Language:
To browser these website, it's necessary to store cookies on your computer.
The cookies contain no personal information, they are required for program control.
the storage of cookies while browsing this website, on Login and Register.

### GDPR and DSGVO law

Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

Custom Search

### Author Topic: Longitudinal Wave Experiment to demonstrate Overunity  (Read 44643 times)

#### magpwr

• Hero Member
• Posts: 1154
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #15 on: August 16, 2014, 01:37:22 PM »
Please put an E element with a gain of one as a differential voltage sense across the light bulb load.  Insert  a current sense resistor in series with the bulb.  Follow that resistor with an E element with gain set to 1/RCURRENT SENSE.  DC couple those E element outputs to your scope  Once the voltage and current look sane, then connect a X1 multiplier to get your instantaneous power.

hi MarkE,

I got even better way to show OU.I am using  "1" Ohms resistor at output instead of using 250watt x3  120volt bulb to demonstrate the capability of this circuit.
The spike is nearly 100volts with 1 ohms resistor.
This time i would have to change fuse to 5Amp so my guess it is somewhere below 5Amps.

6.6volts x 5Amps=33watt input with around 98watt output for 1 ohms load."This is just an estimate"

The interesting part the output frequency did not deteriorate much with 1 Ohms load.

--------------------------
Latest update-I have attached single spike waveform using 100v/div in the 200ns range for reference using 1 ohms resistor.

#### MarkE

• Hero Member
• Posts: 6830
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #16 on: August 16, 2014, 01:47:21 PM »
hi MarkE,

I got even better way to show OU.I am using  "1" Ohms resistor at output instead of using 250watt x3  120volt bulb to demonstrate the capability of this circuit.
The spike is nearly 100volts with 1 ohms resistor.
This time i would have to change fuse to 5Amp so my guess it is somewhere below 5Amps.

6.6volts x 5Amps=33watt input with around 98watt output for 1 ohms load."This is just an estimate"

That simplifies things a bit.  Now what you need to do is Tie an E element across the resistor, and then feed that to both inputs of a X1 multiplier.  Finally, you need to put that through an integrator and watch the level change from cycle to cycle after operation has stabilized.  Multiply the difference in the integrator output over the course of one cycle by the operating frequency to get the true power out.

#### gyulasun

• Hero Member
• Posts: 4078
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #17 on: August 16, 2014, 03:39:41 PM »
Hi magpwr,

Sorry to chime in,  would like to ask whether the Multisim circuit simulator uses ideal coil and capacitor models?
I mean when you pick a coil symbol and place it and its Properties window opens, can you choose a loss factor or series resistor value to approach the properties of a real life coil?

If there is no such feature in the simulator, then you have to connect a low Ohm value resistor in series with each of your coil symbols in the schematic to avoid using ideal coils.  (You do not have supraconducting wire to make your coils, do you?)

Considering a 12 kHz "middle" frequency [(4+21)/2] for your circuit, and considering a practical Q=100 unloaded quality factor for all your coils, the series loss resistances would come as follows:

for the 220 uH coil, XL=16.6 Ohm so r=16.6/100= 0.166 Ohm please connect such resistor in series with each 220 uH coil,

for the 500 uH coil, XL=37.7 Ohm so r=0.377 OHm  please connect such resistor in series with each 500 uH coil,

for the 1 mH coil, XL=75.4 Ohm so r=0.754 Ohm please connect such resistor in series with each 1 mH coil

In fact, if the simulator uses ideal capacitors too, then a loss resistance ought to be introduced for them too, estimating the resistor values similarly for them like for the coils.

I do hope that your actual build will give similar extra performance like the simulator forecasts...

rgds,  Gyula

#### magpwr

• Hero Member
• Posts: 1154
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #18 on: August 16, 2014, 04:30:31 PM »
Hi magpwr,

Sorry to chime in,  would like to ask whether the Multisim circuit simulator uses ideal coil and capacitor models?
I mean when you pick a coil symbol and place it and its Properties window opens, can you choose a loss factor or series resistor value to approach the properties of a real life coil?

If there is no such feature in the simulator, then you have to connect a low Ohm value resistor in series with each of your coil symbols in the schematic to avoid using ideal coils.  (You do not have supraconducting wire to make your coils, do you?)

Considering a 12 kHz "middle" frequency [(4+21)/2] for your circuit, and considering a practical Q=100 unloaded quality factor for all your coils, the series loss resistances would come as follows:

for the 220 uH coil, XL=16.6 Ohm so r=16.6/100= 0.166 Ohm please connect such resistor in series with each 220 uH coil,

for the 500 uH coil, XL=37.7 Ohm so r=0.377 OHm  please connect such resistor in series with each 500 uH coil,

for the 1 mH coil, XL=75.4 Ohm so r=0.754 Ohm please connect such resistor in series with each 1 mH coil

In fact, if the simulator uses ideal capacitors too, then a loss resistance ought to be introduced for them too, estimating the resistor values similarly for them like for the coils.

