A delicate balance

Full size image available through contact

A Department of Energy review of Fermilab's Accelerator Run II produced high grades and several remarks of "excellent progress," but also made clear that results are the ultimate benchmark.

The charge to the review committee from John O'Fallon, Director of the High-Energy Physics Division at DOE, clearly stated that "it is vital to the U.S.program in high-energy physics to maximize the performance of the Tevatron, and therefore, the scientific output of the collider experiments in this critical period before LHC turn-on."

"We are the flagship program for high-energy physics worldwide right now," said Fermilab Director Michael Witherell,"and we're expected to deliver on that."

From that starting point, and with a focus on Tevatron luminosity, the review committee findings included references to excellent progress in pursuit of luminosity goals; especially notable progress on stochastic cooling; confidence in the Recycler's success; impressive combined availability of the Linac and Booster, and very good availability for the Main Injector; and the judgment that the laboratory's technical approach for increasing luminosity is "sound and well-motivated."

To Witherell, that last characterization was critical because of the nature of the investigation.

"The biggest question for the committee was,'do we have a good plan for the future, and are we going to make as much as we can of the scientific opportunity ahead of us?'" Witherell said."This was a good review for us because it made the case that we are doing the right things on a very hard problem."

The Oct. 28-31 review endorsed the lab's "fully resource loaded plan" for FY03. But in another sense, the review was a more intense restatement of the laboratory's ongoing challenge: marshalling resources for immediate needs without sacrificing important longer-term goals. The funding picture is sufficiently clouded that the review specifically stated: "Adequate funding throughout the luminosity upgrade period is not assured and so constitutes a substantial risk to reaching the goals."

"What that's saying," Witherell explained,"is that we're giving extra priority to the accelerator effort, but the budget is not a good one. So we are supporting well the things that are going to get us additional luminosity this year. We're supporting a little less well the things we need to do to get added luminosity for the next year, and probably a little less well the things that are going to get us to the last step. I think that's the right priority, but I think what did come to everyone's mind was that, to do as much as we can for luminosity in 2003, '04, '05 and beyond, it will take somewhat more resources and somewhat more funding than we're seeing in FY03. That's part of the larger problem in high-energy physics now. I hope it will help make the case that a little bit of additional money will have a big return in the science we're able to do."

The lab's plan is divided into "base" goals, with a high degree of confidence, and "stretch goals," relying on everything developing as planned, and assuming a higher payoff in luminosity from improvements in 2003. The base goal for total integrated luminosity is 6.5 inverse femtobarns (fb-1) by 2008; the stretch goal is 11 fb -1 by 2008. For FY03,the base integrated luminosity goal is 200 inverse picobarns (pb -1 ); the stretch goal is 320 pb -1. The Tevatron's initial luminosity has improved by a factor of four since January 2002 (from below 1E31 to 3.7E31), which tangibly improved the tone of the DOE review. The Tevatron is now delivering 5 pb -1 of integrated luminosity in a typical week, or nearly 2.5 times the normal weekly level of Run I.

"If we just ran for the year at the level we have for the last month, we'd make our goal for FY03," said Steve Holmes, Associate Director for Accelerators and interim head of the Beams Division. "But if we did that, we'd be in no position to double that performance in FY04. So we have to find the correct balance between sustaining operations and implementing improvements, and continue to make investments during FY03 for '04,'05 and '06. We've got a reasonable plan for the coming year, we're organizing overall activity on Run II in a project mode, and I think we got an endorsement of that approach."

The perception by experimenters represents another important endorsement. DZero cospokesperson John Womersley, in producing a presentation of the plan for experimenters, described the changes as a delay, resulting from the additional time it took to reach current luminosity levels,and as the difference between a realistic outlook and a "wish list." The key factor, he said, is an improvement in planning.

He compared the lab to a sports team that was always able to win on talent and intuition, but now realizes that the game has changed and requires a more systematic strategy.

"One thing that's come out of the last year has been learning how to improve the accelerator operations," Womersley said."But another thing to come out is an understanding that we need to plan better, and a sense of realism of what we can promise. We have a better understanding of how many people it takes, and how long it takes, and how complicated a job it is, to get physics out of a new accelerator and new detectors."

Schedules, Womersley said, are more important to physicists than they are to physics.

"The point of delivering luminosity is to do the physics,and physics doesn't sit around waiting for a certain amount of data to come in," he said. "The physics is a continuous process. There isn 't a single magic number that defines success or failure for the Tevatron in terms of the physics. There isn't a time that starts and ends when you have enough to do something. You always have enough to do something interesting,and as more data comes in you do more."

Witherell suggested the process of improving luminosity was an example of producing results under what might not be the best of circumstances. He noted that the team leading the accelerator efforts was concurrently formulating the luminosity plan while preparing for first, a director's review, and second, the DOE review.

"These two reviews were very important," Witherell said."Yet while all that preparation was going on, we've also had some of the fastest improvements in accelerator performance, all at the same time. That 's impressive to me."

The reviewers apparently felt the same way.

Alaska earthquake shakes up Tevatron The Tevatron, the world's highest-energy particle accelerator, is so sensitive to its environment that it felt the effects of the Nov.3 earthquake in Alaska -- some 6,000 miles away.

The earthquake, whose maximum magnitude registered at 7.9 in the vicinity of Denali National Park, caused the Tevatron to vibrate and lose its particle beam at around 4:30 p.m. (Central time) that Sunday afternoon. But it took an alert troubleshooter, checking the machine logs from his home computer while watching CNN on TV Sunday night, to finger the earthquake as the cause. Duane Plant, of Beams Division/Engineering and Support, noticed beam loss over a long interval before the machine quenched, or warmed above superconducting temperatures.

"It was similar to a loss we saw back in June, when there was a smaller earthquake down in Indiana," Plant said."That earthquake didn't quench the machine but it made a small loss pattern. This looked similar to me, so I went to some web sites of the U.S. Geological Survey, got the earthquake times in Alaska, and did a 'back-of-the-envelope' guess on the time it would take to get here. That guess was within two minutes of when the logbook showed the quench."

Plant got on the phone with Beams Division Tevatron specialist Todd Johnson. The two spent about an hour on plots and figures, and decided it was time to call the crew chief in the Main Control Room with the earthquake scenario. "Since the crews couldn't find any other cause for the quench," Plant explained,"we thought maybe it was best to just go ahead with the next step in reestablishing the beam, instead of spending another four or five hours looking for the cause. We thought we understood the cause -- as absurd as it sounded."

On Monday morning Plant, Johnson and MCR Operators checked with the seismology center at Northern Illinois University in DeKalb, and confirmed that the wave arrival times from the earthquake coincided with the effects felt at the Tevatron. Instrumentation in the Beams Division had recorded position shifts of 30 microns in the Tevatron and other accelerator components -- about three times the width of a human hair. And just down the road, Argonne National Laboratory reported beam loss in its Advanced Photon Source at virtually the same time as the Tevatron. "So much can affect the beam here at the lab, you have to keep your eyes open 'way outside the box,all the time," Plant said."Then again, with a machine like this, operating out at the edge, there is no box."