If you have a question about this topic, the answer is probably in here!
Post Reply
User avatar
Richard Hull
Site Admin
Posts: 10900
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull


Post by Richard Hull » Tue Nov 19, 2002 7:17 pm

It is very important that newbies and even veterans read this rather long FAQ as many needless questions can be avoided. I have tried to cover most bases. It is highly recommended that you print this out for reading and reference. RH


Neutron counter - any full and complete instrument designed to detect neutrons and present a neutron count or neutron absorbed doseage related measurment when placed in a neutron flux.

Neutron flux - A specific number of neutrons passing through a specific area in unit time. i.e. neutrons per square centimeter per second (standard of the industry)

Neutron count rate - the number of neutrons detected by a counting device in a given period of time.

Absolute neutron count rate - the actual number of neutrons issuing from or passing through a given point or area in a given time. (corrected for inefficiency of a counter, range, solid angle, etc.) The ultimate knowledge of God in the matter of neutrons. We are lucky to know to within +/-10%, the real absolute count.

Absolute,Isotropic neutron emission rate - The total or absolute number of neutrons emitted from a more or less idealized point source of neutrons per unit time. The ultimate quality factor number for the amateur fusorite. i.e. the irradiance or brightness of a neutron source.

Fast neutrons - Neutrons with an energy exceeding 50 kev. The neutron found in the fusion reaction with deuterium in the fusor has an energy of 2.45 mev.

Thermal neutrons - Neutrons with an energy at or below .02ev which puts it in thermal equilibrium with air molecules at room temperature.

Epithermal neutrons - neutrons with energies between 1ev and about 1kev, though there is no hard line drawn. Most important in capture cross sectional studies and neutron resonances within atoms.

Neutron scattering - Neutrons are scattered in most materials of high cross section through the action of multiple short range impacts with atoms of the moderator. After penetrating enough thickness of any moderating material, the neutrons are so scattered that the direction of their original source can no longer be determined as they appear to be coming from everywhare. This is always the case with thermalized neutrons within a moderator.

BF3 tube/counter - a THERMAL neutron detection tube or counter based on the boron tri-floride high neutron cross section reaction (B10 + n = Li7 + a) for detection. Normal efficiencies for these tubes are in the 1-5% range for un-enriched tubes where full thermalization is not possible. With fully enriched B10 to near the 96% level and at atmosphere in larger tubes nearly 90% of all true thermal neutrons (<.025ev) entering the counter will be counted. All thermal neutron detectors MUST be surronded by a THERMALIZING MODERATOR if used to count FAST neutrons. In such situations scattering of neutrons at or near thermal energies can further reduce the net efficiency of the BF3 FAST counter (NOT THE TUBE)! To thermalize a given surface flux of FAST neutrons they are scattered and the net effect is to reduce the thermalized flux intercepted by the tube. It is a statistical thing and tables are available from most manufacturers of moderated SLOW neutron detectors that will let one back interpolate to the fast flux.

Enriched BF3 tube - Normal boron compounds are a mixture of B10 (20%) and B11 (80%) . An enriched counter is one containing Mostly B10 atoms in the boron. These B10 atoms are the only ones which can interact with neutrons for the idealized counter reaction. Istopically enriched boron is very expensive and difficult to make. Such tubes are much higher in efficiency and and price.

He3 counter - A THERMAL neutron detector tube similar to the BF3 tube, but a bit more efficient. (~5-10%) Also more costly due to the isotopically pure helium 3 fill gas. It works off the reaction (He3 + n = H3 + P).

Fission counter - A FAST neutron counter based on The U-238 + n reaction inducing a fission of the U238 contained in the detector tube. A rare tube indeed and very inefficient. Normally used in very high flux measurements.

Boron counter tube - A tube containing a boron coating on the inside wall of the tube to detect thermal neutrons. This tube uses the same reaction as the BF3 tube above. Rare today as the Boron is usually isotopically enriched in these tubes.

Solid angle - the effective angle subtended between a point source of neutrons at the center of an imaginary sphere and the detector's effective cross sectional area at the surface of this same imaginary sphere of radius "r" (the distance from the center of the source to the center of the detector.)

