New developments?

[break]

Hi,

any new developments in the last month?

What are the problems you’re facing now and how is your timetable for this year?

break

[wolfram]

Patience padawan, the polywell crowd has gone something like 2 years without meaningful updates. If they don’t post anything in January, they’ll probably post something delightful in early February.

[DerekShannon]

There will be a January report. Unfortunately we have encountered some new engineering issues, so don’t expect any major breakthroughs, but there will be some good news to share. In the meantime….try to help beyond being an internet commenter, I don’t know how exactly, just figure it out ;-D

[vansig]

any hints on the engineering issues to be solved? we could make suggestions, maybe

[DerekShannon]

A *tougher* (shatter-proof!) insulator material than alumina, but still with as good or better dielectric strength and heat-resistance, pretty pretty please? Oh, and it has to come in the shape of a hat, with no delivery lead time, and people pay to have it taken off their hands ;-D

[wolfram]

Also, the twitter account has a link to the new 2011 report, a delightful summary with some numbers to stew on. Commercial supplier coming up with the next batch of switches could double as being a prudent step 1 to mass roll out, having the specs for important components already worked out and ready to mass produce. You know, should they work as expected.

[Warwick]

I know nothing about this, but graphene?

[annodomini2]

Graphene is a conductor not an insulator.

[jamesr]

DerekShannon wrote: A *tougher* (shatter-proof!) insulator material than alumina, but still with as good or better dielectric strength and heat-resistance, pretty pretty please? Oh, and it has to come in the shape of a hat, with no delivery lead time, and people pay to have it taken off their hands ;-D

I’m not sure about their physical properties (and I’m sure any good one won’t come cheap) but how about some of the insulators mentioned in:
http://www.apelc.com/pdfs/6.pdf

After all, you cannot assume an insulator that has good properties at DC or low frequencies will react in the way you want with the fast rise time in a DPF. So looking at what they use in the Radar & microwave world may be of use.

[asymmetric_implosion]

Sorry to hear the insulator went. I guess it will be 4-8 wks before more data is taken while the insulator is made unless you have a better supplier than I do. I use the hat insulator design as well.

Only one material that I know of has better properties for a plasma focus insulator than alumina and it is far from cheap; diamond. Many other materials have been tried but each has a problem. If you are looking for a cheap solution that works OK, I suggest alumina silicate. I’ve use it in my PF as a cheap alternative to alumina when I’m trying something new. Alumina silicate is easy to machine but it is not as mechanically robust as alumina. You need to be careful about weight distribution on the disk part of the hat. I’ve broken more insulators installing them than by running the machine when using alumina silicate. Other materials that might be useful are zirconia but last I checked no one could make it in the shape you need. Diamond has the same problem. I know people love glass and quartz as an insulator but I’ve never had a good experience with either. Quartz needs to be heat treated properly and no one has gotten it right for me. Glass works OK but more fragile and more expensive than alumina silicate. I can suggest a vendor but again, some change to the cathode might be required because glass cannot be made with a tight corner at the disk-cylinder interface and be strong.

My best suggestion is thicken the alumina wall where it is breaking. I know it’s a pain for the pulse power design but it does help. I’ve run a single alumina insulator at 0.25 MA for over 25,000 shots from ~5000 shots by increasing the wall thickness by 20% in the vertical section of the hat. If you are breaking on the flat disk section, you have a tolerance problem or a weight distribution problem. Both are a pain to resolve. To my knowledge very few machines above 0.5 MA have an insulator lifetime of more than 1000 shots. Nasty plasma environment coupled with shocks is not a recipe for a long insulator lifetime. High grade alumina (>99.6%) might have some benefits as well given that it is closer to sapphire, but I use >93% with great success.

