Steps towards FF-1 Feasibility

[Joeviocoe]

asymmetric_implosion,

I’m quite new to this forum, but have seen many posts describing the experiments you’re doing. But since you have so many posts, I cannot pin down the basics of your experiment. Do you actually have a DPF? Or is it something similar? What have you built, and for what purpose? This would help me understand the context of your knowledge here. Thanks.

[jimmarsen]

What about scaling up the yield per shot? Is that feasible?

[Joeviocoe]

jimmarsen wrote: What about scaling up the yield per shot? Is that feasible?

That is the first thing they are trying to do. But there are many limits to that too. Yield scales to the 4.7th power of Current (I^4.7). But current is limited by MANY parts of the device. Capacitors, switches (low jitter needed), the electrodes especially need to handle the higher current. Once yield per shot is maximized, they can take it to the next limit of cycle rate… which have their own limitations too.

So they are estimating 132 kJ output from the plasmoid for each shot. After conversion losses and recycling 100 kJ for the next shot… that is 24.7 kJ net output for each shot. And they estimate that a production reactor could function at a cycle rate of 200 times per second and only need maintenance once a month.

These are bold claims that they will spend the next few years proving.

[jimmarsen]

Does the physics of DPF prevent designing a larger device; are there engineering limitations?

[asymmetric_implosion]

Joeviocoe: I operate three plasma focus devices at the 60 kA, 140 kA and 0.5 MA level (typically 250 kA at rep rate). Our experiments are focused on non-energy applications of the plasma focus such as active nuclear interrogation, lithography and basic plasma physics related to what is known as high energy density physics in support of larger experiments for our customers. I work for a small business like LPP but we have a much larger scope of plasma applications including thin film coatings, plasma/ion thrusters and plasma diagnostics that are applicable to many areas of plasma physics. Though our customers have specific needs I try to take some time to work on larger problems like neutron yield scaling (don’t buy the 4.7 scaling law, there are many different explanations and supporting data for each. Published numbers exist from 3.5 to 5.5 in the power law. I have my own take on it but I am finishing up the paper on it so I will leave it there). My work involves repetition rate plasma focus devices (>0.1 Hz) at scales ranging from 60 kA (10-100 Hz) to 0.5 MA (~1 Hz). I have encountered gas lifetime issues, materials issues and repetition rate pulse power problems including switch lifetime and switch properties for specific configurations.

[Joeviocoe]

jimmarsen wrote: Does the physics of DPF prevent designing a larger device; are there engineering limitations?

That question has been asked by many, me included. And from what I gather.. there are limitations that make larger devices a problem. 5MW was chosen for many good reasons.

But luckily, these devices will be small enough that you can stack 10, 20, or more in a single area. And as long as you can get rid of the waste heat (or use it separately), then it should be a very efficient use of space.
Still not as good as Gas Turbines or Diesel Generators for MW per sq. meter … but much better than Solar and wind even (and those are intermittently producing power). And when you account for the fuel needed for conventional power plants, fusion devices like these are so much better.

The main point LPP likes to focus on, is the fact that smaller generators can be place closer to the load. So instead of central power plants pushing power over hundreds of miles through transmission lines that waste energy, a single 5 WM DPF could be placed in several locations throughout the city, taking up a small space in existing electrical substations (where High Voltage power lines meet transformers on the ground to step down to neighborhood level voltages).
For large ships (cargo and cruise liners) and locations that require 50 MW – 100 MW in a centralized location, I’m not sure DPF is practical. Perhaps if Polywell works with pB11, that would be a great solution.

[asymmetric_implosion]

jimmarsen wrote: Does the physics of DPF prevent designing a larger device; are there engineering limitations?

Yes and no. The primary limitation encountered to date is the insulator between the anode and cathode in vacuum. Experiments have demonstrated a system without an insulator at 5 MA but the experiments were focused on soft x-ray production. Is the insulator the show stopper to scale up??? I don’t know for sure. Some experiments at 2 MA suggest a solution to the problem but it might drive the fusion yield away from optimum. This is a pressing point in PF physics that is actively researched because scaling up a PF to high current should cost less than a Z-pinch. The problem to date is that PF devices seem to show decreased fusion yield with currents above 1 MA. Z-pinch devices show a fall off at something like 300 kA. See the attached figure with PF devices and radial implosion Z-pinch devices as an example. This is data I’ve compiled from the 1970’s to present from peer-reviewed published lit including data from LPP. If you can scale up the device along the alpha=0.25 curve, you can get more fusion yield per shot. If the fusion yield (DD neutron yield) falls off at 2 MA or more, it seems that a smaller and high rep rate PF is the better option.

