Hello all
After reading about "slow wave" traveling wave tubes for high frequency amplification I came up with the following equations.
f = (.9c/Lc)sin(acrtan(P/D)) where
.9c is the speed of light within a insulated copper wire. This will vary because the speed is influenced by changes in the permittivity around the wire. Your hand, etc.
Lc = length of the coil (or a toroid's major radius circumference)
P = Pitch of coil - distance between adjacent coils Controls the "slow wave" speed in the coil.
D = Diameter of coil
Lw/N = P/Sin(a) where
Lw = length of straight wire used to make coil
N = number of turns on the coil
a = arctan(P/D)
Sin(a) = Lc/Lw
Lc = (lamda)sin(a) where "lamda" is the "free space" wavelength.
Have fun
:)
Hello all
As a corollary to the above post I looked into skin effect. The general rule of thumb is that that the skin effect should be more than 1/4 of the wire diameter. This works out for copper to be:
f = 0.06987/d2 where "d" is the diameter of the wire in meters (m)
So for 20 Gage (0.8128mm) wire the maximum frequency is approximately 105KHz and
for 48 Gage (0.0315mm) wire the maximum frequency is approximately 70MHz.
Looks like Litz wire is the way to go when driving a circuit at higher frequencies to avoid higher resistance and heating problems.
:)
lol
by the way you do a outstanding job
I think This kinds of work is very impotent for refereeing the mind
And you have done it well
I think you should do this kinds of work more and more
thank you very much
I agree, thank-you Montec
Hello all
The approximate inductance (L) of a toroidal coil can be calculated by the following.
L = μoμrr2N2/D
μo = permeability of space = 0.000001256636 or 4π-7 (4"pi"-7)
μr = relative permeability of toroidal core (air = 1, other ferromagnetic cores will be based on core alloy)
r = radius of toroid minor axis (what you wind the wire around) plus 1/2 the wire diameter , both in (m)
N = number of turns (wraps around minor axis)
D = diameter of major axis (distance across toroid, center to center of minor axis) in (m)
A formula derived from the above is:
L = μoμrNA/P (Somewhat more useful)
A = area across the minor axis in (m2) (Ï€r2 of coil cross section)
P = pitch of coil winding (distance between wraps at major axis center line) in (m)
This shows that coil inductance is directly related to the number of turns and area within the turn. But is indirectly related (inverse) to the pitch of the coil. A multi- layer coil increases "N" and "r" is then an averaged layer-radius of the coil.
The inductance of a coil allows the calculation of reactance (resistance) at a desired operating frequency. So a high frequency transformer needs less wire to act as resistance at an operating frequency. This keeps the current flow within the specs for the wire.
@ stevenfrank38, MrMag
Thanks for the encouragement.
:)
Hello all
Here are some simple equations for designing a compact "resonant" toroid.
TR = D1/D2 = 1 + 2d/x (Toroid ratio between major and minor diameters)
D1 = Major diameter
D2 = Minor diameter
d = diameter of wire
x = distance between wire loops on the outside of the toroid ( I assume x < d for a TR > 3 will make for a better coil )
D1 = TR/π√(dLw/(TR-1)) (√ = square root, π = "pi") This single layer toroid approximation is valid when D2 >> d.
Lw = length of wire used to wrap the toroid = .9C/fc
C = speed of light
fc = designed frequency of the toroid.
D2 = D1/TR
N = π(D1 - D2)/d Number of loops
P = d(TR/(TR-1) Average pitch for the toroid
fc = (.9c/Lw)sin(acrtan(P/Ï€D2))
Nn-Nn+1 = 2Ï€ The difference between loop counts (N) where "d" remains constant for a multilayer (n) toroid.
Corrections for a previous post. (need that "pi" in there)
f = (.9c/Lc)sin(acrtan(P/Ï€D))
a = arctan(P/Ï€D)
:)
I like pi.
Thanks for the info. very useful.
Came up with an idea to use(home made) ferrofluid as a core material. The core shell can be made so the fluids can be changed to easily experiment with different materials and mixtures. Would be a real time saver, along with only having to wind it 1 time. ;] Just thoughts
Mags
Hello Magluvin
A TR ratio of "pi" or 2"pi" has crossed my mind.
