All journeys

For the VA Tech fusion team to discuss and display their work

Re: All journeys

Postby Doug Coulter » Sat Mar 26, 2011 2:41 pm

To your first question, yes, D will give a nice reddish color with air (nitrogen). You saw what it looks like when pure here. Sadly, it's hard to go by the camera pix I have up, as the camera messes up the color a little. A hydrogen glow at medium high pressures is kind of red, it gets more purple as the pressure goes down.

You've got more than one thing going on at a time here. Your system hasn't been "conditioned" yet, and so it will act fairly strange until that process is done with, just run the thing and it will get better. I've seen that weird stuff with voltage rise here too -- nothing too weird going on there, just knocking invisible bits of dust and dirt off things.

Once that is complete, you need a vacuum that can completely shut the glow off at full voltage. You should just set the supply for 30 or 40kv and a low current limit, say 5 ma or so, and try to get it where it's drawing nothing at full voltage -- if you can do that, you're there. It will only take a tiny amount of gas over the full-off vacuum level (which you should note on your gage when you find that spot where it just goes out all the way -- it's more accurate and repeatable than most gages!) to be in fusion territory. When you can get that -- try running about 10 ma, which for us is right on a "sweet spot". Remember 10 ma at 40kv is 400w -- so things are going to get hot quick -- pay attention to that so you don't melt things. If you can hold 20kv with D in there, you'll start having enough neutrons for your detector to count, not many, but some. It will go up hugely with voltage (it's still going up fast when I hit 53kv, my current maximum).

Here we put some black hi-temp paint on the tank (sold for hot rod headers) so we can use an IR thermometer to watch that. It's easy to get to a couple hundred C if you're not paying attention, and some things may not like that very much. IR thermometers don't read right on stainless steel -- it has a weird emissivity. They'll read quite low on it. You could also put a thermocouple here and there if that's more to your liking. I like the IR because being able to do it from a distance is kind of neat (and keeps me away from the HV) and you can scan it around and learn things easily.

Once you've had D in the tank, there will always be a little in there -- driven into the various tank metal components, which will come back out when they are heated. I see it on my mass spectrometer here anytime I get the thing hot, even if I haven't run for a long time. Doesn't take much!

Sounds like you're really really close now. I'd start going up in voltage, working your way up slow, which means adjusting the gas pressure till you can get to say 20kv (with it "lit off"), then 30, then 40. You'll see some spitting and sparking -- run at each voltage till that stops, then move up (by lowering the gas pressure). You do want to do that with D or hydrogen. At first, this takes awhile to get done, but after it's been done, it's good till you open the tank again, more or less. It takes my new grids about 30 minutes of conditioning before they get "good" -- but only if that's the only thing that changed, and I didn't have the tank open for more than the few seconds it takes to make the change. Leave it open for a day, and you start all over again, and that takes hours.

You can do this with the supply set for the full voltage, and some low current limit, and get all your levels conditioned by nothing but changing the gas pressure -- that's what I do and it's the quickest and easiest way I've found. You just watch the voltage on the supply while you adjust the gas stuff.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Lee Hall » Mon Mar 28, 2011 10:14 pm

We finally got the parts to finish assembling our fill side today. We have one charging reservoir for D2 only and one for any other gas we want to put in there (Argon at the moment). I installed everything today and let the pump pull it all down to a vacuum (I hope). As i left, the pump looked like it had stabilized around 10.1 mTorr, but I'm hoping its just something outgassing since it hasn't been at a vacuum up to this point.

For more pretty pictures, I included a screen shot of the reddish plasma we've gotten this past week. Earlier, it had a lot more red to it, but as we brought the pressure up, it started to disappear.
Red Plasma_1.JPG
Reddish Plasma (some D2 and Air?)
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby Lee Hall » Tue Mar 29, 2011 5:28 pm

We've got the vacuum and fill system 'fixed.' Once we have gotten a date that works for our adviser and the university's radiation safety officer, we'll run some deuterium through the system.
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby Doug Coulter » Wed Mar 30, 2011 9:11 am

Good news! Looking forward to results! At the point where you can get above very roughly 20kv at a decent current (say, 10 ma) your neutron detector should be counting solidly above background levels, and it gets better quick from there up. If you have access to one, a gamma ray spectrometer head might tell you interesting things as well (not a fancy cryo one, just an NaI:Tl would be fine). One advantage of that is (in general) it will show you lower energy X rays than most other detectors, so you can look for:
Shine through of X rays at the power supply voltage/energy level. This shouldn't be a big problem for you -- 40kv mostly will stay inside the SS tank, but will come out the window.
Some higher energy gammas from fusion reaction byproducts hitting anything high Z in the tank. Some of these might also show up as "bright" pixels in your camera, so look for that at least.
The lens of course won't focus them, but it will look like what we called "snow" back in the days of analog broadcast TV -- and a little means a lot of gammas, CCD's aren't that sensitive to this.

