Pulsed Fusor #3: Case Study of a 1HP Fusor

For posts specifically relating to fusor design, construction, and operation.
Post Reply

Pulsed Fusor #3: Case Study of a 1HP Fusor

Post by guest » Sun Dec 30, 2001 7:46 pm

Read this first>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Warning this is an attempt at a medium power fusor.
Standard shielding will not work at these levels. ( steel vessel)
I would hate to lose friends to lethal doses of x-rays
and neutrons.
Pulse systems are not toys.
A dug out or pit will be needed during even early tests.
If you come up with the right combination rad levels would cause radiation poisoning if conducted in your garage.
You can still use your garage as a instrument shack,
but a cinder block box lined with lead and paraphene
would work... out in the yard. Under would be best.

The whole fusor will have to be housed in a faraday cage. This includes power supply also. About the last
thing this group needs is to have a member destroy
a power grid by emi overload. Law suits galore! Not to mention the possible damages to properties and loss of life. This means battery powered fusor supplies also. A fusion test shot will kill tv reception, burn up comos ic's, kill anything with intelligence in it, cars ,microwave ovens, wireless phones, pace makers and biomedical implants. It needs proper shielding.


This will be my first practical design.
To start with the Nucleonics of the system.
This system will be designed around the catalysed
D-D system (burning up formed tritium.)
My back ground includes engineering design of nuclear items ... breeding blanket... system cooling...shielding....safety ect.
This first design will be bare with out a blanket for boosting output..... if the system is any good at all it should work bare.
The output will be neutrons and heat (xrays).
The design goal is one horsepower or 746 watts (thermal not electrical)
This is a design and a journal of how I came to do it this way. I have not implemented this design yet but I've completed the theory and practical design considerations. Pull up a cool one this takes time.

Input Joules........Cat D-D Reaction.......Output Mev

3.8E-14..............D + D = He + N.................3.3
3.8E-14..............D + D = T + P....................4.0
7.2E-14..............D + T = He + N................17.6
Mev to Joules (2.49E7ev)(1.60E-19 joule/ev)=
3.98E-12 joules
To find the energy released re = output energy - input energy

39.8E-13 joules-1.48E-13 joules = 38.32E-13 Joules

re=38.32E-13 joules per each catalysed fusion

Next multiply 746 watts by 1 second to get 746 joules.

Then divide 746 joules by the 38.32E-13 joules to see how many fusion reactions that is.

When I do that I get 1.946E14 reactions.

Since each fusion reaction makes 2 neutrons then
the 1.94E14 reactions must be doubled to get 3.88E14 neutrons in total.

That ends the nucleonics part.

Now the fun really begins ... The Energy balance

It all begins with my favorite all time thing...... n=i^2.

But at this level of production it's n=i^3 (third power scaling)

so we plug in the # of neutrons to get the # of amps we need.

3.88e14 =i^3

i = 6.52E4 Amps

To find watts multiply by the voltage say 10 kev and the 6.52E4 amps to get wattage required.

It comes out to 6.52E8 watts.

Not enough to run a fluxcapacitor for a time machine but way over 746 watts!

What to do?

Go to Walmart!
Look at the power saving light bulbs 60 watts at
14 watts usage. Some slight of hand ? No ....good physics. It is called Peak power averaging. The flourescent bulb would consume 60 watts of power
if fed continously but if you feed it instead pulses at a rate of 100 times a second at a pulse width of say a millisecond... The input power requirements fall amazingly. A flourescent tube is really a great model for a fusor because both are discharge tubes. It seems the same power averaging trick could be applied to the fusor. Lets go to the math and see.

With 10 kev voltage for example at one horse power level

in the millsecond pulse width
(6.52E8 watts )(1E-3 seconds) = 6.52E5 joules

In the microsecond pulse width
(6.52E8 watts)(1E-6 seconds) = 6.52E2 joules

In the nanosecond pulse width
(6.52E8 watts)(1E-9 seconds) = 6.52E-1 joules

But don't forget we pulse 100 times so

1. (6.52E5 joules)(1E2) = 6.52E7 joules>>746 Joules
This rules out milliseconds

2. (6.52E2 joules)(1E2) = 6.52E4 Joules>>746 Joules
This rules out microseconds

3. (6.52E-1 Joules)(1e2)= 6.52E1 Joules <746 Joules
This selects nanoseconds

To make 746 joules of fusion takes a total of
65.2 joules in pulses of .652 joule at 100hz and pulse width of one nanosecond.

