This device looks really cool from Berkeley lab

For posts specifically relating to fusor design, construction, and operation.
Post Reply
steve_rb
Posts: 133
Joined: Sun Oct 18, 2009 8:18 am
Real name: Steve Robinson
Location: Tehran University

This device looks really cool from Berkeley lab

Post by steve_rb » Sun Jan 17, 2010 3:45 pm

Diging old posts I found this:

http://www.lbl.gov/Tech-Transfer/techs/ ... ylindrical

Looks really cool to me. It worth trying. Anyone has more experimental detailes on this? Or higer resolution photos?

Steve

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: This device looks really cool from Berkeley lab

Post by Doug Coulter » Sun Jan 17, 2010 5:22 pm

I'd seen this too. It's more or less a fusor, inside out. This creates a fairly easy to make beam-on-target device with some desirable properties for that class of thing. As the target is large, and outside (could be the tank walls or tightly thermally coupled to them) you can cool it easily -- most targets lose their embedded fuel if they got hot. However, since it is a beam on target vs a beam on beam device (as fusors *hope* to be), it needs higher voltages to operate well, with the attendant higher energy X ray output.

Due to having a large unfocused beam area, spot heating of the target is minimal -- see above. Most beam on target devices do have higher net Q than fusors because of the higher density of at least one of the colliding entities, in this case the target, so atoms in the target are a lot easier to hit than atoms of a gas at near-vacuum. The fact that the neutrons will come out from "all over" could be either advantage or not depending on what you want.

The design looks adaptable in a few ways, you could probably use other kinds of ion sources for example, but the big deal over a simpler borehole tube here is that it's easier to keep the target cool, so the D or T it is loaded with isn't driven out faster than the beam can drive more in.

But in the final analysis, the thing is just a clever way to do beam on target fusion. If all you want is a ton of neutrons, and don't mind the higher energy X rays, it's probably a good way to go.

Note, even in a fusor, when a lot of fusion is going on, significant amounts of pretty high energy X rays come out that are higher than the power supply volts would indicate, but I think that's not as big an effect compared to just starting with a higher supply voltage in the first place.

For example, this might be a lot happier at the 125kv input region, where a fusor will be fine in the 40-50kv region, over an octave lower energy main X ray output. Of course, it depends somewhat on what the target materials are how many full power supply energy X rays will be produced as well.

We looked into making one here, but decided that our first efforts in this area would be a simpler design ( like the phillips borehole, www.coultersmithing.com/pdfs/Pch8.pdf ) that we could make in a quartz tube and hook up to the main tank via
a tubing coupler off a side leg of the main beam tube. The idea there is that the two ends of a conventional beam on target tube could be opposite polarities of power supply at say plus and minus 50kv or so, eliminating the need for having a 100kv supply exposed to air and feedthroughs in a metal tank, which gets very problematic in practice if you're not willing to do things all under oil (and having done that, we're not).
Why guess when you can know? Measure!

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: This device looks really cool from Berkeley lab

Post by Chris Bradley » Sun Jan 17, 2010 5:40 pm

I think the author of the patent is just a little over-inflating this above the current state of the art. It is focussing on compactness, which is fair enough, but it says; "As a broad general principle, "comparable" conventional neutron tubes (e.g., with diameters of several tens of cm and lengths of up to a few hundred cm) produce 10^6 to 10^8 neutrons per second in D-D operation."

I didn't realise a meter-sized device was needed for 1e6 neut/s!?!

The power on the outer of the device says 650W/cm^2 in the patent. I guess you can figure the power input to the thing - prob no difference to a regualr fusor if you run the numbers....

I suspect that german firm will have something to say about such a claim. I thought they were up to 1e10 n/s in a 20cm device.

Also to note - nowhere does it actually rate the total neutron output, it only says 'are expected to be'.... and ITER is expected to be in the 500MW of neutrons range, oh yeah, we know that!....

steve_rb
Posts: 133
Joined: Sun Oct 18, 2009 8:18 am
Real name: Steve Robinson
Location: Tehran University

Re: This device looks really cool from Berkeley lab

Post by steve_rb » Mon Jan 18, 2010 4:23 am

I have seen they have sold the device to a few countries and I also in Tak Pui Lou thesis it says neutron is measured > 1e9 n/sec for 30cm diameterX40cm long tube. It is obvious neutron count depends on how big is the tube (the bigger the tube the higher HV should be of course). One important problem I see here is optimisation of gas pressure, plasma chamber dimensions, size and numbers of the holes on the plasma electrod and R input power. If optimised efficiency of neutron generation will increase and can go up to 1e10 n/sec even for 20 cm long tube. This optimisation will take a lot of time and effort and need a team of experts on calculations and design and needs good softwares too. If this guy has spent 10 years or so on this with aiming on optimising these factors they may already achived high neutron levels.

Anyway I am seriously looking into this and willing to try it. I was lucky I found two second hand but in a good condition HV power supply (0-120KV and 0-50Kv) purchased for very low price and they are sitting in my lab since 6 month ago. Anyone willing to help please post you comments how and where to start. Specially I need dimensions to start building parts, specially with plasma chamber diameter and lenght and number of holes and their configurations. For start I am thinking about 30 cm diameter and 40 cm long with about 32 holes (4X8) with 1.5 mm diameter of each hole in equal spacing from each other. I am not sure about target but I think I will cylindrical copper (1-2 cm thick) with about 7 cm distance from plasma wall and water cooling. I am not good myself in calculations and I mostly worked on trial and error basis. But I am sure here can get very good advice from experts regarding this.