[Lerner]

We need to be able to accurately calculate how to avoid the breakage of our alumina hat insulator as we move up to higher currents. So we’d like help from all you mechanical engineers out there to get us some data about the impact fracture of alumina under pre-existing stresses.
In the last breakage, we had inadvertently left a stress of about 1 MPA on the insulator. This is of course small compared with the static tensile strength of 267 MPA. During the pulse, the magnetic field imparts and impulse to the upper anode assembly which rises slightly and then comes back down. This then conveys an impulse (cushioned by rubber) to the insulator. Even ignoring the energy absorbed in compressing the rubber the total energy available from the motion of the upp0er assembly is only 0.02 mJ/cm^2, far less than the fracture energy of alumina which is 2.5 mJ/cm^2.
So the question is: is it possible for such a small impact, combined with a small pre-existing stress, enough to cause fracture? Are there tables showing fracture energy as a function of pre-existing stress for alumina? Or am I leaving something out of this calculation?

[Lerner]

Question to assymetric implosion: Did you ever calculate the stresses and impulses invovled with your insulator? We are running at 1 MA and intending to go above 2 MA, so we need to know in advance what tolerance we need to achieve.

[asymmetric_implosion]

Lerner wrote: Question to assymetric implosion: Did you ever calculate the stresses and impulses invovled with your insulator? We are running at 1 MA and intending to go above 2 MA, so we need to know in advance what tolerance we need to achieve.

Simple calculations were done but I wasn’t the one that did them. Our cathode (anode pointing skyward in our case) does not move the hardware at 0.25 MA in an appreciable way as determined by experimental testing. The calculations supported this conclusion.

I would point out that imperfections in the alumina can cause problems. I’ve encountered “bad” alumina that has voids or smaller scale imperfections in the past. I would like a second opinion from one more knowledgeable but I speculate that local defects increase local stress making it easier to break. Once a fracture starts the insulator is broken in a shot or so. This was my hypothesis with quartz. I could fire 1 or 2 shots before it breaks. My best suggestion is to increase the alumina thickness and buy the highest grade alumina which should have minimum defects. I use 1/4 inch alumina without any problems at 0.25 MA. My guess is other 1-2 MA machines use much thicker insulators to avoid the shock problem.

I mention three other options but all have problems and cost money. The first is to tie the anode and cathode together mechanically using nylon or some other insulating ties to limit the motion. The problem is how to tie the two together without flashing over. The second is to make the plasma focus a tri-plate instead of a bi-plate. If you carry the cathode current in two plates that surround the anode (cathode-anode cathode sandwich) you should be able to balance the force on both sides of the anode and it should reduce the impulse. The same problem as the first; how do you get the current to flow through the anode without a short. Techniques like these are used on >2MA Z-pinches which have large impulses. Tri-plates are the norm for large machines to reduce the impulse of the current pulse and reduce inductance. The problem with tri-plates is they tend to flash on large voltage spikes (like the pinch) splitting your current. The third possibility is to change the insulator design from a hat or reduce the radius of disk. I found that reducing the disk radius improved insulator survivability during assembly. I typically make the disk diameter just 10-30% larger than the outer diameter of the cylinder of a hat insulator. It is just enough space to put a thin o-ring in place on both the anode and cathode side to maintain the vacuum seal. I wish I could say more but I am straying into proprietary design information.

[Henning]

Somewhere else in this forum I’ve suggested Kapton (aromatic polyimide) as a possible insulator.

German Wikipedia says aluminium oxide has a dielectric strength of 35 kV/mm.

Kapton has a much better dielectric strength (150-300 kV/mm, table 7 of PDF). Some more details are in http://www2.dupont.com/Kapton/en_US/assets/downloads/pdf/summaryofprop.pdf

But also see figure 8 “Retained Dielectric Strength at 325°C (617°F) for 25 µm (1 mil) Film” and figure 14 “AC Dielectric Strength vs. Temperature”. The temperature was applied continuously, so the aging shouldn’t be such a big problem in a pulsed environment. (Maybe it is, because of mechanical strain.)

Kapton is sold as sheets, but it’s also available as tubes, for example at http://concentrictube.com/ . Maybe they know how to manufacture them as hats.

On the other hand, if it’s such a good material, the other DPF scientists would have used it already. So maybe this suggestion isn’t that useful at all…

[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?

[Author]

Posts