:)
Hello all
Here are some more equations that deal with "slow" waves in coils
Vp = Csin(Ó©) Slow wave velocity in a coil
C = Speed of light
Ó© = "theta" = arctan(P/Ï€D2)
D2 = Coil diameter
P = pitch of coil winding
Some other equations for Vp
Vp ≈ μoμrCND2/4L An approximate value for Vp.
μo = permeability of space = 0.000001256636 or 4π-7 (4"pi"-7)
μr = relative permeability of core
N = number of turns
L = coil inductance This may vary along the length of the coil by winding a non-uniform coil. (Or by placing magnets when using a ferromagnetic core)
Vp ≈ CD2P/4A Another approximation
A = πD2 Area within the coil loop.
Now the big question is what happens when you have two "slow" waves traveling in the opposite directions within the same space.
Time for some experimentation it seems.
:)
Hello all
I should also add that the inductance of coil A can be changed by shorting a contra-wound coil B. (I have measured this with an induction meter.)
:)
Quote from: Montec on April 24, 2011, 02:25:08 PM
Hello all
I should also add that the inductance of coil A can be changed by shorting a contra-wound coil B. (I have measured this with an induction meter.)
:)
Yes but my understanding in that case is that you waste power by using up some part of the magnetic flux for "heating" the counterwound coil wire.
Hello gyulasun
Quote from: gyulasun on April 24, 2011, 03:11:10 PM
Yes but my understanding in that case is that you waste power by using up some part of the magnetic flux for "heating" the counterwound coil wire.
True, but speaking in Amp-turns you could change the inductance of a 10 amp 100 turn coil with a 1 amp 1000 turn coil; or change the resonant frequency of a tank circuit with just a switch (or modulate the tank circuit with an electronic switch).
:)
Hello and Happy Holidays to all
"Slow" wave or Phase coupling between two air coils can be accomplished by matching the phase velocity (Vp) of the coils. This will maximize the coupling. Basically you are just matching the ratio of the pitch and coil diameter.
P1 = Pitch of coil 1 (Distance between adjacent loops in the coil)
D1= Diameter of coil 1
P2 = Pitch of coil 2.
D2 = Diameter of coil 2.
Then
P1/D1 = P2/D2
Since P1/D1 is (when looking at a right triangle) the tangent (opposite over adjacent) of the acute angle then Vp = CsinÓ¨ = Csin(arctan(P1/Ï€D1))
where
Vp = Phase velocity of the signal within the coil
C = Speed of light
Ó¨ = Acute angle
When using two wires of different gages (wire diameter) then the gage of the wire can replace the pitch in the above equation if the coil is wound tightly.
G1 = gage of wire 1
G2 = gage of wire 2
then
D1 = G1((G2+G1)/(G2-G1)) Diameter of coil 1
D2 = G2((G2+G1)/(G2-G1)) Diameter of coil 2
The diameter of the form used to wind the inside coil is
D = G1((G2+G1)/(G2-G1) - 1)
A thin layer of tape may be used to separate the inside and outside coils, or they may be contra-wound.
:)
Hi good day ;D
But This is what Tesla said:
“Today's scientists have substituted mathematics for experiments, and they wander off through equation after
equation, and eventually build a structure which has no relation to reality.â€
But still the formula of the vibration is missing. :P
At least your a scientist :D . joke
But i believe you a little ;)
Quote from: gyulasun on April 24, 2011, 03:11:10 PM
Yes but my understanding in that case is that you waste power by using up some part of the magnetic flux for "heating" the counterwound coil wire.
If the counterwound coil is shorted by a capacitor than this capacitor will give back the energy induced in that coil during the next quarter of the cycle. The resistive RI^2 losses in that coil do not have to be high if the induced current is low and voltage is proportionally high (this is the same voltage that appears across that shorting capacitor).
See http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564 (http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564)
I now have two oscilloscopes and can tune or hunt for resonance with confidence.
Much background and experimental information is in overunityresearch.com under the ltseung888 bench.
Thank you for the helpful information.
