All journeys

[Doug Coulter]

Just a thought. We looked at using a borated shield here too, as we sometimes have pretty serious neutron output. Well...we didn't wind up using it.

Here's why.

When boron captures a neutron, it emits gamma rays, some pretty hot. When you calculate the thickness of lead needed to stop these extra hot gammas (which are hotter than what I've measured from a fusor) -- well, lets say the price of the lead alone is an inhibition. Converting a 2.5 mev neutron to a 4.44 MeV gamma is not exactly "winning the game" or so it would seem.

As Richard H says -- the square law is your friend, and one of the best shields ever in practice, usually. I would postpone that worry until after you have enough neutrons for it to contribute significantly to total exposure. Far more of the input energy comes out as X rays and gammas, so the lead is the place to start. Here we see copious X rays at the PS voltage, plus a bunch of higher energy ones that are evidently created in the fusion process -- by products hitting the tank walls or something similar. Lead of course also captures some neutrons and makes gammas as a result. Check out things that might be in your system here. http://www.nndc.bnl.gov/capgam/indexbyn.html

Hot neutrons are supposedly worse than cold ones -- but that's an easy thing to handle, and only a couple of inches of HDPE (or a little more of wax or oil, or water) will slow them right down.

Here's a link to the multiplier vs neutron energy. As you can see, just making them thermal cuts the effective exposure way down. The multiplier is used in the formula here. Look at Sieverts in this link.

So a rational plan might be to worry more about photons at first, then neutrons once you have more success. I believe in betting on "winning" but unless you are *very* successful, photons are going to be the main issue. Look at it this way -- we put in perhaps 400w, most of which becomes heat and X rays. About one microwatt is fusion output (the run you saw here was about these numbers), and some fraction of that is neutron output (the rest is high energy gammas). Human damage, to a first order, is related to how much sheer energy is deposited in the human. Most of the fusor output energy is not neutrons. In all cases, distance is your friend. So are short exposure times. I don't run my fusor all day or even every day...and I calculate that I am at most doubling my net radiation dose over background levels, which is my personal target, based on nothing but my intuition. The government allows for a heck of a lot more.

Radiation dose chart

For what it's worth, the other day I put our neat little military Canberra integrating rad dose meter (it sees neutrons too) right inside the shield at the fusor during a run. The peak reading was 2.47 mrem/hour, or 24 uSv/hour, but I only run for about 10 minutes in a given session. So 1/6 of 24 uSv is my dose on a day I run -- if I wasn't behind 1/4" lead plus that lead glass window. That measurement was on the hot side of the shield.

A chest X ray on the above chart is 20uSv, but I'm getting something less than 4uSv/run (shielding gives me much less than that). As you saw, it's not that big a task to get exposure down to what most would consider a reasonable level. X ray shine-through the unshielded tank only starts getting going above 40kv -- and you have a 40kv supply...I wanted the lead because I can run 53kv at present, and am planning for higher voltages, so some of the power supply voltage X rays do get through the tank. Yours won't, except at the window(s) and feedthrough(s).

And oh, yes I also tried our dosimeter at my operating position. It read in the 120 micro-rem/hour range there -- quite low, about 1.2 uSv/hour, or 0.2uSv/run. So the shielding I have, minimal though it is, is plenty.

While it's possible to make one of these dangerous, it's hard work to get there! More danger from high voltage or tank implosions, frankly.

[chrismb]

Doug,

From all your squirreled literature, can you tell me more about this 10B(n,g) reaction? The cross-section on nndc says it is less likely than a gamma from a DD reaction - i.e. if you get 1 DD gamma in 10,000 reactions, it looks much the same for an n+10B absorption. Is that right?

[Doug Coulter]

I'd like a link to where it shows that on nndc, that place isn't easy to search. The link I gave shows the most common gamma to be the 4.44 mev one. AFAIK, all neutron captures produce a gamma, and the whole point of using boron is the cross section for thermal neutron capture is very high (among things we can afford).

I did discover when trying to use borated wax in my failed neutron camera that the incidence of gammas produced is very high when there are neutrons.

[chrismb]

AFAIK, all neutron captures produce a gamma

That wasn't my understanding. If a neutron is captured, then if it stays there, stuck in the nucleus, then it might give off a gamma if there is an excess mass energy for that isotope. But n+10B gives off an alpha (for most of the time, like DD gives off n or p), which I assumed it did promptly and this is how the excess energy gets shrugged off.

But, hey, I'm only making interpretations here of what I've read, I'm no professional on this, that's why I was asking.

[Lee Hall]

I passed the information on the to shielding people of our team. In passing, a couple gave me the impression that the likelihood of the gammas from the boron shielding is low. But the have the link to this thread so I'll just have to trust them to make the best decision.

Today we ran some tests on the vacuum system. We disconnected the bypass loop from the system and the fusor seems to pressurize rather slowly when it is isolated from the pump (.1 mTorr every couple of minutes or more). We then re-attached the bypass loop without one of the bellow tubes (it was used with a rotary vane pump and had some oil deposits in it still) and replaced it with the brand new, 4 foot long tube we have. It looks like the system still acts the same way if the fusor is isolated and if vacuum is attempted to be maintained simply through the bypass (pressurizes about 0.1 mTorr every couple of seconds). I'm not sure what this means, but we're going to run it overnight to see if it's a matter of getting some contaminants out. If this doesn't work, we're going to have to decide if we should take the bypass out, apply glyptal to all of the joints, or just do something different. I think either way, we're going to run some D2 through it next week to just see how we are.