Neutron cross-section - a multiple use term with many specific subsets. Basically this refers to the specific ability of a material to intercept neutrons. low cross sectional materials will allow neutrons to pass through large volumes with little interaction. High cross section materials tend to involve their nuclei with neutron collisions at much more frequent intervals per unit distance traveled within them. It is basically a matter of the density of the material (how packed are the atoms) and whether the particular atom "looks big" to a neutron or vice-versa.
Specific cross sections can be defined... such as "thermal neutron cross section". Most substances have their highest cross sections in the thermal neutron range. Here, it is the neutron that looks big to the atom. Cross sections are measured in Barns (1 Barn = 10e-24 sq cm). This accepted scientific term is literally derived from the term "big as the side of a BARN". The study of neutron cross sections is a separate science in itself.

Capture cross section - a series of resonance points at specific neutron energies or narrow bands of energy for specific atoms where neutron capture is virtually assured. Some resonances in some atoms can approach 100,000 barns!! Cadmium and Gadolinium are prime examples. Few such resonances appear over 2kev. Most are in the EPITHERMAL and THERMAL energy range. Another complication that makes cross sectional study a separate science.

Proton recoil - When fast neutrons fly into hydrogen atoms, the hydrogen nuclei, (protons), recoil after absorbing some of the kinetic energy from the impacting neutrons.

Moderation/moderator - neutrons are slowed down and their kinetic energy is reduced by collisions in bulk matter. This process is called MODERATION.
Some materials are better moderators than others and this is based on the material's specific neutron cross section. a MODERATOR is normally chosen that has a high cross section. Hydrogenous materials are the norm here. Water, plastics, oils, paraffin, etc. All such moderation in hydrogenous materials is by proton recoil.

Neutron scintillator - A FAST neutron detection system based on neutron induced proton recoil in a hydrogenous material, (i.e. plastic), creating a light pulse when the recoiling proton strikes an embedded phosphor material such as ZnS:Ag. The Bicron BC-720 neutron scintillator is a prime example.


Neutron counting is, at best, difficult and imprecise. At worst it is a sort of black art with a lot of its input conditions merely guessed at and data pulled out of the muck of electronic noise being massaged into whole form using statistics. (Witness, the recent multiple finger pointing and pot lecturing the kettle in the "sonofusion" furor of 2001-2002 and even back to the Pons-Fleishman cold fusion episode of the early 90's.) No matter who measures low levels of neutrons, there seems to always be points of contention. It is the weak part of any discovery argument and the first one picked at as it heralds true nuclear events taking place.

Why is neutron measurement so tough?

We are attempting to electically count and indicate the passage of particles which have no electric charge! What could be more stupid or daunting a challenge than that?

The particles we are counting look at bulk matter as if it is transparent, for the most part, and one of our neutron particles traveling at near the speed of light can easily whistle through many yards of concrete or stone.

Thus, we are faced with a poser... How to take something that can penetrate most anything, stop it or transfer enough of its energy to a charged particle so that we can detect it readily in a small volume like a geiger tube used for charged particles? Given that we have such a process in mind, and the demand for a smaller, more managable volume of detection, we are doomed to the de facto acceptance of low counting efficiency. Given all this....In small neutron flux fields, counting will be so far down into the noise that the entire process becomes a statistical game of magic wand mathematics with, at best, wide standard deviations for absolute measurments. (Which is, ideally, what we seek)

Morphing Neutrons to Charged Stuff....................

All neutron counting is, at its finest reduction, a secondary process and at its worst embodiment, a tertiary process. All normally considered processes in the real world of today involve getting a neutron to do something to neutral matter which will ultimately spit out something charged that can be used for detection by our coulombically driven instrumentation.

The two most popular processes are

1. charged particle recoil by neutron impact
2. charged particle ejection from a nucleus of a material by nuclear transformation, or fission.

Case #1 is the ubiquitous proton recoil reaction whereby FAST neutrons slam into material of low Z and high cross section. The best material being a dense material containing a lot of hydrogen. (water, plastic, paraffin, oil)

The FAST neutron impacts or glances the hydrogen nucleus imparting a fraction of its energy to the now freed proton which spins off into the material to give up its energy after a very short flight (fraction of a millimeter) in bulk matter or a few centimeters in air.