To make 746 Joules of fusion takes a total of 260.8
joules in pulses of 2.6 joules at 100 hz and a pulse width of 4 nanoseconds

To make 746 Joules of fusion takes a total of 465.4
joules in pulses of 4.65 joules at 100 hz and a pulse width of 7 nanoseconds
( to fit the EG&G HY3202 Thyratron)

I'm a scot on my mothers side so that 65 watts
will be born by a blanket so you get a true 746 watts.
(boric acid in water to minimise Tritium production)

This might be something!
I need help on the electronic part....will accept all comers.
I expect that these machine parameters would make it an ideal arena for the hydrogen thyratron .
Tubes Rule!

Larry Leins
Physics Teacher

User avatar
Richard Hull
Posts: 12295
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Pulsed Fusor #3: Case Study of a 1HP Fusor

Post by Richard Hull » Tue Jan 01, 2002 7:50 am

Good luck Larry.

Getting those deuterons to answer those idealized calcs is going to be a real chore.

The problem comes in getting the bulk of the input energy transferred to make real fusion events. The observed efficiency of conversion to fusions is on the order of .0000000001%.

So the 28+mev per fusion is great but if you have to make 200 billion high speed deuterons to get a single fusion your input energy is mostly going where it always goes.....Electrons smashing into the walls, #1 loser. Deuteron-Neutral collisions, #2 loser. neutral-neutral and wall-neutral collsions #3 loser.

I don't want to rain on any parades, but the doing is tougher than th' calculatin'. Calculating is pure and unsullied. The seething interior of a high pressure fusor is a study in how many species are there and where are they all going. The fusion in a fusor is non-maxwellian, but every thing that isn't fusion is! (especially the neutrals)

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.


Re: Pulsed Fusor #3: Case Study of a 1HP Fusor

Post by guest » Tue Jan 01, 2002 5:45 pm

I know how difficult it will be .
I know these are just numbers.
But it will not deter me from trying.
I will keep in mind all you have brought out.
The losses will be substantial.
I'm gambling that that the electron losses will be minimised by reducing voltage.
If it is not so... so be it.
Also the voltage at this level should reduce on a per deuteron basis the energy losses due to near misses and neutral interaction. .
Even though the crossection is greater at high voltage
, the losses overide this advantage.

I've even considered ion pumping for neutral removal.

But like everything else in life it must be built first and tried. I have no illusions.

I appreciate your input after all I'm still a newbie.
If you have thin skin take up latch hook rugs not fusion.
I will be in the trenches soon.

I would be happy to get even close to the calcs.

Larry Leins
Physics Teacher


experimentaion...Re: Pulsed Fusor #3...

Post by guest » Wed Jan 02, 2002 12:00 am

From L. Liens and R. Hull parameter data:

4E14 n for 750J pulse x 0.0000000001% efficiency
= 4E2 n/ pulse
Assuming 20ns pulse,
4E2/2E-10 = 2E12 n sec-1

Capacitor design:
Energy stored in a capacitor: U = 1/2 * C * V2
so for a 0.3x1x.0003 meter 2 plate mylar capacitor (~.03mfd) charged to 10kV
U=1.5J and for same capacitor at 22.4 kV~75 J (1/10 required by L. Leins).

Typical rise time for flat capacitor of these dimensions is 5-10 ns rise time with pulse duration 10-30ns. This is possible since the impedance of the capacitor is << 0.1 ohm.

Coupling capacitor bank to device has been an issue. 0.1~1 J discharges into chamber with low impedance coaxial electrode configuration have been accomplished. However, with spherical inner electrode, as posted previously, discharge evades the sphere and only occurs at the point where the standoff meets the sphere (this is the point where the impedance of the transmission line begins to increase). Under steady state conditions, an even glow discharge is noted around the inner grid.

Thus conclusion is that inductance of the spherical grid is nontrivial, and impedance reduced configurations are where my work now centers.


Posts: 2346
Joined: Sat Jun 30, 2001 5:13 am
Real name:

Re: Pulsed Fusor #3: Case Study of a 1HP Fusor

Post by DaveC » Wed Jan 02, 2002 8:11 am

Larry -

Was just mulling over your calcs... and several things occurred to me
(1) For these rather short duration pulses, presumably the rise time will be under 10% of the duration(??) If so, then impedance matching the fusor to your pulse line will be critical to prevent the energy from scattering back to the source.