Quote from: ltseung888 on January 04, 2012, 09:52:28 AM
See http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564 (http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564)
I now have two oscilloscopes and can tune or hunt for resonance with confidence.
Much background and experimental information is in overunityresearch.com under the ltseung888 bench.
Thank you for the helpful information.
I looked at the experimental data at overunityresearch.com (http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564) but could not register to reply there.
I noticed that you are multiplying the instantaneous voltage and current on the input and output side of your devices.
I applaud you on this correct and accurate method of measuring power. If everyone was measuring power like this then there would be much less confusion.
Do you know if your oscilloscopes multiply Ch1 * Ch2 every sample by every sample before decimation and display?
...or do they multiply Ch1 * Ch2 after decimation/averaging and display?
This is an important question to answer from the standpoint of the power measurement accuracy. Unfortunately many oscilloscopes do not multiply all of the the data that is sampled. Instead they multiply only the data that is decimated/integrated and displayed.
If your oscilloscope does not multiply every sample then it is useless for measuring high frequency power.
Also, I noticed that in that thread at overunityresearch.com (http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564) you used the phrase "Power RMS".
If you are calculating a Root-Mean-Square of a power waveform (after instantaneous multiplication * voltage) then this RMS is a wrong function to use at that stage.
You should calculate the regular
arithmetical mean (average, AVG) of the power waveform to obtain the average power (RMS is a mistake here).
RMS values are useful to obtain e.g. the equivalent RI^2 heating value of current or as arguments to multiplying RMS(current)s * RMS(voltage), if the waveforms are sinusoidal and the phase relationship between them is known.
In other words:
WRONG:RMS(voltage_inst * current_inst) = Average Power
AVG(voltage) * AVG(current) = Average Power
CORRECT: voltage_inst * current_inst = Instantaneous Power
RMS(voltage) * RMS(current) * cos(Phi) = Average power (but only for sine waveforms)
AVG(voltage_inst * current_inst) = Average power (for any waveforms)
Quote from: verpies on January 04, 2012, 11:50:29 AM
I looked at the experimental data at overunityresearch.com (http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564) but could not register to reply there.
I noticed that you are multiplying the instantaneous voltage and current on the input and output side of your devices.
I applaud you on this correct and accurate method of measuring power. If everyone was measuring power like this then there would be much less confusion.
Do you know if your oscilloscopes multiply Ch1 * Ch2 every sample by every sample before decimation and display?
...or do they multiply Ch1 * Ch2 after decimation/averaging and display?
This is an important question to answer from the standpoint of the power measurement accuracy. Unfortunately many oscilloscopes do not multiply all of the the data that is sampled. Instead they multiply only the data that is decimated/integrated and displayed.
If your oscilloscope does not multiply every sample then it is useless for measuring high frequency power.
Also, I noticed that in that thread at overunityresearch.com (http://www.overunityresearch.com/index.php?topic=1171.msg19564;topicseen#msg19564) you used the phrase "Power RMS".
If you are calculating a Root-Mean-Square of a power waveform (after instantaneous multiplication * voltage) then this RMS is a wrong function to use at that stage.
You should calculate the regular arithmetical mean (average, AVG) of the power waveform to obtain the average power (RMS is a mistake here).
RMS values are useful to obtain e.g. the equivalent RI^2 heating value of current or as arguments to multiplying RMS(current)s * RMS(voltage), if the waveforms are sinusoidal and the phase relationship between them is known.
In other words:
WRONG:
RMS(voltage_inst * current_inst) = Average Power
AVG(voltage) * AVG(current) = Average Power
CORRECT:
voltage_inst * current_inst = Instantaneous Power
RMS(voltage) * RMS(current) * cos(Phi) = Average power (but only for sine waveforms)
AVG(voltage_inst * current_inst) = Average power (for any waveforms)
Thank you for your excellent comments. I shall clarify the following points.
1.
My Atten Oscilloscope calculates the Ch1*Ch2 values at every sample. Thus the Instantaneous Power value at any instant is equal to the instantaneous Voltage*Instantaneous Current.2.
The Instantaneous current is actually the instantaneous voltage across a one ohm resistor.3.
