armature, and a commutator. Unlike a conventional DC electric motor, however, the Newman motor has no iron or other ferromagnetic materials in the magnetic circuit. In fact, the presence of any ferromagnetic materials except for the magnetic armature severly degrades the performance of the machine.

A Newman motor is assembled sort of "inside out" when compared to a regular DC electric motor; that is, the coil is wound around the magnet, and the magnet rotates, while the coil remains stationary. A commutator is necessary to perform the dual function of reversing the polarity of the voltage applied to the coil as the magnet reverses position twice per revolution, and to interrupt the current flow through the motor coil many times poer revolution according to Newman's theory. The design of this commutator is quite critical to the proper operation of the motor, and is covered in a seperate paper written by this author.


The coil is usually a simple solenoid design, with multiple layers of wire wound on it. Depending on the applied voltage, the wire gauge will vary from 8 gauge to about 32 gauge. Thge lower voltages use the larger diameter wire, and the high voltage machines will use the finer wire. Newman has used both extremes on his various designs. Note that while Newman prefers the high voltage designs (he feels the high voltage devices have less loss because of the lower current in the windings) he has successfully demonstrated a machine operating on 12 volts DC power input.

My suggestion is to use a voltage no higher than 300, due to the problems with the very high back voltage generated by the device. Output voltages of 50 times the input voltage are not uncommon with the larger units. These great voltage spikes are difficult to control, and tend to destroy test equipment connect- ed to the Newman motor*. Also, high voltage machines require many more turns of fine wire, with a rather rapid increase in construction effort and cost.

*Note: the voltage spiking problem has been solved with the latest commutator designs. This permits the utilization of higher voltages without the earlier back-emf problems.


I have been asked many times about sources for magnets for Mewman motors. My recommendation is to try surplus houses, such as Fair Radio, Jerryco, or suppliers such as Edmund Scientific Co. These folks usually have surplus magnets in various sizes at reasonable prices - at least when compared to new magnets.

What is the best type of magnet*? Well, for the experimenter, it's most probably whatever you can get at a good price. Newman motors have been built with everything from Alnico (C) magnets to the latest super-powered rare-earth magnets. A popular material is ferrite composition, of the kind commonly used in loudspeakers. These magnets are usually readily available in surplus catalogues, and are not too unreasonably priced. They also are usually made available in large quantities on the surplus market, which is a good thing, since you will probably need quite a few of them, depending on the size of the motor you are building. [Note: neodymium magnets have been used]

If you use magnets such as ferrite loudspeaker magnets, they are usually stacked end to end and covered with something such as epoxy or fibreglas to prevent the assembly from flying apart due to centrifugal force while in high-spoeed operation. If a single stack is not as powerful as you would like, you can place several stacks side-by-side to increase the magnetic field. The magnets may also be placed inside a non-metallic tube to hold them in place.

How large should the magnet be? I suggest that the weight of the magnetic material in the rotor be made about 1/4 the weight of the wire used in the coil of the motor. This is not an absolute rule, just a first approximation for testing, but it has worked well in previous designs.


What about the coil size? Remember that as the machine grows bigger, everything interacts to cause the price of the parts needed to increase! Design the coil so that it's axis is about 3/4 to 4/5 as long as the rotating magnet assembly. The coiul should be close in dimensions to a so-called "square" coil design; that is, a coil which is as wide across its diameter as it is long. This design comes close to giving the greatest inductance with the smallest mass of wire, and also keeps as much of the wire as close to the magnet as possible.

Since the magnet rotates end-over-end inside the coil, the length of the assembled magnetic rotor determines the inside diameter of the coil. Let's take a few figures as an example. The following is not necessarily a recommendation, but just serves as an example...

Note: in the newest designs, the magnetic rotor configuration is designed differently.

Suppose the magnet when assembled is 11 inches long. If we allow 1/2 inch clearance between the ends of the magnet and the inside of the coil form, that will make the coil form inside diameter about 12 inches. Allowing 3/4 of that size, the coil would be about 8 inches long.

