Bob_the_lost
Elsewhere
Gods, i did this at university. I even passed these modules despite them being the most boring topic ever invented. It won't work.
Diode Array Superior New Energy Source
- Thread starter Kauai Inventor
- Start date
Crispy
The following psytrance is baṉned: All
I don't want to insult you - that would be mean. Yes, you are correct that there are always mavericks - that the accepted truth sometimes gets turned on its head. However, you must also accept that for evey reveolutionary maverick, there are 999,999 cranks, weirdos, fools, or otherwise mislead people. Therefore, you'll just have to grin and bear my skepticism.
I don't believe that useful energy can be extracted from the random movement of air molecules. Seeing as there have been zero recorded instances of the 2nd law of thermodynamics being broken, the probability of there being such an instance are very very very very small.
I personally don't have the scientific knowledge to give your idea a thorough theoretical analysis - I'd wager that you don't either. The people who showed that the earth orbited the sun performed experiments and collected data to confirm their theory. They had additional knowledge that let them make new conclusions. All I see from you, is a lack of knowledge and a lack of experimental data.
Add all that up, and I have to give a mark for the probability of your device working of 0.00 lots of zeroes 1. I'm sorry, that's just the way I see it.
I don't believe that useful energy can be extracted from the random movement of air molecules. Seeing as there have been zero recorded instances of the 2nd law of thermodynamics being broken, the probability of there being such an instance are very very very very small.
I personally don't have the scientific knowledge to give your idea a thorough theoretical analysis - I'd wager that you don't either. The people who showed that the earth orbited the sun performed experiments and collected data to confirm their theory. They had additional knowledge that let them make new conclusions. All I see from you, is a lack of knowledge and a lack of experimental data.
Add all that up, and I have to give a mark for the probability of your device working of 0.00 lots of zeroes 1. I'm sorry, that's just the way I see it.
Kauai Inventor
New Member
1] Thanks for the honesty about your skepticism. Its true that a lot of fools need to be endured but the only way up is past the normals and escaping the conformity police.
2] I believe that the nanonozzle array will work. Someone at a free energy website reported that it failed when they tried it with a color T.V. shadow mask in a vacuum so high that the free mean path was long. I believe that it might have worked with higher gas pressure so the mean free path was short enough to provide many collisions within the individual nozzles.
3a] I don't know when a proposition has enough support or refutation to be certified true or false. I have explored the topic for a long time without finding tight arguments supporting the second law though I don't understand higher math. Sometimes a proposition can be consistent over a wide useful range but fail beyond it like geometry. Instead, I find the second law supported by several loose analogies. The second law may break down in the nanometer realm that we are newly entering.
3b] Some critical experiments are newly possible. The experiments with appropriate diodes are expensive.
4] The payoff if the second law can be broken through is so high that it may be rational to explore the possibility. The payoff alternative may be explored in fiction.
Aloha
2] I believe that the nanonozzle array will work. Someone at a free energy website reported that it failed when they tried it with a color T.V. shadow mask in a vacuum so high that the free mean path was long. I believe that it might have worked with higher gas pressure so the mean free path was short enough to provide many collisions within the individual nozzles.
3a] I don't know when a proposition has enough support or refutation to be certified true or false. I have explored the topic for a long time without finding tight arguments supporting the second law though I don't understand higher math. Sometimes a proposition can be consistent over a wide useful range but fail beyond it like geometry. Instead, I find the second law supported by several loose analogies. The second law may break down in the nanometer realm that we are newly entering.
3b] Some critical experiments are newly possible. The experiments with appropriate diodes are expensive.
4] The payoff if the second law can be broken through is so high that it may be rational to explore the possibility. The payoff alternative may be explored in fiction.
Aloha
Kauai Inventor
New Member
Come to http://www.freewebs.com/diodearray Many diode array applications are listed as terse list items there. If these inspire you to think of more application please tell us. This may inspire yet further applications or development in an open process worth following.
Aloha
Aloha
Kauai Inventor
New Member
A Debate
Here is a debate on the diode array. The full thread is at http://pub6.bravenet.com/forum/show.php?usernum=487525627&msgid=754498
Kauai Inventor:
Aug 29th, 2006 - 4:56 AM
Diode Array statement of knowlege
I haven't learned of anything preventing the success of Johnson noise being rectified by nanometer scale diodes where the output of one diode aggregats with a plurality of others in consistent alignment parallel to produce scalable electrical power with matching refrigeration.