I do hope that your actual build will give similar extra performance like the simulator forecasts...

rgds,  Gyula

hi gyulasun,

Thanks for your invaluable input.I have applied the losses as you suggested for the inductors by connecting resistors in series.

I'm not sure how to apply loss for capacitors but i simply put it as 5% tolerance.There is a remark mentioned as mica/teflon capacitor in properties.

This time the maximum observed spike is reduced from 180volts which was near the bulb filament breaking point to around 140volts spike for the 120volts bulb.

The output frequency is reduced as well to around 3x of input.

#### poynt99

• TPU-Elite
• Hero Member
• Posts: 3582
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #19 on: August 16, 2014, 05:10:21 PM »
hi MarkE,

I got even better way to show OU.I am using  "1" Ohms resistor at output instead of using 250watt x3  120volt bulb to demonstrate the capability of this circuit.
The spike is nearly 100volts with 1 ohms resistor.
This time i would have to change fuse to 5Amp so my guess it is somewhere below 5Amps.

6.6volts x 5Amps=33watt input with around 98watt output for 1 ohms load."This is just an estimate"

The interesting part the output frequency did not deteriorate much with 1 Ohms load.

--------------------------
Latest update-I have attached single spike waveform using 100v/div in the 200ns range for reference using 1 ohms resistor.
How are you coming up with your input and output numbers? It's not a good idea to estimate power from spikey wave forms.

Doesn't MS have a power probe?

I would urge you to watch this video and perform the same power measurement both on your input and output.

#### TinselKoala

• Hero Member
• Posts: 13968
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #20 on: August 16, 2014, 05:46:18 PM »
hi TinselKoala,

Thanks for spotting that one.I have re attached the waveform shown in scope with AC setting.I was playing around with settings and forget about it.
Damm i got have that lunch-I need to charge.

can you draw me basic capacitor connection to probe.I am unable to think clearly now.

I think you may be misinterpreting  me.
You are presenting data to be used in power calculations from scope channels that are _AC coupled_.  This is an error.

When you select 'AC-coupled' on a scope channel, the scope uses a relay or a switch or a software item that puts a capacitor in series with your scope probe, between the BNC jack and the scope's vertical amp/attenuator stage.

This is what I meant by asking you if you would measure an unknown FE device by putting a capacitor in series with your probe... because that is what selecting "AC-coupled" on a scope channel does.

In some cases the coupling may not make a difference: If you are dealing with pure sine signals that are truly vertically symmetrical about the channel zero baseline. But in most cases, selecting AC-coupled destroys the vertical information in the trace, except for p-p amplitudes.  By switching in the capacitor, you have blocked any DC component in the signal from reaching the scope, and the _average_ of the signal is moved down, or up, to the channel baseline by the AC averaging function of the coupling capacitor.

DC-coupling should always be used for total power measurements and in fact for most scope measurements. AC-coupling is only appropriate for a few types of measurements, such as the AC ripple sitting on top of a large DC offset.

Below I show pix of an antique scope's AC coupling caps, and the slightly more modern Link DSO's AC coupling caps.

#### magpwr

• Hero Member
• Posts: 1154
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #21 on: August 16, 2014, 05:49:22 PM »
How are you coming up with your input and output numbers? It's not a good idea to estimate power from spikey wave forms.

Doesn't MS have a power probe?

I would urge you to watch this video and perform the same power measurement both on your input and output.

hi poynt99,

I attached the power meter for your viewing pleasure.

No readout and not implemented in the first place because i'm not surprised why the watt meter can't handle pulse current.

It is the primary reason why i was sticking with a primitive fuse to easily show everyone the estimated current needed without blowing the 2Amp fuse.

The scope shot via 0.1 ohms resistor to measure current is even harder for many to see and understand in pulse and gauge current spike.

It's nearly midnight now.Oh dear my Saturday just flew by.

#### MarkE

• Hero Member
• Posts: 6830
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #22 on: August 16, 2014, 11:00:55 PM »
Magpwer, if you connect up the E elements a multiplier and an integrator as I suggested then you will get the true energy and power delivered to the load.

Capacitors all have internal inductance and resistance.  Those values can be derived from manufacturer supplied S parameters.

#### Marsing

• Sr. Member
• Posts: 300
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #23 on: August 16, 2014, 11:31:02 PM »
Magpwer, if you connect up the E elements a multiplier and an integrator as I suggested then you will get the true energy and power delivered to the load.

Capacitors all have internal inductance and resistance.  Those values can be derived from manufacturer supplied S parameters.

But what is the E element, can it be found on earth ?

#### poynt99

• TPU-Elite
• Hero Member
• Posts: 3582
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #24 on: August 17, 2014, 12:14:47 AM »
hi poynt99,

I attached the power meter for your viewing pleasure.

No readout and not implemented in the first place because i'm not surprised why the watt meter can't handle pulse current.

It is the primary reason why i was sticking with a primitive fuse to easily show everyone the estimated current needed without blowing the 2Amp fuse.

