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

User Menu

Custom Search

Author Topic: Diode Array, many small parallel diodes to aggregate rectified Johnson Noise  (Read 21863 times)

Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
The diode array and photocells that yield electricity from ambient heat by absorbing 10 micron infrared will probably ultimately become the best energy sources in the world. They may be thousands of times cheaper than present forms of energy and have fewer direct side effects. Ihis would have enormous social impact.

The diode array consists of billions of C60 buckyball anodes on a N type InSb substrate which becomes very many uniform very small diodes in consistant alignment parallel. Each diode contributes 1/2 kTB watts times the device efficiency to the array. k = Boltsmann's constant, ~ 1.38 !0^-23; T = absolute temperature, degrees Kelvn; B = the natural thermal bandwidth, ~10^12 Hz. I have a rough expectation of 50% efficiency so each diode should yield ~1 nanowatt. The aggregated power may be hundreds of watts per sq cm or sq inch which is a unit ~6.5cm? times larger @ 20C / 68F.

I hold U.S. patent 3,890,161, DIODE ARRAY filed 1973, granted 1975, Reverted to the public 1992.
On the web at? www,diodearray.com and http://peswiki.com/index.php/OS:CBC:Main_Page

College nanofab lab people with a supportave academic atmosphere can check out the diode array concept and, if encouraging results emerge, build start up practical chips. These people can start with a brief discription of the concept then progress, reinforced by each other in the global research network.

Frequently Raised Objections:
1) The load will nullify the effects of the source. Diode arrays may run at 1/20 volts matched load. This voltage would be produced by a 10 meg ohm resistor. The load resistance is typically much lower.
2)There is a minimum foward voltage reqirement:
No, the current tunnels through the depletion region circumventing this abstract voltage. The depletion region changes width dynamically; it is narrow and highly permiable in the foward direction and wide and obstructive in the reverse direction. Any shift perferring one polarity over the other will be aggregated. The diode array prototype succeded in being an over kTB machine.
3)Maxwell's demon needs more energy to work than is gained by his work:
The demon can be entirely passive. It does not have to illuminate the moving charges to work with them. Diodes will slowly deteriorate to non rectifying blobs in a very long time but they will handle much more energy than is needed in their manufacture in that time.
4)It works for full wave rectification but not half wave:
It is easier to see that it works with full wave rectification but the key to its operation, asymetrical conductivity, is present in the half wave variation. The full wave version uses 4 diodes per element and a source energy resistor while producing 2 times the power. furthermore, the 4 diodes and the resistor have to be carefully arranged. Its harder to make and is half as good.


Aloha, Charlie

Edit concens inserting my picture which I haven't suceeded in doing yet.
« Last Edit: April 16, 2005, 07:41:48 AM by Charlie Brown ARN »

andreas_varesi

  • Newbie
  • *
  • Posts: 33
Hi Charlie,

your Diode Array sounds very convincing, much more than all of those magnetic overunity toys. Do you have any prototypes running or is there anyone who made use of your patent?

Thanks for the information

Andreas

Kysmett

  • Full Member
  • ***
  • Posts: 101
If I remember correctly, you were looking for manufacturers for a prototype.  Am I correct?
If so, I may be in a position to help.  Please send me (or post)the specs and the processes that you think are needed and I will talk to a small lab that I have connections with and see if I can get it done. 

There is one question that I have been meaning to ask.  If these diodes extract power from the ambient, then at what point in the manufacturing process is a potential voltage created (an of course how much)and how does one deal with that.

Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
Kysmett,

Nano Lab went next door to a neighbor among the hi tech companies near MIT for a spin casting machine to stretch a film of C60 buckyballs in plastic monomer evenly on the surface of the 2" InSb wafer  I sent them from www.universitywafer.com . Carbon is so light that it is hard to see with x rays. A scanning electron microscope can probably show where the C60s are. I decided to relax the spacing spec from 9nm to 30 nm, There are now expected to be 10^11 C60s / cm2.

The diode array will produce its open circiut voltage of ~1/10 volts as soon as its made. This voltage may serve to repel the C60s into an even dispertion. An insulative cover slip may be needed.
Applying the metal top conductive layer is one of the last steps in manufacture; the diodes will be separated before the metal is applied.
 The wafer may be cut into dies before he top metal layer is applied and the dies coated seperately. It is important not to short out the output with bridges from the top metal to ground on the bottom of the substrate. This won't harm the chip but it will hide the output. Electricity will be produced and cooling will occour as soon as an electrical load in value between a short and an open is attached between the metal layers.

