Newman -- point by point Part 1, Currents and field strengths http://www.phact.org/e/skeptic/newman.htm
Prepared by Tom Napier. Copyright © 1999, All rights reserved.
Introduction:
This document is one of a series which address specific errors made by Joseph Newman in his book, "The Energy Machine of Joseph Newman."
This section examines the connection between the current flowing in a solenoid, its number of turns and the resulting magnetic field and power consumption. It relates to the section of Newman's book which runs from page 297 to page 302 and, in particular, to the table on page 299. (Page numbers refer to the 8th edition.)
Newman claims:
1) That the current required to generate a given magnetic field can be reduced by increasing the number of turns in the solenoid.
2) That the power required to generate a given magnetic field falls in proportion to the number of turns and tends to zero as the number of turns increases.
3) That this proves that the magnetic field "emanates from the atoms of the conductor and not from the current."
My response:
Conventional physics states that the magnetic field intensity generated in the center of a hollow cylinder of a given length is a linear function of the total current flowing round the cylinder. It is not a function of the diameter of the cylinder provided its length is many times its diameter.
Physics also states that it is immaterial how the current flowing in the cylinder is generated. The total can consist of a high current flowing round the cylinder once or a small current flowing round the cylinder many times. Which is preferred is a practical matter, not a physical one. This has been known to be true for about 180 years. Thus Claim 1) is true and is in complete accordance with conventional physics.
No power is required merely to maintain a static magnetic field. To establish a field in the first place requires a momentary input of power. This power is stored in the field and can, in principle, be recovered when the current is turned off. Until then, power is only needed to force the required current through the winding resistance of the solenoid. This power generates heat in the solenoid but does nothing else useful.
It can be shown, see below, that this power does not depend on the number of turns of wire but only on the dimensions of the solenoid. It can also be shown that there is an minimum power required for any given field, no matter how big the solenoid is made. This shows that Newman's Claims 2) and 3) are incorrect.
The analysis:
To make life simpler I'm going to assume square section wire of side "T" inches. The results can easily be corrected for round wire if one wants. (I am using "*" as a multiplication sign and, to avoid superscripts, I am representing exponentiation as repeated multiplication.)
Let's call the inside radius of the solenoid "P" and the outside radius "Q." Its height is "H." The cross-section of one side of the solenoid is thus a rectangle H by (Q - P) inches. Since the wire has an area of T*T the number of turns "N" in the solenoid is H*(Q - P)/(T*T).
The volume "V" of the solenoid is pi*H*(Q*Q - P*P). Its mass is this number times the density of copper which is 0.320 lb/cuin. Since the area of the wire is T*T its length "L," assuming no wasted space, is V/(T*T). This length of wire has a resistance "R" given by L/(T*T) times the resistivity of copper which is 0.63 micro-ohms per inch. The power needed to drive a given current "I" through the solenoid is its resistance multiplied by I*I.
Provided the diameter of the solenoid is smaller than its height the field strength at its center is proportional to N*I/H.
From these results one can show, see the appendix, that the power required to maintain a field strength of F Webers is:
W = 808*F*F*H*(Q + P) Watts.
-----------------
(Q - P)
That is, the power required to maintain the given field is a function only of the field strength and the dimensions of the solenoid. It is completely independent of the number of turns. As Q becomes much greater than P, (Q + P)/(Q - P) tends to unity. That is, for a given height of solenoid the power tends to a constant no matter how much copper one adds. Adding to the height increases the power needed. This directly contradicts Newman's Claim 2).
A worked example:
One can increase the number of turns in a solenoid in two different ways. One can use thinner wire so that the new solenoid has the same dimensions and mass as the old one. Alternatively. one can add more turns of the same wire. This gives a solenoid having a much larger diameter and mass than before.
In both cases the current can be reduced in proportion to the increased number of turns and the magnetic field will remain the same. Let us suppose that all solenoids are 10 inches high and have an internal diameter of 2 inches. That is, H = 10 and P = 1. We'll start with a reference solenoid which has an external diameter of 3 inches (Q = 1.5 inches) and which is wound with 0.1 inch wire. (Roughly equivalent to 9 AWG wire.) Obviously it has five layers of 100 turns each. That is, there are 500 turns. We are going to put 2 amps through it to get a field of 1000 A.turns. (Just for the record, the central field intensity will be 49.5 gauss or 0.00495 Webers.)
The volume of the solenoid is 39.270 cuin and its mass is 12.57 lbs. The length of the wire is 3927 inches and its resistance is 0.2474 ohms. Putting 2 amps through the wire requires 4 x 0.2474 Watts, that is 0.9896 W. That's our reference figure.
Case 1)
Keep the same solenoid dimensions and use thinner wire.
Let's use 0.025 inch square wire. (Roughly equivalent to 21 AWG wire.) Its side is a quarter of the previous wire so its area is a sixteenth. Since the dimensions of the solenoid are the same the number of turns has gone up by a factor of 16, that is we now have 8000 turns. To get the same magnetic field we only need a sixteenth the current or 0.125 Amps. The volume of the wire hasn't changed so its length must have gone up by 16. We now have 62832 inches of wire or almost exactly a mile.
The length has increased by 16 and the area has decreased by 16 so the resistance has increased by 256. We now have 63.335 ohms with 0.125 amps flowing through it. That's still 0.9896 watts or exactly as much power as before.