Don't tell whoever you borrow an NaI head from it's going to be close to neutrons....if abused, you can activate the I in there and make the scintillator noisy for awhile.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Lee Hall » Mon Apr 04, 2011 8:51 am

Alright so we finally have some results. We ran deuterium through the reactor for about 2 hours on Saturday, April 2nd. I was told to deviate from our original test plan, so we had a couple of overcurrent trips as we tried to start a plasma at -19 kV. Finally we were able to go with our original test plan, which was to power up to -10 kV or so and slowly leak in deuterium. When we did this, we finally got a sustained plasma. However, the closer we got to -20 kV, the worse off our current got. The thing was lighting off around 27-28 mTorr at -10 kV. Around 14 mTorr, I was supplying about -22 kV, but the current was not easily controlled. We definitely got some x-rays out of this thing, but our highest neutron count over a 3 minute time range was 1148, where our highest background reading was 252 over 3 minutes. It was told to me that we needed to get to 20 mA to get some decent neutrons out of this thing. I was under the impression that we were aiming for higher voltages and currents less than 10-15 mA.

Problems I had during the test. At some point, the pressure would spike without any adjustments to the fill needle valve. I believe this continuously happened at currents higher than 14 mA. This pressure, I believe, resulted in higher currents being drawn and eventually caused the power supply to overcurrent trip. I'm not sure how to keep this in check. Another problem is the fact that we were attempting to get to 20 mA of current. This resulted in our spiral glowing bright red and its eventual failure, as seen below. What is the deal with this thing? Are we aiming for this high current or are we looking for higher voltages at lower pressures? Either way, I need to present some sort of 'scientific' argument to convince our adviser over one way or the other. I have some videos on youtube now with "VTec2H Fusion" as part of the title.
Bent spiral_small.JPG
The results of the currents we drew through our 0.045" diameter Ti spiral

glowing spiral_1.JPG
A picture of the cathode during the test. Bright red wire.

So here is a link to one of the videos from our test. This is the point in time where we had the most activity from our geiger detectors. In the background, you can hear me calling out some of the voltages and also hear the beeps of the geiger detectors. The glow fades in and out as I save it multiple times, but eventually i saved it too much and drew too much current. General numbers from our test: highest voltage was -26.2 kV, highest current was ~30 mA, pressures ranged from 28.5 mTorr to 13.5 mTorr.
http://www.youtube.com/watch?v=qMtOFYQeNvY
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby Doug Coulter » Mon Apr 04, 2011 10:18 am

Great! You're on the road now :D

Here's what most people don't understand. You're not in a pure-plasma environment, and the fusor isn't a resistive load that follows ohm's law with a fixed resistance. In general you have a lot of neutral gas atoms/molecules in there, up to any current you could actually run with materials that exist (even if you could get them).

So, at some gas pressure, you try run in say 20kv. It draws a lot of current. You try adding more current to get the voltage to come up. When you do that, more of the neutrals are ionized, so you have a better "conductor" in there -- the effective resistance goes down! And trying to pour the coals to it just makes it worse, and at some power level in put, things melt.
You've now experienced that first hand. My latest runs show the exact same thing.

This is good -- replication is the basis of all progress in science. It validates the idea that neither of our measurements are flukes.

Further, getting things in there really hot has a way of making them release any little pockets of trapped gas (inter crystal boundary, any other little places that can trap gas, like under a screw thread, and so on). So now you've seen that too. Various things can trap a little gas, any void in any material in there. The good news is that, unless that gas is being created due to reactions with the deuterium, or something decomposing due to heat, once you get it out, it's out (until the next time you let it up to atmosphere). This is part of what I've been calling the "conditioning" process.

So, what you need to do is run with higher starting volts and less gas. The thing has less "negative resistance" at that point, and as we've all found out, it's the voltage that makes the neutrons -- high current can come later on when everything else is right -- you need to go "smart" before you go "big".