The scope shot via 0.1 ohms resistor to measure current is even harder for many to see and understand in pulse and gauge current spike.

It's nearly midnight now.Oh dear my Saturday just flew by.
Perhaps that meter only works with sine waves, not sure.

Ultimately what does the wave form look like going into the first components after the IGBJT's? That is across L4/C9 and L6/C10? Levels, PW, and Period? You could eliminate the IGBJTs and FG circuitry and replace it with one PWL signal generator. If you can provide me with a good scope shot of the input, then I can replicate this in PSpice and show you how it can compute average power on any wave form.

#### gyulasun

• Hero Member
• Posts: 4078
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #25 on: August 17, 2014, 12:19:30 AM »

...
I'm not sure how to apply loss for capacitors but i simply put it as 5% tolerance.There is a remark mentioned as mica/teflon capacitor in properties.
...
The output frequency is reduced as well to around 3x of input.

Well,  you mentioned that you are going to use good quality capacitors for this circuit, if you could find a data sheet for that type which may include ESR (Equivalent Series Resistance), then you could use that value, also inserting resistors in series with each capacitor in the LC chain with that value.  IF you cannot find data sheet, you may use (as a worst case) 0.1 Ohm resistors in series with all caps in the LC part.
The 5% tolerance you chose has no relation to the losses in the capacitor, it simply means how many percent the actual cap value may differ from its printed value (for a printed label of a 100 pF cap the 5% tolerance means that the cap can have any value between 95 pF and 105 pF).
The remark for teflon/mica cap selection may mean that the simulator treats them with a certain loss which is less than that for the non teflon/mica types, I guess.  Perhaps in the user manual all this is discussed?

Gyula

#### MarkE

• Hero Member
• Posts: 6830
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #26 on: August 17, 2014, 12:47:47 AM »
But what is the E element, can it be found on earth ?
The E element is the SPICE dependent voltage source:

http://bwrcs.eecs.berkeley.edu/Classes/IcBook/SPICE/UserGuide/elements_fr.html

#### magpwr

• Hero Member
• Posts: 1154
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #27 on: August 17, 2014, 03:50:08 AM »
I think you may be misinterpreting  me.
You are presenting data to be used in power calculations from scope channels that are _AC coupled_.  This is an error.

When you select 'AC-coupled' on a scope channel, the scope uses a relay or a switch or a software item that puts a capacitor in series with your scope probe, between the BNC jack and the scope's vertical amp/attenuator stage.

This is what I meant by asking you if you would measure an unknown FE device by putting a capacitor in series with your probe... because that is what selecting "AC-coupled" on a scope channel does.

In some cases the coupling may not make a difference: If you are dealing with pure sine signals that are truly vertically symmetrical about the channel zero baseline. But in most cases, selecting AC-coupled destroys the vertical information in the trace, except for p-p amplitudes.  By switching in the capacitor, you have blocked any DC component in the signal from reaching the scope, and the _average_ of the signal is moved down, or up, to the channel baseline by the AC averaging function of the coupling capacitor.

DC-coupling should always be used for total power measurements and in fact for most scope measurements. AC-coupling is only appropriate for a few types of measurements, such as the AC ripple sitting on top of a large DC offset.

Below I show pix of an antique scope's AC coupling caps, and the slightly more modern Link DSO's AC coupling caps.

hi TinselKoala,

I understood what you meant by adding coupling caps now.

I'm glad this device just managed to "pass" the virtual ou test i think after being scrutinize by many engineers which i gladly accept within 24hours of this newly created topic.

#### magpwr

• Hero Member
• Posts: 1154
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #28 on: August 17, 2014, 04:15:49 AM »
Perhaps that meter only works with sine waves, not sure.

Ultimately what does the wave form look like going into the first components after the IGBJT's? That is across L4/C9 and L6/C10? Levels, PW, and Period? You could eliminate the IGBJTs and FG circuitry and replace it with one PWL signal generator. If you can provide me with a good scope shot of the input, then I can replicate this in PSpice and show you how it can compute average power on any wave form.

hi poynt99,

The igbt is the requirement for actual experiment which i have shown how the waveform would look like in the opening comment for this topic.

There are older versions of this experiment which i was using signal generator which i have posted in Dally kapanadze thread.

I would suggest you start to find out by experimenting.I don't wish to spoil the fun.

In actual experiment H-Bridge is a requirement.I have not experiment with half bridge yet.

#### d3x0r

• Hero Member
• Posts: 1417
##### Re: Longitudinal Wave Experiment to demonstrate Overunity
« Reply #29 on: August 17, 2014, 07:34:08 AM »
@magpwr
I don't suppose your simulation shows any phase difference between the end and the start of the coils?
In my testing, it actually seemed that the phase of the output was actually ahead of the signal driving it... ( although it could have been 330 degrees retarded ) though it did seem to advance further the longer it went instead of being more and more retarded....

I also had at one point my signal generator was in the middle, and one side would get the normal sin wave character, and the other side would get a back-pulse only... which I never figured out a justification for.