How We Do Fusion in a Fusor

A hypothetical situation and a discussion about "Breakeven"

Instrument Rack of Richard Hull's Fusor IV

Fusion is Easy!    Fusion Main   What It Takes  Ask a Question about the Fusor   Main Home Page

Introduction

"Extraordinary claims require extraordinary evidence," so the saying goes. I can't and don't expect you to believe me when I say that fusion can be done on your kitchen table with a Farnsworth Fusor. Therefore, I need to provide you with some sort of hard proof in order to convince you.

There are many ways to prove that fusion is taking place in a fusor. The easiest and most common method is to measure the neutron flux being emitted from the core during the reactions. This process, described below, is the generally accepted methodology used in almost every area of fusion research. For those of you doing cold fusion or sonofusion experiments, this is also one way of measuring any neutron emission that takes place.

I will take you through a hypothetical situation below involving a typical amateur fusor. Then we will discuss what all of these numbers mean.

The Neutron Counter

While many types of neutron counters exist, I chose to use an Eberline PNC-1. It is an older model, dating from the 1960's and 70's, but it received a calibration from Ludlum Measurements back in June of 2004, and should be accurate for at least 2 years. Ironically, the counter was within their calibration tolerance range when I sent it to them!

The PNC-1 consists of a tube filled with Boron Trifluoride (BF3), which is connected to a meter. When ever a neutron enters the tube, it induces a reaction in the BF3 fill gas, creating an electrical pulse, which then registers on the meter. However, the BF3 tube only counts thermal neutrons, which are neutrons that have an energy of 0.025 eV. They are in thermal equilibrium with the surrounding environment. Since the neutrons produced in fusion are fast neutrons with an energy of 2.45 MeV, we need to surround the BF3 tube with a material that slows the neutrons down until they reach 0.025 eV. This material is called a moderator, and any hydrogen-rich medium can be used as one. A typical moderator would be about 3-4 inches thick, and be made of water, wax, or plastic. The PNC-1 uses wax as a moderator.

How We Know We are Getting Fusion Neutrons

The following was written by Richard Hester on Fusor.net explaining how we know that the detected neutrons are indeed the result of fusion reactions:

"At ~10^6 reactions/sec, no one will be seeing any measurable temperature rise or pressure rise from the fusion reactions.

In response to the original post, there is no doubt that there are fast neutrons coming out of the fusor, though, as they have been measured 6 ways by different researchers.

They have been measured using BF3 and He3 proportional counters, which detect neutrons very well, but cannot differentiate between fast and thermal neutrons.

They have also been detected using a Bicron BC-720 scintillator, which readily responds to fast neutrons via signals generated by proton recoil, but cannot see slow neutrons well at all, except perhaps to an extremely small extent via the capture gamma ray that occurs when a proton absorbs a fast neutron that has been slowed almost to a stand still due to elastic collisions inside the detector. Since the scintillator is designed specifically to have only a low-level response to gamma radiation, this contribution will be extremely minor, especially since the discriminator in a detector system using this scintillator is deliberately set high to reject low level output pulses that can result from gamma excitation, PMT noise, etc.

Fast neutrons have been indirectly observed by one researcher back in 1999 using a diffusion cloud chamber via the recoil protons generated by collision with hydrogen nuclei in the alcohol vapor inside the chamber.

Fast neutrons have been detected using a plastic scintillator, which has a good detection efficiency for fast neutrons via recoil protons generated inside the scintillator by elastic collisions between fast neutrons and hydrogen nuclei in the scintillator host plastic.

Two researchers here have also observed unmistakable signs of neutron activation in Indium and Manganese by using a moderator between the operating fusor and the sample being irradiated. In each case, the activated element was identified via its characteristic gamma spectrum using a multichannel analyzer. Activation can also be used to distinguish whether low or high energy neutrons are emitted from the fusor, as activation for Indium and Manganese only occurs due to thermal neutrons. Removing the moderator between the fusor and test sample stops activation, proving that the neutrons coming out of the fusor are relatively high energy.

More recently, fast neutrons have been detected by two researchers here using a bubble detector, which only responds to fast neutron flux.

Researchers at the University of Wisconsin have also detected the fast protons from the alternate branch of the D-D reaction using a silicon PIPS detector. The observed energy correlates closely with the expected proton energy resulting from the D-D reaction.

So - we know for sure that neutrons are coming out of the fusor, and fast neutrons at that. They occur at applied potentials of as low as 10kV, ruling out such exotic reactions as stripping, which starts occurring at a threshold energy of several MeV. The reaction cross-section for D-D fusion has been extensively documented all the way down to 10kev and below, so we know that there is reaction down at those energies, even though the cross-section is small. No other neutron-producing D-D reaction of any sort has been documented for low incident energies. Fusion is being done, even though the reaction rate is miserably small compared to the input power."
 