The Atten Oscilloscope also allows me to capture the actual time interval, Ch1 and Ch2 values in a CSV file. The CSV file can be manipulated via the Microsoft Excel Program.4.
To get a real understanding of what is happening, I usually compare the Output waveforms with the Input (Instantaneous voltage, current and power).5.
In many instances, the Output Power Curve resembles a Standing Wave. Standing waves are characteristics of resonance. In all cases of Standing Waves at Output, the area within the standing wave (energy) is much higher than the Input Area.6.
For accurate measurement, I normally use the arithmetic mean of the Ch1*Ch2 from the CSV file. In addition, I can calculate the average positive area and/or the average negative area of the power curve. 7.
For a perfect Standing Wave, the positive area will be identical to the negative area. Thus the mean power value will be zero. This is often confusing to engineers not familiar with Pulse- resonance circuits. Sometimes, we can treat the positive area as energy going out and the negative area as energy feeding back.8.
You can forget about my use of pp and rms calculations. Such calculations are for training purposes only. Instead of spending an hour on average manipulating the CSV, I let the students do simple comparisons with pp or rms values.9.
In some early posts, I use the term Tseung FLEET Comparison Index pp or rms. Such values are useful for comparing different FLEET prototypes or tuning on the same prototype. You are right is saying that such results are not meaningful outside such comparisons. They are useful specifically for my comparison purposes.10.
When I or my students do the tuning, we use two oscilloscopes. One displays the Input Instantaneous values and the other displays the Output Instantaneous values. The shape and the amplitude easily tells us whether we are near resonance.11.
The tuning for a given toroid can be based on varying the holes on the breadboard, the spacing of the wires, adding capacitors, resistors, LEDs, Diodes or other electronic components. So far, my best prototype results have COP > 100 based on the accurate CSV file calculations. I shall provide more such data at overunityresearch.com under the bench of ltseung888. The plan is to train as many person as possible on the basic technique and then go to the 100 watt range with different electronic components and larger toroids.
That pretty much clears up my concerns about your power measurement techniques.
I understand the concept when you write "Output Power Curve resembles a Standing Wave" but this phrase somehow rings wrong, because the Power Output Curve is a synthetic wave (a result of a calculation) that does not exist as an actual wave in your circuit because it does not represent something real, such as current or charge distribution in a helical winding or a transmission line.
I think it would be more correct to write "Output Power Curve is indicative of a Standing Wave in ...".
I hope you don't get offended at this semantic nit picking ;)
Another subject:
After all of your experimentation, can you elaborate on your observations about the relationship between:
1) Bulk/Lumped LC resonance.
2) Reflected wave resonance (a.k.a. standing wave)
As an example, notice that there is only one frequency at which the inductive and capacitive reactances are equal in lumped LC resonance HOWEVER there are many frequencies (harmonics) at which the reflected waves form standing waves.
Note for newbes: See the attachment and the high school video demonstrating mechanical standing waves (caused by reflection) at different frequencies :
http://www.youtube.com/watch?v=4vdSP-580Vw (http://www.youtube.com/watch?v=4vdSP-580Vw)
Also, see the electric standing waves in a real coil, visualized by a neon screwdriver:
http://www.youtube.com/watch?v=Jxp6wrh2Pqo (http://www.youtube.com/watch?v=Jxp6wrh2Pqo)
Quote from: verpies on January 05, 2012, 07:05:44 AM
That pretty much clears up my concerns about your power measurement techniques.
I understand the concept when you write "Output Power Curve resembles a Standing Wave" but this phrase somehow rings wrong, because the Power Output Curve is a synthetic wave (a result of a calculation) that does not exist as an actual wave in your circuit because it does not represent something real, such as current or charge distribution in a helical winding or a transmission line.
I think it would be more correct to write "Output Power Curve is indicative of a Standing Wave in ...".
I hope you don't get offended at this semantic nit picking(https://overunityarchives.com/proxy.php?request=http%3A%2F%2Fwww.overunity.com%2Ffile%3A%2F%2F%2FC%3A%2FUsers%2FJen%2FAppData%2FLocal%2FTemp%2Fmsohtmlclip1%2F01%2Fclip_image001.gif&hash=d120e86edf8aa597491650a1f5340fe264629a7b)
Another subject:
After all of your experimentation, can you elaborate on your observations about the relationship between:
1) Bulk/Lumped LC resonance.