[Doug Coulter]

Chris -- second request for a link.

You could be right for all I know. I do know that in actual tests, having boron around made my "directional" neutron detector (plastic scint with borated wax collimator) not directional anymore, since the plastic scintillator would detect either neutron-proton knockon, or gammas about the same. So the next rev of that will use a plastic (not borated) moderator, and I'll still have issues with gammas from hydrogen neutron capture, just not as bad (I hope).

I believe there is the same issue with Cd absorbing neutrons and giving off gammas (which is all it does, the next isotope up is otherwise stable). Even alphas etc can make photons when they hit things -- maybe not so energetic though. Guess we have to fish around for the ratio of gamma to alpha decay, I don't think I've seen a number. Either one would trip a B10 type detector tube at any rate, though I suspect the pulse for an alpha is a lot larger because as CharlesW said, they plow the gas better. I do see varying pulse heights out of mine (A B10 lined tube), and that might be the cause?

Since "hot" neutrons are much worse in terms of Sieverts than thermal ones, it seems enough for me for now to just slow them down, and not sit too close to the screen. After all, if it's a microwatt of fusion, it's less than half that in neutrons. AT any rate, the dosage in Sieverts is pretty low from neutrons, and not so low from gammas on my machine.

Now, if these guys get to much more success than anyone else -- it's another story, but so far no neutrons at all. Their vacuum isn't good enough to get their voltage up to the "threshold" where you start making enough to count decently (about 16-18kv in a fusor, and no point trying to push volts up with sheer current, tried that -- you just melt things and the voltage goes *down* as you head out of gas discharge and into "arc"). Theoretically, there is no threshold, I'm just going on when my 3He tube starts counting noticeably above background.

Then, neutrons go up real fast with applied voltage, much more than linear, as one might suppose looking at the DD cross section. Which tends to explain JonR's high numbers somewhat -- he's running something like 80kv (I'm 50, Tech is 40). It could also be a focus issue, and the higher voltages get better focus with conventional grid designs. So far though, with a few pretty different designs tried here, that curve vs voltage is about the same with all the geometries I've tried -- some are better at any voltage, though the shape of the curve is about the same in all cases.

[chrismb]

Chris -- second request for a link.

Oh.. I thought you were saying "I'd like a link" rhetorically as you don't like the fiddly interface. There is a screen with drop down menus that you have to select what you want to see. It says (n,g) is a little under 1 barn for thermal neutrons, and 10,000 barns for (n,a).

neutrons go up real fast with applied voltage, much more than linear, as one might suppose looking at the DD cross section. Which tends to explain JonR's high numbers somewhat -- he's running something like 80kv (I'm 50, Tech is 40).

The DD reaction curve says that 80kV reaction rate should be around 4~5 times higher than at 50kV.

[Doug Coulter]

So, how about a link to that screen? I couldn't find it. My CRC book doesn't have it, but does list the slow neutron cross section as some 7k barns or so.

Yeah, it's going up much faster than linear with voltage here. JonR is claiming >10x what I do, but I'm darn conservative and using calibrated sensors (calibrated by someone else), in this case I'm using BTI and Canberra sensors. They seem to agree with one another, which is always nice. I then use those to get a feel for my two neutron detector tubes.

Also, things get much touchier as you're on the steeper part of the Paschen curve at higher volts. Of course, some of this has to depend on whether you're doing fast on neutral, or fast on fast too -- it would move you to a different place on the cross section curve, depending. I'll get around to trying higher volts -- that big power supply project is somewhat on hold for the moment, and I'm working the standard counter project right now. It will take more work on the feedthrough stuff to get up to the 120kv I should have when I do that PS project. And on other things! That's getting to the point of being real hard to do in air.

[chrismb]

http://www.nndc.bnl.gov/exfor/exfor00.htm

[Lee Hall]

Would having a mixture of D2 and air in the fusor at low voltages (4 or 5 kV) result in a reddish plasma? We got a reddish-pink plasma lately that disappeared as I increased the pressure and went over to our normal purple glow. The purple glow then continues as I took more gas out of the system. When I went all the way back to a 0.8 mTorr vacuum and slowly got another plasma at 15 mTorr, it was purple again. I can't really explain the anomaly except if somehow some D2 was leaked into the system.

However, I really am having conflicts with how low our vacuum is. To me, either we are RIGHT on the border for pressure or just a touch too high. Here's the basis for my reasoning. At the moment, we can get down to our 1 mTorr target vacuum. We ramped the power supply up to -20 kV to see how high we could go without seeing a glow. At voltages around 11 kV or so, there would be intermittent glows that would start as the voltage was increasing, but then disappear after the voltage stabilized. The currents during this time were all less than 0.5 mA.

In an attempt to 'practice' operation, I have been supplying a -3 kV voltage to the fusor and slowly letting air leak in through out fill system. At about 10 or 11 mTorr, we start to draw some current. Right now as I'm 'operating' this thing, I'm trying to slowly ramp the voltage up as I pinch off some of the air flow. At this point, I am pretty stable at 7.6 mTorr of air, 9.51 mA of current, and -6.05 kV. This low voltage is really making me uneasy. I'm really thinking we need to get another vacuum pump.