7 ? well, ok, you need a few logic gates to control the switching!
After experimenting with switching inductors with transistors, on and off (excuse the pun) for two years and making little progress, i started experimenting at the beginning of March this year with switched charge circuits using capacitors and inductors - similar to techniques used in some switching Power Supplies
i noticed that i was getting more total Coulombs 'charge' in the capacitors at the end of a test than i had at the beginning
this led me to investigate whether the charge gain could also lead to an energy gain
early tests showed overall system efficiencies around 70 - 80% which surprised me because the 'text-book' equations were showing a maximum 50% efficiency for charging capacitors
i realised after a while that, although i was achieving a high efficiency in charging my output capacitor, i was wasting the energy which was being expended in charging it in the first place
by re-arranging my circuit so that the load resistance was included in both the charging AND the discharge path of the output capacitor i was able to achieve overall system efficiencies around 125%
since then, with lower DC resistance inductors and lower ESR capacitors i've been able to increase the performance to achieve around 145% efficiency at the load wrt the input
the components are all off-the-shelf apart from the inductor, which have been hand-wound for my experiments
i don't believe there is anything special about the inductor - mine are a few mH, DC resistance between 0.5 to 2 ohm, enamelled wire, 0.45mm diam., random wound, solenoid style, on ferrite tube cores about 12mm diam. x 30mm
MOSFET Q1 is switched to give a pulse width which (for my inductor & supply volts) can be in the range, say, 50 - 200us, repetition rate (again for my system) from, say, 0.5ms & greater
i apply of the order of 10 - 30 pulses to charge C2, then (when Q1 is off) switch MOSFET Q2 to fully discharge C2 back through the inductor and resistive load
to check the efficiency at the load, i run the test for one cycle: ie. charge C2 once with a pulse-train, followed by a full discharge of C2
i measure the energy supplied to the circuit by starting with a known voltage on C1, after adjusting either the pulse width or the number of pulses to achieve a 1V discharge from C1, and then calculating the difference in energy states of C1 at the start and end of the test (both C1 and C2 have been measured)
i measure the energy output at the load resistor R1 by recording the voltage waveform across it and exporting the data to Excel, where i calculate the instantaneous power for each data sample (Vr * Vr / R1), then sum all the powers for whole number of samples from the start of the first pulse to the end of the final discharge and divide by that number of samples to get the average power
i then multiply the average power by the time period for that number of samples to get the energy
to check the results, i note the final stored voltage reached on C2 and find the energy it represents for that value of capacitor - this value agrees closely with the discharge energy value obtained from the scope waveform
as a rule-of-thumb, the total energy converted by the circuit will be twice the final amount stored in C2 because the same value of work is expended in charging a capacitor as gets stored by that process - this cross-checks with the total value of energy measured at the load
these are the values for the circuit and scope trace (hopefully) loaded with this post:-
input energy supplied by C1 (from 8V to 7V): 1.42mJoules
measured energy charging C2 through R1: 1.12mJ
measured energy discharging C2 through R1: 0.97mJ
total energy converted through R1: 2.09mJ
load efficiency: 2.09 * 100 / 1.42 = 146%
output energy check:-
final C2 voltage: 3.12V
final stored energy on C2: 0.95mJ
(agrees with measured discharged value shown above)
C1 measured as 190uF (220uF nominal; 100V; non-polarised)
C2 measured as 196uF (220uF nominal; 100V; non-polarised)
R1 measured as 10 ohms
D1 is 1N5817 Schottky diode
Q1 is FDN304P MOSFET (charge switch)
Q2 is IRF540N MOSFET (discharge switch)
progress of these experiments has been posted on a couple of threads on Overunity.com since March '08 and more background and intermediate data is given on my website:
Doc Ringwood's 'Free Energy' page
all the best
sandy
After experimenting with switching inductors with transistors, on and off (excuse the pun) for two years and making little progress, i started experimenting at the beginning of March this year with switched charge circuits using capacitors and inductors - similar to techniques used in some switching Power Supplies
i noticed that i was getting more total Coulombs 'charge' in the capacitors at the end of a test than i had at the beginning
this led me to investigate whether the charge gain could also lead to an energy gain
early tests showed overall system efficiencies around 70 - 80% which surprised me because the 'text-book' equations were showing a maximum 50% efficiency for charging capacitors
i realised after a while that, although i was achieving a high efficiency in charging my output capacitor, i was wasting the energy which was being expended in charging it in the first place
by re-arranging my circuit so that the load resistance was included in both the charging AND the discharge path of the output capacitor i was able to achieve overall system efficiencies around 125%
since then, with lower DC resistance inductors and lower ESR capacitors i've been able to increase the performance to achieve around 145% efficiency at the load wrt the input
the components are all off-the-shelf apart from the inductor, which have been hand-wound for my experiments
i don't believe there is anything special about the inductor - mine are a few mH, DC resistance between 0.5 to 2 ohm, enamelled wire, 0.45mm diam., random wound, solenoid style, on ferrite tube cores about 12mm diam. x 30mm
MOSFET Q1 is switched to give a pulse width which (for my inductor & supply volts) can be in the range, say, 50 - 200us, repetition rate (again for my system) from, say, 0.5ms & greater
i apply of the order of 10 - 30 pulses to charge C2, then (when Q1 is off) switch MOSFET Q2 to fully discharge C2 back through the inductor and resistive load
to check the efficiency at the load, i run the test for one cycle: ie. charge C2 once with a pulse-train, followed by a full discharge of C2
i measure the energy supplied to the circuit by starting with a known voltage on C1, after adjusting either the pulse width or the number of pulses to achieve a 1V discharge from C1, and then calculating the difference in energy states of C1 at the start and end of the test (both C1 and C2 have been measured)
i measure the energy output at the load resistor R1 by recording the voltage waveform across it and exporting the data to Excel, where i calculate the instantaneous power for each data sample (Vr * Vr / R1), then sum all the powers for whole number of samples from the start of the first pulse to the end of the final discharge and divide by that number of samples to get the average power
i then multiply the average power by the time period for that number of samples to get the energy
to check the results, i note the final stored voltage reached on C2 and find the energy it represents for that value of capacitor - this value agrees closely with the discharge energy value obtained from the scope waveform
as a rule-of-thumb, the total energy converted by the circuit will be twice the final amount stored in C2 because the same value of work is expended in charging a capacitor as gets stored by that process - this cross-checks with the total value of energy measured at the load
these are the values for the circuit and scope trace (hopefully) loaded with this post:-
input energy supplied by C1 (from 8V to 7V): 1.42mJoules
measured energy charging C2 through R1: 1.12mJ
measured energy discharging C2 through R1: 0.97mJ
total energy converted through R1: 2.09mJ
load efficiency: 2.09 * 100 / 1.42 = 146%
output energy check:-
final C2 voltage: 3.12V
final stored energy on C2: 0.95mJ
(agrees with measured discharged value shown above)
C1 measured as 190uF (220uF nominal; 100V; non-polarised)
C2 measured as 196uF (220uF nominal; 100V; non-polarised)
R1 measured as 10 ohms
D1 is 1N5817 Schottky diode
Q1 is FDN304P MOSFET (charge switch)
Q2 is IRF540N MOSFET (discharge switch)
progress of these experiments has been posted on a couple of threads on Overunity.com since March '08 and more background and intermediate data is given on my website:
Doc Ringwood's 'Free Energy' page
all the best
sandy
Comment