Aloha
Kauai Inventor:
Aug 30th, 2006 - 2:31 AM Talk to us Casual Observer
It is not my presentation policy to state the case for the second law when I believe that the second law can be evaded; I concentrate on the brief and solid merits of my case. I invite Casual Observer to present important relevant material on the 2nd law of thermodynamics, quantum mechanics, semiconductor physics, or electronics for the audience of this forum.
Aloha
Casual Observer:
Aug 30th, 2006 - 4:49 PM
Okay, Charlie, here it is.
It's a waste of time to try to educate you, but I will present an analysis to other interested readers to show why your free energy from noise idea cannot work. I'm going to avoid the thermodynamics argument, as well as any argument involving the discussion of diode junction physics, although either one of them would be sufficient. Instead I'm going to present an idealized noise generator circuit model.
Let's assume a circuit composed of a thermal noise generator (i.e. a resistance) in series with a noiseless diode with a forward bias voltage of 0.3 V. I'm assuming a Schottky diode instead of a silicon diode to reduce the turn-on voltage. This model represents one of the proposed diodes in your diode array. For simplicity I will only consider one diode generator at this point in time. In addition, a load resistor is connected across the resistor-diode combination to form a complete circuit loop.
Thermal noise across the source resistor must have a peak-to-peak value of at least 600 mV in order to turn on the diode during the positive conduction phase, i.e. +300 mV to -300 mV, and dump energy into the load. So how practical is it to generate this amount of noise voltage?
As it turns out, the total rms value of noise voltage generated by any source is completely independent of the value of the source resistor. The shunt capacitance across the source resistor forms a low pass filter to limit noise bandwidth. As the source resistance increases, noise bandwidth decreases but noise voltage per root Hz increases. As source resistance decreases, noise bandwidth increases but noise voltage per root Hz decreases. The consequence is that total rms noise voltage is strictly a function of the shunt capacitance, i.e.
Enoise = sqrt (kT/C), where k = Boltzmann's constant and T = absolute temperature.
Now assuming an 8 sigma peak to rms noise value, 600 mV peak-to-peak of noise will require an rms noise voltage of 75 mV. Plugging that value into the kT/C equation gives us a maximum value of Cshunt = 736E-21 F, or 0.736 attofarads. That's how small the shunt capacitance must be just to allow the source resistor to occasionally forward bias the diode, much less supply any power to the load.
Now 0.736 attofarads in an incredibly small capacitance. Just how small? Let's consider a typical capacitance value between two layers in an IC process that are separated by 1 um of oxide. Checking a standard IC text, we find that C = 5E-17 F / square um.
From this value, we can calculate that Cshunt only requires an area of 0.0147 square um. That represents a square capacitor that is only 121.3 nanometers on a side.
So here's problem #1, Charlie. You must construct your diode AND your load resistor inside a region that is 1 um x 0.121 um x 0.121 um. Note that the shunt capacitance across the load resistor will limit the noise bandwidth just as effectively as the shunt capacitance across the source resistor, so the ENTIRE circuit must fit in that volume. We're talking about a volume that is less than that of the smallest transistor with gate plus source / drain diffusions that we can possibly construct today. If you decrease the oxide thickness, it gets even smaller.
Here's problem #2, Charlie. You can't possibly connect anything to this circuit and extract the power. Any attempt to place the load resistor off-chip, or even make a contact to the noise generator, will immediately increase the shunt capacitance, lower the sqrt (kT/C) value, and render your circuit inoperative.
Here's problem #3, Charlie. Connecting two noise generators in parallel gains you nothing. The rms noise voltage is strictly a function of Cshunt, and connecting two of them in parallel doubles Cshunt and again reduces Enoise below the critical limit. So your idea of connecting diodes in parallel will not work. In fact, more diodes in parallel only makes the problem worse.
Note that the kT/C limit applies to noise voltage or noise current. I can apply a Norton transformation to a noise voltage circuit, get a noise current circuit, and nothing changes. You cannot get around the kT/C value no matter what you do. It is a fundamental noise limit that is factored into the design of high-speed switched-capacitor circuits. It has been verified countless times by working circuit designs and noise measurements.
The only possible solution is to somehow invent a diode with zero turn-on voltage, i.e. an ideal diode. Tell me how to do that, Charlie, and I'll build you a free energy generator. Unfortunately, QM and solid-state physics can easily show that such a thing is impossible.
There you go, Charlie. A simple circuit model that proves the free energy idea that you've spent 33+ years pursuing doesn't work and cannot possibly work. Not that it will matter to you, as you are utterly ignorant of physics and modern electronics, and will ignore everything I've just written.