Aloha, Charlie

Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
andreas varesi, 

The prototype was made from a diode patch which is almost a diode array. As diodes became smaller to handle higher frequencies They became smaller and harder to find so a leading innovator came up with the idea of making the diodes in an easy to find patch where one diode would be selected at random. Dr. Thomas W Crowe of the University of Virginia, Charlottsville went to the National Nanofabrication Facility at Cornel U for masks to make evenly spaced Au anodes embedded in glass on N type GaAs to form these chips which were used in satellite transponders untill fabrication techniques became good enough to use one diode. Dr. Crowe and others from the solid state device lab U of V, C spun away to form Virginia Diodes Inc where draw down stocks of diode patches as dies on wafers are availiable. Dr. Crowe does not support the diode array but VDI will sell diode patches to experimenters. Sole inventors get a 50% discount.
I invite people to conduct experiments on these diode patches which are converted to diode arrays by applying a conductive layer to bring all the diodes into consistent alignment parallel. The diodes are too large to be practical but the lab I commissioned  got somewhat more than 1/2 kTB watts from a chip immersed in pure inert vegetable oil for temprature uniformaty in a well shielded box. They applied ferromagnetic conductive paste to the face of the chip. The ferromagnetic quality makes the paste easy to handle.

It is bolder to go for the C60 version of the diode array which should lead to excellent practical use. I think that many developers should work together on the web to realize this wonderful energy prospect.

Aloha, Charlie 

Kysmett

  • Full Member
  • ***
  • Posts: 101
C60 Buckeyballs are hollow and usually used to encapsulate something or at least have that ability if I am not mistaken....  are the ones you use empty?  If so, would it pay to place an atom it it to function as a more  efficient P material?

Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
 Kysmett,

 Undoped C60 is metallic in conductivity. The C60 is is similar to Au in Au / semiconductor diodes notably the Au / GaAs diodes in the satellite transponder diode patch. Metal recombines with injected electrons quicker than a P type semiconductor would. C60 dis more durable than Au, it will not diffuse or electromigrate. Dopped shell or interior C60 can be tried later. Larger C buckyballs like C84 are tempting because the greater diameter gets the embedding insulating layer further from the lower limit of thickness of ~6 angstroms.

Aloha, Charlie 

Kysmett

  • Full Member
  • ***
  • Posts: 101
The diode array will produce its open circiut voltage of ~1/10 volts as soon as its made. This voltage may serve to repel the C60s into an even dispertion. An insulative cover slip may be needed.
Applying the metal top conductive layer is one of the last steps in manufacture; the diodes will be separated before the metal is applied.
 

So then it is assumed that one must be very careful when handling as they are always "live".  What is the energy density at your new dispersion?

Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
Being live but locally shorted indefinately with a non damaging amount of power involved is acceptable. This property can be used to trim the voltage of a lot of DAs in series. This is simpeler than non shorting methods. Inefficient but smooth and cheap methods of power control are acceptable because heat dissipated in analog power transistors can be cheaply recovered. Good thermal engineering is needed for a reliable device. A pump in a cooling loop can always be running.

At 30nm spacing in a tesselation of equlateral triangles there will be ~10^11 diodes / cm2 this implies ~100 watts / cm2 @ 20C. The first approximation is that the power is proportional to the square of the absolute temperature. The diodes are spaced much wider than the diameter of the C60s so depletion regions can grow to cut off the reverse polarity Johnson Noise excursion. The depletion regions can be crowded somewhat because losses due to leakage can be made up for in diode density. I do not know the exact size of the reverse polarity depletion region.

Aloha,

Charles M. Brown

andreas_varesi

  • Newbie
  • *
  • Posts: 33
Hi Charlie Brown,

I have also tried to built a prototype that uses diodes to convert brownian movement directly into electric energy. My main problem is, that all diodes I did use had a too high threshold voltage, e.g. Germanium with 0.2 - 0.4V. The voltage I recieved per micro element was only some mV, so I wasn't able to rectify and collect the current. Also using a bias voltage to overcome the threshold didn't work, because I always had a leakage current from the battery that was higher, than the current I could get from one micro element (not the thing we call overunity). The only kind of diode I didn't try was a nano tube diode. Do you know, whether they have also a threshold voltage? How do you solve this problem in your system?

