pulsed fusor power issues

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

pulsed fusor power issues

Post by Richard Hull »

Pulsed power fusors have to deal with the terrible and rather unpredictable non-linear plasma impedance monster.

When we operate a fusor at say 25kv and 8ma to get neutrons, we have nursed and coaxed it to this level. Using ohms law, the fusor appears to have an impedance of 3.125 megohms and at that QUIESCENT point, it does. If the voltage is imediately lifted by only 1 kv to 26kv, the device wants to draw about 100ma for a Z of 260k ohms. If we now instantly lift to 30kv about 1 amp will be drawn for a Z of 26k ohms. the process folds over increadibly rapidly at the fixed pressure of the original 25kv stable 8ma operation to 100's of amps then 1000's of amps and so on. (Townsend discharge finally folding over to an arc discharge.) This point is reached much sooner than most on this discussion group think. So, the fusor goes from a high resistance device with stable operation to a device with an impedance of a fraction of an ohm in only about 20kv lift from the original voltage. It really winds up, at some point in applied voltage and internal pressure, as a gas regulator tube. As in all regulators with no ballasting, the tube would be instantly destroyed or damaged as it tumbles from townsend discharge into arc discharge.

The upshot of the above is that if you plan to lift a fusor from 20kv to say 60kv you might need a pulsed supply impedance under 1 ohm!...... Not easy to do

Let us say that we have a pulse supply of 500 ohm output impedance capability. Let us further say that we have a stable plasma and light fusion established at 18kv (keep alive). Now if we plan to lift this system to 60kv in a pulse and the plasma impedance at this voltage is 40 ohms.
We will only lift the system to a maximum of 24-25kv for the impedance at or near the 60kv level is so low that 55kv or more is disappated in the pulse transformer. These are crude figures, but give the general implication of the oddball dynamic impedance changes in such systems.

This is why I militate for an extensive series of experiments by every would-be fusioneer with a demo fusor. Play with it. Massage its gas system. Fart with the pressure vs. voltage characteristics and you will see how truly different and non-intuitive the system is and how it can foldover into nasty modes real easily.

Naturally, reducing the pressure will change all of this, but reduction of the pressure by an order fo magnitude to allow this idea to become valid would mean at minimum an order of magnitude reduction in fusion.

Remember also that fusion scales rather directly with current, BUT, You still must meet certain voltage potential criteria ACROSS the fusor, gas load, AT THIS CURRENT, to warrant that you are on a good part of the cross section curve to do fusion.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
guest

Re: pulsed fusor power issues

Post by guest »

I agree totally with Richard .
The non linearities are what you want to exploit in a pulsed fusor.... without blowing up the grid!
The pulsed regime is a crap shoot without
continious fusor time and some experience of what does what and where. Only then will the nonlinear nature be revealed to you.
I'm testing the pulsed power system as we speak but it's going to be some time before an actual pulse shot
will be under taken.
My supply has an impedence of .25 ohms at pulse time by using the thyratron impedence drop ( yet more nonlinear effects) in paralell
with a 1 turn primary and 30 turn secondary with huge
insulation. I can dial down the impedence by simply reducing the output turns. Less high voltage but less impedence as well. By using fractional primary turns
I can reduce the number of secondary turns keep the voltage the same and still drive down the impedence.

The point of all this chicanery is to create a high current monster that will stomp around like godzilla
...the trick is to get it to stomp around the empty space
of the inner grid rather than romp all over your fusor.

EYE that is the Rub!

I'm really serious about the work
so I'm learning all I can about the demo and continious stuff first. Time spent in the begining will actually speed up the work later and save buku bucks
on blown up thyratrons and burnt grids.

Larry Leins
Fusion Tech
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: pulsed fusor power issues

Post by DaveC »

One would expect the gas disharge within the fusor to bear some resemblance to the standard gas discharge curves.. (like the ones' we plotted as an electronics lab exercise for a neon lamp.) .. IF the supply current is limited through an appropriate sized impedance.

In fact it will be a useful exercise to make a Current vs Voltage plot for the fusor varying the voltage from zero up to a high enough value to go through the negative resistance regime.. and on to the constant voltage drop regime.

(Depending on how robust your HV supply is, you may need to put a current limiting impedance in series with the HV supply to the fusor to avoid excessive current. If your supply is not too stout, it will limit current quite naturally.)

If you plot the Fusor voltage on the horizontal axis and the Fusor current on the vertical, the curve should go outward from the origin toward increasing negative voltage moving downward at a slight slope, until a maximum voltage is reached. Attemping to increase the supply voltage after this causes the fusor current to increase so much that the internal voltage drop in the transformer or the current limiting resistor, causes the actual fusor voltage to drop. This is the point where you begin to see the glow in the atmospheric gas, demo mode.