NIGHT RIDER ! ! !
Aug 30th, 2006 - 6:47 PM
Re: Okay, Charlie, here it is.
Wow , C.O. , VERY IMPRESSIVE ! SERIOUSLY .
Kauai Inventor:
Aug 30th, 2006 - 8:52 PM
Thank you Casual Observer
Your commentary is excellent and sophisticated. It is too rare that someone presents state of the art arguments against the diode array proposal. I hope that we will both try to find the truth.
I agree that capacitance is an undesirable parasite. When the diodes enter the nanometer scale realm which is appropriate to their low power level when handling Johnson noise and the very high current carrying capacity of C nanotubes than the anode face area for a unit diameter single wall C nanotube is less than 3 sq. nanometers. On the other hand two factors leading to greater capacitance are [1] the cathode is an extended plane and [2] the spacing is smaller than 1 um; it is the thickness of the depletion region. This is unknown now. Will you allow that the capacitance will be small enough to not limit the upper frequency severely below the 1 THz upper limit that is already there for thermal quantum reasons? W tip/insulator/Ni base rectifiers were used to rectify light to get a frequency determination for the (1971) National Bureau of Standards.
I have poor understanding of the .3V forward voltage of a Schottky anode diode. The value is part of an exponential curve forward and thermal carrier quantity, proportional to junction area and very low, determined constant current flatness in reverse. Any A.C. voltage will shift the diode resistance lower in the forward direction and higher in the reverse direction. The change in resistance can be considerable. The resistance ratio is the efficiency.
I saw a treatment of a single diode feeding a single resistor. The argument there was that the resistor's Johnson noise power canceled the diode's Johnson noise power. The energy transfer impasse is overcome if many diodes feed a matching load resistor because the value of the matching resistor decreases as the number of diodes increase which decreases the noise voltage of the load resistor. The arbitrary simplicity lead to a faulty conclusion. The thermal noise analysis in Donald Fink's book Radar Engineering assumed that the load did not feed the source. In the case of the diode array, I question an analysis where an outside series resistor supplies the signal. I treat the depletion region crossed by tunneling as a dynamic resistance which changes with the strength and polarity of the electrical power flowing through it. Next, I treat the diodes in consistent alignment parallel as phases in a polyphase rectifying circuit where the ripple decreases as the number of phases increases. The ever smoother D.C. output becomes insensitive to capacitance. Therefore, capacitance beyond the junction will not filter the frequency rectified by the diode.
Aloha
Kauai Inventor:
Aug 30th, 2006 - 9:08 PM
Calling Site Administrator
Help. I would deeply appreciate editing capability. Two green heads mysteriously dropped into my text, More usually, I often see writing bloopers when I read the text after posting it.
Aloha
Kauai Inventor:
Aug 30th, 2006 - 11:45 PM
Eratta
[eratta repaired and reference removed except the following:]
And to modify an invalid argument, the power level of the stabilized cw lasers was greater than Johnson noise. However, I have uncertain memory that there was a statement in the paper that the diode would work at much higher frequencies.
Aloha
Kauai Inventor:
Aug 31st, 2006 - 5:22 AM
Notification
I copied the posts in this thread between Aug 29th, 2006 - 4:56 AM and Aug 31st, 2006 - 5:22 AM to other forums with an introduction and eratta correction of my posts.
Aloha
Here is a debate on the diode array. The full thread is at http://pub6.bravenet.com/forum/show.php?usernum=487525627&msgid=754498
Kauai Inventor:
Aug 29th, 2006 - 4:56 AM
Diode Array statement of knowlege
I haven't learned of anything preventing the success of Johnson noise being rectified by nanometer scale diodes where the output of one diode aggregats with a plurality of others in consistent alignment parallel to produce scalable electrical power with matching refrigeration.
Aloha
Kauai Inventor:
Aug 30th, 2006 - 2:31 AM Talk to us Casual Observer
It is not my presentation policy to state the case for the second law when I believe that the second law can be evaded; I concentrate on the brief and solid merits of my case. I invite Casual Observer to present important relevant material on the 2nd law of thermodynamics, quantum mechanics, semiconductor physics, or electronics for the audience of this forum.
Aloha
Casual Observer:
Aug 30th, 2006 - 4:49 PM
Okay, Charlie, here it is.