Multi-atmosphere, pressurized, methane filled counter tubes can be used similar to a geiger tube where the electronic charge of multiple ionizations due to the high speed recoiling proton stoping in the gas is collected on a highly charged central wire. Such tubes are virtually unused today as their efficiency is very low. (<01%)

The Bicron BC-720 FAST neutron scintillator is a bit more efficient, much smaller in volume and easier to deal with in modern FAST neutron counting efforts. It turns a fast neutron into a recoiling proton which hits a phosphor whose excited elecron orbital falls back to a lower energy releasing a photon in the blue end of the spectrum. This photon now flys into a photocathode of a photomultiplier tube which turn the photon into an electron or two. These are accelerated towards a succession electron emissive dynodes. At each impact the electrons release more electrons until electron multiplication is so great that a huge shower of electrons winds up hitting the anode of the tube creating a large, amplified current pulse heralding the passge of the original neutron through a horrendous sequence of events and multiple transformations.

Lets wrap it up..... neutrons to recoiling proton to excited phosphor atom to photon to electrons to shower of electrons to current pulse. Whew!

This is the most realizable neutron counter to the underfunded do-it-yourselfer in the fusion effort.

** NOTE*** There is another method of neutron detection by proton recoil that is strictly qualitative in nature. This is the wilson type cloud chamber or a continuous action diffusion type cloud chamber. Here, the supersaturated air in the chamber along with the alcohol vapor sees the hydrogen within yield the proton recoils when neutrons traverse the chamber. Again, the efficiency is low and detection of neutrons is heralded by the production heavy broad tracks due to recoiling protons.

Amateurs can and have made simple cloud chambers so this is another proton recoil option.

Case #2 above is the bulk of other neutron detection schemes from the sublime to the ridiculous.

The first subgroup is a good one "tranformation". In most all tranformation devices, only slow neutrons will work or be detected as this involves neutron capture. Neutron capture almost always leaves an atom radioactive or excited. In the idealized transformation or capture neutron detectors, the transformed radioactive nucleus needs to have an ultra short half life. (sub microsecond) upon decay, we now detect the charged particle spit out of the decaying nucleus.

The BF3 and He3 detectors are prime examples of transformation THERMAL neutron counters.

Such counters can have true THERMAL neutron counting efficiencies of up to 90%!!! (multi-atmosphere, isotopically pure versions).

In all these counters we are detecting electronic charge created by alpha particles or protons ionizing the internal gas as they plow to a stop following the decay reaction.

So.....it is a neutron melding into a new unstable atom to ejected high speed decay particle to ionization electrons collected on a positively charged wire.

The other process, FISSION, mentioned above under case #2, is where we can detect neutrons by doing what is done in an A-bomb. Inside an argon filled tube with a highly charge central wire we can place a thin layer of U235 on a large surface area metal plate or on the walls of the counter. A THERMAL neutron hitting this deposit may fission one of the atoms and the 200+ mev nuclei ejected create the ultimate ionization event in the tube with the good old electronic charge, again, collected on the central wire.

The same counter can be made with a deposit or actual depleted uranium, (U-238), metal surfaces covering large areas within the tube. Here a very FAST neutron of at least 1 mev has a reasonable probability of fissioning a U-238 atom with the same huge release of energy in departing fission fragments and accompanying electronic charge collection due to ionization in the tube's gas.

Fission tubes are rarely used today as the fast fission tube is very low in efficiency and the slow neutron fission tube contains licensable amounts of U-235, a controlled fissile material.

There are other processes in neutron counting but are left for the eager dilettante to ferret out of the literature (see the books and reference forum).


For the purpose of this FAQ we will limit the detection schemes to scintillation counting (Proton recoil counters) and the gas proportional counters or transformation counters such as the BF3 and He3 counter detectors.

These are the ones that 100% of the successful, neutron club, amateurs on this list have used thus far.

It is not within the scope of this FAQ to discuss contruction of counters. It is assumed you are in possession of one of the above counter systems that is in good working order.

FIRST - calibrate your instrument yourself with a known neutron source or have your counter calibrated by a professional NIST traceable establishment. Ludlum has been found to do a good job at a very reasonable price.

Taking background counts.............