I believe the extremely voltage sensitive region that Richard describes is where the I - V curve folds back ( goes into a negative resistance regime).

Incidentally, since all gas discharge devices can be operated as relaxation oscillators, why not operate the fusor this way? The oscillation rate (frequency) will be voltage dependent. The circuit is simplicity itself. No need for a pulse transformer. Depending on your output capacitor, you can make the pulse width small or large. Has anyone tried to do this? It might have advantages over steady state operation.



At any rate as Richard has suggested, exploring the operating modes of your fusor will reveal its basic nature and the somewhat perplexing behavior becomes at least predictable. Keep a good lab book.

Dave Cooper
guest

Re: pulsed fusor power issues

Post by guest »

I have incorporated negative resistance into the pulse supply's operation. I am working on several modes of operations as each set of equipment becomes availible. The relaxation mode of the fusor could be used that way, but the main concern is to increase the
operating voltage and raise the crossection of fusion
without building really hairy dc supplies. One trick I want to try out is the trick that Dr Bussard came up with in conjuction with his IEC concepts. I want to build the
output transformer into the vacuum chamber and use vacuum as the insulator on bare transformer windings.
A small ten turn coil made out of copper tubing would form the coil secondary. It is a cross between the Farnsworth rectifier and a 1930's electron accelerator
plan I found in the nuclear section of my library. A 1/4 turn of copper would serve as the primary. It would be
operated in forced oscillation mode using pulsing negative dc. At 15kv the coil would output around 600 kv at 15 ma. By attaching the output coil dirrectly to the fusor grid the onerous chore of a ultra high voltage feed thru is avoided. Safety would be easier due to the fact that the other end would be tied to the case. All ultra high voltage would occur inside the fusor walls.
Using a neg dc pulse also eliminates the Achilles heal of the Farnsworth rectifier the output diode. No filaments to worry about and no battery operated supplies either.
I dub this concept Franken Fusor.
It is the upgrade to my high road fusor.

Larry Leins
Fusion Tech
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: pulsed fusor power issues

Post by DaveC »

Larry -
Putting HV components under high vacuum is a good concept. We are using this method in our E Beam hardware. But...there is an important caveat: The vacuum needs to remain high. Generally pressures must be below 10^-6 Torr for good HV behavior.

Several other challenges occur too.

Flashovers and surface tracking will still be with you...Impact ionization of dielectrics can create runaway conditions in the direction of E field lines, leading to flashovers. With strong XRay illumination, many quiet surfaces become electron emitters.

Then there is outgassing... This is always a challenge anyway, but to minimze its effects, you do need to be under high vacuum for a few hours, and preferably a few days. High temp bakeouts, are useful ...IF.... you first remove the delicate insulations... but then those delicate insulations need rather long times to reach their ultimate vapor pressures... which are usually in the 10 ^-5 to 10^ -7 Torr at best.

In our e beam apparatus, the system is continually pumped and we never add any other gasses.

I am not sure exactly what to expect when there is a necessary partial pressure of D2.

Finally, there is the challenge of induced desorption of gasses by any surfaces that become bombarded by electrons, photons or ions. Several years ago, because of marginal vacuum levels in e beam tubes, we learned a lot about transient pressure rises due to temperature and, electron bombardment. A system under high voltage can develop substantial presure rise as the filament comes up to temperature and often developed major internal flashovers when electron beam current increased. Our measurments of dynamic internal pressure were rather indirect, but ion currents often would increase nearly a decade in magnitude, indicating similar pressure increases.

But you are right on the money as to the enticing advantages of eliminating HV feedthroughs. That's what we like about the idea, too.

Dave Cooper
guest

Re: pulsed fusor power issues

Post by guest »

Yeah I wondered about out gassing and d2 gas screwing up the works. I guess a Ln2 trap with sorbant before the tranni coils is in order. I can get around the x - ray issue by bending the output 90 degrees before the attachment to the fusor grid. The model I'm basing the concept on used relatively high vacuum ( around 10 microns) . The electron accelerator used copper pipe
turned into a helix with turn seperation in the 1" to 1 1/2 " range. By pulsing full wave on a push pull oscillator at 15 kv, the coil output 750,000 volts.
It used a 1/4 turn primary. The coil was attached to an
accelerating dee. To obtain ions or electrons a filament
is biased to the proper polarity depending on the charge of the desired particle. As the dee was pulsed
the particle would be attracted when the polarity was correct for attraction.

