Can we demonstrate over unity energy?

[partzman]

I like the simulation but its still not the same circuit, you need to add D1 and C1 as I have it in the last circuit diagram I post before yours here, I would like to see that simulation and have a copy to run if that is ok.  But this information so far is helpful, please not that D1 and C1 in the last circuit model are a part of what I have been doing with this circuit.

OK, here is your latest circuit with several changes to the sim.  R2 was added to help the sim converge thus speeding it up plus, I removed the "skip initial operating point" (uic) in the simulation command as it produced a large current transient at startup due to the charging of C5 from V1 which then reduced the plot viewing amplitude of any subsequent current traces.  I have also added the initial command statement (.ic) however so C1 starts with no charge.  Also, V(tp1) is offset 10v to separate it from V(vc1) for ease of viewing.  These changes in no way affect the overall energy measurements of the simulation.

The sim plot has both cursors on V(vc1) which indicates the voltage change on C1.  Cursor1 is placed at the point when the voltage on C1 is at it's peak and cursor2 is placed at the end of the sim.  The energy lost In C1 between these cursors is (22.712025^2-21.424129^2) *1 * .5 = 28.42J.

The plot math shows the input energy over the 60s period to be 273.5J and the output energy produced in R1 is 27.955J.  So, the COP = 27.995/(28.42+273.5) = .093.

For observation, the large amount of energy consumed from start to 6.46s is mostly to charge C1.  From that point on, C1 supplies nearly all the energy to the circuit and is depleted over time.  We have also neglected the energy to initially charge C5 to 12v which would amount to 7.2mJ.

I have attached the .asc file for you to play with.

Regards,
Pm

[D.R.Jackson]

OK, here is your latest circuit with several changes to the sim.  R2 was added to help the sim converge thus speeding it up plus, I removed the "skip initial operating point" (uic) in the simulation command as it produced a large current transient at startup due to the charging of C5 from V1 which then reduced the plot viewing amplitude of any subsequent current traces.  I have also added the initial command statement (.ic) however so C1 starts with no charge.  Also, V(tp1) is offset 10v to separate it from V(vc1) for ease of viewing.  These changes in no way affect the overall energy measurements of the simulation.

The sim plot has both cursors on V(vc1) which indicates the voltage change on C1.  Cursor1 is placed at the point when the voltage on C1 is at it's peak and cursor2 is placed at the end of the sim.  The energy lost In C1 between these cursors is (22.712025^2-21.424129^2) *1 * .5 = 28.42J.

The plot math shows the input energy over the 60s period to be 273.5J and the output energy produced in R1 is 27.955J.  So, the COP = 27.995/(28.42+273.5) = .093.

For observation, the large amount of energy consumed from start to 6.46s is mostly to charge C1.  From that point on, C1 supplies nearly all the energy to the circuit and is depleted over time.  We have also neglected the energy to initially charge C5 to 12v which would amount to 7.2mJ.

I have attached the .asc file for you to play with.

Regards,
Pm

Ok I will look at the simulation and thank you I appreciate the file.

[TinselKoala]

The plot math shows the input energy over the 60s period to be 273.5J and the output energy produced in R1 is 27.955J.  So, the COP = 27.995/(28.42+273.5) = .093.

Beautiful work as usual pm. 

Where does the excess input energy go, then? Is it dissipated in the internal resistances of components such as the transistor and R2? Can we be sure that these resistances aren't dissipating more than they "should"?

I haven't looked at the sim files myself so I don't know where all the power-dissipating components are.

[partzman]

Beautiful work as usual pm. 

Where does the excess input energy go, then? Is it dissipated in the internal resistances of components such as the transistor and R2? Can we be sure that these resistances aren't dissipating more than they "should"?

I haven't looked at the sim files myself so I don't know where all the power-dissipating components are.

Thanks TK!  The energy is dissipated in the load, circuit switching losses, coil resistances, etc.  Actually DR's circuit is not as inefficient as it appears in the last sim.  If C1 is replaced with a 10ufd cap and the sim is then run until the circuit stabilizes, the COP is ~ .93 and this could be improved upon with mosfet switching and some timing.  Unfortunately it is still conservative.

Regards,
Pm

[partzman]

Ok I will look at the simulation and thank you I appreciate the file.

You're welcome.  As I pointed out above, your circuit is not as inefficient as it might appear in my last sim.  If the sim was allowed to run long enough to consume the energy in C1 that you've already paid for, the efficiency would show a marked improvement. Try changing C1 to 10ufd plus you could also measure the "on" time of the 2N5550 to determine the pulse width required to drive a mosfet.  These changes should increase your present efficiency.

Regards,
Pm