It's a waste of time to try to educate you, but I will present an analysis to other interested readers to show why your free energy from noise idea cannot work. I'm going to avoid the thermodynamics argument, as well as any argument involving the discussion of diode junction physics, although either one of them would be sufficient. Instead I'm going to present an idealized noise generator circuit model.
Let's assume a circuit composed of a thermal noise generator (i.e. a resistance) in series with a noiseless diode with a forward bias voltage of 0.3 V. I'm assuming a Schottky diode instead of a silicon diode to reduce the turn-on voltage. This model represents one of the proposed diodes in your diode array. For simplicity I will only consider one diode generator at this point in time. In addition, a load resistor is connected across the resistor-diode combination to form a complete circuit loop.
Thermal noise across the source resistor must have a peak-to-peak value of at least 600 mV in order to turn on the diode during the positive conduction phase, i.e. +300 mV to -300 mV, and dump energy into the load. So how practical is it to generate this amount of noise voltage?
As it turns out, the total rms value of noise voltage generated by any source is completely independent of the value of the source resistor. The shunt capacitance across the source resistor forms a low pass filter to limit noise bandwidth. As the source resistance increases, noise bandwidth decreases but noise voltage per root Hz increases. As source resistance decreases, noise bandwidth increases but noise voltage per root Hz decreases. The consequence is that total rms noise voltage is strictly a function of the shunt capacitance, i.e.
Enoise = sqrt (kT/C), where k = Boltzmann's constant and T = absolute temperature.
Now assuming an 8 sigma peak to rms noise value, 600 mV peak-to-peak of noise will require an rms noise voltage of 75 mV. Plugging that value into the kT/C equation gives us a maximum value of Cshunt = 736E-21 F, or 0.736 attofarads. That's how small the shunt capacitance must be just to allow the source resistor to occasionally forward bias the diode, much less supply any power to the load.
Now 0.736 attofarads in an incredibly small capacitance. Just how small? Let's consider a typical capacitance value between two layers in an IC process that are separated by 1 um of oxide. Checking a standard IC text, we find that C = 5E-17 F / square um.
From this value, we can calculate that Cshunt only requires an area of 0.0147 square um. That represents a square capacitor that is only 121.3 nanometers on a side.
So here's problem #1, Charlie. You must construct your diode AND your load resistor inside a region that is 1 um x 0.121 um x 0.121 um. Note that the shunt capacitance across the load resistor will limit the noise bandwidth just as effectively as the shunt capacitance across the source resistor, so the ENTIRE circuit must fit in that volume. We're talking about a volume that is less than that of the smallest transistor with gate plus source / drain diffusions that we can possibly construct today. If you decrease the oxide thickness, it gets even smaller.
Here's problem #2, Charlie. You can't possibly connect anything to this circuit and extract the power. Any attempt to place the load resistor off-chip, or even make a contact to the noise generator, will immediately increase the shunt capacitance, lower the sqrt (kT/C) value, and render your circuit inoperative.
Here's problem #3, Charlie. Connecting two noise generators in parallel gains you nothing. The rms noise voltage is strictly a function of Cshunt, and connecting two of them in parallel doubles Cshunt and again reduces Enoise below the critical limit. So your idea of connecting diodes in parallel will not work. In fact, more diodes in parallel only makes the problem worse.
Note that the kT/C limit applies to noise voltage or noise current. I can apply a Norton transformation to a noise voltage circuit, get a noise current circuit, and nothing changes. You cannot get around the kT/C value no matter what you do. It is a fundamental noise limit that is factored into the design of high-speed switched-capacitor circuits. It has been verified countless times by working circuit designs and noise measurements.
The only possible solution is to somehow invent a diode with zero turn-on voltage, i.e. an ideal diode. Tell me how to do that, Charlie, and I'll build you a free energy generator. Unfortunately, QM and solid-state physics can easily show that such a thing is impossible.
There you go, Charlie. A simple circuit model that proves the free energy idea that you've spent 33+ years pursuing doesn't work and cannot possibly work. Not that it will matter to you, as you are utterly ignorant of physics and modern electronics, and will ignore everything I've just written.
NIGHT RIDER ! ! !
Aug 30th, 2006 - 6:47 PM
Re: Okay, Charlie, here it is.
Wow , C.O. , VERY IMPRESSIVE ! SERIOUSLY .
Kauai Inventor:
Aug 30th, 2006 - 8:52 PM
Thank you Casual Observer
Your commentary is excellent and sophisticated. It is too rare that someone presents state of the art arguments against the diode array proposal. I hope that we will both try to find the truth.