That is, changing the wire size has reduced the necessary current but has made no difference at all to the power. What has changed is the voltage required. In the original solenoid we needed less than half a volt to put enough current through the coil. The new coil takes a sixteenth the current and hence sixteen times the voltage or 7.9 V to be exact. In practice it is more convenient to drive a solenoid with 8 volts at 0.125 amps than it is to use a 2 A, 0.5 V supply.
Note particularly that Newman's prediction that using more turns would reduce the required power did not hold up. In his table he assumes that the voltage driving the coil is a constant 10 V and derives the power from 10 times the current. As can be seen above, this is absurd. If you applied 10 V to the reference solenoid it would pass about 40 amps and probably melt.
Case 2)
Let's keep the wire size the same but reduce the current to 0.125 A. We still need 8000 turns to get the same magnetic field. That is, at 100 turns per layer we need 80 layers of wire. That's 8 inches worth, the solenoid is now 18 inches in diameter. (2 inch central hole plus two 8 inch windings.) Its volume is now 2513.27 cuin and its weight a whopping 804 lbs, 64 times the mass of the reference solenoid. This is beginning to sound like a real Newman coil.
The wire is now 251327 inches long, that's almost 4 miles. It has a resistance of 15.834 ohms. If we put 0.125 A through it, it will dissipate 0.2474 Watts. This is a quarter of the dissipation of the reference coil.
A summary:
In case 1) where we went from 500 to 8000 turns but used the same mass of copper the power required was identical. In case 2) we went from 500 to 8000 turns. We used 64 times as much copper and reduced the power to a quarter. This obviously contradicts Newman's contention that the power required for a given field falls directly in proportion to the number of turns or to the mass of copper used. Thus both his Claim 2) and Claim 3) are incorrect.
For completeness I should point out that a solenoid 18 inches in diameter requires a somewhat larger current to achieve the same magnetic field intensity as would one only 3 inches in diameter unless the length of the solenoid is increased in proportion.
A final confirmation:
I predicted that the power needed for a given field is:
808*F*F*H*(Q + P) Watts.
-----------------
(Q - P)
I claimed that the field was 0.004949 Weber in a solenoid 10 inches high with an outside radius of 1.5 inches and an inside radius of 1 inch. The above equation gives 0.990 Watts which agrees with the figure found above for both the 500 turn and the 8000 turn solenoids.
In the second case H and P remained at 10 and 1 inches but Q became 9 inches. We calculate the power to be 0.247 Watts which also agrees with the result above. If P is made negligibly small compared to Q then the minimum power required is 0.198 Watts, not zero as predicted by Newman.
Appendix:
1] Volume of copper, V = H*pi*(Q*Q - P*P) cubic inches
2] Length of wire, L = H*pi*(Q*Q - P*P)/T*T inches
3] Resistance, R = 0.63*H*pi*(Q*Q - P*P) ohms
---------------------
T*T*T*T*1,000,000
4] Number of turns, N = H*(Q - P)/T*T
5] Field, F = 0.00004949*N*I/H Webers
combining 4] and 5] we get
6] Field, F = 0.00004949*I*(Q - P)/T*T Webers
which leads to
7] Current, I = F*T*T/0.00004949*(Q - P) Amps
8] Power = I*I*R Watts
= F*F*T*T*T*T*0.63*H*pi*(Q*Q - P*P)
-------------------------------------------------------
0.00004949*0.00004949*(Q - P)*(Q - P)*T*T*T*T*1,000,000
9] This reduces to: 808*F*F*H*(Q + P) Watts
-----------------
(Q - P)
Newman -- point by point
Part 2, Commutators and coils
Prepared by Tom Napier. Copyright © 1999, All rights reserved.
Introduction:
This document is one of a series which address specific errors in the book, "The Energy Machine of Joseph Newman."
This section examines the construction of Newman's "Energy Machine" and attempts to make sense of his performance figures. It relates to the section of Newman's book which runs from page 60 to page 70. (Page numbers refer to the 8th edition.)
The early models of the Newman Energy Machine consist of a rotating magnet placed near or inside an air-cored solenoid. A commutator attached to the magnet shaft switches the current from the battery through the windings of the solenoid.
The principle of the motor is that the interaction of the field generated by the current flowing through the coil and the field of the permanent magnet causes the rotating part to make almost a half turn. By reversing the direction of the current flow at the end of the first half turn one can cause the rotor to make a second half turn. The current is then switched back to its original direction for the third half turn, and so on. The commutator and brushes act as a reversing switch to change the current direction twice per turn. This automatically causes a series of impulses which tend to turn the rotor in the same direction, resulting in continuous rotation.
This form of motor dates back to the earliest days of the electric motor. It is still used today as a laboratory demonstration or a toy, though more commonly in the form in which a coil rotates within a fixed magnet.
The advantage of using a rotating coil is 1) the moving part can be much lighter since the heavy permanent magnet is stationary and 2) fewer brushes are needed since the coil being driven rotates with the commutator and can be directly wired to it. The chief disadvantage of this simple motor is that its torque varies considerably throughout the cycle and drops to zero twice per cycle. Should a motor stop near this dead-point it will not start again without being pushed. For these reasons practical motors are built with many more windings on the armature and hence many more commutator segments. This means that the active winding is always the one with the most torque being exerted on it. This gives both a smoother and a stronger torque characteristic for a given input current. Multi-pole motors always self-start.
