Rosemary Ainslie circuit demonstration on Saturday March 12th 2011

[Rosemary Ainslie]

Sorry about that screw up.  It's unlikely that Harti will amend those posts so I may redo them all.  Just needed to get onto another page here.  Abject apologies guys - I sized that report wrongly.  What I'll do is repost the report and then Harti can simply delete the previous to get it back in order. 

Again.  Sorry.  I'll start it all again.

REPORT ON TWO TESTS OF A SWITCHING CIRCUIT FOR DEMONSTRATION AT CPUT on 12TH MARCH 2011.

Prepared by Rosemary Ainslie, Donovan Martin, Evan Robinson, Mario Human.

The following tests were designed to evaluate some aspects of a thesis that predicts a potential for the conservation of potential difference at a supply. This thesis is based on a non-classical magnetic field model and what is demonstrated here is a non-conservative field condition on a circuit, as required by that model. While this may confront Kirchhoff’s Laws, the experimental results are in line with Faraday’s Laws of Induction. This may suggest that Inductive Laws supersede the conservative field requirements. It is proposed, therefore, that the results are in line with classical requirements albeit that they seemingly contradict the results determined by the Second Law of Thermodynamics.

1 FIRST TEST

1.1 Circuit description
The experimental apparatus comprises a simple switching circuit (see Figure 1). 6 x 12 volt lead acid batteries are in series with both a heating element (RL1) and 5 MOSFET transistors (Q1) in parallel. The transistors are driven by a functions generator. A current sensing resistor (Rshunt) on the source rail of the supply determines the rate of current flow both to and from the battery supply source.

[Rosemary Ainslie]

2 - continued - revised report

1.2 Schedule of circuit components
1.2.1   Resistor element RL1 - Incoloy alloy air heating rod element threaded with nichrome resistive wire. Resistance = 11.11Ω, L = 2.23μH. 200 watts. Supplied by Specific Heat
1.2.2    Current sensing resistor Rshunt - 4 ceramic wire wound 1 watt resistors 1Ω each, placed in parallel. Resistance therefore = 0.25Ω. L = 220nH
1.2.3   MOSFET transistor Q1 - 1 x IRFPG50 with Zener body diode
1.2.3 MOSFET transistor Q2 â€" 4 x IRFPG50 with Zener body diode
1.2.4   Function generator
1.2.5   6 x 12v batteries - Raylite silver calcium

1.3 Schedule of measuring instruments
1.3.1   Le Croy WaveJet 324 200 MHz Oscilloscope (DSO) - 2GS/s 400 Vpk tolerance. Sample range maximum 500 000 samples
1.3.2   Tektronix MSO 3054 Mixed Signal Oscilloscope (DSO) - 500 MHz 2.5 GS/s. Sample range maximum 1 million samples
1.3.3   FLUKE Digital Multimeter TopTronic T48 True RMS with thermocouple measuring to 400°C (rated at ±1%+4).

1.4 Circuit operation
The circuit is designed to allow a secondary current flow that is induced from the collapsing fields of RL1 during the OFF period of the duty cycle as a result of counter electromotive force (CEMF).  This reverse current path is enabled by the body diode in the transistors as well as the eccentric positioning of MOSFETs (Q2) that are configured to enable a negative current flow driven by a negative charge applied to the Gate of Q2.  This allows a current flow that returns to the battery supply source to recharge it.  Small adjustments to the offset of the functions generator enables the generation of a ‘burst oscillation’ mode that is triggered when the gate voltage defaults below zero.  This oscillation occurs at a naturally resonating frequency determined by the impedance of the circuit components.  The adjustment to the offset also requires careful tuning to regulate the level of power required to be dissipated at the load.  See Figure 3 for typical gate voltage setting.

1.5 Measurement of wattage dissipated
Measurement of the energy dissipated at the resistor element (RL1) was determined by comparison with results from a control to avoid the complexity of factoring in power factor corrections. A constant voltage was applied from a DC power supply source in series with RL1. The voltage was then steadily increased in increments of 1 volt each from 1 volt through to 22 volts. The wattage was then determined as the squared product of the voltage over the resistance of RL1,

(1)


The temperature of the resistor was then recorded against the applied wattage and the temperature difference above ambient determined the level of wattage as represented in Table 1 and Figure 2.


edited.  corrected the inductance measured over the shunts.

[Rosemary Ainslie]

3 - continued revised report

1.6 Measurement of wattage delivered by the battery supply
Power is calculated as vi. The flow of current (i) is determined by the voltage measured across Rshunt over the resistance of Rshunt.


(2)


Typically the battery supplies a direct current. Therefore, voltage that is measured above zero, is considered to result in a current flow delivered by the battery. And, conversely, voltage that is measured below zero is considered to result in a current flow delivered to the battery. The instantaneous wattage delivered to or by the battery is then determined as the product of the voltage across the batteries and the current.

2. SECOND TEST

2.1 Circuit description
The circuit is configured identically to the diagram in Figure 1 but with a reduction in the number of batteries applied to three, supplying approximately 36v. All other parameters are identical to the First Test.

2.2 Circuit operation
With a reduced supply voltage, the voltage across Rshunt increases, corresponding to the increase from the positive applied voltage signal from the gate during the ON period of the duty cycle and as determined by the offset. This results in an increase in current flow from the battery. This increase is commensurate with an increase in temperature rise that is measured to be dissipated on RL1. The rate of temperature rise depends on the offset adjustment and the applied source battery voltage during this ON time. At its highest setting, this results in an excess of 44 watts being dissipated. It has not been possible to test this to higher temperatures and for extended periods, as the there is a limit to the voltage tolerance of the DSOs.

