[asymmetric_implosion]

A company called Science Research Labs (SRL) built a solid state DPF back in 1992 and upgraded it for years to reach 230 kA. The pulse power alone cost $1M. The operating voltage was 8 kV at most. It used saturatable magnetics. The problem is using solid state in series. SRL mastered the technology at the 230 kA level. A 10X scale up would be daunting at best. A 1 MA PF typically operates at 40 kV or more so that’s many switches in series and parallel. For a 6 kV switch unit, you would want to have 8 switches in series. With a 100 A switch you need 20,000 switches groups. A 2 MA PF with no room for upgrades in current requires 160,000 switches. Solid state if built correctly can last for 10^10 shots.

Spark gaps and railgaps can achieve both the rise time, operating voltage and current carrying capability to drive most pulse power devices. Rail gaps can operate up to 20 Hz at ~100 kV and 0.75 MA per switch if done properly. Switch life time is a problem for spark gaps and railgaps with ~10^7 shots at most.

[asymmetric_implosion]

In the case of plasma shock behind the insulator, it is more like taking a hammer to the insulator than a heat stress. The impact happens on a time scale much faster than the time to transfer heat into the material. We saw evidence of plasma chemistry on the inside surface of the insulator. The alumina discolors slightly from pristine white to a stained yellow. It remains a hypothesis but the data at the 0.25 MA scale seems to support it.

[asymmetric_implosion]

You could try putting samples parallel to the plasma flow between two cathode rods near the top of the cathode rods. You could mount 4 samples and try the three proposed insulators and alumina. It might not be the fairest test but it would be apples-to-apples. If you test it with the ion beam or electron beam you need to use samples at least 0.25″ thick. I put ceramics in the base of the anode for some experiments and thin sheets will crack easily under the stress of either beam. Alumina fairs very well in the tests I’ve completed.

[asymmetric_implosion]

In my limited experience with it, boron nitride works well in quasi steady state plasma conditions like glow discharges. When I used it in pulse discharges (~100 A, a few microseconds) it visibly eroded in a short period of time and nitrogen showed in interesting places.

[asymmetric_implosion]

Be wary of Boron nitride. I’ve used it in plasma facing conditions before and it turns to powder pretty easily. I have no experience with silicon nitride or zirconia. Best of luck.

[asymmetric_implosion]

Francisl wrote:

A plasma is forming behind the insulator. The rapid heating of the gas to plasma increases the pressure and shocks the insulator.

Can the gas behind the insulator be displaced by more insulation as either a solid or packed in powder? Can some type of pressure relief be formed in the insulator or the anode to accomodate the shock wave?

I have not tried either solution. In my case, the problem is the thermal expansion of the anode. You need to leave a little room for the anode to expand. LPP likely has tight tolerances for other reasons. My concern with powders is that gas can get into the powder. The plasma forms and explodes the powder. After some number of shots the powder is all over the vacuum chamber. Good, bad, other…I don’t know. The another insulator would still have to be plasma facing so it would likely need to be alumina or some other ceramic material. I don’t know of a flexible ceramic.

To be fair, the insulator problem is a very hard problem and with limited resource it hasn’t been fully explored. Packed powder or another solution might work. In the case of limited funds, I typically suggest the tried and the true…add more of what you need. It comes at a cost, but I think that cost can be handled more easily than venturing into the world of materials development. I might not have pointed it out, but many machines at the >1 MA level have insulator lifetimes of hundreds of shots at most with tens of shots being more common. There are ideas on how to increase the insulator lifetime but there are tradeoffs. It depends largely on your pulse power requirements and goals. I know for my work, a thicker insulator was an acceptable solution. It cost me 5-10 kA at peak current but I can easily compensate with a slightly larger charge voltage. I’m sure these discussions are going on at LPP.

[asymmetric_implosion]

willit wrote: Dow corning 1540-20p is a pourable silicone that can be formed in a mold any shape. Mix it with alumina or other insulator dust and you have a tailor made insulator with a lot of flex, insulating properties thermal capabilities, its off the shelf and inexpensive. 1 mold can make as many as you need and can easily be modified. Samples are easily obtained and would most likely cover the first few trials.

Silicone is like a polymer. It has many of the same problems of the polymer including UV sensitivity. Can alumina be mixed with it and will the silicone wet to the alumina? Wetting is key to building a strong compound. There are alumina and binders that can be molded into any shape but they are prone to cracking in the presence of intense plasma discharges like the plasma focus breakdown.

[asymmetric_implosion]

pulser wrote: Do you think glass filled Torlon would work? Much more shock proof.
http://www.boedeker.com/torlon-5530.htm

Torlon is bulk kapton. Same problems as Kapton. Been tried and it was found to pollute chamber and reduce neutron yield.