It is vital that you take at least an hour long background count just before making a data run with your fusor or any neutron source! Likewise, once you have recorded your fusor data, you should again take a full hour long background count. Sum the total of the two background counts (before and after) and divide this total by two to establish a mean HOURLY back ground count. (Naturally divide this result by 60 to determine the CPM or count per minute rate) Next, subtract this from your data run CPM count to arrive at the actual counts returned due to the neutron source. This is important regardless of what type of counter you use.

Using a FAST neutron scintillator. (BC-720 or its

The background count in these scintillation detection devices is high compared to the BF3 or He3 SLOW neutron detectors. This is a necessary consequence of cosmic rays creating pulses not only in the scintillator itself, but the photomultiplier tube as well. Depending on the scintillator discriminator setting, 1-3 CPM is not too great a background. As always, a strong gamma radiation source should be placed as close to the scintillator head as possible and the discriminator set to the point where the unit will not count any higher with the source in place than with it in another room. This is very important. By doing this you allow the background count to be very low due to a properly set discriminator. NORM or standard radioactive particles in your environment will not be counted. I have setup my BC-720 system to where it produces a normal background count of about .5 cpm.

With repeatable background readings, and a good digital counter to log the pulses from your counter, you may now take timed count readings of a running fusor. As most fusors are in the 6-8" diameter range and most Scintillators have just a sheet of metal infront of the PMT housing. I would suggest placing your counter so that the center of the scintillator crystal is about 15cm from the center of the poissor. This will normally mean about 1-3 inches from the end of the PMT housing to the outer shell body of the fusor.

Now for the data collection procedure...................

You must know and record the precise distance from the center of the crystal to the center of the poissor as it is a major factor in the total isotropic emission rate calculation later. Errors in this measurement will alter your count by the SQUARE of your error in measurment! Be real tight assed here about precision. Record and log this value as radius of solid angle sphere.

Radius of solid angle sphere = 15 cm.

Your Scintillator has a surface area. For most, this is a 2" diameter PMT and crystal or just 3.1416 sq inches which is 2.54^2 X Pi or 20.2 sq cm ( I like metric). Log this value

Area of detector = 20.2 sq cm.

Bicron supplies a chart which plots efficiency of the detector versus energy of the neutron to be detected.

For our fusion neutrons (2.45mev), this plotted figure is 0.6% or it is the same as saying that for every 167 fusion neutrons passing through the detector we will expect only one single count to be recorded. Record and log this 167 number as the efficiency multiplier.

Efficiency multiplier = 167

You have now taken your data, it is assumed, and subtracted the background and have a neutron count cpm value in hand. Log and record this as corrected instrument count. Use CPM

Corrected instrument count = 400 cpm (typical of a low level run)

You now have all the required data in hand to run the numbers on your fusor.

Doin' th' math........................

1. Figure the surface area of the solid angle sphere

4 x pi x r squared

4 x 3.1416 x 15 ^2 = 2844 sq cm

2. What fraction of the area of the sphere is the detector area? We need to generate a solid angle multiplier factor.

area of solid angle sphere / area of detector

2844/20.2 = 140.8 Thus, whatever count we really come up with in CPS in our detector we must multiply this to find the total point source emission. Log this as solid angle multiplier factor.

solid angle multiplier factor = 140.8

3. How many actual fusion neutrons each second went through our detector?

(corrected instrument count / 60) X detector efficiency multiplier

(400 / 60) x 167 = 1,113 neutrons/second

4. Finally, what was the total number of neutrons emitted per second by the fusor?

total neutron/sec count in the detector X soild angle multiplier factor

1,113 X 140.8 = 156,757 neutrons per second

You are now done if you have the BC-720 type scintillator detector.

It is obvious, and a mathematical trifle, that if you fix the detector forever at this 15cm distance and forever use the same detector, all the above mathematical gyrations need never be done again. You can collect terms and constants to boil the above down to where there is only one unknown and you would only need to record a total count minus background in cpm. It would look like this....


total corrected count in cpm x 392 = total isotropic emission in neutrons/sec.


Pretty slick, huh?

Likewise a million neuts/sec would demand a total corrected count of....

1,000,000/392 = 2551 cpm

Thus endth the lesson for this detector...................

Using a moderated BF3 or He3 Slow neutron detector

Here is where it can get sticky.. So very much depends on calibration and manufacturers charts being in hand to really make these counter do the fine job that they are readily capable of.