Larry Leins
Fusion Tech
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: pulsed fusor power issues

Post by Richard Hull »

I talked at length on the old songs list in places (various times) about the relaxation oscillator mode of the fusor. I first noticed it in the demo fusor fusor II. I actually show a segment on the first fusor video where a friend and I are triggering the system to flash while right at the ragged edge with a charged teflon wand snap waved at about 2 foot range. The glass bell jar allowed this E-field shock triggering.

With the right capacitor-resistor limiter combo, the fusor will self oscillate. A delicate condition to say the least, but it will work just like the little blinky Ne-2 lamp experimental relaxation oscillator demos that have fascinated electronic neophytes for years.

I think now, that if I were to attempt a major pulsed fusor, this might be the way I would hit it rather than pulse elevating the voltage externally. A rapid gas pulse as might be had from a reworked car injector might just be a key component.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
guest

Re: pulsed fusor power issues

Post by guest »

It seems like an eternity since I've seen a blinkey demo. Build one at age 12, but mine didn't act like an astable oscillator but rather like a flip flop. Could I get
a diagram sometime if possible? I think it would be a neat thing to try out. Any math to this device at all?
ie frequency, duration ect?
The car injector would be a really good Idea if I could get the fusor to oscillate at regular intervals then synchonize the injected puff with the on condition.
I could then modify a turkey baster to act as a electrolytic cell. Draw up the deuterium oxide into the baster syrenge then flow current to make d2 gas.
let it electrolyse until the syrenge is full of gas . Humm
need a three way valve.... one way to draw heavy water... another position to move the now meatered gas into a dehydrater into a storage manifold.
Could be that simple system for model power I keep seeing.

Thanx Richard,


Larry Leins
Fusion Tech
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: pulsed fusor power issues

Post by DaveC »

Considering how little D2 gas is needed for a fusor run, it could be practical to electrolyse it from heavy water... assuming you had some. But I think the gas form is much easier to handle.

Also.. for HV pwr supplies in vacuum, the absolute pressure has to be way below 10 microns - more like 10^-3 microns!! Much above this pressure and your supply will be bathed in glow discharges and probably wont work well.

This is not too much of a problem for real, high power fusor operation. It is a contradictory requirement though for the demo fusors which run at 10 microns or possibly higher pressure.

Dave Cooper
guest

Re: pulsed fusor power issues

Post by guest »

I happen to have 700 g on hand of heavy water.
Not a heaping helping to be sure but enough.
My low end stuff is now coming on line.
But unfortunately the low end stuff is too dangerous
to use at the levels demanded by neutron production.
My metal chamber is just now coming together.
I have all the pieces just got to get it welded by a good
welding shop.
I intend to run three lines of fusion work.... continous....
pulsed.... power conversion when it becomes possible. Already gathering materials for the pulsed and power conversion.

Larry Leins
Fusion Tech
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: pulsed fusor power issues

Post by Richard Hull »

The schematic for the Ne2 blinker is simple. A resistor in series with a capacitor to ground. The Ne2 is placed across the capacitor. A DC voltage in excess of 110 volts is placed across the RC charging network and as the cap charges the voltage on the neon climbs until it discharges, shorting out the cap and starting the whole process over again. The blink time is based on the potential of firing of the lamp 65v -80v highly variable and the RC time constant to hit the firing voltage. Typical values are 1 megohm and 5ufd @200v.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: pulsed fusor power issues

Post by DaveC »

Thinking a bit more about relaxation oscillator modes of fusor operation....

First a non-electrical one.

Consider if D2 is injected as gas pulses. The high vacuum pump will respond with a slightly delayed proportional pressure pulse in foreline pressure. If the foreline has a bypass over to the gas inlet valve, AND, if the system is well pumped, clean and tight, then the bulk of the rough pump gas load in each pulse will be D2 neutrals. The bleed line thus recycles these for subsequent shots.

Since Diffusion and Turbo pumps produce compression ratios of 10 to 1000, there is easily enough pump "gain" to efficiently inject the gas.

Haven't worked it out theoretically, but it seems the high vac pump with a foreline bleed to the gas injector as described here comprises a rudimentary gas "oscillator" .

As an upper limit, the frequency could not be more than molecular velocities divided by feedback (bleed line ) lengths, or something in the KHz range. This would be maximum... Given the thermal distribution of gas velocities, it could be somewhat lower.
.
Might be a useful operating mode, although this is not the same as Farnsworth's concept of internal recirculation.