[asymmetric_implosion]

I’ve seen that before. 🙁 I thought it was the disk part shattering but it is the cylinder. A plasma is forming behind the insulator. The rapid heating of the gas to plasma increases the pressure and shocks the insulator. I’ve cracked many insulators at the 0.25 MA level because of this. You need to increase the wall thickness of the alumina. It increases the inductance but you gain in lifetime by a great deal. It should take minor changes to the anode to make this work. The rest of the hardware can remain as is.

Also, the disk diameter seems to be overkill unless you’ve had arcs in the air before.

[asymmetric_implosion]

A high band gap semiconductor would be interesting if it was held in reverse bias. People tried it with barrier discharges and found some interesting results. It is purely speculative and probably not appropriate for the LPP experiments, but it could lead to some interesting results.

[asymmetric_implosion]

The big problem with coating is the low temperature tolerance of the polymers. In most dense coating applications like you would need for a plasma focus insulator, you need to be coating at 300-400C. Kapton can’t survive it. CVD methods require high temp (>500C). PECVD can work at lower temp but you are coating a dielectric so charging is an issue. Energetic methods like cathodic arc can work, but the coating thickness is limited to ~100 um because of internal stress. ALD would do a beautiful job but it will take a lifetime to produce a coating of consequence. Other methods based upon atmoshperic pressure deposition are low density or require extreme temperature. Polymers are unlikely to survive explosive cladding. I can’t think of another type of coating method.

[asymmetric_implosion]

I’ve looked into Zirconia before but the problem was finding a vendor to make the part. I asked for quotes on a hat with dimension about ~1/10 scale that LPP would need and no one could make it near theoretical density (the key to getting the other properties). That was a couple years ago so a vendor might exist now. It has many desirable properties compared to alumina other than the sensitivity to temperature. The data I found was zirconia cost more than alumina as well.

A liquid dielectric could be used outside of vacuum to insulate the parallel plates that feed the anode and cathode. In vacuum, how would you maintain the sleeve shape? I guess you could make a water fall but a liquid is more volatile and less survivable than a solid in the presence of a plasma.

Pulser: How thick can you coat the polymers? Are we taking microns or hundreds of microns? It would probably take mm to protect the underlying polymer.

[asymmetric_implosion]

benf: Macor is an interesting material but it is more prone to shattering than alumina. I’ve used it in other plasma facing applications and it doesn’t do as well alumina or alumina silicate.

pulser: A coating is probably not enough to protect against the plasma and UV. The shock from the breakdown can shatter the coating leaving the underlying material polymer exposed. NASA has looked into this problem for spacecraft that use Kapton as a protective coating. Cracking the tough outer coating leads to failure of the polymer underneath. Coating would likely enhance the lifetime of a Kapton insulator but I would guess the lifetime increase would be some fraction of the Kapton only lifetime instead of a factor of 10 or 100.

The insulator is generally one piece (machine or fired) to prevent any potential arcing at the joint. Sapphire would be ideal but it is difficult to make in a hat design. People make sapphire cylinders large enough in diameter and long enough but they cost ~$10K per unit (Saint Gobain) and they need to be made into a hat. Machining adds a great deal of stress to ceramics if not done properly. Few people have done it properly for me, but I might not be working with the right folks.

The insulator-metal-insulator solution is used a great deal for high voltage but the problem is mechanical. 🙁

[asymmetric_implosion]

Kapton is ideal in many respects except for being a polymer. Polymer bonds are susceptible to damage under the intense UV of a plasma focus and the plasma facing during breakdown. The surface will turn into a powder mess that is mainly carbon. Carbon starts moving around the vacuum chamber and hurts the base vacuum. Tests conducted back in the 1990s showed that polymers hurt neutron yield and reduced the gas lifetime for folks that fire more than one shot on a charge of gas. I’ll try to find the reference but I don’t think I have it any longer. The ideal material can resist UV bombardment and plasma facing (alumina, diamond, or other ceramics). It would also need to resist mechanical shock (a polymer would work pretty well) and hold off high voltage (polymer or ceramic). I don’t think a perfect material exists. People have looked for this insulator material like tokamak folks have looked for a first wall material. It might be beyond our abilities at this time. The insulator is the common problem cited by technical reviewers for scaling up the DPF beyond 2 MA. Designs without an insulator exist but they have other problems. Where is unobtainium when I need it?

[asymmetric_implosion]

delt0r wrote: This is not the same as claiming D+D->4He by a long shot. A very long shot.

D+D-> He-4 does happen in fusion. It is a rare branch of the reaction even in plasma. They detect it on NIF as a minority event. I was surprised to see it at the last APS- Division of Plasma Physics meeting.

[Author]

Posts