One interesting feature of the motor is that it has a well defined maximum speed. As the commutator switches on the battery voltage to the coil the current starts to rise at a rate which is inversely proportional to the coil's inductance. However, as the current increases the drop across the coil's resistance reduces the effective voltage and hence the rate of rise of the current. As a result the current rises more slowly as time passes and, given long enough, reaches a fixed value V/R. Normal motors have low inductance windings so the current only reaches the resistance limited value at very low motor speeds. This is why DC motors take a large current when they start or are stalled.
However, at high motor speeds a second effect takes over. The rotating magnet induces a voltage in the coil which opposes the input voltage. At a high enough motor speed this induced voltage cancels out the driving voltage and the coil current drops to zero. This is a good time to break the commutator connection. Firstly, since the coil current is zero, no large voltage spikes will be induced. Secondly, it stops the induced voltage driving a reverse current through the coil and slowing down the motor.
Newman's motor has a single stationary winding and a rotating two-pole magnet. Since he uses very big and heavy coils this makes sense. Because the coil is stationary his commutator has four brushes, two to carry the current in from the battery and two more to carry the current out to the coil. His motor has two other odd features. One is that the two halves of the commutator are not continuous, each is cut into 10 sets of three segments. During each half turn of the magnet the coil is in turn powered, unpowered and short-circuited. The other strange feature is that the coil has a much higher resistance and inductance than is commonly found in electric motors. If the concepts of resistance and inductance are unfamiliar to you see the Appendix.
To make a magnetic field change rapidly, as you must do to reverse the field as the rotor turns, you must either apply a very high voltage or use a low inductance coil. Newman has chosen to use a high voltage, everyone else wants motors which run from 12 volts or 110 volts so they use a lower inductance.
When you apply a voltage to a coil the current through it starts to rise. If neither the coil or the power source had any resistance the current would rise for as long as you applied the voltage. In practice both the supply and the coil have a considerable resistance. The rate of rise of the current, initially high, falls exponentially with time. It eventually settles down at a constant value controlled only by the total resistance and the voltage. If you cut off the voltage soon enough only the initial rise will occur.
Appendix 1. Resistance:
All normal conductors have some resistance, that is, when current passes through them some electrical energy is converted into heat energy. Except in special cases, such as water heaters, resistance is a bad thing. When one wants simply to generate a magnetic field by passing a current through a coil of wire its resistance is a nuisance. Without it no energy would be needed to maintain the field.
The power lost to heating the wire can be calculated from I*I*R watts where I is the current in amps and R is the resistance of the wire in ohms. However, when the current is varying with time, as it is in a Newman motor, measuring or computing the total power lost becomes rather complicated.
A conventional current meter will only give the correct value for the current when the current is constant. If the current is pulsing on and off the meter is probably going to give an incorrect value. However, even if the meter does read the mean current correctly despite the pulses, this does not allow one to calculate the mean power! Because the peak power depends on the square of the current the ratio of the mean power to the mean current depends on the length and shape of the pulses.
As a simple example, suppose 1 amp is passing continuously through a 1 ohm resistor. The mean current is 1 amp and the mean power is 1 watt. Now pulse the current so that 10 amps flows for one tenth of the time and no current flows the rest of the time. The mean current remains 1 amp. However, during the pulse the peak power is 100 watts. The other nine tenths of the time the power is zero. That is, the mean power is 10 watts. Even though the mean current is the same the input power has risen by a factor of ten. The power being dissipated in the resistor has also gone up by a factor of ten, possibly making it dangerously hot. Thus, the mean power cannot be calculated from the mean current. This, unfortunately is what Newman and Hastings do regularly.
Appendix 2. Inductance:
When you change the current flowing in a coil of wire you also change the magnetic field surrounding it. This changing field induces a voltage across the coil which acts against the source providing the input current. The faster the current changes the higher the voltage must be applied to make it change. The extent to which a coil resists a change in current is its inductance. Inductance is simply defined as the ratio between the rate of current input change and the voltage required to make the current change. If a coil's current increases at one amp per second when one volt is connected to it, it has an inductance of one henry.
Note that a pure inductor does not limit the current. If the source could supply a constant one volt for any output current the current could increase at an amp per second for ever. All the power supplied by the source would be stored in the magnetic field. This power can be recovered when the field is reduced to zero again.
Of course in real life the resistance of the coil would limit the current to one volt divided by the coil resistance. If this was 0.1 ohms no more than 10 amps would flow into the coil and the magnetic field would stop increasing at that point. The stored energy is I*I*L/2 joules. The coil had an inductance of 1 henry so it would be storing 50 joules or enough energy to light a 1 watt bulb for 50 seconds.
Unfortunately, input energy is needed to keep the current flowing through the resistance of the coil. This energy would be I*I*R or 10 watts in this case. That is, the resistance would be wasting as much energy as the inductance has stored every five seconds. That's why inductors are not generally used to store energy except for very short periods. For example, if we tried to keep the current flowing by shorting the ends of the coil all the stored energy would be turned into heat in tens of seconds.
If you suddenly stop the current flowing, for example, by opening a switch, the magnetic field collapses very quickly. This generates a very large voltage across the ends of the coil. This voltage can be high enough to create sparks across the switch contacts. Not only is this damaging, it also wastes the energy which was stored in the coil.