Doing Fusion and Gathering Raw Data

"The first step in operating a fusor is to not operate the fusor." Before you turn anything on, you must first measure your neutron background with the fusor off. Turn on the neutron counter, let it stabilize for a few minutes, then do a gross count over a period of 15 minutes to an hour. To do this, you must hook up a digital counter to the pulse output of your neutron counter. You can see this in the picture above. Once you have gathered your data. Divide your total count by the number of minutes you counted for to get the rate in "counts per minute," or CPM.

Ex: (3 counts) / (15 minutes) = 0.2 CPM

The neutron background is typically very low, usually under 1 count per minute. This makes detecting fusion fairly easy, unless you are running at really low power levels.

The next step is to do a dry run with no fusion fuel in the system. This is usually unnecessary once you have a stable fusor system, but it is a good idea for your first few fusion runs. Turn the fusor on and adjust it to the Voltage, Current, and Pressure at which you will be operating. Start a similar gross count with your neutron counter as before. The neutron count should not increase at all, since no fusion is taking place. If your count rate increases significantly, you may have some fusion fuel left in there from a previous run, or your fusor may be producing electrical noise. When the latter occurs, you will know it, since the neutron meter will go off scale and sparks will be flying between pieces of equipment. Electrical noise is either non-existent or a real pain in the neck with fusors. It is almost never subtle and you will know when you are getting it.

Finally, you get to do fusion. Using whatever method that you have established, set your desired voltage and slowly admit deuterium to the fusor chamber until you reach the desired pressure. At this point, the neutron counter will start to click away almost immediately. It will be very obvious that the neutron background has increased. When you can get the Deuterium flow stabilized, start your stopwatch and do a gross neutron count for as long as possible. I try to go for 15 minutes, but overheating can become a problem at higher voltages. If at any point you experience electrical noise, reset the counter and start over. You only want to count neutrons. A typical count rate showing obvious fusion would be anywhere from 20 to over 2000 counts per minute, depending on your voltage and power level.

After you finish your fusion run and have turned the fusor off, it is time to do one more, post-run background count. This should be almost identical to the background count you did before the fusion run, and the count rate should be back to the original low levels.     

Data Reduction

Now it is time to take your raw data and convert it into a "Total Isotropic Emission Rate," that is, the number of neutrons your fusor emits every second.

Usually, you can tell that you got fusion just by looking at your data. If your average background count rate was 0.2 CPM and your count rate during the fusion run was 40 CPM, then it is pretty obvious that fusion reactions took place. The exception to this is when you run at really low, bleeding edge power levels, as you would find at maybe 15,000 volts. In this case, you will need to do many (5 or more) long fusion runs to show that the count rate is statistically higher when the fusor is on. Plus, I like to say at least a fivefold increase in the count rate shows fusion is taking place, even though my neutron counter has a 10% or better accuracy.

Anyway, back to the data reduction. First you determine your average background count rate. Take your total counts and divide by the total time to get your average neutron background in CPM. Then do the same thing for your fusion run data. Total counts divided by the total time gets you your average fusion count rate. Now, subtract the background count rate to get the number of counts per minute that were attributable to fusion.

Example:

Background rate = 0.2 CPM

Fusion Count Rate = 500.0 CPM

(500.0)-(0.2) = 499.8 CPM from fusion neutrons

Now, divide that number by 60 to get number of fusion neutrons counted per second:

(499.8)/60 = 8.33 CPS from fusion

Here is where things get a bit trickier. A BF3 counter is usually only 1% efficient when counting 2.45 MeV fusion neutrons. So now we need to multiply our (8.33 CPS) by 100 in order to get the number of neutrons entering the volume of the detector per second, since only 1 out of 100 of those neutrons will be counted.

(8.33)*(100) = 833 neutrons/detector volume/second 

Finally, since the neutrons are emitted isotropically in all directions, and since the detector is a certain distance "r" away from the fusion reaction zone, we can consider the detector to be a small section of area on a much larger sphere.

Fig 1. Demonstration of the above concept

Now we need to multiply by the ratio of the area of the total sphere (A_s) to the frontal area of the detector (A_d). This multiplier is known simply as (A_s)/(A_d). To find (A_s), we use the formula for finding the surface area of the sphere: A=4πr² where "r" is the distance from the axis of the detector to the center of the fusor's inner grid. Let's say that distance is equal to 20cm.

So:

(A_s)=4*π*(20)²

(A_s)=5027cm²

Looking at the data sheet for the PNC-1's BF3 tube, we find (A_d) to be equal to 15cm²

Therefore:

(A_s)/(A_d)=(5027)/(15)=335

So to finish it off, we multiply our 833 neutrons/detector/second by 335 to get our total isotropic emission rate, which is the number that tells us how much fusion we are doing.

