Grid Shielding - Testing with Results

It may be difficult to separate "theory" from "application," but let''s see if this helps facilitate the discussion.
Post Reply
davidgrzan
Posts: 11
Joined: Tue Aug 28, 2012 8:55 pm
Real name:

Grid Shielding - Testing with Results

Post by davidgrzan »

Today was the award ceremony of the regional science fair I competed in so I decided to share what I found during my research. I built a demo fusor and tested grid shielding methods as it was one of the few areas that could be tested without fusion occurring. I received a couple awards and was selected as an alternate for the research paper category but was not selected for the project category. I guess I was slightly disappointed that I didn't get to move on to the next stage because the judges and people there seemed to be very interested in it which got my hopes up.

I wanted to use the anode to deflect ions from impacting the grid and while searching through the forums, I found that Steven Sesselmann had came up with a good idea, using baffles lined up with the grid wires to deflect the ions along its whole path. I constructed it and cranked up the voltage to see if it had any visual differences compared with the control (you can see it in the first picture). The control grid is just a regular copper tubing grid with three loops. To test the temperature of the grid quantitatively, I came up with the idea of attaching solder to the grid to see how long it would take for it to melt. Given the melting points of the soldering wires, I could compare how the control and shielding methods worked.

Control time to reach 577 degrees Celsius: 101 seconds
Plate or baffle method: 87 seconds

The control worked better as the inner grid had a slower rate of heat transfer. I think the plate method heated up faster because of the stronger electric field in between the cathode and anode. This would cause more field emission or electrons coming off of the cathode and heating it. The plates could be deflecting ions but it still is wasting energy on heat.

The second method to shield the grid is one I came up with. I decided to try to use neodymium magnets to distort the plasma around the grid wires so that almost none of it would impact the grid. You can see the pattern of the plasma caused by the configuration of magnets I chose in the second picture. It was distorting it how I wanted but the grid wires were now in the zones that were void of plasma. This means that the ions were still heating the grid and the results showed that.

Control time to reach 577 degrees Celsius: 101 seconds
Magnetic field method: 62 seconds

I ended up trying to move one of the magnets closer as I could not move all of them closer because of the way the apparatus was set up and of how strong the magnets were. You can see the picture of the righter most magnet being moved closer to the right part of the grid (picture 3). It is at a lower voltage and the plasma can be seen going around the grid wires. Increasing the voltage decreases the zone where there is no plasma to outside the grid. No matter where I put the magnets, the zone void of plasma hovered just outside the grid (at the higher voltage, 7kv, that I was doing the tests with). I tested it anyway and the magnet that was closer did better than the magnet that was farther away.
I hypothesized that with further tests, if stronger magnets were moved closer to the grid, there would be a decrease the rate of temperature change.

Basically that's just a brief summary of the important results so if any of you have any questions I'd answer them or if you found something I might have missed or done wrong let me know. I'm most likely willing to do another test if you have a suggestion.

Also I'd like to add that this forum and the people here were very helpful to me when doing my project and I thank you guys for that.
Attachments
lowervoltage2.jpg
magnets2.jpg
plates3.jpg
Tyler Christensen
Site Admin
Posts: 551
Joined: Mon Feb 23, 2009 9:08 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by Tyler Christensen »

Was cathode current relatively unaffected between control and test? If it was affected, since you were running constant-voltage it sounds like, you might want to normalize the times to wattage. I wonder if that would change the outcome.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Grid Shielding - Testing with Results

Post by Dennis P Brown »

Extremely good experimental work and a really clever approach to non-fusion work. I'd love to see your paper. Sorry you didn't move to the next stage but as an experimenter you really demonstrated fantastic abilities.
User avatar
Jim Kovalchick
Posts: 717
Joined: Wed Apr 13, 2011 8:00 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by Jim Kovalchick »

Trying something new is important, and you should be proud of your work. The reality of science fairs often means that adventurous but non-mainstream projects suffer in the judging. Putting cow poop in a bucket and observing it give off methane while claiming that it is green energy will often beat nuclear fusion depending on the judges. Last year, my son's neutron producing fusor made it to ISEF, but only after narrowly winning his high school fair to a project that tried to use rotten pumpkin rinds as a water filter. The pumpkin project, which was well written, came across as very 'green.' A fraction of the judges will understand your work. Don't let that part ruin your accomplishment. Very few students are willing to step up to scientific challenges like you have. Be very proud of yourself! Congrats and well done.
davidgrzan
Posts: 11
Joined: Tue Aug 28, 2012 8:55 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by davidgrzan »

I've attached my research paper if anyone wants to see.

In response to Tyler Christensen, I set up a voltage meter but not a current meter. So are you saying that the temperature of the cathode affects the current if I were to measure it? If that's the case it would be a great way to measure the temperature indirectly.
Attachments
Effects of Cathode Shielding Methods in an Inertial Electrostatic Confinement Fusion Reactor 2.pdf
(716.3 KiB) Downloaded 538 times
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Grid Shielding - Testing with Results

Post by Dennis P Brown »

Just skimmed through your paper - very impressive. Thanks for putting it up. I think a number of students should read this for a "How to do a good research study" and writing up the results. I hope to have time to read it in detail later this week.
Tyler Christensen
Site Admin
Posts: 551
Joined: Mon Feb 23, 2009 9:08 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by Tyler Christensen »

Not the temperature affecting current. My thought is what if the shields or magnets affect the current draw (which it probably does). I don't have time to read through the paper right now, but it looks like you concluded your ideas failed to improve confinement because the grid got hotter even faster. But what if the magnetic system draws four times the current of the control. Then it actually would be dramatically helping confinement away from grids, per watt, even though the grid heats up even faster due to the higher wattage.
kcdodd
Posts: 28
Joined: Thu Sep 20, 2012 12:29 am
Real name:

Re: Grid Shielding - Testing with Results

Post by kcdodd »

If the magnetic field is mostly trapping electrons, making a better ionizer.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Grid Shielding - Testing with Results

Post by Dennis P Brown »

Really liked your method to measure temperature - highly creative! Sometimes lack of money works in very novel ways.