(2) Also, for < 10 ns wide pulses, I would expect ion propagation times to figure in, since they will move about 100 times slower than electrons, which are not all that fast either, at 10 to 20 keV.

(3) The effect of a cloud of slow moving ions, is to reduce the net field at the ion gun exit aperture... effectively raising its impedance - a variant of space charge effects - and reducing the energy coupled into the system.

(4) It seems that accelerating voltage and ion density (pressure) are closely coupled. At higher ion densities the MFP of the ion is reduced meaning lower energy between collisions, requiring higher accelerating potential to obtain similar fusion probability. But, since the fusor normally would operate to the left of the Paschen minimum, increasing pressure means reduced breakdown voltage.

Not sure any of this is fatal... but some of these factors figure in significant ways.

Dave Cooper


Re: Pulsed Fusor #3: Case Study of a 1HP Fusor

Post by guest » Wed Jan 02, 2002 5:39 pm

I know isn't it neat.
The impeadence issue is a serious one for sure.
The plasma focus folks talked of using copper sulphate in water to help adjust the impedence by adjusting the column of copper solution.
They used 18 coaxial cables of RG 58 Precisely cut to the same length coupled to a brass O ring.

The slow moving ion part is dealt with under "snowplow effect" equations .

PF=Plasma Focus.

In fact all of Mr Cooper's objections come up under PF
,Dr Mather spent twenty years dealing with them.

Here's a book that details forein experiments in PF.
All under UAEI (1) funding I might add.
I forget.
(1)***People under forty might have heard of UNESCO.
The unit of the UN that allowed nuclear prolifertion.
It was one of Ronald Reagan's favorite units.**

Small Scale Plasma Experiments- World Scientific Inc

I need to go back to the library to get the isbn #

To answer Mr Megley's question yes that design was a stab in the dark .

Was I too conservative ?...certainly.

Oh yeah about the sphere... You must glow before you discharge. Uof I showed this worked best.
The discharge of current will take the easiest path always.
If you don't provide a nice path through the discharge, it will climb down your insulator to the "grounding" of the sphere. (without a discharge path the inner sphere
will fill to it's capacity then block off all attempts at pulse.)

About the high yields in my case studies
I used numbers from plasma focus.
That machine has numbers in the 10^14
neutron region.

People all over the world have replicated those numbers.
Places like Mayalsia , Saudi Arabia, Pakistan and Iran.
My best advanced nuke teacher was Iranian.
His name was Dr Hamadi.
He worked on lightning research at Arkansas Tech.
He's the one that gave me the info on plasma focus.
It is a hot issue in Iran.
They have built a 10 MJ facility.
It only maintains a plasma for only five to 10 nanoseconds! It generates all the neutrons during that brief time.
It is an example of how brief a pulse can be and still make something happen.

Rain makes the flowers grow.
(The observed efficiency of conversion to fusions is on the order of .0000000001%. Richard Hull )
The efficiency when I did this calculation by I^3
scaling is 746 watts/ 6.52E8 watts.=1.144 x 10^-6/100=
0.000000001144% when run continously.

Dear Mr Hull:

I agonized about bringing this stuff up for 2 years.
It is not fun being a bomb thrower.
I still intend to build a continous fusor.
I find it a critical step in doing this other stuff.
I'm about half done on the vacuum side.
Got my PMT Tube.
Just closed the deal on a bell jar.
I must acknowlege the work of Dr. Miley and his team.
I'm a New Frontiers kid.
Space travel has fascinated me since 1962.
( I remember when Kennedy was shot and watched the launching of John Glenn into orbit on black and white tv.)
When I found a book that had both space travel and fusion, I went bananas. (When I was a junior at Ole Miss in Spring 1996)
In that book " Fusion Energy in Space Propulsion", chapter 7 was titled "Inovative Technology For an Inertial Electrostatic Confinement Fusion Propulsion
Unit" By GH Miley and company. ISBN #1-56347-184-1
It took me about a year to reseach and verify this stuff of pulsed fusion.

Then I discovered the fusor forum in 1998.

I did not take this course lightly or without consideration of how the group would take it.

Larry Leins
Physics Teacher

Post Reply