Newman -- point by point Part 3, Roger Hastings' errors
Prepared by Tom Napier. Copyright © 1999, All rights reserved.
Introduction:
This document is one of a series which address specific errors in the book, "The Energy Machine of Joseph Newman."
This section examines the arguments used by Dr Roger Hastings to demonstrate an over-unity motor performance. It relates to Chapter 5 of Newman's book which runs from page 22 to page 35. (Page numbers refer to the 8th edition.)
In this chapter, written in 1982, Dr Roger Hastings attempts to show from measurements and calculations that Newman's motor has an efficiency much greater than unity. He makes several elementary mistakes in his reasoning.
Driving the oil pump
In the first section Hastings compares the performance of a normal DC electric motor driving an oil pump with that of the Newman motor driving the same pump. We are not told much about the normal motor except that it is claimed to have 80% efficiency and that, when driving the pump at 1 Hz, it consumed 2 amps at 12 volts. From this Hastings deduces that it takes 19 watts (12 V times 2 A times 0.8) to drive the pump and therefore, if Newman's motor can drive it with less input power, its efficiency must be greater than 100%.
This calculation is completely incorrect. The efficiency of an electric motor does not have a fixed value, it varies from its peak value to zero as the motor speed decreases. A typical 12 volt motor would exhibit its rated efficiency at a speed in the 3000 to 8000 rpm range. Unless the DC motor was geared down to drive the low speed load, and Hastings makes no suggestion that it was, it was running virtually stalled and had almost zero efficiency when running at 1 Hz (60 rpm). Almost all of the 24 watt input would have been heating the motor windings.
From this we can deduce that it takes very little power to drive the oil pump at 1 Hz. Newman's motor is inherently a low speed, high torque device. It would have little difficulty in running at 60 rpm when driving a pump. Based only on the battery life, Hastings estimates that the Newman motor is consuming 2.4 watts. This is consistent with an efficiency of well under 100%.
Power to torque ratio
Hastings then reports on a comparison of the static torques of the two motors and uses this to compare the ratios for torque per watt of the two motors. This comparison, taking the ratio of input power to output force, is dimensionally nonsensical. One can easily generate a static torque from two permanent magnets (or a spring) without using any input power at all. Does this make the magnets infinitely efficient?
As mentioned above, Newman's motor generates a substantial torque when stationary. Normal electric motors do not generate as much torque but they can continue to supply it when rotating at high speed. Since output power is torque times rotation rate, a normal electric motor has a substantial power output at high speeds. Newman's motor has a higher initial torque but is incapable of high speeds; it can manage perhaps five turns per second. Thus the power output of the conventional motor is much higher than the power output of Newman's motor. The faster a conventional motor runs the less input current it takes thus, as mentioned above, its efficiency rises with speed.
Starting power equals running power?
In his section C) Hastings makes a massive blunder. He computes the energy required to spin up the Newman motor from the moment of inertia of the rotor. This calculation is correct as far as it goes. It shows that the motor requires an input of 13 watts to accelerate it from 0 to 6 Hz in 21 seconds. Hastings then states, "This yields a minimum energy to keep the rotor spinning at 6 Hz of 13 watts. Therefore the batteries must be supplying at least 13/70 = 190 mamps."
This again is total nonsense. All he has demonstrated is that the batteries had to supply 190 milliamps during the 21 seconds it took to spin up the motor. Once the motor is spinning the power needed to keep it spinning could be quite low. Hastings supposes that the 21 second initial battery drain continued during the entire ten hour test run. He then expresses amazement that the batteries have lasted longer than the two hours predicted from the 190 milliamp current. Thus is like wondering why your car battery doesn't run down during a 100 mile trip since the starter takes 100 amps.
Motors and currents
In any normal DC motor the input current starts off high. It is limited only by the resistance of the windings which is made low to avoid losses. However, the current drain drops enormously as the motor gets up to speed because of the back EMF generated by its motion. For example, a quick test on a Radio Shack 12 V motor running on 5 V (to avoid damage) showed a drain of 1.7 A when stationary, falling to 0.27 A when running with no load. The induced EMF in the motor coils subtracts from the applied voltage, leaving less voltage to drive the current. Of course, reducing the current also reduces the field and hence the motor torque.
Since the current through each winding is being turned on and off by the commutator the inductance of the winding contributes to the reduced current drain. This inductance makes the coil current take a finite time to rise to its static value. If the commutator switched the current on and off rapidly enough the current would always remain low. In practice the winding inductance is low enough for this not to be a significant effect. For example, a three pole motor from Radio Shack which is rated at 8300 rpm at 3 V has a coil inductance of 275 uH and a resistance of 0.77 ohms. At 8300 rpm each pole is active for 2.4 mS. However, 3 V applied to 275 uH gives a current rise rate of 11 amps per mS. Even though the effective coil voltage is less than 3 V, due to the back EMF, it is apparent that it is the coil resistance which dominates the current.
