particle velocity vs energy in volts

The title says it all. This is for established factual (we hope) things.

Re: particle velocity vs energy in volts

Postby Doug Coulter » Mon Oct 10, 2011 5:40 pm

On looking, it looks like later numbers for masses are indeed updated a little bit, and continue in the direction of electrons being a little heavier, so that 5.93 constant might have to reduce a little bit - the data below shows a slightly heavier electron than Terman knew about (even his later version). But the discrepancy is in the 3rd place after the decimal, and this stuff is mostly tolerant to 1% error, so we should be fine. I gathered all this data below for the writing of a program to calculate drift tube accelerator dimensions, given a frequency, peak RF voltage, known injection voltage, and ion mass - I'll post the program somewhere else when I get it going. I think I'm going to try and work all in eV for the program purposes, as those numbers are easier to find and seem to have more precision.

From Wikipedia (mostly)

For example, an electron and a positron, each with a mass of 0.511 MeV/c2, can annihilate to yield 1.022 MeV of energy. The proton has a mass of 0.938 GeV/c2, making a gigaelectronvolt a very convenient unit of mass for particle physics.

1 GeV/c2 = 1.783×10−27 kg

The atomic mass unit, 1 gram divided by Avogadro's number, is almost the mass of a hydrogen atom, which is mostly the mass of the proton. To convert to megaelectronvolts, use the formula:

1 amu = 931.46 MeV/c2 = 0.93146 GeV/c2
1 MeV/c2 = 1.074×10−3 amu

Mass of proton:
1.007276466812(90) amu[1] spin 1/2 isospin 1/2 parity 1

Mass of neutron:
1.00866491600(43) u[3] spin 1/2 isospin 1/2 parity 1

Mass of H:
1.00782503207(10) amu spin 1/2+ (I assume these are with the electron)

Mass of D:
2.01410178 amu spin 1+ parity 1 (mostly)
Mass of T:
3.0160492 amu spin 1/2+
Mass of 3He:
3.0160293 spin 1/2+
Mass of 4He
4.00260325415(6) spin 0
Mass of electron
5.4857990946(22)×10−4 u[7] amu
9.10938291(40)×10−31 kg[7]
0.510998928(11) MeV

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: particle velocity vs energy in volts

Postby Starfire » Tue Oct 11, 2011 6:14 am

The answer is not fusion - an annihilation reaction looks more like the route to go - if only :?:
Starfire
 
Posts: 143
Joined: Thu Aug 05, 2010 4:26 pm
Location: North Ireland

Re: particle velocity vs energy in volts

Postby Doug Coulter » Tue Oct 11, 2011 8:43 am

Well, it'll be interesting to see what I can learn with a couple small mass-selective accelerators which I'm designing now. One reason I'm listing the spins etc is that in a head-slap or face-palm moment, I realized that even with "non-thermalized" fusors, we are trying to "cheat" or at least "gamble with" some very well accepted quantum conservation laws, and I'm trying to learn all the implications of that. While there are reports of tiny CPT (charge, parity, time) violations in exotic conditions the norm is that those things are conserved. Bashing reactants against one another randomly naturally results in low effective cross sections, since the chances that all the conservation laws can be conserved and still get the desired reaction is very low with random inputs.

It seems worthy of study. One relatively simple target might be to try and affect the relative frequency of the three possible DD reactions deliberately. As far as I've been able to find out, no one has, and in fact, it seems no one has tried. It appears to be a "low hanging fruit" everyone has missed. Not quite Nobel prize turf if we pull it off -- it's too embarrassing for science in general that no one thought of this half a century ago. Looking into the high cross section DT reaction -- these things are more easily satisfied, and we see this much larger than usual cross section as a result of this, and apparently a "resonance" in an intermediate reaction product, 5He. It seems an obvious way to proceed, for me, at this point.

