Back to the mess

I have to finish a paper for a conference by Wednesday night. It is on the measurements relating to Touschek lifetime and momentum acceptance. There's an analysis of data to be done, and general writing and preparation of the paper. I really don't want to work on it.

Why don't I want to do it? Its late, I know. A paper needs to be written, and doing an analysis at the last minute can derail me from the process of fixing figures, adding references, and putting a clear narrative together. But its also that I've gotten myself out of thinking of this topic, and am wary of going back. It was my compromise. I will not leave accelerator physics entirely. I will do some work in this field, but move outward at the same time. But the topic is a mess for me. It is a personal mess in that my own files and documents and the relevant equations are not in such clear order. And a general historical mess in that it relates to the topic of dynamic aperture and sextupole optimization which is an unsolved problem. That question of dynamic aperture and stability has been the piece that I have slowly worked on, and tried to lay out a personal groundwork, so I don't feel so lost working in that area. Maybe this is a reflection of the fact that I didn't really finish this process.

So, my own angle on Touschek lifetime and measurement that I would like to get across is that the measurements are a diagnostic for the various lattice optimizations. There is both vertical emittance reduction goals, and increase of momentum acceptance via sextupole optimization. Stating clearly what these mean, and having measurements to ground discussion and results puts this other more nebulous "accelerator physics" activity onto a ground that relates to the goal of the machine- production of stable, long lived synchrotron radiation.

wreckage?

I look around accelerator physics and see that there are so many tried and aborted projects.
I think of this process of working on an open source project, and I see its been tried before with the UAL framework. Every idea one has has already been tried and failed.

Perhaps I should try to put a more positive spin on this, and list the open source beam physics projects. We have Zgoubi, and the previously mentioned UAL. Then there is XAL, which is an offshoot of UAL developed for the SNS. Each of these is rather project related, I believe.
UAL is an attempt at generality, but from a physics perspective seems to be more about absorbing existing codes, than developing new algorithms. It is used in the RHIC online model.

Also, I think beam physics should look to other fields such as HEP, with GEANT, and on the synchrotron optics side with codes such as shadow.

beam optics codes

So I'm continuing on with my conversion of the Excel optics code into Open Office Calc. I've been calling the code Open Optics. I've almost finished the part where a lattice is copied together with the appropriate formulas for calculating some uncoupled parameters by adding them up around the ring. Hard to keep up momentum on this project, because I think that no one will really use it.
But its really not that hard of a job to do, so I can keep moving forward slowly.

For a second project, I have two different possibilities. One is working on the idea of developing a TPSA library to work with Accelerator Toolbox (AT), so that the one turn map can be computed and used to optimize various quantities.

Secondly, there is the need for more collaboration on Accelerator Toolbox. Many different people use it at different labs and they are repeating the work. I can try to put the code in a central repository so that people can work together on it. This is probably the first thing to do before working on the TPSA project. Once more people are working together, it may be easier to use existing code and get more ideas about the right way to combine AT with TPSA. Of course, then, something like a normal form algorithm should be written- or at least something to find tune shift with amplitude.

For the moment, I need to learn better how to use Sourceforge and SVN. I did this with the Tracy code, but my collaborator knew more about it than me. I should set it up myself, and figure out how to do it from behind the firewall, etc.

Finally, there is MAD-X and PTC. This is the existing rich TPSA, normal form code. But its not used that much for light sources, and the tracking routines are somewhat obscure- buried in the PTC Fortran 90. Getting this physics (well, sort of- symplectic integration) out in the open, would be part of the point of collaborating on AT. The tracking routines are very easy to find and read there.

map analysis

In a circular accelerator, if you look at a single beam position monitor (BPM), or say, a couple of them, to give you phase space, you essentially looking at the properties of the one and many turn behavior of electrons. The value of the Hamiltonian -free description, the map description, is that it treats the machine as a whole, without necessarily thinking of how that map comes about.
This is analogous to solid state physics and scattering analysis. You define various quantities of the material that can be probed via scattering stuff off it. You don't need to know the position of every atom in the material in order to define these quantities.