I agree that capacitance is an undesirable parasite. When the diodes enter the nanometer scale realm which is appropriate to their low power level when handling Johnson noise and the very high current carrying capacity of C nanotubes than the anode face area for a unit diameter single wall C nanotube is less than 3 sq. nanometers. On the other hand two factors leading to greater capacitance are [1] the cathode is an extended plane and [2] the spacing is smaller than 1 um; it is the thickness of the depletion region. This is unknown now. Will you allow that the capacitance will be small enough to not limit the upper frequency severely below the 1 THz upper limit that is already there for thermal quantum reasons? W tip/insulator/Ni base rectifiers were used to rectify light to get a frequency determination for the (1971) National Bureau of Standards.
I have poor understanding of the .3V forward voltage of a Schottky anode diode. The value is part of an exponential curve forward and thermal carrier quantity, proportional to junction area and very low, determined constant current flatness in reverse. Any A.C. voltage will shift the diode resistance lower in the forward direction and higher in the reverse direction. The change in resistance can be considerable. The resistance ratio is the efficiency.
I saw a treatment of a single diode feeding a single resistor. The argument there was that the resistor's Johnson noise power canceled the diode's Johnson noise power. The energy transfer impasse is overcome if many diodes feed a matching load resistor because the value of the matching resistor decreases as the number of diodes increase which decreases the noise voltage of the load resistor. The arbitrary simplicity lead to a faulty conclusion. The thermal noise analysis in Donald Fink's book Radar Engineering assumed that the load did not feed the source. In the case of the diode array, I question an analysis where an outside series resistor supplies the signal. I treat the depletion region crossed by tunneling as a dynamic resistance which changes with the strength and polarity of the electrical power flowing through it. Next, I treat the diodes in consistent alignment parallel as phases in a polyphase rectifying circuit where the ripple decreases as the number of phases increases. The ever smoother D.C. output becomes insensitive to capacitance. Therefore, capacitance beyond the junction will not filter the frequency rectified by the diode.
Aloha
Kauai Inventor:
Aug 30th, 2006 - 9:08 PM
Calling Site Administrator
Help. I would deeply appreciate editing capability. Two green heads mysteriously dropped into my text, More usually, I often see writing bloopers when I read the text after posting it.
Aloha
Kauai Inventor:
Aug 30th, 2006 - 11:45 PM
Eratta
[eratta repaired and reference removed except the following:]
And to modify an invalid argument, the power level of the stabilized cw lasers was greater than Johnson noise. However, I have uncertain memory that there was a statement in the paper that the diode would work at much higher frequencies.
Aloha
Kauai Inventor:
Aug 31st, 2006 - 5:22 AM
Notification
I copied the posts in this thread between Aug 29th, 2006 - 4:56 AM and Aug 31st, 2006 - 5:22 AM to other forums with an introduction and eratta correction of my posts.
Aloha
gentlegreen
I hummus, therefore I am ...
.
Kauai Inventor
New Member
Diode Array Round 2
Second debate between Casual Observer and Charlie:
CASUAL OBSERVER:
Charlie, you are completely ignoring the kT/C limit to thermal noise generation. The rms noise voltage value of any Johnson noise generator is determined strictly by absolute temperature and shunt capacitance. You must explain how your proposed system is going to be constructed with a capacitance less than 1 attofarad. You CANNOT avoid this minimum capacitance limit.
BANDWIDTH (i.e. frequency response) HAS NOTHING TO DO WITH IT! You are hung up on the fact that very high frequency diodes exist, hence (in your mind) your idea should work. But Cshunt is independent of bandwidth. A very high frequency diode will have an internal capacitance far in excess of the minimum value that would permit spontaneous forward-biasing via internal thermal noise generation.
Furthermore, it is trivial to spot the fallacies in your reasoning just by considering what would happen in a diode assuming thermal noise voltages actually could overcome the potential barrier in the depletion region. First, consider a load capacitance between the anode and cathode. Now assuming that you could prevent any portion of this load capacitance from coupling to the interior of a nanoscale diode structure and lowering the kT/C value (impossible, but this is a thought experiment), then if your idea worked researchers could measure a reverse bias voltage across the diode. The combination of thermal noise generator and diode structure would behave like a charge pump, charging up the load capacitor across the anode and cathode.
But there's a problem. If the load capacitor were to somehow accumulate a voltage of (say) 100 mV, then the amplitude of the noise generator must now overcome not only the forward-bias voltage, but also the voltage across the capacitor. The required positive peak voltage is no longer 300 mV; now it is 400 mV. Cmin must therefore decrease even further in order to generate any sizeable potential across the anode and cathode.