2) Reflected wave resonance (a.k.a. standing wave)
As an example, notice that there is only one frequency at which the inductive and capacitive reactances are equal in lumped LC resonance HOWEVER there are many frequencies (harmonics) at which the reflected waves form standing waves.
cstanding waves (caused by reflection) at different frequencies :
http://www.youtube.com/watch?v=4vdSP-580Vw (http://www.youtube.com/watch?v=4vdSP-580Vw)
Also, see the electric standing waves in a real coil, visualized by a neon screwdriver:
http://www.youtube.com/watch?v=Jxp6wrh2Pqo (http://www.youtube.com/watch?v=Jxp6wrh2Pqo)
Thank you once again for the excellent comments and information.
I want to share the actual experimental observations with you in this post.
Please examine the four pictures. We shall discuss the implications in thwe coming posts. For now, do you agree that the Output Power Curve as shown on picture 2 contains more area (energy) as compared with the Input Power Curve?
Looking forward to your comments.
It doesn't matter how much good advice you offer Lawrence he just carries on believing that he has OU
he's been doing it for years, his blind faith in what he does is only giving the free energy community a bad name.
I don't know if he enjoys making people angry when they discover that nothing he has will give them free energy,
but I suspect he loves it, after all he's been doing it for long enough now to know better.
at best we have a lesson in how not to do things at worst we are supporting a con man.
Quote from: ltseung888 on January 05, 2012, 09:41:31 AM
do you agree that the Output Power Curve as shown on picture 2 contains more area (energy) as compared with the Input Power Curve?
I cannot say yet.
What is the meaning of the ~5.00V and ~500mV legend at the bootm of the traces. Does it mean 5V and 500mV per division?
I am not sure about the vertical offsets of those traces. Where is the zero on the Y axis? Can it be trusted? Does the adjustment of the vertical position (vertical offset) of Ch1 or Ch2, change the result of the Math Ch. ?
What is the arithmetical mean of the Output Math trace? I am unable to average it by eyeballing it and I don't feel like counting pixels ;)
Quote from: powercat on January 05, 2012, 03:21:02 PM
It doesn't matter how much good advice you offer Lawrence he just carries on believing that he has OU
he's been doing it for years, his blind faith in what he does is only giving the free energy community a bad name.
I don't know if he enjoys making people angry when they discover that nothing he has will give them free energy,
but I suspect he loves it, after all he's been doing it for long enough now to know better.
at best we have a lesson in how not to do things at worst we are supporting a con man.
I am not a big believer but if I find no errors in his power measuring methodology, then I will have no choice but to accept his OU claim.
Of course any claim can be confirmed by replication or refuted with fraud, but that's not a scientific argument anymore. More like a sociological one...
Quote from: microcontroller on January 05, 2012, 02:43:23 PM
And then still this is not the right way to measure input vs output energy !
He is simultaneously sampling the instantaneous current and instantaneous voltage at high frequency at the
input of the DUT, multiplying the samples and averaging the results to obtain average input power (or integrating it to obtain input energy)
He is simultaneously sampling the instantaneous current and instantaneous voltage at high frequency at the
output of the DUT, multiplying the samples and averaging the results to obtain average output power (or integrating it to obtain output energy)
Dividing these two averages (powers) yields In/Out Power ratio (or COP)
Am I missing something?
Are you suspecting that he is not sampling both channels simultaneously or loosing data between samples?
What would be the correct power measuring methodology according to you?
Quote from: verpies on January 05, 2012, 06:04:06 PM
I cannot say yet.
What is the meaning of the ~5.00V and ~500mV legend at the bootm of the traces. Does it mean 5V and 500mV per division?
I am not sure about the vertical offsets of those traces. Where is the zero on the Y axis? Can it be trusted? Does the adjustment of the vertical position (vertical offset) of Ch1 or Ch2, change the result of the Math Ch. ?