Attached files

[Joeviocoe]

asymmetric_implosion wrote:

Does the physics of DPF prevent designing a larger device; are there engineering limitations?

Yes and no. The primary limitation encountered to date is the insulator between the anode and cathode in vacuum. Experiments have demonstrated a system without an insulator at 5 MA but the experiments were focused on soft x-ray production. Is the insulator the show stopper to scale up??? I don’t know for sure. Some experiments at 2 MA suggest a solution to the problem but it might drive the fusion yield away from optimum. This is a pressing point in PF physics that is actively researched because scaling up a PF to high current should cost less than a Z-pinch. The problem to date is that PF devices seem to show decreased fusion yield with currents above 1 MA. Z-pinch devices show a fall off at something like 300 kA. See the attached figure with PF devices and radial implosion Z-pinch devices as an example. This is data I’ve compiled from the 1970’s to present from peer-reviewed published lit including data from LPP. If you can scale up the device along the alpha=0.25 curve, you can get more fusion yield per shot. If the fusion yield (DD neutron yield) falls off at 2 MA or more, it seems that a smaller and high rep rate PF is the better option.

I thought that it was higher voltages that brokedown the insulator, not higher current. And that higher current, with higher cycle rates, causes the skin effect to heat the cathode unevenly causing damage.

And that the DPF fusion yields scale higher with increased current ( I^4.7 ) but that 3 MA would be optimal.

Am I completely wrong here?

[asymmetric_implosion]

Joeviocoe wrote:

Does the physics of DPF prevent designing a larger device; are there engineering limitations?

Yes and no. The primary limitation encountered to date is the insulator between the anode and cathode in vacuum. Experiments have demonstrated a system without an insulator at 5 MA but the experiments were focused on soft x-ray production. Is the insulator the show stopper to scale up??? I don’t know for sure. Some experiments at 2 MA suggest a solution to the problem but it might drive the fusion yield away from optimum. This is a pressing point in PF physics that is actively researched because scaling up a PF to high current should cost less than a Z-pinch. The problem to date is that PF devices seem to show decreased fusion yield with currents above 1 MA. Z-pinch devices show a fall off at something like 300 kA. See the attached figure with PF devices and radial implosion Z-pinch devices as an example. This is data I’ve compiled from the 1970’s to present from peer-reviewed published lit including data from LPP. If you can scale up the device along the alpha=0.25 curve, you can get more fusion yield per shot. If the fusion yield (DD neutron yield) falls off at 2 MA or more, it seems that a smaller and high rep rate PF is the better option.

I thought that it was higher voltages that brokedown the insulator, not higher current. And that higher current, with higher cycle rates, causes the skin effect to heat the cathode unevenly causing damage.

And that the DPF fusion yields scale higher with increased current ( I^4.7 ) but that 3 MA would be optimal.

Am I completely wrong here?

Voltage is a secondary concern. Insulators for the Z-machine operate at up to 6 MV without any problems. The insulator failure seems to come in three forms: debris build up leading to bad initiation which is current driven by cathode erosion; mechanical failure from shocks due to drive largely by magnetic pressure i.e. current squared; and UV irradiation leading to materials damage and possibly re-strike again dominated by current. Kies has a nice paper on the topic and how to mitigate some of the problems up to 300 kV charge voltages. It is not clear to me if the tungsten pins used by LPP can get around the problems by localizing the initiation a small distance from the insulator. Using a thicker insulator tends to eliminate the problems due to shock unless the cathode and anode plates start to move due to magnetic pressure like on the LPP device but again, Z-pinch folks solved this problem long ago as they are operating at the 20 MA level in Z-machine. The UV problem is the most difficult to deal with as the plasma must flash the insulator. Higher Z gases produce far more UV than hydrogen and deuterium so boron could be a problem. There are alternative ideas like using gas puffs to supply the necessary fuel on the PF axis and let the axial phase be dominated by hydrogen. Again, Kies has a paper on this at the 1.5 MA level for the SPEED-2 PF. Banks exist that could drive PF devices at 10 MA with reasonable rise times. The high risk is getting the insulator to survive more than 1 shot. Everything else can be done at a few shot a day level. Moving to even 10 Hz operation at 2-5 MA has never been done. To my knowledge, the Italians were working on a 1 MA, 1 Hz bank for a PF. I don’t know if the machine was ever built. Discussions have come and gone about using LTD systems to power PF devices but no one has tried to my knowledge. They are 1 Hz, 1 MA systems that will sit in most living rooms. Beyond 1 MA is a few shot a day territory for most pulse power systems.