Why has this approach not developed further in accelerator physics? Because there aren't enough accelerators. Each one has its own peculiarities, and so people think more about how to change those peculiarities and effect certain global behaviors rather than be more creative in defining the global behaviors. It would be like analyzing ten space ships, each of which was rather different from the other. One would probably not extract a general theory of space ships out of this.

Tower of Babel

At some time, I'd like to try to lay out the genealogy of particle tracking codes in accelerator physics. Looking about the CERN site for MADX, we find a presentation on the Universal Accelerator Parser for the Accelerator Markup Language. Here, we see the status of accelerator codes described by Sagan as a "Tower of Babel". But, to be honest, at this point, I would not say that this accelerator markup language has caught on that widely.

I don't know why, but I keep trying to construct some kind of epic narrative about this topic. If the field of particle tracking codes is the tower of Babel, then we might refer to the shutdown of the SSC as the "hand of God" coming in and crushing the unified human effort to pierce Heaven for the glorification of Man.

Yes... dangerous waters. There is already far too much implicit religious metaphor pervading this field.

In any case, let us hope that something like the UAP/AML can help put the pieces back together again.

I should also mention another effort in this direction, the UAL, Unified Accelerator Library. I don't know the best link for this, since the BNL site seems to not be online anymore. Here's a paper, or here for a SPIRES search.

pictures of dynamics

ssc aftermath

I'm still interested in this question of how the fact that the SSC (see here or here) wasn't built influenced the future of accelerator physics. Check out this list of technical notes and imagine the amount of work contained, much of it quite painful. Would these people ever have been motivated to do this had they known that the thing wouldn't have been built? The tools and insights forged in the process were quite valuable, but the whole issue of credit and respect seems to have become particularly skewed as the motivations shifted. Just my somewhat outsiders perspective.

more on Wendell Berry

I've been really appreciating this book of essays by Wendall Berry called "The Way of Ignorance". His basic examples come from agriculture and farming which he has experience with, but I can't help finding many parallels to physics.

His essay "Local Knowledge in the Age of Information" really solidified for me a good part of the reason (in a roundabout way) why I am in a national lab instead of trying to go directly to a university. He describes the need for conversation between the center and the periphery. He says that the periphery needs the center as much as the center needs the periphery. At the center, inormation is collected and efficient ways of representing and presenting that information is found. Life occurs on the periphery. Contact with the center can give one access to a wealth of experience from other places. But without conversation, the center loses its understanding that it owes its existence to the periphery, and the periphery loses its ability to ask for help and to gain useful information that will work within an appropriately local context.

Here's a (perhaps obscure) example given by a certain physicist that I work with. I think it provides a physics example Of Berry's point about the need for conversation and an illustration of what local knowledge looks like. Consider non-linear single particle dynamics in accelerators. The needs of this discipline involve predicting whether a particle in a certain non-linear system will stay within a certain region for very long or not (the dynamic aperture). Non-linear dynamical systems have been studied in other contexts as well. The gravitational n-body problem is an example from astrophysics. In that context, one wants to know, for example, whether our solar system will be stable over long time periods. The knowledge from this and other fields requiring the understanding of long term dynamics of non-linear systems has made it to the universities. Dynamical systems research collects together such knowledge. Now, certain theorems exist, such as the KAM theorem in which long-term stability criterion have been found. The problem is that the perturbations are just too big in an accelerator or a solar system for the theorems to apply. So, it may be that people in the universities think that in some sense the problem has been solved, wheras in "the field", the solutions are useless, and the very sense of arrogance implied in the attitude that the problem is essentially solved may at times do more harm than good.

Yes, this idea of conversation between the center and the periphery is very interesting to me, even if my example and explanation is rather murky at the moment.

synchrotron light sources

I suppose there's a time when you should do something with what you've learned. You can't be a student your whole life! (Well, of course you can, and I fully plan to, but I should also do something useful.)