The advantages of these detectors is ultra low background count. Often so low that it need never be a factor in final calculation on a fusor churning out over 50k neuts/sec.

These counters are a minimum of ten times more efficient than the Scintillators. However, they demand more loose interpolation based on assumptions and presumptions not found in the scintillator systems, often making the neutron count look a little fluffier than it really is.

In virtually all factory assembled, moderated, BF3 or He3 FAST neutron systems, the actual data given in mr/hr is back figurable (interpolated) to a FLUX. This flux is assumed to exist at the center of the detector. Most of the neutron dose rate meters are working the flux off of an RBE curve that is hopefully, but not quite linearized for neutron energies between near thermal to about 3mev.

In general...(your gotta' get your curves) this reduces to a flux of 7 n/cm sq/sec /mr/hr. WHEW! So, if you are reading 2.5mr/hr the flux in every square cm of the solid angle sphere's surface whose radius is that distance between the center of the fusor poissor and the center of your remball, or moderator is 17.5 n/sec/sq cm. This may sound ridiculously simple to neutron metrologists, but many amateur neutron heads have trouble following the thread. Nonetheless, all must understand and have a working concept of what is really said in the foregoing.

Carrying the logic forward.....A solid angle radius of 25 cm (remball very close to the fusor shell) would yield a total area of about 7800 sq cm so we would multiply 7800 X 17.5 to get 136,500 n/sec total isotropic emission. Very,very simple calc right? But, the big assumption was the interpolation based on the broad swath RBE curve for the remball with a specific area detector of a specific type set to factory spec. (Got Spec?)

For those BF3 and He3 detectors that have only CPM readouts, The norm, (Manufacturer special spec) can be from 20-60 cpm /mr/hr conversion factor, cpm to mr/hr. This can now be back referenced to the stock RBE curves. (Got Curves?) Here you are double interpolating! More room for mis-steps. This is why it is very important to have these instruments in perfect calibration for the curves to mean anything. This was not the case for the scintillators.

Manufacturers do not all use the same geometry for their moderators, but all attempt to mime the standard PE moderated RBE curves for the 8" bonner sphere (sort of an older de facto standard).

When using these units, try your best to obtain every scrap of manufacturer info and data on your device. Have the device checked and calibrated before any serious use.

As noted above, the mathematics reduction to absolute neutron numbers in a fusor for these factory made devices is really quite simple as these units are, for the most part, working off the idea of neutron FLUX measurement.


Beware of noise. It is a killer. A noise pulse can't be differentiated in simple systems from a real neutron pulse.

Attempts to rid oneself of noise pulses in the data are rarely effective. The proper method is to never let the noise into the data.

Particularly afflicted here is any attempt to get meaningful neutron numbers from a pulsed fusor. Any such effort is doomed to being plagued by measurment problems. A proton recoil cloud chamber with camera timed to the pulse that was pre-calibrated in a known continuous neutron field would be ideal here.

Fusors with big filter caps on their DC supply are noisey when they are at the plateau of relaxation oscillation. Megawatt peak pulse energies can be radiated in the power wiring when this happens. As most electronic neutron detection schemes have amplified front ends that look for signals in the 2-10 millivolt range, it is easy to have pulses this big radiate into the front end due to any number of local noise sources, yielding false counts.

A good test is to run your fusor at full power, but with no deuterium and take neutron counts. Compare the results with no operation background counting. They should be the same to within a 5% differential or better!

Know your noise situation and shield accordingly. DO NOT let noise be your neutron count! The instruments aren't lieing to you. You are letting them lie to you. If you report noise numbers + neutron numbers, then you are lieing to others and kidding yourself.

Wrapping it up.......................

Neutron counting is tiring only because the preparation and setup required to make the data viable requires a bit of planning and fore-knowledge of the source, the instrument and the measurment environment. Due to the nature of the beast, neutrons are illusive and efficiencies of the detectors are also correspondingly low, especially as regards FAST neutron counting. Small fluxes associated with small fusion sources often throw the measurment data near the noise floor of the instrument. This condition is hampered even more by necessary assumptions needed to be made on behalf of the metrologist which, if in error, will often be compounded in the final result.

This concludes a rather massive opus on neutron counting. Any suggestion for expansion or correction would be appreciated.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Post Reply