What is the arithmetical mean of the Output Math trace? I am unable to average it by eyeballing it and I don't feel like counting pixels(https://overunityarchives.com/proxy.php?request=http%3A%2F%2Fwww.overunity.com%2Ffile%3A%2F%2F%2FC%3A%2FUsers%2FJen%2FAppData%2FLocal%2FTemp%2Fmsohtmlclip1%2F01%2Fclip_image001.gif&hash=d120e86edf8aa597491650a1f5340fe264629a7b)
The scale for Ch1 is 5.00V per division. The scale for Ch2 is 500mV per division. I deliberately set the Input and Output traces to the same scales including the horizontal time scale. The zero position for the Atten Oscilloscope for Ch1 is the arrow next to the number 1. The zero position for Ch2 is the arrow next to the number 2. The product (Ch1*Ch2) is indicated by the symbol M (Maths function).
I shall provide both the raw CSV file for Input and Output. You can work out the artithmatic means from the raw data. I shall also provide the enhanced CSV file giving the mean value, the positive only mean value and the negative only mean value.
I shall also provide the full circuit diagram both as circuit diagram drawing and from the breadboard. The possibility of experimental error is practically zero.
I can also swap the oscilloscopes and the probes to further reduce possibility of equipment error.
In addition I can provide similar Standing Wave patterns from at least 3 other prototypes already in my possession.
By the way, the method of using the Instantaneous Voltage * Instantaneous Current to get the Instantaneous Power was introduced to me by the Electrical Engineering Professors at the Hong Kong Universities. It was also confirmed by Electrical Engineers trained at MIT and UCLA. It is an established academic technique.
May the Almighty guide all the researchers to the proper path.
I have converted the CSV files into XLS files because this forum does not allow downloading of CSV files
ADS00026.XLS is the Output and ADS00027 is the Input raw data. You will find surprises in these two files.
Have fun.
I shall redo the experiment to absolutely confirm that the CSV (XLS) files correspond to the DMP files.
*** The chance of the ADS00026 and ADS00027 corresponding to the above DMP files is very high. The Ch1 voltage pp from the file is 7.2V (7.6 on the DMP file) and the Ch2 voltage pp from the file is 280mv (280mV on the DMP file).
Quote from: microcontroller on January 05, 2012, 07:40:03 PM
You not missing something, your missing a lot more then that.
That's not a scientific argument.
What am I missing?
Quote from: ltseung888 on January 05, 2012, 08:53:18 PM
I have converted the CSV files into XLS files because this forum does not allow downloading of CSV files
ADS00026.XLS is the Output and ADS00027 is the Input raw data. You will find surprises in these two files.
Have fun.
I shall redo the experiment to absolutely confirm that the CSV (XLS) files correspond to the DMP files.
*** The chance of the ADS00026 and ADS00027 corresponding to the above DMP files is very high. The Ch1 voltage pp from the file is 7.2V (7.6 on the DMP file) and the Ch2 voltage pp from the file is 280mv (280mV on the DMP file).
Looks like you have vertical resolution problems.
Look at the attached graph made from the ADS00027.XLS file. The Ch1 is quantized to 2bits ( less than 4 levels - 3 actually). Such large quantization error makes your captured data unsuitable for further calculations.
The vertical resolution of other channels is only a little better:
ADS00026 Ch1: 38 levels (59% of
6 bit resolution)
ADS00026 Ch2: 17 levels (53% of
5 bit resolution)
ADS00027 Ch1: 3 levels (75% of
2 bit resolution)
ADS00027 Ch2: 13 levels (81% of
4 bit resolution)
In ADS00026 you should increase the analog amplification (sensitivity) on Ch1 so it takes the full advantage of your scope's maximum vertical resolution (e.g. all 8-bits, 256 levels).
In fact all the channels should be amplified as much as possible but without incurring clipping by the A/D converter (sampler). Some people call it "normalization to the A/D limit"...
P.S.
Digitizing at 500Ms/s is fine as long as the sampling of the two channels is simultaneous (not interleaved!)
Quote from: verpies on January 06, 2012, 05:51:14 AM
Looks like you have vertical resolution problems.