Thank you for all your answers

Andreas

Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
I stuffed a perfboard with ~2,000 cheap diodes like the 1n914 in ~1992 too and had Forrest lab test it. They found 10^-15? watts or so which is less than the output of one ideal diode, 1/2 kTB or ~2 x 10^-9 watts. The problems here are limited frequency response, a THz is 1,000 GHz, high junction parasitics of capacitance and leakage ,and large size limiting the diodes/ cm2. The forward voltage isn't an issue because a nonconductive diode will have high voltage Johnson noise. I see the operation of a diode array where the diodes share a smoothed common voltage as each diode's depletion region having a different width according to the local situation of the electrons in and near this volume. The depletion region is organized a little so it has a response as a whole. This degree of organization is similar to that of a bubble which is organized into a curved film around its contents. When the electrons are temporarly dense the depletion region will be narrow and some of the electrons will leak accross the depletion region and deliver a unit of current to the anode (the multiple anode side of the array is negative). The forward voltage represents the resting size of the depletion region and isn't important.

I considered carbon nanotubes for a while but they vary in conductivity according to their helicial pitch so I went for the uniformaty of buckyballs and will accept the 1.2 nm width of the monolayer of buckyballs which implies using ~ 1 nano gram of embedding plastic? / cm2.

Aloha,

Charles M. Brown
« Last Edit: July 14, 2005, 11:11:00 PM by Charlie Brown ARN »

kenbo0422

  • Jr. Member
  • **
  • Posts: 73
Charlie Brown,

The 9nm thickness you spoke of is realistically measureable with the testing equipment that I manufacture.  We have come up with a way to even scan the surface of chips, etc. with a micronewton force on a diamond tip giving a 3d picture of the surface as a test for accuracy in manufacturing.  Many of the chip manufacturers use these machines to keep getting smaller and smaller tolerances and higher densities in chips.  I'm sure that your diode array should be able to be manufactured with great quality and high performance with what knowledge I have gathered from that part of the business.  I would really love to see these ideas come to fruition.


Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
Prototype the Diode Array which aggregates rectified Johnson Noise
« Reply #12 on: April 26, 2005, 05:59:36 AM »
By all means see which of the nanofabricators you know or can know are willing to roughly evenly disperse C60 as a monolayer spaced ~30 nm in offset rows like trees in a monoculture orchard in a plastic monomer which would be ~1.2 nm thick, the diameter of the C60 on a 100 micrometer? metal substrate (say CrAu) (this thickness is needed if current has to move laterally and excessive if the prime diode arrays can be stacked on top of each other alternated with a nano pillar structure that lets heat bearing fluid through laterally while conducting electricity transversely (through the pillars). The lateral current path can be minimised if the diode array active area is a stripe and the lateral current moves crosswise, a short distance IN PRODUCTION UNITS (in a prototype the current has to be able to go to pins bypassing any possible failed section getting in the way of a high power design). N type InSb would be deposited on top of the C60 / set plastic embedding layer. The process is discribed elsewhere on this site: (www.overunity dotcom The international free energy research forum / Discussion board help and admin topics / Links to other energy research sites / www.diodearray dotcom for the diode array / Reply 5). I neglected to bring up the need of a short current path configuration (crossing a stripe) there though.

Aloha, Charlie
« Last Edit: April 26, 2005, 06:15:37 AM by Charlie Brown ARN »

kenbo0422

  • Jr. Member
  • **
  • Posts: 73
Charlie Brown,

I believe the places you are needing for fabrication are called forges or foundries, like in the old days but with a new purpose.  They specialize in one of a kind setups.  A guy from IBM (former employee) told me that is what you need to find.  IBM did their own stuff and I think its done by someone else now.  The guy who invented the inkjet technology lives in Louisville, KY.  Bill (Mr. inkjet to the Japanese) used to work for IBM as well.  I believe the micro switching and pumping foils were developed at a local foundry.  Give Google a search....


Charlie Brown ARN

  • Jr. Member
  • **
  • Posts: 66
{Hypothetically} A small radio, PDA, or light could be powered by a diode array containing 100 million diodes in 100 sub assemblies in series each containing 1 million diodes in parallel. It would operate
at 5 volts, produce 100 milliwatts, and feed 20 milliamperes through a 250 ohm load. The array area of the chip would be 1 / 1000 cm 2, ~0.32 mm square. Voltage stabilizing circuitry may be added. The noise voltage of a 250 ohm load producing a full 4 nanowatts of noisepower, noise up to ~1 Thz (easy for the diodes, hard for the load), is 1 millivolt A.C..

Aloha, Charlie
« Last Edit: July 14, 2005, 11:12:41 PM by Charlie Brown ARN »