This is a fairly complex cause-effect relationship. At longer mean free paths (less gas), you can, and must run higher voltages to make it draw current at all. But with less gas, you also have a lot fewer energy-wasting low speed collisions with neutrals before the ions get up to full speed. This might explain the more or less exponential nature of increased fusion with higher voltages.
EG, with the higher voltages, the ions would be going faster due to the bigger field you apply, and ALSO more of them get to full speed without hitting trash on the way into focus. Further, at lower currents, you have more of the field you think you applied -- less cancellation of it due to space charge implied by the higher currents.

As far as grid heating goes, that's very dependent on grid design details. To the extent there is "recirculation" of ions, or (more likely) more ions are being created by collisions with other ones, the ideal grid design has those happening in places where on the next pass they don't hit the grid wires. When you accomplish that, the grid doesn't heat so much anymore. I ran a kilowatt input this weekend without anything more than the middles of the rods getting to dull red heat. No other part of the grid assembly got visibly hot at all. FWIW, I never got my spiral to work all that well, and the grid design I now use makes a lot more neutrons per watt input, due to this effect -- for whatever reason, it intercepts fewer incoming ions in my particular overall geometry.
In your geometry, you'll have to fool with what works there, it probably won't be the same as what works in my tank shape. The other main players in IEC who are doing well are using the 3 ring geometry for their grids, last I heard. If I were doing that, and seeing how theirs all get hot and fail at the place where those wires all intesect at the end, I'd make them come to a circle there (a hole in the end) rather than all intersect at that point, but that's just a guess on my part. Obviously, when the wires get real close together at the end, the incoming ions there have no way to miss them, so opening up a hole at that spot (with a ring to tie the wire ends together) should help a lot with heating -- if there's nothing there, it can't heat up.

Some spitting and sparking is normal, especially at first. With a proper ballast to limit peak currents, as I think you now have, the result is that the high spots and loose dust particles just get blown off, and things just get better with run time. Too much peak current blows little pits at those points and makes things worse (as opposed to average current -- important).

In my experience, the best operation comes right at the gas pressure level where the thing will barely keep itself lit at all. This means there are less neutrals (less of everything) floating around to create that divergent current draw - as current goes up, current goes up more, out of control.

Your reported numbers are "in range" for the speeds and feeds you're reporting, this is very good news, actually. As you improve things (mostly vacuum stuff) you'll be able to run higher voltages, and as I've reported, each increased KV makes the neutron output go up faster than the previous one did, it's at least exponential after the threshold. Once you do that -- then and only then, worry about getting more current going through there.

Higher current has some negative effects other than melting things. Remember, you're applying a field, but that's not the sole determinant of the actual field gradient -- the ions themselves are charged and create a field that tries to cancel the one you've applied. The result is that with say 20kv applied, the ions don't actually reach a 20k electron volt energy/speed, because they see a somewhat canceled out applied field, due to the other charges in there. Further, since like charged ions repel one another, they become harder to get focused on a tight interaction zone, the beam "blows up" just like an old time CRT does when you try to increase the brightness too far.

When there is too much gas (and too many neutrals) another bad thing happens -- ions hit neutrals on the way in to the reaction zone, and lose energy doing that. So when they arrive, they might not still have enough energy to fuse. So a very big key issue is having the correct "mean free path" so that the energy you invest in each ion makes it to the reaction zone, rather than being thermalized in low speed collisions with neutrals on the way in.

I'll add more later, but here the news is -- Congratulations, guys! We're not yet doing the "clubs" like over at fusor.net, but clearly you'd be "in" if we were.

Here is how to embed a youtube video: (I put this inside code tags so it wouldn't be interpreted as instructions to put the video in here as well as below)


Code: Select all
[youtube]qMtOFYQeNvY[/youtube]


Which results in:




Note you just put the url after the v= part in between the youtube tags. See the button above the post editing window.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Doug Coulter » Mon Apr 04, 2011 12:16 pm

I notice your grid jumping around in the video. That's due to electrostatic forces not being the same everywhere (and every-when) and spatially symmetric. Seen that here too, they can be amazingly strong, even make loosely built stuff fly apart (it's happened here). This could be either due to gas pulsing, or ??? and is exacerbated by the grid not being in the exact middle from a field point of view, or not perfectly symmetric. It doesn't always ruin you, but it's not something you especially want. That's one expensive electrostatic speaker!
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Lee Hall » Mon Apr 04, 2011 1:45 pm

I'm hoping it was our vacuum pump vibrating the cart our fusor is on. We have put a whisk design in there for the next test (it's made out of tungsten) and it's hardly moving if at all. I've been looking more at the counts we were getting per 3 minutes during our test. Our highest was 1184 and the rest of the time we had a plasma, they were 450-550 / 3 minutes. As far as the clubs you referred to, are the neutron counts on the magnitude that we could say we definitely got fusion? We didn't really look enough at the portable neutron detectors we had around when we got the better counts, so all we were going on is our He-3 detector. The x-rays that can be heard from the detectors in that one video make me think we might have briefly gotten something, but I'm really not sure.

