Heat produced by Focus Fusion and cooling

[Aeronaut]

Vansig, we’re most likely going to need a fixed anode at the system design level, since the electrical and cooling connections are going to have to pass through the baseplate. My earlier sketches assumed 90 degree bends, but hopefully a 10 to 15cm baseplate could run the plumbing in a series of arcs past the cap(s) and switching gear. Seems like just about everything needs access to the space between the switch(es) and baseplate. :-S

For mass production we’re going to need as simple an anode design as we can get while the materials science is being proven experimentally. This is another reason for an all beryllium anode, at least until the onion’s yield approaches 4 or 5 MW, or at least shows a promising improvement curve.

[vansig]

the gallery photos reveal a lot, but
this would be easier to visualize if there were a 3D model of FF-1 around?

[Aeronaut]

I don’t know of any 3D models of any FF design. But if we can develop a single cap production model similar to the UN University machines, the cap, ground plane, switch, and anode can almost be bolted together with little in the way of flange design. At least until we need a cooling system. I’ll sketch one up when we’ve worked up some rough ideas of the plumbing’s dimensions.

[Allan Brewer]

Aeronaut wrote:
For mass production we’re going to need as simple an anode design as we can get …

Sure, but it has to work before it can be mass-produced, and there are some serious worries about the cooling feasibility (2 kg/s of helium gas would need to be pumped) (~175 atmospheres) (supersonic)….

Can I suggest a helpful experiment for LPP (which they have probably scheduled anyway?) – When the Be anode is mounted, assuming it has the same dimensions as the copper electrode, and other things being equal (gas, pressure etc.), How much extra power is required to get the same current as with the copper electrode? This extra power represents the extra heating in the Be electrode to overcome its extra resistivity over Cu, and by comparing the relative resistivities of Cu and Be it should represent 0.53 of the total power loss in electrode resistance at about room temperature. Applying the temperature coefficient for Be resistivity for the expected running temperature of the electrode (which might triple the power loss) and multiplying by the number of shots per second should give a reasonable estimate of the electrode cooling requirement. Then we can either relax or worry/think with a target.

[Brian H]

Color me confused, but I thought the primary heat source/problem to be addressed was the plasmoid’s output, not the application of charge to the plasma through the birdcage.

[jamesr]

Brian H wrote: Color me confused, but I thought the primary heat source/problem to be addressed was the plasmoid’s output, not the application of charge to the plasma through the birdcage.

Most of the energy from the plasmoid will be in the form of X-rays or the electron & ion beams – and hopefully more than was used to make it due to the excess of some fusion 😉 But most of the direct heat, will be from the resistive joule heating due to the current flow in the anode & cathodes. The total surface area of the ~12 cathodes is much larger than the anode so their proportion of the heating problem will be less – hence why we are concentrating on the anode.

A proportion of the X-ray energy will inevitably end up as heat, but that will be distributed throughout the volume of the anode and the whole spherical shell of components in the firing line.

[Brian H]

jamesr wrote:

Color me confused, but I thought the primary heat source/problem to be addressed was the plasmoid’s output, not the application of charge to the plasma through the birdcage.

Most of the energy from the plasmoid will be in the form of X-rays or the electron & ion beams – and hopefully more than was used to make it due to the excess of some fusion 😉 But most of the direct heat, will be from the resistive joule heating due to the current flow in the anode & cathodes. The total surface area of the ~12 cathodes is much larger than the anode so their proportion of the heating problem will be less – hence why we are concentrating on the anode.

A proportion of the X-ray energy will inevitably end up as heat, but that will be distributed throughout the volume of the anode and the whole spherical shell of components in the firing line.
Current causes heat; the only link in a circuit between the cathodes and anode is the plasma. It is still not clear to me that that circuit “closes”, resulting in current through the anode.

[jamesr]

Brian H wrote:
Current causes heat; the only link in a circuit between the cathodes and anode is the plasma. It is still not clear to me that that circuit “closes”, resulting in current through the anode.