But there's more. Ultra-wideband thermal noise would by definition include RF frequencies. You seem hung up on the idea of building nanoscale diodes. If a carbon nanotube Schottky diode could spontaneously forward bias itself via internal noise generation, it would become a wideband emitter of random radio frequency energy bursts, and grow colder than its surrounding environment. For that matter, if the world behaved the way you thought it did, microwave diodes would spontaneously emit bursts of microwaves (with the open anode/cathode connections acting like antennae), and grow colder over time.
And that's another problem, Charlie. Researchers have been constructing carbon nanotube diodes for several years now. NO SUCH BEHAVIOR HAS BEEN OBSERVED. They do not spontaneously emit bursts of high-amplitude RF energy, nor do they grow colder as they do. That's in no way surprising, since such behavior would violate the 2nd law of thermodynamics. Even nanoscale semiconductor diode structures just a few atoms wide do not exhibit spontaneous reverse entropy in the laboratory. Engineers would also be constantly observing such effects in deep-submicron transistor structures, and integrated circuits would cease to operate. But clearly they do work.
Now I've entirely avoided any discussion of the thermodynamics and physics of the depletion region in a diode, which would further negate your arguments. Other people have undoubtedly explained them to you before, far better than I could. But my original point stands. You must explain how you are going to construct a physical structure (with interconnects!) that will have a total capacitance (internal plus coupling to stray external) that is a fraction of an attofarad in value. You cannot avoid the minimum capacitance limit imposed by kT/C if you intend to generate sizeable thermal noise voltages!
The ONLY way you could ever construct a 2nd law violater would be to build a diode with a 0 V forward-bias voltage, i.e. an ideal diode that acts like a perfect conductor in one direction, and a perfect insulator in the other. This would be the electronic equivalent of Maxwell's Demon, but solid-state physics prevents it.
There you have it, Charlie. You asked for an explanation, and you got it. I'm not the first person to try to explain to you why your idea won't work. I probably won't be the last. But as I said, you will not listen. You've been chasing a phantom for 33+ years, and you can't let it go. You will take your obsession to the grave. You may keep arguing away, since you clearly get such a kick out of it, but I am done with you.
CHARLIE:
Casual Observer ----- quote:
Charlie, you are completely ignoring the kT/C limit to thermal noise generation. The rms noise voltage value of any Johnson noise generator is determined strictly by absolute temperature and shunt capacitance. You must explain how your proposed system is going to be constructed with a capacitance less than 1 attofarad. You CANNOT avoid this minimum capacitance limit.
-----
With a anode face less than 3 sq nm, ~4800 times less than the lower limit of ~120 nm on a side square for a process that need not be used, mitigated by the spacing being less than 1 um, sub attofarad capacitance seems possible. The output may be frequency limited at the thermal quantum upper frequency limit rather than C shunt limited. It may be a close enough call to make discussion less straightforward. If the call is close than the power involved is similar anyway.
The capacitance is most relevant when the junction is reverse biased at the maximum power transfer voltage. The thickness of the depletion region under this condition iterates back to determine the capacitance.
The junction capacitance can be reduced by breaking the extended plane cathode into abutting needle cathodes. This alternative is drawn in the patent.
Reasonable goals here are to maximize power density or minimize power cost. If the device with a simple structure is frequency limited, than it is adequate.
kT power always escapes diodes though at low efficiency at low power. The mobile electron current carriers at the junction's stratified edge inside an n type InSb / metal anode Schottky diode are in positional equlibrium between their tendency to leave any particular place and an electrostatic attraction to trapped dopant ions with no associated conductivity which with inert atoms comprize the depletion region. This equilibrium can easily be shifted. kT power can shift it. When kT power is moving electrons towards the anode the depletion region becomes thinner and more conductive; when kT power is moving electrons away from the anode the depletion region becomes thicker and less conductive. Even kT times a capacitance limited frequency band of 0-1 GHz will preferentially rectify some of the originally random power. The net forward current will establish a low voltage as it merges from its source diodes and continues to flow to a load. If the load is an open circuit then the voltage will climb to a value that will repel electrons in the diodes so there is no net flow to the load. If the load is a short circuit then a lot of current will flow from the diodes but there will be no voltage so there will be no electrical power produced. A load that pulls the power down to half of the open circuit value will also conduct half of the short circuit current. The product of the current and voltage will be maximised here meaning that the diode array will release the most electrical power. Any capacitor can be added to the circuit in parallel. If none of the capacitor plates are near the diode junction than the capacitor will not disturb the junction.