How Newman's motor differs
Experiment with a small scale Newman-style motor showed that if the supply voltage is applied to the coil for a complete half rotation the current flowing is initially limited by the coil inductance. At low speeds the coil resistance eventually limits the current to a fixed value. However, as the speed increases the voltage induced in the coil by the rotating magnet becomes more important. This voltage drives down the input current. The resultant is a current pulse which rises slowly because of the coil inductance then falls slowly because of the induced voltage. The motor settles down to a natural speed at which the coil current is zero when the commutator breaks it. If you try to make the motor go faster, e.g. by changing the commutator phase and length, the induced coil current actually reverses, slowing the rotor. Incidently, even when the coil current is quite small the voltage pulse across it as the commutator breaks the current can be hundreds of volts. This pulse oscillates at the self-resonant frequency of the coil which is some 40 kHz in my case.
In the Newman motors the commutator is segmented and it has coil shorting segments. The latter allow the induced voltage to pass a current through the coil resistance, heating it and wasting input power. This also acts as a brake on the rotor.
The time available to charge the coil is much smaller than the time constant given by its resistance and inductance so the mean running current is lower than the static current. Twenty times per revolution the segmented commutator allows the field to build up then it shorts the coil so that the field collapses . This stops the current and field ever reaching their static values but is wasteful since all the energy which was just put into the field is lost to heat.
Just before the coil is shorted there is a blank section on the commutator. On reaching this section the field starts to collapse, generating a very high voltage across the coil. This voltage generates sparks, Newman sometimes uses it to light fluorescent lamps. However, all that is happening is that some of the energy which the battery supplied at a lower voltage is being returned at a much higher voltage. (This effect is used in the fly-back transformer used in every TV set.) The rest of the energy goes into heating the coil when the coil is shorted. The whole process then begins again.
By the way, the fact that the magnetic field drops during rotation makes Hastings' computations of the motor torque misleading.
Some real numbers
To quantify this effect, assume the motor is turning at 300 rpm. There are twenty commutator segments per turn, each consisting of 50% on, 30% open circuit and 20% short circuit. Ignoring the width of the brushes, each segment makes contact for 5 mS. Taking Hastings figures for the coil resistance (13 ohms), inductance (23 H) and applied voltage (200 V), we have an initial current rise of 8.7 amps per second. Thus in 5 mS the current will rise to 43 mA. The mean current will be a quarter of this.
Under static conditions the coil would pass some 15 amps if 200 V were applied from a low impedance source. That is, it would be dissipating over 3 kilowatts. Presumably the battery resistance is limiting the voltage to some more reasonable value but this makes one doubt Hastings' figures for the power consumed during start-up. If one assumes the battery resistance is about 0.5 ohm per cell a 200 V battery would have a resistance of around 70 ohms, much higher than the coil resistance of 13 ohms. Thus initially much more power is being dissipated inside the battery than is going into the motor. (Total resistance 83 ohms, therefore current = 2.4 amps. Battery dissipation = 406 watts, motor dissipation = 75 watts. Motor efficiency < 15.7%.)
Running on empty
Hastings makes one curious statement, that "the old batteries have worn down to a point at which they will not even run a 1 1/2 V small toy motor." It is not clear whether he applied the entire battery to the motor or just one or two cells. Small toy motors take an amp or two to get started and old batteries just cannot supply that much current, their internal resistance is too high. However it is not surprising that they can still supply the 100 milliamps or so needed to spin up Newman's high impedance motor.
Disbelieving your own answers
Hastings makes a complicated estimate of the inherent motor efficiency and demonstrates that it cannot be higher than 56%. He then says, "It is clear that measured efficencies for the Newman motor are far in excess of predicted efficiencies." Why, then did he trouble to prove that its efficiency could not exceed 56%. Does he not believe his own calculations?
Newman -- point by point Part 4, Roger Hastings makes more errors
Prepared by Tom Napier. Copyright © 2000, All rights reserved.
Introduction:
This document is one of a series which address specific errors in the book, "The Energy Machine of Joseph Newman."
This section examines further arguments used by Dr Roger Hastings to demonstrate an over-unity motor performance. It relates to Chapter 6 of Newman's book which runs from page 36 to page 39. (Page numbers refer to the 8th edition.)
In this chapter, written in 1984, Dr Roger Hastings attempts to show from measurements and calculations that a small version of Newman's motor has an efficiency much greater than unity. He once again makes several elementary mistakes in his reasoning.
The performance
Hastings claims that the motor has an output "in excess of 10 watts" for an input of less than 0.5 watts. If true this would be a remarkable achievement. However there are extensive errors in the way inputs and outputs are measured and compared. In particular, Hastings seems to be unaware of how powers and currents must be computed in pulse circuits. His 10 watt output is not the mechanical output of the motor he is testing. It is actually his erroneous estimate of the energy being wasted in the resistance of the solenoid.
The motor configuration
The motor tested by Hastings consists of a solenoid about 13 inches high by 11 inches diameter. This is wound with 30 gauge wire and is claimed to weigh 145 pounds. Its resistance is claimed to be 50 kilohms and its inductance is claimed to be 16,000 H. Sitting alongside this is a cylindrical magnet mounted on a bearing. It has a commutator attached to its shaft which switches the current being delivered to the solenoid. There is about a 11.5 inch separation between the center of the solenoid and the magnet bearing. From the scale on the photograph the magnet appears to be 4 inches in diameter and 8 inches long. It is alleged to weigh 14 pounds. Elsewhere in Newman's book (page 65) there is a further reference to this magnet as weighing 14 pounds and having a 4 inch diameter. There is a discrepancy here since a solid cylindrical magnet of these dimensions would weigh over 28 pounds, not the 14 pounds reported.