So. What am I doing? Well, I'm part of a group designing a third generation synchrotron light source: the NSLS-II.

Ok, so what is a synchrotron light source?

Check out this photo of a machine called the NSLS VUV ring:
See the brownish tube in the center going all the way around? (Just above the dark blue things, and with some yellow, orange and light blue things here and there.) Its a copper pipe, called the beam pipe and for this machine its 51 meters around. Electrons travel down that tube at close to the speed of light.

All right. Nice. Three questions:
1) How'd the electrons get there in the first place?
2) Why do they stay in the pipe and not slam into the walls?
3) Why on earth would we want to do this?

Question 1 is why this is really a branch of the field of accelerator physics. We need a particle accelerator to get the electrons going fast enough. The electrons in this machine have an energy of 800 thousand electron volts. That's the energy one electron would have after going through 800 thousand 1 volt batteries.

By the way, in case you forgot what an electron was... we're mainly made of them. They are negatively charged particles that orbit the nucleus in atoms.

The answer to question 2 involves the yellow, orange, and light blue things surrounding the beam pipe. They are magnets. The light blue one is a dipole magnet, meaning it has two poles, north and south. The magnetic field bends the charged electrons. So that's how we can get the bunch of electrons going in a circular shape. The problem is that with just the bending, the beam would be unstable- if it was just a bit off the "perfect" trajectory, it would very quickly die a copper death. So the yellow magnets are there. They are called quadrupole magnets, four poles. They are like lenses for the particles. The orange magnets are sextupole magnets, six poles. They're also necessary, but its a bit technical as to why.

Ok. So, on to question 3. Why?

Well, for those of you who have taken an electricity and magnetism class, you might remember that charged particles create electric and magnetic fields. And remember that light is just an oscillating electric and magnetic field? Look around you. Yep. That's what's hitting your eyes: electric and magnetic fields. Basically, whenever a charged particle changes state, it gives off light. So, the electrons running around the ring here are having their path bent by the dipole magnets and thus giving off light. Because these machines are also called "synchrotrons", the light is called synchrotron radiation.

So that's it. Get some electrons stored in a big circle and whenever they bend, they give off light. That ought to be good for something.

Actually, these machines, synchrotrons, were originally designed to smash particles together. The higher the energy the particle, the more interesting things that might pop out from the collisions. Unfortunately, the higher the energy, the more of this synchrotron radiation gets produced which saps energy from the system. So, originally synchrotron radiation was seen as a big nuisance! Until some biophysicist (and probably some others) came along and said: hey, we'll take the light! So they started running the particle smashing synchrotrons parasitically, using the synchrotron light for whatever they needed it for. It turned out that this source of light was so useful that it was worth building these machines solely to get the light out! Second generation light sources were redesigned so that the electron beam was more suited for radiation purposes rather than particle smashing purposes. Then, third generation light sources were designed where the beam was really allowed to bend. In fact these devices were put in that caused the beam to wiggle back and forth. They were called wigglers. They are also sometimes called undulators. Here's a picture of one of these beasts:


So that's what I've been up to. The work I do, mainly relates to Question 2 above.
In particular, now that we add these crazy undulators and wigglers to the ring, we have to really be sure that the beam will be stable. So we need to write some computer code that tracks the electrons around to see what happens. To do this, we need to understand Hamiltonian dynamics, a formulation of classical mechanics. In my own skewed world, this is the reason I got into this field, and I still actually find this aspect quite interesting! Remember the craze surrounding chaos theory? One of these days, I need to try to really understand the KAM theorem. Well, this one of the fields where some of the math was developed, or at least was a major recipient. There are still plenty of open problems here! As for whether quantum mechanics is relevant for these beams of electrons in these machines, the answer is that it is and it isn't (and it might be). But that's a story for another day.