Theren is also an interesting electrical mode:

At fusor pressures, the mean free paths of deuterons are many times the dimensions of the fusor itself.
This gives rise to the Internal recirculation of the deuterons.

Because the internal potential fields are conservative, it also suggests some of these ion paths (trajectories) will align with the ion gun source.

These paths will have distinct cycle times, thus implying resonant frequencies.

By "tuning" the rep rate of fusor voltage pulses, to match these cycle times, some enhancement of the basic fusing process, may occur, since the ions get a well timed kick to boost their velocity.

There's a bit more to this which I will try to develop in a later post. But you can get the basic idea ...

Dave Cooper
guest

Re: pulsed fusor power issues

Post by guest »

Thanx Richard:
I couldn't remember whether the neon tube went in
parrallel or series with the capacitor. My dad showed me one later that day I posted it.
Wow that goes back to the days when I had a Knight kit!
Humm I've got everything to try it on a demo fusor.
I think I can get it to work ok.
I'm waitng on more vacuum grease and vacuum epoxy.
The injector will have to wait for a minute.
My design trial could be at 15 kv.
Or I could incorporate that godzilla size cap to run at 10 kv.... it would only draw only enough current and then extinct .... self regulating .... pretty neat.
I will keep good lab notes.
The cool part is I can finally use that dog of a solid state neon trani as something besides a doorstop.

Larry Leins
Fusion Tech
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: pulsed fusor power issues

Post by Richard Hull »

Dave made mention of 'at fusor pressures the ion mean free path is many time the fusor dimensions'. For the simple fusor this is not really correct. As a matter of fact the mean free path in a 6" fusor is barely the RADIUS of the device!!! This is why, as I have said, no simple fusor should exceed about a 10" diameter as a maximum. (Simple fusors run in the 1 micron and above regime.)

Recirculation is a 10e-4 torr or lower event. This mode uses ion guns or special ionizer systems internal to the device.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: pulsed fusor power issues

Post by DaveC »

Richard - I think we are both correct here. You are certainly with regard to mean free path at 1 micron pressure...about 5 cm... and at gas temps around 300 K.

I was tacitly assuming the lower pressure regime, without actually saying so. There, at pressure in the 10-6 Torr, the mfp gets proportionally longer, reaching 10's of meters..

But ... there is also the caveat that for higher temperature molecules, (and the fusor deuterons are such), the mfp increases dramaticallyeven at 1 micron pressure. Complicating things though is the increased gas density at the poissor. So... I am not certain where things end up, and what an effective mfp is in that case.

From the modelling I have done of electrons (much smaller of course, and much faster) in a spherical shell, at very low pressures,.they loop around for several hundred times, before colliding with something. Seems plausible that to a lesser extent the Deuterons and D2neutral molecules should do the same. What do you think?

Dave Cooper
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: pulsed fusor power issues

Post by Richard Hull »

The mean free path for ions at 10e-6 torr is just a bit under a kilometer! Lots of recirculation here. Remember, "mean" means just that. There are a significant fraction that go a bit farther. Tom Ligon and I farted with the numbers and figured that about 20% of the deuts in a simple fusor might make a full recirculatoy path. No very encouraging.

At lower pressures recirculation is almost total and unending except for the poissor density which has been calculated to be perhaps a 10X pressure above the chamber with a few reserachers claiming as high as 50X. this would defintiely impact recirculation, but who cares, that is where you want collisions...Right?

Electrons can easily have 10X or more of an ion's mean free path, so 20 kilometers is reasonable for electrons at 10e-6 torr. Electron mean free path is also a function of energy. I gave the precise equation in one of the old songs posts.

Farnsworth originally wanted only electrons to do the work in the fusor, but he could never make it work out.

The snag of operating at 10e-6 torr is that you only have 1/1000th of the deuts on hand compared to a simple fusor. Fusion is, and forever will be, a function of the density of the reactants in any given volume at any given energy. Ya' just can't escape that one.

This is one reason I made such a bold and sweeping statement in my post "on to every parade a little rain must fall". (Fusion powered future forum)

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
guest

Re: pulsed fusor power issues

Post by guest »

I've now modified my front end on my pulsed fusor supply to incorperate the relaxation mode of fusor operations. The power supply design now charges
the capacitor with pulsating negative dc then the fusor then decides when it is ready. I can now run my charger in the 1 Khz range without the impedence nightmare of a forced fusor device. I have yet to find out if I still will have to use the keep alive current under this mode. I 've designed a copper sulfate water resistor to provide the resistance at the fusor head.

Fusion is fun!

Larry Leins
Fusion Tech
Post Reply

Return to “High Voltage - Fusor Input Power (& FAQs)”