Cube fusor build

For posts specifically relating to fusor design, construction, and operation.
Jon Rosenstiel
Posts: 1391
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Cube fusor build

Post by Jon Rosenstiel » Thu Sep 12, 2019 5:52 pm

A lab get-together with Joe Gayo and 13” length of 4” x 4” 6061-t6 aluminum courtesy Bruce Meagher was the impetus I needed to embark on this cube fusor build. Start to finish took a little over one month. The following images document the build.

4 x 4 x 13 6061.jpg
4-inch square x 13” chunk of 6061-t6 aluminum courtesy fellow forum member Bruce Meagher. Thanks-much, Bruce!

Boring.jpg
Boring the 1.875” diameter through hole. Lathe is a 14.5” swing Logan.

Tapping.jpg
Tapping 10-32 thread for KF-40 viewport bulkhead clamp.

Cathode 1.jpg
I could not figure out how to make a one-piece cathode using the equipment I have on hand so I built it out of two pieces which were shrink-fit together. (Shrunk-fit??) The “torus” shape was machined using a ½” corner radius end mill mounted in my lathe’s boring bar holder.

Cathode 2.jpg
Shows the cathode in position. Shortly after this photo was taken the cube went back to the mill to have its scratched surfaces cleaned up with the fly cutter.

Cathode 3.jpg
Cathode and endcap. I initially thought that the endcaps were parabolic in shape so I didn’t even consider trying to machine them myself. I ended up getting a set of endcaps from Joe, in trade for a pulse height analyzer. Turns out the dish in the endcaps is not parabolic but spherical with a radius (as near as I can measure) of around 1.75”. Still kind of difficult to machine on manual equipment, but not impossible.

Assembled.JPG
Surfaces have been cleaned up, the cube has been washed in Citranox and then assembled. To the right of the cube are some test cathodes… it was a bit tricky getting the shrink fit just right. After assembly I sat at the workbench with the cube in front of me and wondered what the heck had just happened. Kind of threw me into a funk that the building part took a good month while the assembly part was over in a few minutes. That was it?

Under vacuum 1.JPG
I tried to configure the various connections (vacuum pump port, pressure measurement, gas feed and power) to match that of my normal fusor so swapping fusors would not be too difficult.

Under vacuum 2.JPG
Another view. Under the fusor bench is a smallish squirrel cage blower (about 1/20 HP) with its outlet directed up through an opening in the benchtop. The 10” x 10” x 3/8” aluminum plate the cube is bolted is meant to act as a radiator. It appears that the majority of the heat produced by the cube is produced where the plasma strikes the endcaps, so the best bet may be to directly cool the endcaps, either cooling fins with forced air or better yet, liquid cooling.

Plasma, 40 kV, 10 mA.jpg
Plasma image at 40 kV, 10 mA.


Jon Rosenstiel

User avatar
Richard Hull
Moderator
Posts: 12204
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Cube fusor build

Post by Richard Hull » Thu Sep 12, 2019 7:59 pm

Fabulous effort Jon! We will be interested to hear of the numbers that you get from this plus ultra little assembly. I just gotta' do a cross. I have all the stuff on hand. This maybe the last HEAS event where fusor IV will be the show. Naturally, I'll hold on to it just in case I am not pleased with the future cross effort. It has been a real trooper since 2004.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Rex Allers
Posts: 431
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Cube fusor build

Post by Rex Allers » Thu Sep 12, 2019 10:39 pm

Jon,

Congratulations (again).

I recall that you originally posted this before today (Sep 12). I assume that post was one of the recent ones that was lost when the site moved.

Just for history, do you recall the approximate date you made the original posting?
Rex Allers

Jon Rosenstiel
Posts: 1391
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: Cube fusor build

Post by Jon Rosenstiel » Fri Sep 13, 2019 5:28 pm

Hi Rex,

My original posting went up around the 15th or 16th of August.

Jon

Jon Rosenstiel
Posts: 1391
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: Cube fusor build

Post by Jon Rosenstiel » Thu Oct 03, 2019 11:19 pm

Cube fusor update… NEUTRONS!

First light occurred on the afternoon of August 15th. It was a short 3 to 4-minute run to make sure everything worked as it should. Thankfully there was no magic smoke but there were lots of x-rays and a few chirps from the Ludlum 12-4 rem-ball neutron detector. Awesome!

I managed to get in four runs the following day, total run time was 23-minutes.

Run #1: 4-minutes at 40 kV, 10 mA. (For me 40 kV @ 10 mA is somewhat of a standard as my regular fusor, when fully conditioned, will put out right at 1.0E+06 n/s TIER at a pressure of around 11 mTorr) At this same power level TIER for the cube was 2.8E+05 n/s at a pressure of 16.3 mTorr. X-rays a few inches from the cube’s endcap were well over 5 R/hr. (Measured on an Eberline RO-3C “cutie-pie” ion chamber)

Run #2: 10-minutes at 40 kV, 10 mA. Max TIER for this run was 5.3E+05 n/s at 20.6 mTorr.

Run #3: 3-minutes at 40 kV, 10 mA, then 3-minutes at 50 kV, 10 mA. Max TIER at 50 kV was 9.0E+05 n/s.

Run #4: 3-minutes at 40 kV, 10 mA, max TIER was 6.0E+05 n/s at 20.6 mTorr.