2.3 Measurement of wattage dissipated at the load
The applied protocol is consistent with that described in 1.5 of Test 1. 

2.4 Measurement of wattage delivered by the battery supply
The mean average and cycle mean average of the voltages measured across Rshunt now default to positive. Instantaneous wattage analsys is based on para 1.6 above.


3. RESULTS

3.1 First test
The temperature over RL1 indicates that about 6 watts is being dissipated as heat. However, the instantaneous wattage analysis indicates that more energy has been returned to the battery than has been supplied resulting in a net zero loss of potential difference from the supply.  Of interest is that the mean and cycle mean average voltage across Rshunt are consistently negative. 

More wattage returned to the battery than was delivered.
Wattage dissipated at RL1 = 6 watts.
Sustained periodic condition of oscillation enabled for 2.7 minutes to the limit of the intervals allowed by the function generator

3.2 Second test
The mean average and cycle mean average voltage across Rshunt indicates that some current has been discharged by the battery to the source rail. However, instantaneous wattage analysis applied to the voltage measured across the battery and Rshunt indicate, here too, that the battery supply source has had more energy returned to recharge it than was first applied to the circuit.  When this is applied to each sample from a spreadsheet analysis across the 500 000 to 1 million samples supplied by the digital storage oscilloscopes, then the product of this and the battery voltage represents the instantaneous wattage. The sum of these values, divided by the number of samples, represents the average wattage delivered over the entire sample range. This results in a negative value indicating that more energy is still being returned to the battery than was delivered. This is in line with the math function of the DSOs where it, too, indicates an increase of wattage back to the battery supply over the amount of wattage initially delivered from that supply.

More wattage returned to the battery than was delivered
Wattage dissipated at RL1= 44 watts
Switching results in the generation of extreme spiking at the transitional phases of the switch. 

4. ANOMALIES

4.1   It is understood that during the ON time the applied signal at the gate will enable a current flow from the battery supply. With the application of more than 36 volts from the battery supply, the circuit can be tuned so that there is no measured voltage or consequent flow of current through to the source rail of the supply during this ON period. The precise cause of this restriction has not been identified and requires further research. Nor can this condition be simulated.

4.2   When the offset of the function generator is adjusted (see Figure 3), the falling edge of the pulse results in a burst oscillation mode. Parasitic inductance is a well-known consequence of MOSFETs placed in parallel. It is undesirable for switching applications and is therefore, traditionally, factored out of the circuitry. On this application we have enabled that oscillation to the limit of the function generator’s slowest switching speed at 2.7 minutes or 6.172mHz. No material or evident variation or decay of that resonance throughout that entire period, is observed (see Figure 4). This results in a measured increase of recharge at the battery supply as well as sustaining the temperature over the resistor. It would be desirable to extend this period of oscillation to see whether decay in this oscillation, eventually takes place. These results may warrant further research, as the implications are that the current flow may be perpetuated through this self-oscillation.

[Rosemary Ainslie]

4 - continued revised report

4.3   Also apparent is that the oscillation is required to retain the temperature measured at the resistor at approximately 40°C above ambient. This temperature rise corresponds to a dissipation of approximately 6 watts at RL1 (according to Figure 2). The fact that it retains this heat is not a result of any unique properties to RL1 as the temperature is seen to fall steeply over a 3 minute period, when it is disconnected from the supply.

4.4   At these slowest switching speeds, at 6.172 mHz, and during that burst oscillation mode period where the frequency is measured at close to 1.5 MHz, the battery supply source is seen to recharge. The same oscillation amplitude is evident at all higher frequencies with the same attendant benefits.

4.5   The voltage across the shunt is at 180 degrees in anti phase with the voltage across the battery (Figure 5) and the voltage across the Drain (Figure 6).  While this is repeatable in simulations it is not evident that the oscillations can be sustained at the same amplitudes over an extended period.

[Rosemary Ainslie]

5 - continued revised report

4.6   Typically, and as can be seen from the oscilloscope screen shots, it is possible to tune the circuit through adjustments to the offset and the duty cycle, to obtain a negative mean average and cycle mean average voltage measured at Rshunt. This indicates that there is more current being returned to the battery supply than was first delivered. This is confirmed by detailed analysis of data downloads to spreadsheets.

4.7   There is evidence of approximately 6 watts of energy dissipated at RL1, and upwards of 40 watts on Test 2, at no measurable cost of energy delivered from the supply. As this heat is not at the cost of energy from the supply it suggests that there is an alternate energy supply source or classical prediction errs in its assumption of equivalence in the transfer of energy.

4.8   Measurement of battery voltage was determined by the mean average voltage on the digital storage oscilloscopes, as well as from the digital multimeters, with probes placed directly on the positive and negative terminals of the battery supply. These battery voltages fluctuate in line with the evident voltage variations of the waveforms displayed. What is shown is that there is a recharge period after the discharge of current from the voltage during the ON period of the duty cycle. It is more clearly evident at the slowest switching speed. This indicates that there is a battery recharge during the period when the switch is in burst oscillation mode that occurs when the gate voltage is negative. Therefore is there evidence that the oscillations resulting from this negative triggering, are indeed recharging the battery.


5. SIMULATION

5.1   The circuit was setup in Simetrix version 5.4 (Figure 7) and simulated in correlation with the above tests (Figure .