Look at the attached graph made from the ADS00027.XLS file. The Ch1 is quantized to 2bits ( less than 4 levels - 3 actually). Such large quantization error makes your captured data unsuitable for further calculations.
The vertical resolution of other channels is only a little better:
ADS00026 Ch1: 38 levels (59% of 6 bit resolution)
ADS00026 Ch2: 17 levels (53% of 5 bit resolution)
ADS00027 Ch1: 3 levels (75% of 2 bit resolution)
ADS00027 Ch2: 13 levels (81% of 4 bit resolution)
In ADS00026 you should increase the analog amplification (sensitivity) on Ch1 so it takes the full advantage of your scope's maximum vertical resolution (e.g. all 8-bits, 256 levels).
In fact all the channels should be amplified as much as possible but without incurring clipping by the A/D converter (sampler). Some people call it "normalization to the A/D limit"...
P.S.
Digitizing at 500Ms/s is fine as long as the sampling of the two channels is simultaneous (not interleaved!)
The scales were set for eyeball comparison purposes. I shall repeat the experiment to try to get the largest sensitivity. Prof. Steve Jones uses the better oscilloscope at his BYU University for accurate quantitative measurements.
I shall try to test the limit of the Atten Oscilloscope in the coming experiments. I do believe the Atten Oscilloscope sample the two channels simultaneously and have common ground.
Thanks for the useful scientific comments. God Bless.
Quote from: ltseung888 on January 06, 2012, 10:20:37 AM
Prof. Steve Jones uses the better oscilloscope at his BYU University for accurate quantitative measurements.
Even the best digital scope will capture inaccurate data if it's using only 2bits of its A/D converter ;)
Your scopes have 8 bits of vertical resolution, so they are theoretically capable of capturing up to 256 levels of voltage.
It's just a matter of setting the vertical sensitivity (analog amplification) appropriately to the signal amplitude.
Quote from: verpies on January 06, 2012, 04:05:13 PM
Even the best digital scope will capture inaccurate data if it's using only 2bits of its A/D converter ;)
Your scopes have 8 bits of vertical resolution, so they are theoretically capable of capturing up to 256 levels of voltage.
It's just a matter of setting the vertical sensitivity (analog amplification) appropriately to the signal amplitude.
I played with the probe settings and the scale settings to get more sensitivity. Here are the results.
I look forward to your comments and analysis.
*** All data are stored in overunityresearch.com. The most likely correct one is shown here.
Quote from: ltseung888 on January 06, 2012, 05:12:14 PM
I played with the probe settings and the scale settings to get more sensitivity. Here are the results.
I look forward to your comments and analysis.
A little better this time:
ADS00042 Ch1: 99 levels (77% of
7 bit resolution, 39% of max resolution)
ADS00042 Ch2: 35 levels (55% of
6 bit resolution, 14% of max resolution)
ADS00046 Ch1: 43 levels (67% of
6 bit resolution, 17% of max resolution)
ADS00046 Ch2: 7 levels (88% of
3 bit resolution, 3% of max resolution)
...but the sub 3 bit vertical resolution of the data captured on Ch2 is still unsuitable for further calculations.
7 levels of Ch2 constitute
only 3% of your scope's maximum vertical resolution, and that multiplied by the respective 17% of Ch1 yields less than 0.5% power calculation accuracy that would be possible if both channels were taking advantage of their full 8-bit max resolutions.
Also, the summing of the instantaneous power values (Ch1*Ch2) in column E, does not yield average power.
In this case the summing corresponds to integrating power and the result of this operation is
energy.
I don't know if your intention was to calculate the In/Out power of the DUT or its energy.
In case it was the former - in order to obtain average power, the formula in Cell E11253 should be =AVERAGE(E3:E11252)
Of course dividing output energy by the input energy is also valid and meaningful - I just don't know if this is what you had intended.
Experiment on Jan 7, 2012
This experiment focused on getting the best sensitivity from the Atten Oscilloscope. It was different from others focusing on getting resonance.
Here are the results:
Hope this provides better sensitivity or more accurate calculations. The future experiments may focus on both tuning and senisitivity. I expect the more accurate results to confirm Output Power greater than Input Power also. The raw data file is provided here. With the raw data, I can get the average, the total positive only values etc. The analysis of the Standing Wave will be much more meaningful.