Since this is a small motor, we might want to make the coil a bit longer, perhaps a full 12 inches. This will allow us to have a bit more copper wire in the magnetic field of the magnet. The extra wire won't be as effective as the wire near the center of the coil, but every bit helps.


The thickness of the wire wound on the coil ,depends on the size of the motor, and the strength of the magnets. The bigger the motor, naturally, the bigger the magnet, so the more wire is required. I suggest making the wire thickness about 1.4 to 1/3 the inside diameter of the coil. In this example, this would make the winding thickness about 3 to 4 inches. This makes the outer diameter of the coil about 16 to 18 inches in diameter, with a winding thickness on each side of the form.

You can calculate the amount of wire needed by computing the area which will be occupied by the windings. To do this, take the length of the coil, in this case, 12 inches, and multiply it by the winding thickness, which is 4 inches in this example. So, 12 X 4 = 48 Square inches.

The wire will not occupy the entire volume, since the wire is round, and when wound on the form, will not fill the entire volume. About 70% of the space will be filled by the wire. A table of wire data, such as the one found in the Radio Amateur's Handbook, will allow you to figure how many turns of wire will be required.

Then, you can calculate the length of an "average" turn on the coil by figuring the length around the coil when the coil form is half full, which, in the case of our example here, will be about 16 inches. (12 inches for the inside of the form, plus 2 inches of wire on each side of the form when it is half full). So, 3.1415926 X 16 = 50.26 inches per turn.

Let's suppose the wire we have chosen measures 0.05 inches in diameter. If we were able to wind it evenly so that each turn were side by side, we could get 1 inch / 0.05 inches per turn = 20 turns per inch. So, 20 TPI X 48 square inches = 960 turns on the coil. Since we won't be able to get all those turns on the coil so neatly, we can assume between 70-80% of them will fit. Therefoire, 960 turns X .75 = 720 turns expected. Always buy a bit more wire than you figure you'll need, just in case your calculations are a bit off, or in case you really can wind the wire really neatly!

Figure how much wire is needed - 720 turns needed; lets allow an extra 15%, so 720 X 1.15 = 828 turns. 828 turns X 50.25 inchges per turn = 41615 inches, or 3468 feet of wire required. The wire table will tell you how many feet of wire are in a pound for the size wire you have chosen.

A suggestion at this point - It will probably be cheaper to buy a 50 pound spool of wire then to buy only a couple of smaller spools of wire if you need only 25 pounds or so.. check with several wire suppliers before buying!!


Beware of winding a coil for a motor which will operate on high voltage without using insulation between layers of wire in the coil. It is entirely possible to have a flashover between windings when the motor runs, due to the very high pulse produced by the motor. This is the reason I suggest starting with relatively low voltages. It also makes the commutator design easier.*

*The latest commutator design enables higher voltages to be utilized. Note: The above article was written several years ago. The principles described above are generally applicable "across the breadth of the technology." However, considerable improvements to the commutator design have been made in the recent past. These improvements are intended to actually reduce the intensity of the sparking by distributing the physical connections over a wider area. The reader should bear in mind that there are TWO totally different design systems (but many sub-configurations within each basic design): there is one commutator design when the energy machine is intended to function as a GENERATOR and a totally different commutator design when the energy machine is intended to function as a MOTOR. The latest design improvements to the commutator system apply to the machine operating as a MOTOR. Subsequent torque can be utilized for mechanical systems or can be used in conjunction with a conventional generator.

Copyright 1991-1996, Ralph M. Hartwell, II

Evan Soule
(504) 524-3063

The latest Wiring/Construction Diagram is available to any truly curious/rational individual** wishing to send a SASE to:

Evan Soule
Post Office Box 57684
New Orleans, LA 70157-7684

**I say this because I honestly ask that only people --- who are truly curious about the technology and who may be sincerely interested in building a prototype for themselves --- should respond. Thanks!

Return to the INE Main Page