Boltzmann established the science of thermal equilibriums. His constant, k, applies to the amount of thermal energy exchanged between participants at equilibrium at one macroscale temperature. A diode can absorb or emit thermal radio noise in due measure. This type of interaction is too normal to notice.
Aloha
Second debate between Casual Observer and Charlie:
CASUAL OBSERVER:
Charlie, you are completely ignoring the kT/C limit to thermal noise generation. The rms noise voltage value of any Johnson noise generator is determined strictly by absolute temperature and shunt capacitance. You must explain how your proposed system is going to be constructed with a capacitance less than 1 attofarad. You CANNOT avoid this minimum capacitance limit.
BANDWIDTH (i.e. frequency response) HAS NOTHING TO DO WITH IT! You are hung up on the fact that very high frequency diodes exist, hence (in your mind) your idea should work. But Cshunt is independent of bandwidth. A very high frequency diode will have an internal capacitance far in excess of the minimum value that would permit spontaneous forward-biasing via internal thermal noise generation.
Furthermore, it is trivial to spot the fallacies in your reasoning just by considering what would happen in a diode assuming thermal noise voltages actually could overcome the potential barrier in the depletion region. First, consider a load capacitance between the anode and cathode. Now assuming that you could prevent any portion of this load capacitance from coupling to the interior of a nanoscale diode structure and lowering the kT/C value (impossible, but this is a thought experiment), then if your idea worked researchers could measure a reverse bias voltage across the diode. The combination of thermal noise generator and diode structure would behave like a charge pump, charging up the load capacitor across the anode and cathode.
But there's a problem. If the load capacitor were to somehow accumulate a voltage of (say) 100 mV, then the amplitude of the noise generator must now overcome not only the forward-bias voltage, but also the voltage across the capacitor. The required positive peak voltage is no longer 300 mV; now it is 400 mV. Cmin must therefore decrease even further in order to generate any sizeable potential across the anode and cathode.
But there's more. Ultra-wideband thermal noise would by definition include RF frequencies. You seem hung up on the idea of building nanoscale diodes. If a carbon nanotube Schottky diode could spontaneously forward bias itself via internal noise generation, it would become a wideband emitter of random radio frequency energy bursts, and grow colder than its surrounding environment. For that matter, if the world behaved the way you thought it did, microwave diodes would spontaneously emit bursts of microwaves (with the open anode/cathode connections acting like antennae), and grow colder over time.
And that's another problem, Charlie. Researchers have been constructing carbon nanotube diodes for several years now. NO SUCH BEHAVIOR HAS BEEN OBSERVED. They do not spontaneously emit bursts of high-amplitude RF energy, nor do they grow colder as they do. That's in no way surprising, since such behavior would violate the 2nd law of thermodynamics. Even nanoscale semiconductor diode structures just a few atoms wide do not exhibit spontaneous reverse entropy in the laboratory. Engineers would also be constantly observing such effects in deep-submicron transistor structures, and integrated circuits would cease to operate. But clearly they do work.
Now I've entirely avoided any discussion of the thermodynamics and physics of the depletion region in a diode, which would further negate your arguments. Other people have undoubtedly explained them to you before, far better than I could. But my original point stands. You must explain how you are going to construct a physical structure (with interconnects!) that will have a total capacitance (internal plus coupling to stray external) that is a fraction of an attofarad in value. You cannot avoid the minimum capacitance limit imposed by kT/C if you intend to generate sizeable thermal noise voltages!
The ONLY way you could ever construct a 2nd law violater would be to build a diode with a 0 V forward-bias voltage, i.e. an ideal diode that acts like a perfect conductor in one direction, and a perfect insulator in the other. This would be the electronic equivalent of Maxwell's Demon, but solid-state physics prevents it.
There you have it, Charlie. You asked for an explanation, and you got it. I'm not the first person to try to explain to you why your idea won't work. I probably won't be the last. But as I said, you will not listen. You've been chasing a phantom for 33+ years, and you can't let it go. You will take your obsession to the grave. You may keep arguing away, since you clearly get such a kick out of it, but I am done with you.
CHARLIE:
Casual Observer ----- quote:
Charlie, you are completely ignoring the kT/C limit to thermal noise generation. The rms noise voltage value of any Johnson noise generator is determined strictly by absolute temperature and shunt capacitance. You must explain how your proposed system is going to be constructed with a capacitance less than 1 attofarad. You CANNOT avoid this minimum capacitance limit.