Looking forward to more valuable comments.
Experiment on Jan 8, 2012
This experiment used two 1 ohm resistor connected in series as the current voltage for both Input and Output. Thus for comparison purposes for COP, we do not need to do the division first.
The Output Power waveform has both positive and negative parts. We can interprete the positive power as power going out to do useful work. The negative power is the feedback. For a perfect standing wave, the positive power is equal to the negative power. Thus the net power required to maintain a standing wave is zero.
In this particular experiment, the average output power is negative. This may be interpreted as more power is fed back than consumed.
See the data below.
I believe the csv files are now good enough to provide the needed COP (mean) calculations/estimateions.
Praise the Almighty.
The Input and Output xls has been updated with analysis.
The COP from average is 2.2.
The Output Power average value is negative. That implies feedback is higher than supplied power.
In addition, both the positive and negative areas of the Power Curve are much higher than those of Input. This implies that we must examine such values when we deal with Standing Waves or similar.
More details will be available at overunity.com under the ltseung888 bench.
Overunity or Lead-out energy is confirmed by such results.
Dr. Tseung,
Please confirm the model number of your scopes is ADS1102, 100MHZ bandwidth. Then the probes to be used are P6100 100MHz, x1x10 passive probes. Thanks.
Quote from: aaron5120 on January 09, 2012, 09:04:54 PM
Dr. Tseung,
Please confirm the model number of your scopes is ADS1102, 100MHZ bandwidth. Then the probes to be used are P6100 100MHz, x1x10 passive probes. Thanks.
Confirmed.
Here is a close up shot of the prototype tested on Jan 8 that produced a COP of -2.2.
It was basically a Joule Thief Circuit with two secondary windings. The number of turns was selected to completely surround the toroid. The actual diameter and characteristics of the toroid did not make much difference. In the experiment on Jan 8, we aimed for maximum sensitivity and any result with COP > 1 was acceptable.
The COP was calculated from the average of the Output Power divided by the Input Power. This is the so called true COP.
The actual of -2.2 was obtained. The negative sign came from the Average Output Power. There can be different interpretations of the negative sign. One was - more feedback than supplied power.
The additional calculation in the ADS00057 and ADS00058 files focused on the positive and negative values separately. Such calculations highlight the effect of the "standing wave". In the particular experiment, the result was not a good standing wave and the COP value was low. This will be improved in the coming experiments.
Please go to overunityresearch.com, under the ltseung888 thread for update and details.
Experiment on Jan 11a
The air coil toroid was used. We introduced unecessary wires. Here is the result.
The result should be compared with shortening the wires to the minimum. in the next post.
Experiment of Jan 11b
The wires were reduced to the minimum length.
Dear Verpies,
I have updated ADS00067 and ADS00070 to include the positive and negative area only calculations.
Such calculation is important to understand the possible "standing wave" effect.
Even though the waveform might not be a standing wave, the positive and negative area by themselves are significant.
Using only the average or the mean is not sufficient.
Any comments?
A possible application of the FLEET technology.
The Company in the Far East has mastered the FLEET technology. They have FLEET prototypes with COP > 20.
They already give out promotional LED hats without the FLEET technology.
Can they make the LED hat last much longer without changing batteries?
The Divine Wine is being bottled.
Hello all
Here is an equation for a three coil phase-velocity (Vp) matched air transformer.
D1 = Diameter (at center-line of coil) of coil 1
D2 = Diameter of coil 2
D3 = Diameter of coil 3
G1 = Gage (diameter of wire and insulation) of wire in D1
G2 = Gage of wire in D2
G3 = Gage of wire in D3
D1/G1 = D2/G2 = D3/G3 = (G3 + 2*G2 + G1)/(G3 - G1) where G3 > G2 > G1.
(G3 - G1) influences the resultant coil diameters so the closer G1 is to G3 the larger the resultant coil will be. (G3 + 2*G2 + G1) is a distance along a section of the coils diameter and (G3 - G1) is the change in pitch (gage) for said section.
The above equation assumes a tightly wound construction.
:)