The motor was driven by a 304 V DC supply. A passing reference makes it clear that this came from 9 V transistor radio batteries connected in series. (In 1984 these would have been carbon-zinc batteries considerably inferior to today's alkaline batteries.)
The commutator is apparently 5 inches in diameter and must generate considerable friction. It differs considerably from the simple two pole design which would normally be used in a rotating magnet motor to reverse the current flow twice per rotation. In Newman's design the commutator contains twenty sections each comprising a through connection, an open circuit and a short across the coil. This has several consequences, not the least of which is that it makes the motor's operation difficult to analyze! The main consequence is that the current into the coil is interrupted before it has risen to its full value. During the open-circuit section of the commutator the collapsing field generates a large voltage spike across the coil. This is followed by a current pulse as the coil is short-circuited. This series of operations can be categorized as energy input, high voltage output and energy dissipation.
Each through section is effective for 0.0156 of a revolution, the open-circuit for 0.0094 and the short-circuit for 0.0063 of a revolution. At the 136 turns per minute quoted by Hastings this corresponds to about 7.5 mS, 4.5 mS and 3 mS respectively.
It should also be noted that, as the brushes pass from the shorting position to the through position, a short-circuit path exists across the battery. In the photographs on page 64 there appear to be gaps or insulating sections in the commutator at this point but it is not clear whether these are wider than the brushes, as they would have to be to avoid a short. If there is a short-circuit condition a high current will flow briefly through the battery twenty times per turn. This is not a high efficiency design.
Taking a somewhat naive approach it can be argued that the commutator applies the 304 V supply to the claimed 16000 H inductor for 7.5 mS twenty times per revolution and that the shorting sector resets the solenoid field to zero at the same rate. The inductive rate of rise of the current would be 19 microamps per millisecond. At the end of each through sector of the commutator the peak current would be 143 uA. The mean current would be 71 uA for 31.25 % of each rotation, giving a mean current of 22 microamps. This is significantly smaller than either the 1.2 milliamps or 14 milliamps claimed by Hastings.
Hastings' test results
The photograph on page 37 shows the current waveform allegedly measured with a 1 ohm shunt in series with the solenoid while the commutator was turning. On the timescale of the photo, 2 mS per division, the risetimes of the stepped waveform are invisible. The fall times have slopes of 0.5 amps per mS. The peak current is 1.5 A and the entire pulse occupies about 5 mS.
As they say in the puzzle books, "What is wrong with this picture?" For a start, it cannot possibly represent the current flowing through an inductor. 300 volts applied to 16000 H would cause the current to rise from zero at about 19 mA per second. Neither the apparent virtually instantaneous rise nor the 500 amps/S fall can occur in a circuit containing a 16000 H inductor. A substantial capacitance across the inductor would be required to pass currents having these rates of change.
It is much more likely that the scope picture arises from the common-mode voltage generated across the commutator than from the differential-mode voltage across the 1 ohm shunt. The schematic on page 37 shows shunts on both sides of the commutator with connections from both to the scope. Unless a scope with a true differential input was being used meaningless results would likely arise with this configuration when a 300 V supply is being switched.
One odd aspect of Hastings reasoning is that he attempts to show that the battery is supplying a large current. For example, he claims that the circuit breaker in an ammeter opened when the ammeter measured the input current. This is supposed to confirm that large current pulses were flowing. Since ammeters, if they have circuit breakers at all, tend to drop out only if an overload current persists for a second or so it is difficult to see what was happening here. Of course, as mentioned above, the commutator may be applying a short-circuit to the battery. The point of this test appears to be to show that at least 150 mA is flowing for long enough to trip the meter.
Hastings then measures the input current by measuring its heating effect in a 500 ohm resistor. He measures a 1 degree C temperature change in 15 minutes and concludes that a mean power of 0.1 W is being dissipated. I doubt if any competent scientist would have drawn such a conclusion from such a small temperature change unless he were using a very elaborate calorimeter but let that pass. Hastings then commits a blunder which crops up again and again in his writing. He assumes that you can derive the average current in a resistor from its average power dissipation and thus derives an average input current of 14 mA. This result is just not true.
Let's take an example. Suppose that the current through the resistor is flowing 30% of the time, not untypical of the current into a Newman motor. If the resistor's average dissipation is 0.1 W it is dissipating 0.333 W when the current is on and, of course, 0 W the other 70% of the time. To dissipate 0.333 W in a 500 ohm resistor requires a current of almost 26 mA. Since this is on for 30% of the time the average current is 7.75 mA, much lower than the 14 mA claimed by Hastings.
Just to complete the picture let's look at the average voltage across the resistor. To pass 25.82 mA through 500 ohms takes 12.91 V. Since this is only on for 30 % of the time the average voltage is 3.87 V. If we applied DC we would have needed 14.14 mA at 7.07 V to get 100 mW. To calculate the true average power we have to multiply the peak current by the peak voltage and then average over time. Multiplying the average voltage by the average current makes it look as if we got 100 mW out for only 30 mW in. Making this error has led a number of people to think they have got power for nothing.
Hastings then measures the DC input as 1.2 mA using a meter. You would think that getting such values as 1.5 amps, 14 mA and 1.2 mA by different measurement methods would have led him to think again. He then argues that the mean current must be lower than 14 mA since there is no apparent battery voltage drop after four hours; what then was the point of the measurement with the resistor which, erroneously, showed that the input current was 14 mA?