-----
With a anode face less than 3 sq nm, ~4800 times less than the lower limit of ~120 nm on a side square for a process that need not be used, mitigated by the spacing being less than 1 um, sub attofarad capacitance seems possible. The output may be frequency limited at the thermal quantum upper frequency limit rather than C shunt limited. It may be a close enough call to make discussion less straightforward. If the call is close than the power involved is similar anyway.
The capacitance is most relevant when the junction is reverse biased at the maximum power transfer voltage. The thickness of the depletion region under this condition iterates back to determine the capacitance.
The junction capacitance can be reduced by breaking the extended plane cathode into abutting needle cathodes. This alternative is drawn in the patent.
Reasonable goals here are to maximize power density or minimize power cost. If the device with a simple structure is frequency limited, than it is adequate.
kT power always escapes diodes though at low efficiency at low power. The mobile electron current carriers at the junction's stratified edge inside an n type InSb / metal anode Schottky diode are in positional equlibrium between their tendency to leave any particular place and an electrostatic attraction to trapped dopant ions with no associated conductivity which with inert atoms comprize the depletion region. This equilibrium can easily be shifted. kT power can shift it. When kT power is moving electrons towards the anode the depletion region becomes thinner and more conductive; when kT power is moving electrons away from the anode the depletion region becomes thicker and less conductive. Even kT times a capacitance limited frequency band of 0-1 GHz will preferentially rectify some of the originally random power. The net forward current will establish a low voltage as it merges from its source diodes and continues to flow to a load. If the load is an open circuit then the voltage will climb to a value that will repel electrons in the diodes so there is no net flow to the load. If the load is a short circuit then a lot of current will flow from the diodes but there will be no voltage so there will be no electrical power produced. A load that pulls the power down to half of the open circuit value will also conduct half of the short circuit current. The product of the current and voltage will be maximised here meaning that the diode array will release the most electrical power. Any capacitor can be added to the circuit in parallel. If none of the capacitor plates are near the diode junction than the capacitor will not disturb the junction.
Boltzmann established the science of thermal equilibriums. His constant, k, applies to the amount of thermal energy exchanged between participants at equilibrium at one macroscale temperature. A diode can absorb or emit thermal radio noise in due measure. This type of interaction is too normal to notice.
Aloha
free spirit
more tea vicar?
all this is way over my head, but just on the off chance you might be the nutty inventor, absent minded professor type who's got the solution to global warming etc. in his head but no idea how to get the funding to prove it...
you could have a word with the nesta people who I'm pretty sure have funds to prove and develop new technological innovations. They may take a fair amount of convincing though - good luck
you could have a word with the nesta people who I'm pretty sure have funds to prove and develop new technological innovations. They may take a fair amount of convincing though - good luck
Shippou-Sensei
4:1:2.5
1. DIODES
2. ???????
3. PROFIT!
Shippou-Sensei
4:1:2.5
ah damint
you know just for once i would like an energy source that doesn't risk the destruction of the physical form of humanity
you know just for once i would like an energy source that doesn't risk the destruction of the physical form of humanity
Kauai Inventor
New Member
Once a diode array prototype of useful persuasive power works then many people will swarm in to develop it. The buyers quest for a low price will keep profits modest. Wonderful applications will greatly improve the lot of mankind.
Aloha
Aloha
Shippou-Sensei
4:1:2.5
are you alowed to use adult sissors yet?
Kauai Inventor
New Member
Diode array powered scissors may have a Do-Not-Cut-Live-Flesh feature.
Kauai Inventor
New Member
Please help me with all these possibilities. Some are serious.
Shippou-Sensei
4:1:2.5
yes
please do tell me more about these new scissors of yours
please do tell me more about these new scissors of yours
Kauai Inventor
New Member
A simple way is to have access openings too small for people's parts to fit in. An easy but fussy way is to train to color. Another way is to detect bloodflow pulsations. Infra red detection is prommising. people shed micro fragments that can be detected in a rapid air stream. With compact and reliable energy we can open our imagination. Diode array application development needs sophisticated open management.
Aloha
Aloha
Kauai Inventor
New Member
Word salad lists can be processed into writing of higher utility. Language is valuable.
Aloha
Aloha
Kauai Inventor
New Member
I invite people concerned with sustainable prosperity for a large population to consider the diode array. If it works: [1] The energy footprint is near zero. [2]Diode arrays can be used anywhere.
Aloha
Aloha
Similar threads