Can't help if an experiment produces somewhat mixed results - that is science. Very clear, well written paper and logical - as the saying goes: say what you mean and mean what you say. That works. Following an outline like you did pays. You should be proud of your research and the paper - few in HS do such excellent work.
ab0032
Posts: 86
Joined: Sun Oct 23, 2011 2:50 am
Real name: Alexander Biersack

Re: Grid Shielding - Testing with Results

Post by ab0032 »

How beautiful.
davidgrzan
Posts: 11
Joined: Tue Aug 28, 2012 8:55 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by davidgrzan »

I think I get what you are saying. Basically if there is more current being drawn there will be more ions, making the grid heat up faster. But more ions means more fusion, helping efficiency.

On another note, I was moved up to the regional JSHS (Junior Science and Humanities Symposium) for my research paper and will be competing there in April.
davidgrzan
Posts: 11
Joined: Tue Aug 28, 2012 8:55 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by davidgrzan »

Are there any simulators out there that could replicate the setup I have here with the magnets? I've tried using a few ones found on the internet but they don't have the flexibility of making something similar to this. I think it would help a lot to try different things on a simulation before implementing them in my demo fusor. I found equations for the B field of the magnet at a distance extending from the poles but not from any other point around the magnet and the lorentz force equations. I didn't have any luck with determining the ion's path with the equations I found though.

I did notice however, that the magnets behaved very similarly to solar wind hitting the Earth's magnetosphere. The magnetic field traps the charged particles creating that protected "zone" seen in the first picture and in my setup. Where the solar wind enters the cusps in the picture can also be seen on the magnets in the fourth picture.

I did another test with a tighter arrangement of magnets. The next two pictures shows the visual results (it looks white because of something burning off of the metals which usually goes away after a few runs). I figured that if I moved the magnets closer to the grid wires, the magnetic field would be strong enough to get that zone around the wires. The area void of plasma seems to hover just outside the grid wires though. I'm thinking it's because of how fast the ions are moving at that point or because the plasma needs to act as a conductor. Does anybody know the exact reason of this?

I'm thinking about insulating the magnet holding apparatus from the grounded plate and testing this method at a lower voltage next. Any suggestions are welcome.
Attachments
IMG_0775.jpg
IMG_0773.jpg
SolarWind.gif
IMG_0771.jpg
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Grid Shielding - Testing with Results

Post by Chris Bradley »

There is a programme called Maxwell which is free as the Student Version ('Maxwell SV'). It will do 2D DC fields.

However, if I may ask whether you have a main objective, or is it just to 'look around' and understand a bit more about plasmas near magnets?

The reality is that a configuration such as this is not likely to improve a fusor. The reasons are that
a) what you are seeing as glow discharge are the trapped electrons and slow ions, whereas fast ions will scoot straight past those magnetic fields almost as if they were not there, or worse still they are deflected as they slow down and reverse direction, leading to drifting out of the area of maximum electric field potential,
b) field lines from solid magnets always start and finish at the magnet, so any charged particles will flow 'along the field lines' and intersect the magnet, which will heat it up markedly and result in the biggest losses.

As has been speculated above, what you may be doing is simply diverting losses from one place to another by redirecting those electrons.

The presence of magnets will cause higher levels of ionisation, as you have mentioned, but it is not easy to understand where, in the voltage potential gradient, those ions form. This is important because it will determine the peak energy of the ions once accelerated. With magnets, you will also be able to continue with a glow discharge at pressures lower than without, which again is due to the trapping of electrons rather than their 'immediate loss', although they will still be lost in quick order once they work their ways along the field lines to the magnets. Lower pressures mean fewer fusible collisions, in time.

Plasmas are unpredictable at the best of times, though, because of all the potential complex interactions (electrons follow ions, then the ions follow the electrons, like a big conga dance, then some drop out of the dance while others join). So you will never get bored with trying different arrangements, and maybe you'll find something unexpected.
davidgrzan
Posts: 11
Joined: Tue Aug 28, 2012 8:55 pm
Real name:

Re: Grid Shielding - Testing with Results

Post by davidgrzan »

Chris Bradley,
Thank you for the simulation idea I will download it soon. To answer your question, the experiment was to test different methods of shielding the grid from ion impacts. The magnets were supposed to create an area void of plasma around the grid wires, decreasing the amount of collisions. It obviously had some problems like you mentioned above unfortunately.

Even if I could get a little less heating to the grid at lower voltages I would have to conclude it to be not worth implementing in a real fusor.

I insulated the apparatus and the pattern became much more clean so I took another picture for those who want to see. The zones which are hovering outside the grid wires are more noticeable in this picture too.
Attachments
IMG_0781.jpg
Post Reply

Return to “Fusor and/or General Fusion Theory (& FAQs)”