Fast-forward a month and a half and I’m still getting 5 to 6E+05 n/s at 20.6 mTorr out of the cube. I was kind of bummed about the neutron count not improving further, but a look back through my notes from the above runs revealed some real positives that I had previously overlooked.

1. 2.8E+05 n/s on the first run is very impressive. In an unconditioned state my stainless-steel shelled fusor will only do half of that.
2. 5.3E+05 n/s on the second run of the day was really, really impressive considering total run time up to this point was under 10-minutes.

Due to D2 wall loading it takes some time for the pressure in a normal stainless-steel shelled fusor to return to baseline after a run whereas the cube’s pressure drops like a brick. So next step for the cube is a titanium cathode. Let’s see if we can encourage a little wall loading.

Jon Rosenstiel

User avatar
Richard Hull
Moderator
Posts: 12204
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Cube fusor build

Post by Richard Hull » Fri Oct 04, 2019 5:29 am

As usual a great report from a seasoned, trusted veteran among us! Your numbers kind'a bummed me a bit, too, as I trust your numbers sometimes more than anyone else. Your work is always first rate. You might look into the hydrogen absoption rate in aluminum compared to the 304 SS alloy metals. It may not load well or unload easily.

I am a little shaky now on s--- canning fusor IV, but will still press forward with an SS cross fusor as per my recent posting in construction forum.

I will watch intently, you work..

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Jim Kovalchick
Posts: 438
Joined: Thu Apr 14, 2011 12:00 am
Real name:

Re: Cube fusor build

Post by Jim Kovalchick » Fri Oct 04, 2019 10:21 am

I have held skepticism for some of the neutron numbers people have reported using bubble detectors for the small cross fusors because the geometry doesn't make sense to me. So, when Jon reports these numbers using his instruments my skepticism of other people's numbers is further supported.

I think that assuming point source geometry when a bubble detector is parked that close to the shell is flawed. Further we are not even fully certain where all the fusion events are happening. I think smaller fusors compound the error.

Also, are big numbers really that important? I think it depends what one wants to do with the fusor. For example, if activation is the goal, perhaps a smaller fusor allows placing the target with moderator and/or reflector closer such that a greater fraction of the total neutrons are intercepted. In a bigger fusor, the total neutrons are bigger but the target is necessarily further from the poisser, and at that point the flux is lower. Geometry is important.

I haven't seen much activation data from the small crosses reported here. Maybe that's something left to test.

Jim K

Jon Rosenstiel
Posts: 1391
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: Cube fusor build

Post by Jon Rosenstiel » Fri Oct 04, 2019 10:12 pm

Jim,

I think the neutron numbers Joe Gayo has put forth are the real deal. https://www.fusor.net/board/viewtopic.php?f=6&t=12826

Joe and I have gotten together here in my lab on a couple of occasions and I feel he is on the up and up. (And as I recall, Joe does have two other neutron detectors that backup the BTI results)

The cube I built was based off of what was presented in Joe’s video and what little he told me during our visits. (Basically, what I built is what Joe started with, not what he’s currently running) Understandably, I think, Joe is playing it close to the vest and doesn’t want to spill all of the beans. (If it were me I would be doing the same)

In an attempt to determine exactly where the neutrons originated in the cube (center or endcaps) I tried applying the inverse square law to neutrons using a BC-720 replica fast neutron detector, but my results were all over the place. I’m not sure, but I think the poor results may have been due to inelastic scattering reactions, prompt neutron-gamma reactions, absorption in the atmosphere, etc., etc. In any case, I need to look into this a little further.

Jon Rosenstiel

User avatar
Jim Kovalchick
Posts: 438
Joined: Thu Apr 14, 2011 12:00 am
Real name:

Re: Cube fusor build

Post by Jim Kovalchick » Fri Oct 04, 2019 10:48 pm

Thanks for the response Jon.

I think your numbers being all over the place as you say illustrate my concern about inverse square simplicity. I think the relative size of the source, its non-point source nature, and the detector geometries confound it. I suspect MCNP would probably help but only after the true nature of the source is understood.

Regards,

Jim

Jon Rosenstiel
Posts: 1391
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: Cube fusor build

Post by Jon Rosenstiel » Fri Oct 11, 2019 10:36 pm

Jim,
I think I may have figured this out, please let me know what you think.

Off-axis tests using my BC-720 replica fast neutron detector showed that the faint circles seen on the cube’s endplates (image below) roughly define the outer limits of the cube’s neutron emission cone. Plugging the circle’s diameter (1.1”) and the distance to the cathode’s center (1.9”) into an online isosceles triangle calculator gave an apex angle for the emission cone of about 34 degrees.
Endcap, 800 x 480.png
Endcap, 800 x 480.png (884.11 KiB) Viewed 2009 times
So anyway, while looking at one of the endcaps the thought occurred to me that if one halved the emission cone’s 1.1” diameter (through clever cathode design and/or magnets) the neutron output would, if everything else stayed the same, quadruple. (Same number of neutrons packed into a much smaller area)

And one more thing… Using TIER (Total Isotropic Emission Rate) to quantify the cube’s neutron emission rate is not correct because the cube is not emitting neutrons in all directions. Question is, how should we go about comparing the traditional fusor to something that’s anisotropic?

Jon Rosenstiel

Post Reply