Right now I'm running a plasma to heat things up in there and condition some things. I'll work on the higher voltages in the next couple of days, but right now we are planning for another test Friday.
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

Re: All journeys

Postby Doug Coulter » Mon Apr 04, 2011 2:22 pm

It could also be pump vibrations -- simple test -- turn off the HV and is it still there. I guess I'm spoiled, I have all that vibration eliminated to almost what you'd want for an electron microscope on both my systems.

Sounds like you were at least doubling the background, so yes, I'd say you got some fusion, maybe not a lot, but "at all", which of course is that first step. If you can do that over and over (so it's not just the possible background itself varying, which I do see here sometimes) then you're rock solid. The only deal breaker might be if you find that your 3He tube responds to X rays too, which it won't if set up properly. To do that you use an X ray source (say Cs-137 or just some uranium ore or a radium watch dial) and set it so it (just) doesn't count on that -- threshold just above where the hottest gamma source you have makes it count. I set my threshold a little higher yet, but that's me -- the only reason to push it as low as possible is to count "every single neutron" which at some point just isn't going to matter much to you. When you get real hot fusion, you'll be seeing kilo-counts per second anyway.

If you need to borrow a source to check that, I can lend you one, our you can bring your thing over here and check it against our sources of gammas, betas, and neutrons -- we could cross calibrate so we're all on the same page. Our very nastiest source for doing things like this is a chunk of torbermite from that "natural" nuclear reactor in Africa, it's hotter than, well, various unprintable things you can guess at. I set up my tube so it doesn't count that at all, and that's a good spot to be at. That one, we keep in a lead pig in another building that I don't hang around in (it's giving off a lot of radon), and we will soon be moving that kind of stuff to a cave in the back yard...

You'll have some X rays just from electrons getting to the power supply volts and slamming the tank. The tank should stop most all of those (they do leak out the window and scatter around in the air, though). Those would be at or under your power supply voltage, and many X ray detectors (for example, most geiger tubes) won't detect those low energy guys anyway -- and they don't hurt humans much either, just not enough energy. But even low energy X rays will light off almost any scintillator detector.

We have seen here, when we know we have a ton of fusion, that we also have a ton of gammas in the 1-2 MeV region as detected on an NaI head that was calibrated on our Cs-137 source ( 511 KeV gammas). So those can mean fusion too, it seems. It's pretty hard to figure out any other way of getting megavolt gammas out of a system with only kilovolt inputs otherwise, eh?

You'll do far better at somewhat lower gas pressures due to the effects I mentioned in a previous post -- and heck, I prove that one every single time I make a run, as I start with too much gas and then take it out till ideal conditions obtain. Keep an eye on the grid temps, and adjust current accordingly. I was able to do "low orange" hot on my Ti grids no problems here (but not orange-yellow). Thinner wire works better for me - less to intercept ions. Where it's getting hotter, they are basically too close together (very rough rule of thumb) so ions have more trouble hitting the space between grid elements. There is more to it than that, of course, because where they hit depends at least some on where they came from, or where they went after the previous pass through. That's part of the game here.

I just set my current limit at some safe target, set the supply volts on "max" and then adjust the rest via gas pressure -- it's the easiest way to make it fly well. In your case, try for ~10 ma and try to get the voltage up via less gas -- you'll see! 10 ma might melt your grid, it was about my limit for my wire versions, but that's in the range where things are good -- say 8-12 should be a good target region for your setup, depending on how much heating you have with a particular grid.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: All journeys

Postby Lee Hall » Thu Apr 07, 2011 5:40 pm

Just wanted to post a picture from our fusion test on April 2, 2011. I'm the person on the right messing with the D2 needle valve and the power supply below the table top. I'm wearing a personal TLD and the person to the left of me is operating the data acquisition program and running some neutron counts from our He-3 detector. We use the right computer screen to display the video from the camera looking into the fusor so the rest of the viewers can see what's going on.
4-2-11 Fusion Test_1.JPG
Left: Adam J.; Right: Lee Hall
Lee Hall
 
Posts: 24
Joined: Sat Jan 29, 2011 8:51 pm

PreviousNext

Return to VA Tech Fusion Project

Who is online

Users browsing this forum: No registered users and 1 guest