That’s how a DPF works! The current starts to flow when the plasma arcs across the base between the cathodes & anode. This radially inward current sheath of plasma creates a magnetic field around the cylindrical axis. The cross product of the current and field causes a force along the axis, sweeping the sheath of plasma down to the end where it filaments & and goes kink-unstable forming the focus.

Its all about the current and how we get that up to the 3MA or so needed to put enough energy into the magnetic field at the focus.

[psupine]

Is there any reason why the sides of the anode have to be parallel? I understand that the cathodes in the cage have to be parallel to the anode surface, but what would happen if the whole thing was kind of conical? One thing is that it would allow more surface area and more internal plumbing volume but still let the business end stay on a tight radius? Or would the conical shape do something nasty to the plasma run down?

[Brian H]

jamesr wrote:

Current causes heat; the only link in a circuit between the cathodes and anode is the plasma. It is still not clear to me that that circuit “closes”, resulting in current through the anode.

That’s how a DPF works! The current starts to flow when the plasma arcs across the base between the cathodes & anode. This radially inward current sheath of plasma creates a magnetic field around the cylindrical axis. The cross product of the current and field causes a force along the axis, sweeping the sheath of plasma down to the end where it filaments & and goes kink-unstable forming the focus.

Its all about the current and how we get that up to the 3MA or so needed to put enough energy into the magnetic field at the focus.
Yes, of course. My blind went mank on it for a mo’, but the toroid and filaments must be re-directing the current away from the anode once they form.

[Aeronaut]

psupine wrote: Is there any reason why the sides of the anode have to be parallel? I understand that the cathodes in the cage have to be parallel to the anode surface, but what would happen if the whole thing was kind of conical? One thing is that it would allow more surface area and more internal plumbing volume but still let the business end stay on a tight radius? Or would the conical shape do something nasty to the plasma run down?

The patent allows for all sorts of shapes or profiles, and there are supposedly some advantages to a narrowing taper towards the tip.

[jamesr]

Brian H wrote: but the toroid and filaments must be re-directing the current away from the anode once they form.

Indeed… the pinch is characterised by a sharp drop in current to around half the peak value over a few 10s of ns. This is assumed, I think, to be when the magnetic field changes topology and the separate plasmoid is formed at the heart of the focus. Although the exact mechanism for this is little understood.

[vansig]

Aeronaut wrote:

Is there any reason why the sides of the anode have to be parallel? I understand that the cathodes in the cage have to be parallel to the anode surface, but what would happen if the whole thing was kind of conical? One thing is that it would allow more surface area and more internal plumbing volume but still let the business end stay on a tight radius? Or would the conical shape do something nasty to the plasma run down?

The patent allows for all sorts of shapes or profiles, and there are supposedly some advantages to a narrowing taper towards the tip.

i’m thinking that taper would become a really blunt cone, and the dimple facing the plasmoid could then grow to quite a large void.

[Allan Brewer]

Allan Brewer wrote:

[Graphene electrical resistivity may be less than 5e-4 nΩ·m; …Shall we put a graphene coating on everything, then? ..

I have spoken with a graphene expert and unfortunately the extremely high conductivity is only (presently) attainable on the nano-scale. Although coating techniques are available on the scale of our electrodes, the result is molecularly non-contiguous flakes which results in a much lower conductivity than Copper and Beryllium.

So the only way I can see of drastically reducing the electrode resistive heating would be to have an ultra-slim “high-Tc” superconductor inside the Beryllium anode connecting at many points to the anode, and cooled by liquid nitrogen.
Does anyone know if liquid nitrogen has ever been used as a primary coolant?
How bad would it be in blocking X-rays?
Even if not possible as a primary coolant it still might be possible to cool the superconductor through an inner channel that way and have helium gas as the main coolant?

[vansig]

part of me wants to go ultra-low tech, here, and only use parts for this that were available to Benjamin Franklin in 1752.

a high Tc superconductor could have interesting effects on the plasmoid’s magnetic field, pushing it away from the anode; will a proper plasmoid even form?

[Author]

Posts