Hastings mentions that he has observed negative current spikes and that these cannot come from the battery. Naturally, when you short out a huge inductor you get negative current spikes. Why is Hastings surprised? In any case, rotating a large magnet near a solenoid induces a reverse voltage and current. The spikes disappear when the magnet is mounted on top of the coil instead of alongside it. Again, why is Hastings surprised?
The bottom line
Now comes the number juggling. The 14 mA which was incorrectly computed as being the battery input current now magically becomes the current flowing in the coil. The lowest of all the currents measured, 1.2 mA, is picked as the battery input current. Since both numbers were mismeasured, why not use them the other way around? Because that doesn't give the answer Hastings wants.
He argues that 1.2 mA times 304 V gives an input of 360 mW. Luckily, since the battery voltage is fixed, multiplying the mean current by the voltage gives the correct answer in this case, assuming that the measurement of 1.2 mA means anything at all.
An average of 14 mA is flowing into the coil, says Hastings, and it has a 50 kilohm resistance. That means that it is dissipating 10 watts. Wow! 27 times the input power! So what went wrong?
For a start, we are using a 304 V input. As Hastings has already computed, at most that can drive 6 mA through the coil. Where did that 14 mA come from? Then, as I mention above, the battery is being applied only briefly to the coil. Because of its huge inductance the current never rises to anything like 6 mA. The peak current is 143 microamps, a factor of 10,000 lower than the peak current Hastings claims. The power being dissipated is miniscule.
The real efficiency
Of course there is another factor which Hastings and Newman completely overlook. The efficiency of a motor is measured as the ratio between its output power in mechanical form to its electrical input power. Ignoring the mechanical output and counting the resistive losses in the windings as "output" is not playing fair. Still, if the loses truly were greater than the input power this would be remarkable in its own right. One wonders why Hastings didn't measure the temperature rise of the solenoid to confirm that excess power really was being generated. After all, 10 W input to 145 pounds of copper will heat it at almost 1.5 degrees C per hour. After the four hour run mentioned in the report the temperature rise would have been significant. How odd Hastings didn't notice it.
There will still be lots of scientists and fringe inventors still looking for cold fusion or over unity machine or space energy and the following is a great summary of current free energy claims. Before you get too excited about any of them, consider that none have made a serious effort to go for my $10000 prize for proof of a free energy machine.
Another account:
For those who believe Joe Newman has invented a free energy machine, let me give you the conclusion I came to after spending hundreds of hours experimenting with his ideas and studying his books. I was once very impressed after hearing about the motors Joe Newman claimed he had been building. Joe claimed his motors were able to produce more energy output than input. Most of the people I knew trained in physics laughed at me for believing that such a motor could be built, but having an open mind, and little previous knowledge on the subject of electric motors, I was determined to see if I could build a motor as Joe described. After reading his book over and over, and watching some of Joes video tapes, I could see Joes basic idea for building better electric motors was quite simple. One point Joe use to emphasize was that in his opinion, motor manufacturers have been misled into believing the electro magnetic fields from a coil of wire came from the amount of current passing through the coil, and that the wire it self had nothing to do with the strength of the field produced. and as a result, Joe claimed they have been building motors improperly with a minimal amount of wire, and much potential energy gets wasted because of this, Joes belief is that the magnetic fields produced in a coil of wire can be made much more efficient by increasing the amount of wire used in his stator coils, Ideally, Joe says the stator coil should have many miles of wire in them so as to create the maximum amount of atom alignment with in the coil when electricity is applied, and for the least amount of input current. In one of Joes video he does a demonstration which shows the difference in magnetic field strengths between a small coil and a large one to prove his point, and it is clearly evident there is some truth to what he says, but if you lack a firm understanding regarding the use of electro magnetism in motors, you will not realize there is a catch which he does not mention. I'm referring to the problems you encounter when you try and make use of this larger magnetic field in a motor. It is called counter electromotive force, or EMF for short. I never understood what that term meant until I built my first scaled down model of Joes motor. What I discovered was that Joe was correct about more wire in a stator coil being an advantage over a small amount of wire, but only to a point, what Joe did not mention was the fact that as you increase the size of your coil, there comes a point where the addition of extra wire does not help produce greater power, not only because of the extra resistance from the wire, or the fact that as your coil becomes larger your wire becomes further away from the rotating magnetic armature, but because of an internal counter force which develops in a coil as the armature turns, and it works against your armature turning beyond a limited speed. I did not understand why it was when I added additional wire to my coil my motor kept turning slower. This went against everything Joe had taught in his video tape, and I was at a loss to understand what was happening. A gentleman with a good understanding in motor design explained this problem to me, He said this counter force I observed is a characteristic of any motor, and especially one with a large coil like mine. and this counter force would prevent me from ever building a motor which was able to produce more energy output than input, he told me because my motor was working much like a generator, as soon as the generated voltage reached the same level as my power supply voltage, this counter force would all ways keep me from seeing more energy output than input. This is called Lenz Law, a well known principle to motor builders, but I was still convinced Joe had something, so I built a special commutator which was shown in Joes book. It was a tough one to build on account of the fact that it was built in a way which could reverse the current to my coil every 180 degrees of rotation, and it also had to be made with special contact segments which when rotating would