My takeaway from fusors, is that they're done - stick a fork in it, take it out of the oven. Now that we have some pretty decent cross lab data, it appears that, well, everyone has found the same spot, same Q, for the conventional fusor. I don't believe that this is "the end", just that with the conditions in a static/DC fusor - we're all "there" as we're going to get without a lot more understanding of the dynamics. While simple in appearance, the fusor is actually a hotbed of complex emergent behavior, which no one has satisfactorily explained or simulated (except with the real thing). My take from what I've seen so far is that this spot, or set of operating conditions that all successful fusors run at is what a mathematician would call an "attractor" - an equilibrium that the system is attracted to by local gradients in all the "error surfaces". My own observations are that (almost) anything that perturbs a fusor off this stable attractor actually increases the Q and reaction rate quite a lot, for a little while, and when the perturbation is removed, the system quickly assumes this low-output dynamic stability around this attraction point. In other words, absent some perturbation the system actually finds just about the worst possible state for producing fusion! The system is simply too complex for me to model in my head fully, or in any existing computer. Seems to me the best way to proceed is to go with a simpler system and investigate some of these details in isolation before returning to what can be done in the "more simple" fusor.

Here's an example of this. In one post, I showed what happens when I drive an auxiliary electrode with the output of a neon sign transformer. During one half cycle, the fusor broke into a pulse mode at around 2khz, and during those pulses the Q was astonishingly higher than normal, and the main fusor power consumption went way down (except presumably during the pulses). It very much seems like only during the onset, before dynamic equilibrium was achieved, were the good conditions for fusion achieved, then it more or less quit being efficient. This tells me that the normal equilibrium in a fusor must be tending to align things in ways that prevent, not enhance, fusion probability - almost anything the knocks it off that stable spot seems to improve things for awhile (and what a strange time constant that is). Further attempts to perturb the fusor, while having some positive effect, kind of showed me I'd been lucky with that first attempt/experiment, and that things were more complex in the parameter space than any reasonable attempts to sweep the space were going to explain. Just pure luck that the effective impedance of the NST interacting with the induced charge from ions flying by the secondary electrode happened to resonate (in the more electronic sense) in a way to make things momentarily work a lot better (in this case, Q 500 times normal, roughly). But I can't see how to find the best conditions in that setup - too many things to try and most of them are difficult and expensive. So I'm going to a simpler system for awhile -- while the hardware isn't as simple, what takes place inside should be, and it will allow fine control over things like spins...where the charge is when, time as well as space (focus) bunching...

And yeah, this belongs elsewhere, maybe...I'll move it. But this is the reason behind wanting to have this thread, the underlying driving philosophy. Sometimes it seems best to take a step back before moving forward. I think I need to learn what I actually want a fusor to do before trying to make it do that. Just making one and firing it up, tuning a little, has taken us as far as that seems to be able to go. To manipulate the conditions wisely, it seems to me I need to know some more about what we even want!

Yes, in a sense, annihilation is what I'm after here. Suppose I want to see a reaction product that has zero net spin. I have reactants that have spin going in - I need to cancel, or more loosely, annihilate that. This would seem to mean that I need to put in reactants with opposing spins, rather than random ones. There's probably a way with the right electromagnetic fields (time varying of course) to get that in the fusor. But it's going to be a lot easier to try with beams and no dependence on recycling things that have gotten into unknown states.

Seems to me the loop of:
1. build fusor
2. run, observe
3. tune based on 2
4. go to 2 unless completely satisfactory
5. ???
6. profit
only allows one to follow large scale monotonic gradients in the error space, and we've all found the one obvious local minima there - we're done. This technique would NOT find all possible minima in a complex error surface/topology, so the loop must be broken in some fashion to keep from continuing to find the same local minimum. This is analogous to what is done when training a neural network, or solving the traveling salesman problem. Local minima that are lousy solutions abound, but the best solution can often only be found by "shaking up" the system (simulated annealing) and following each local gradient thus detected until one can be more sure the real global minimum has been found. The "attractor" that is observed in a fusor is so strong that a big step back seems called for, a large "simulated annealing" input required, to avoid being "trapped" over and over in that same spot.

So, I'm gonna do just that.
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

Previous

Return to Formulas and data constants

Who is online

Users browsing this forum: No registered users and 4 guests