I won’t speak to an optimum because different people define optimum differently. I don’t know if there is a true optimum for a fusion system. My guess is that there is a mix of cost, engineering risk and possible physics limitations that will decide on how a repetition rate PF will be built. I can tell you this much, a 60 kA PF firing at 10 Hz sounds like an automatic weapon firing. A 3.5 MA Z-pinch causes the floor to move noticeably. A 20 MA Z-pinch feels like a small earthquake. Firing a 2-5 MA PF at 200 Hz will be an experience for those near by without some sort of seismic isolation. I feel for the operator.

[Joeviocoe]

asymmetric_implosion wrote: My guess is that there is a mix of cost, engineering risk and possible physics limitations that will decide on how a repetition rate PF will be built. I can tell you this much, a 60 kA PF firing at 10 Hz sounds like an automatic weapon firing. A 3.5 MA Z-pinch causes the floor to move noticeably. A 20 MA Z-pinch feels like a small earthquake. Firing a 2-5 MA PF at 200 Hz will be an experience for those near by without some sort of seismic isolation. I feel for the operator.

Does the DPF produce force alone only one direction? If so, stacking 2 in opposing directions could dampen the seismic vibrations.

[jimmarsen]

What about increasing the duration of a shot to increase yield? Is that feasible?

[Joeviocoe]

jimmarsen wrote: What about increasing the duration of a shot to increase yield? Is that feasible?

I don’t think so, the Dense Plasma Focus is basically a collapsing knot of plasma filaments. It is inherently unstable (in contrast to trying to make it stable like Tokamaks) which is good since the instability allows the plasmoid to fuse the fuel. But its “duration” is a physical, parameter really… and I think is a product of everything else. I don’t think they can change that independently.

You’ll hear them talking about nT (n tau) which is the product of density and confinement time. If by, “duration” you mean confinement time which allows for more fuel to be fused, and thus increased fusion yield… then “duration” has always been a factor that LPP has been trying to maximize. Along with density of the plasma, that makes up a critical part of attaining a net power output.

[asymmetric_implosion]

The direction of the force on the anode is largely 1D but I will let LPP speak about their experiment in detail. I know it was discussed a month or more back when they broke an insulator. I think the force depends on the machine configuration but the shaking comes from magnetic pressure on the electrodes outside of vacuum, switch closure which can be very loud depending on the current in the switch and the plasma shock hitting the vacuum vessel wall. At 60 kA, our switches are nearly silent. The vacuum wall makes most of the noise. Z-pinch machines use more complex pulse power with multiple stages of operation to achieve the 100-200 ns pulse required to drive the Z-pinch.

As far as increasing the duration of the pinch goes, there is one PF-like experiment that demonstrated >1 us confinement. The hypocycloidal pinch, a modified form of a PF, was developed by a researcher at NASA (NASA report TN D-8116). He didn’t use D2 as fuel but found an x-ray pulse that lasted for >1 us. No one picked up the research but it shows promise for stability well in excess of any other Z-pinch device at useful densities. I’d love to build one and test it but the not in the cards for me.

[break]

asymmetric_implosion wrote: Anode erosion does not have to be a show stopper.

Thank you for your backround information. It makes hope again… 🙂

[asymmetric_implosion]

I have hope for the anode because you can use the physics of a highly directional beam to help you. The switches are the problem. I guess it is good for creating jobs and manufacturing to replace switches every few days if not every day. The switches are going to require a heavy investment to realize a 5 MW power plant operating at 200 Hz for more than a few days.

[Author]

Posts