send short pulses of voltage from my batteries to my coil as the commutator turned, and in between each input pulse to my coil, one more segment was needed to short out my coil. Everything was made to very tight tolerances, but I still had no luck when it came to getting more energy out of my motor than I put in to it, and the short segment of my commutator made the motor run even slower. What really surprised me was when I attached a few more very strong rare earth magnets to my armature expecting it to turn much faster, and it slowed down again. I knew It was not from the additional weight of my magnets since they did not amount to much. Once again I was up against the counter EMF problem. I became discouraged, and put everything aside while pondering where I might have gone wrong. Then I happen to find out about a mail list on the internet with many other Newman experimenters I could join in conversations with, and among the list members were some people who had pretty close contact with Joe, One gentleman I was able to speak with, Evan R Soule',Jr The editor of Joes book, and the man I would consider Joes chief PR man would often debate with us about Joe credibility, or the test results on his motor. I spent about a year on the list chairing information with others in hope of having better success with my experimental motor, and over time I kept seeing some pretty nasty arguments between MR Soul and other list members who started to feel Joe was con man. At first I thought the people who attacked Joe were pretty close minded, and Just brain washed by conventional electronic theory. but as I grew in my understanding of electronics and electromagnetism, I realized there was some merit to what others were saying, and I found much of what Joe taught did not make sense. I noticed many of his test results on the internet list seemed to emit important data one could use to judge if his motor did indeed put out more energy than it used. At first I assumed the missing data on Joes motor was just a mistake, and when I ask Evan why Joe left out such vital data, I got some pretty strange answer about Joe not being able to disclose proprietary contractual information, this made no sense to me because the information I was asking for would not have disclosed private data about Joes construction details, I just wanted him to clarify what Joe had all ready stated. To me it just appeared that Evan was covering for Joe, and coming up with a pretty poor excuse not to answer my simple question. At times when Joe did include all his motors performance data, his math was off. So no one really knew what to think about Joes test results. This started to anger many of the list members. Some of the other tests Joe conducted to prove his motors efficiency were quite strange and abstract. It seemed as if he would run every kind of test except the simple and obvious method for showing his motors true efficiency, and some times he would say things that made very little sense to any one. I have seen several occasions where some one made a suggestion about some of Joes theory being incorrect, and Joe would go into a revenge mode calling them stupid liars, or accusing them of being tied in with the "power brokers of the world", and trying to undermine him. Often the debates would wind up in some pretty bad language when some stated that Joe was wrong about something, and this would give Evan grounds to kick them off the list if they did not leave of their own free will. Joe also claimed to have a great many enemies. I started to feel very perplexed about all I saw, because here was this man who at one time was a star guest on the Johnny Carson show on account of his controversial motor, and he had several engineers backing his work, but I don't know where they all are now. Some of them Joe now considers as enemies who have stolen his ideas. in fact, Joe has a large list of people he said have stolen his ideas. but I have yet to see any free energy motors on the market for sale, including anything made by Joe. as I continued to read many unfavorable opinions about Joe, one in particular really brought questions to my mind about what Joe was claiming. A gentleman named Norm Biss who was a professional motor builder was hired to help Joe build his latest motor design, according to Norm, Joe had some pretty mixed up ideas as to how a motor should be built, and Norm said Joe had stolen a key to the door of his company so he could steal the motor one day before the company planed to test the finshed product, Norm said Joe was afraid to let any one test his motor, but Joe claimed that Norm had not built the motor correctly, and after he rebuilt it, he brought it to be exhibited in public, and told everyone he would have proof that his motors worked as he claimed. Many who did attend this show left in an outrage because Joe did not have the motor runing. and those who stayed an extra two days did get to see the motor finally run, but their was no equipment attached to the motor that would give a quantitative analysis of the motors performance. Joe was said to have boasted about all the torque available at the motors output shaft, and that no one could stop it with their bare hands, but for those who were a bit more scientificly minded, trying to stop a motors shaft from turning was not the proof they had traveled many miles to see. Does Joes motor produce free energy? I use to think so after seeing one of Joes videos which showed dry cell batteries showing a greater charge after being used in Joes motor, and I still wonder how this was possible, but when it became obvious Joe was unwilling to do a reasonably simple test in public, and seeing he was holding back important details in his test results, I could see something was very wrong with the whole story. A very simple test Joe could have conducted if he really wanted to prove his motor put out more energy than it used would have been to attach a generator to the output shaft on his motor, and this could have pumped enough power back into his motor to keep it runing with out the use of any batteries. But Joe only does very abstract tests which few can follow the results of. I have never seen him do a simple easy to follow test which would be hard to rig, and I doubt I ever will. Dave Maltz Dave's Alternative Energy. http://www.homezen.com/powersys.html , Amazing 3700 hour white LED flashlights! http://www.homezen.com/powersys/minilite.htm
Get Power From a Resistor
More Free Energy and Hot Air - by Milton Rothman ERIC'S OPEN OFFER TO VALIDATE CLAIMS OF FREE ENERGY Joe Newmans Free Energy Claims - are they valid? Free Energy FAQ page free energy with wires and magnets Ocean Thermal Energy Conversion in Hawaii Electricity out of thin air
pages exposing Joe Newman and Dennis Lee who some people suspect of leading a nationwide scam. Also, Amin, Mills (who may be legit?) and Tewari 
Click to subscribe to free_energy More rants by Eric
Eric