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: This device looks really cool from Berkeley lab

Post by Chris Bradley » Mon Jan 18, 2010 6:13 pm

steve robinson wrote:
> I have seen they have sold the device to a few countries and I also in Tak Pui Lou thesis it says neutron is measured >1e9 n/sec for 30cm diameterX40cm long tube.
Well I never!..

Much the same, then, as NSD's high end DD product of a slightly smaller size, and *after only 10 years of research*!!

http://www.nsd-fusion.com/2.5mev.php

Maybe cheaper and quicker to have bought one of NSD??...

steve_rb
Posts: 133
Joined: Sun Oct 18, 2009 8:18 am
Real name: Steve Robinson
Location: Tehran University

Re: This device looks really cool from Berkeley lab

Post by steve_rb » Tue Jan 19, 2010 4:11 am

What is this? Is it a Bazooka ? With a device this long Leung team could generate 10^12 n/sec. Even more neutrons with nested option.

Any idea what is inside this?

lutzhoffman
Posts: 188
Joined: Tue Dec 08, 2009 1:59 am
Real name:

Re: This device looks really cool from Berkeley lab

Post by lutzhoffman » Tue Jan 19, 2010 10:54 pm

Hello:

Actualy this long device mentioned by Chris, is closer to a fusor, with a cylindrical grid if I remember correctly from researching neutron sources a while back.

Edit: Here is a cut and paste of the technology, from the link provided by Chris, I went back and looked after posting, so I remembered correctly:

(Our Core Technology
The NSD neutron generator is an improvement to the spherical Inertial Electrostatic Confinement (IEC) device. It provides a linear geometry source.
IEC is the simplest way to achieve sustained nuclear fusion. While the possibility to scale the technolgy up to achieve more power output that input can be debated, there is no doubt that the technology provides a credible neutron generator without the disadvantages of a solid target.)


This is even more interesting than the Berkley device. The Berkley DD device is not very impressive when viewed in the context of efficiency. To get 10 to the 12th N's they needed something like 1 amp at 100-150KV. So in effect they are getting only 10 to the 10th N's per kilowatt.

Now compare this to existing mature technology like Carl's really cool toy, the 5MEV accelerator for example, if it were run with D+

5MEV D+ on Be, yield = 10 to the 10th N's per uA! So at 1 ma beam current from Carl's liniac, this would yield 10 to the 13th N's at only 5KW, thats 10 times the Berkley figure at 5KW. Or 2 times the Berkley figure of N's at 1KW at 1/100th of the power level of the Berkley device.

I am sure that Carl would run for the hills, or "hit the red button", if the accelerator he works on had an "1ma current excursion", with D+ on a Be target, or for that matter any low atomic number target : )

Just out of curiosity what is the max beam current on the 5MEV RF liniac accelerator? (In case Carl is reading?) It would be good fun to fire the 5MEV D beam into a vat of heavy water, and video the result. Sorry its just a personal thing I enjoy the appearence of ion beams in air, and other matter.

Yes I know a 5MEV liniac is not 100% efficient, but I would bet that a RF liniac can reach 1 ma beam current, with under 50-100KW of average power input. I will stand corrected by Carl, if I am wrong about this. For the same N output the Berkley device would need 1MW. Plus to equal the Berkley D-D neutron output, Carl's liniac would only need 100uA of beam current.

I think the achievment of the Berkley device is in how the ions are produced, it would be very hard to make a 1000ma ion source for 100-150KV. Their work on RF ion sources is impressive. The rest is however pretty mundane. The other plus point of the Berkley work is in tiny N sources for cancer therapy for example, where the source is inside of the patient.

The linear neutron source with the cylindrical grid from NSD in Germany is however very interesting since it can be duplicated by folks in this forum

PS: The NSD web site has a neat demo of the concept, if you click on the technology on the provided link from Chris. Its a linear fusor in a basic sense, but it may be more easy to build, and to scale, due to it being very hard to find large spherical vacuum fittings. Linear fittings are easy, you even have large 4-6" x 48" Quartz tubes, with vacuum fittings, on ebay from the semiconductor industry from time to time.

steve_rb
Posts: 133
Joined: Sun Oct 18, 2009 8:18 am
Real name: Steve Robinson
Location: Tehran University

Re: This device looks really cool from Berkeley lab

Post by steve_rb » Wed Jan 20, 2010 5:33 am

As I remember Berkeley figure was 10 to the 12th at 2.5 KW. Correct me if I am wrong?

Is it possible to have a drawings of NSD inside grids or any photos of it? I don't think it is simpler than Berkeley device. When diameter decreases that makes it a hell of a difficult to build.

Steve

User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 11:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: This device looks really cool from Berkeley lab

Post by Carl Willis » Wed Jan 20, 2010 10:02 pm

Hi Lutz,

I think all these technologies have their place.

RF linear accelerators offer a lot of benefits for low-energy / high intensity neutron sources, since they can put a lot of current into endothermic high-yield reactions like Li-7(p,n) at energies not too far above threshold. They'd be well-suited to a high-energy neutron source based on Li(d,n) reactions as well. Large linacs are also used in spallation neutron sources.

Cyclotrons are well-suited to producing high intensities of high-energy neutrons from Be-9(d,n) or C-12(d,n) or Be-9(a,n) using low beam currents. Linacs can be competitive here too, depending on the specifics.

DD and DT neutron generators are compact and efficient sources of monoenergetic fast neutrons. This is in opposition to linacs, which either need insulation for high DC voltages or require big RF powerplants, and cyclotrons, which need bulky, inefficient magnets as well as RF.

Ka Leung's background is ion sources and his neutron generators are based on very intense, magnetically-enhanced RF discharges. All the ones I've seen are of the ion source-target variety, implemented in various geometries.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277

Post Reply