(833)*(335)=279055 neutrons produced per second, +/- 10% (accuracy of calibration)

Discussion

Restated, that is 2.8x10⁵ neutrons produced per second. This is fairly typical of an amateur fusor running at maybe 25-27 kilovolts. This number is high enough to allow for simple neutron activation experiments with Indium and Silver.

Since we are doing D-D fusion and neutrons are produced in only 50% of the reactions, we can multiply the isotropic emission rate by 2 to get the total number of fusions per second.

The current amateur record is 5x10⁶ neutrons per second and the record for any fusor was 10¹² neutrons per second by Bob Hirsch (former DOE chairman) using Deuterium-Tritium fuel in the late 60's. *This is roughly 30 watts of fusion power*; a record not matched using any method of fusion until recently and very good when you consider that the total input power was 4,000 watts. That fusor was roughly a factor of 150 away from breaking even, assuming the energy could be extracted with 100% efficiency.

*Derived from: 10e12 (total fusions=the number of neutrons in a D-T reaction) * 1.7X10e7 (average eV/fusion) * 1.6X10e-19 (joules/eV) = 27.2 joules or 27.2 watt-second

****************************

**New note from Richard Hull**

" ...I know Bob Hirsch and he never claims more than 10 e10 neutrons per second from his best run. (This is consistent with real devices operated similarly with d-d versus d-t seeing no more than a 100 fold increase in reaction rates for the same energy input.) Gene Meeks, Hirsch's assistant, constantly claims the higher 10e12 figure. This higher figure is often seen in the literature.

If the principal investigator is to be believed this puts the power out backwards to ~.03 watts, but the input was the same...

These figures are from Bob in several interviews."

*********************************

What does this mean in terms of breaking even?

By the same train of logic discussed above, a fusor will "break even" at an emission rate of roughly 10¹⁴   to 10¹⁶ neutrons per second (one assumes 1-10 kilowatts of input power). This is not a trivial number at all, and you can see that we have quite a ways to go. We need to amplify our current fusion outputs of 10⁶ by literally 100 million to 10 billion times! You can also see that the D-T reaction is much easier and more energetic than the D-D reaction that we are doing here.

Suffice to say, anybody who makes a fusor that can put out 10¹² neutrons per second will be dead very quickly unless he or she uses the appropriate shielding, which would be several feet of Boron-loaded concrete. Hirsch had his fusor in a "cave" made of this special concrete. He used a 50/50 Deuterium-Tritium mix at a pressure of about 10^-4 Torr, and had his master power supply set to 80,000 volts and 20 milliamps. This is only 1600 watts, but he also had four 600 watt ion guns injecting these gases into the fusor. Thus, his input power was 4000 watts, plus maybe another 1000 if you take into account the power usage of things like pumps and instrumentation.

This was the last fusor ever built by Philo T. Farnsworth's team. When he died of pneumonia, the fusor effort died with him, and the members of the group went their separate ways. Hirsch, who was a member of the team, went on to become the head of the Department of Energy, but could never get a fusor program off the ground. Only today is the concept starting to become resurrected.

It is somewhat interesting to note that this kind of fusion output was not matched until relatively recently, when the first big Tokamaks came online. To compare, MIT's Alcator reactor produces an output of 10¹⁴ neutrons over a 2-second pulse. This is 100 times higher than Hirsch's fusor, but you must also take into account the fact that a Tokamak uses literally megawatts of input power to produce fusion, and it cannot be operated continuously without overheating.

In terms of fusion power out per unit power in, it is quite possible that the fusor has never been matched. It certainly has not been matched in terms of "fusions per dollar." However, it is unlikely that it will ever be able to ignite and self-sustain on its own, due to the nature of its design. An electric field must always be present in a fusor, or the reactions will cease. Furthermore, half of the input energy is lost in the form of electrons being accelerated away from the inner grid. Hirsch used some novel design techniques to get around this, though.

With D-T fusion, extracting the charged Alpha particles may be away around the self-sustaining problem. Their charge can be converted directly into electricity with ~95% efficiency. If you can put that energy back into the fusor, then you can run it off its own internal power. All this requires an even greater amount of fusion output, since the alphas only carry away a small fraction of the energy. 

Simply stated, a fusor makes a hell of a cheap neutron source, but it is unlikely to break even in the near future unless some serious research is conducted. It may happen someday, but for now, amateur scientists can just have fun and play around with it. Everybody is capable of making a contribution.   
 

Fusion is Easy!    Fusion Main   What It Takes   Main Home Page