Here’s an announcement from the CERN DG Rolf Heuer sent out to CERN employees today:
The foreseen shutdown work on the LHC is proceeding well, including the powering tests with the new quench protection system. However, during the past week vacuum leaks have been found in two “cold” sectors of the LHC. The leaks were found in sectors 8-1 and 2-3 while they were being prepared for the electrical tests on the copper stabilizers at around 80 K. In both cases the leak is at one end of the sector, where the electrical feedbox, DFBA, joins Q7, the final magnet in the sector.
Unfortunately, the repair necessitates a partial warm-up of both sectors. This involves the end sub-sector being warmed to room temperature, while the adjacent sub-sector “floats” in temperature and the remainder of the sector is kept at 80 K. As the leak is from the helium circuit to the insulating vacuum, the repair work will have no impact on the vacuum in the beam pipe. However the intervention will have an impact on the schedule for the restart. It is now foreseen that the LHC will be closed up and ready for beam injection by mid-November.
This is an extra two week or so slip with respect to the latest draft schedule I’d seen. In addition, the question of how to deal with defective splices remains open. Efforts now are directed towards determining what the maximum safe energy is, assuming that the cold sectors are not warmed up, with the plan to have an answer to this question by the second week of August. Part of this effort involves study of possible changes in the parameters that determine how quenches are detected and dealt with, in order to optimize the maximum safe energy.
Update: The latest CERN Bulletin is out, with more about this.
I’m beginning to think that the LHC will never run as advertised. It’s too big, has too many single points of failure, takes far too long to repair. It’s like a giant software project, where a single tiny bug can crash the system, but that takes six weeks to recompile from source, even if only the slightest change needs to be made.
I don’t think these issues are teething troubles. The design of the LHC was probably too optimistic about the potential for errors. There has been a years delay already, and the machine hasn’t even done anything yet.
Is it reasonable to say at this point that this device will at some point be in continuous operation for, say, five years? How many potential gremlins will have to lie dormant for that long? Hundreds? Thousands? More than a million? Has there ever been a single project in human history that has run so smoothly for so long?
Admittedly, I’m being very pessimistic. I’m sure that these issues have been discussed, and indeed have been surmounted in other colliders. But the LHC is over 4 times larger than the next largest particle accelerator. By my offhanded reckoning, with the same instruments, wouldn’t that mean that it is only likely to run for a quarter the length without failure. Have our instruments become four times more reliable in the meantime?
The cooldown time is not so big a problem as the fact that almost every component in the machine is a single point of failure. I honestly don’t think that humans can reliably operate such delicate devices on such a large scale.
OMF,
I don’t think we should be so pessimistic. Yes, it is a complicated project, hence many unanticipated delays, screw-ups, etc. The people who built this machine are smart and dedicated, however, and the set-backs don’t imply it will never work.
Obviously the previous poster decided that his own personal paranoia is important to the rest of the readers here, and also to the running of the LHC.
Amyway, since some of the sectors to be warmed up because of the leaks can then also be checked for bad splices, the reliability of the whole machine should increase considerably.
Charles
Whoa, man, ” the fact that almost every component in the machine is a single point of failure.”? That’s a bit much, but I do hear you that there is an outside chance (very small, I hope) that the damned thing will never start up.
The fact straight up though regarding that is: we don’t know. Experimental Physics, Applied Ahysics, and Engineering are extremely well-defined fields. I say let’s give the Euros in those fields a break, and hope, rather reasonably hope, they slay their dragons and get the sucker up and running, later if not sooner. We really don’t want the machine to go the way of the SSC, right?
In the meantime … there much work to be done. We have a sweet opening between the death (or dimishment) of String Theory, and the LLC producing particles that raise more questions that answer.
Peter is working on BRST Cohomology. Good for you, Peter. Go for the gold.
OMF,
I think it’s way too early to start worrying. In a huge, complicated project like this, you’re going to run into all sorts of problems. So far, they’re all things that are readily fixable, it just takes time (except maybe the magnets not getting to 7 TeV, it’s possible the machine will end up being a 6-6.5 TeV machine). Getting the LHC up and running, and working reliably at design luminosity may very well end up taking a few more years. The publicity campaign that encouraged people to think the machine was going to be working and producing (safe) black holes any day now may have been a mistake. Realistically, this is a process that takes a lot of patience.
We all apreciate optimism. But this project is as dead as a can of spam. Notice too that the latest delay (in a long series to come) is getting leaked now after Heuer announced that his new budget was aproved. Yes it is huge and complicated. But it is years over schedule, it will not run at design energy for the foreseeable future . . . and there is no credible start date at any energy. You can define failure more acurately.
Sorinis and greg, you’re so pessimistic it’s hard not to take it as personal paranoia. As PW himself stated (and you don’t think PW is especially naive don’t you?) you’re going to run into all sorts of problems when you build a new collider. “So far, they’re all things that are readily fixable, it just takes time.”
An extra 5 years before full luminosity at 7 Tev is unlikely but would not be so surprising (remember the now glorious Tevatron?), and it doesn’t matter. You’re talking about a 40 years project. Who cares if it takes 45?
Anyway don’t worry, don’t change anything in your attitude. That explains a lot about why the LHC is not in the USA and why the SSC went the way it did.
Loco, the basic point made by Sorinis and Greg is correct. If LHC is delayed,then people will start losing interest (esp. grad. students/postdocs) and will gravitate towards particle astrophysics/cosmology/gravitational waves. In fact AFAIK the top cited experimental HEP paper is not aparticle physics experiment, but WMAP. In fact in the last 8 months or so most hep-ph papers seem to be PAMELA rather than TEVATRON or LHC
(of course this is anecdotal and I have not done a count).
I think the tevatron is due to announce some higgs results towards the end of august. Peter, are you covering htat?
OMF,
your analysis is very superficial.
A vacuum leak is a trivial problem which has been overcome during the last cool down. Remember, the machine was already completely ready once proving you wrong.
The bad splices are annoying because completely fixing them would require a complete warm up of the machine. They just don’t want to do it now just because they want the machine be running now so everything can be tested including the detectors.
Once the bade splices are fixed, they are just fixed and do not interfere with anything.
There may be a couple of other problems ahead, but their number is surely finite.
Shantanu, I’d say you have a generous way of extracting “the basic point” made by some folks here.
Sorry Mikael,
we all saw one week after what it was ready for. And please do not talk of bad luck; in the past they had a similar accident on a test bench, but it looks like the potential consequence of it on the final machine was totally neglected.
Sorinis,
If the Tevatron has new results on the Higgs this summer, I’m sure I’ll post about them when they come out. If someone at CDF or D0 has preliminary results now they’d like to share, feel free to e-mail me…
Shantanu, who cares if some marginal hangers on are discouraged that experimental physics is hard? That’s a good thing. It leaves more money for those that actually care.
Certainly the string guys typing arbitrary ‘science free’ formulas (IMHO) into computers could use some company on World of Warcraft. That’s (probably) the more useful activity they perform for society, after all the Orcs aren’t going to defeat themselves.
Greg, one counter-example to your point is “Gravity Probe B” which was launched after almost 40 years. Are you really looking forward to their results?
Also if an experiment keeps getting delayed
it risks not getting funded (rather than the opposite way) (another example been the DUMAND experiment
Shantanu,
As a student on ATLAS who is waiting for data to come in to complete a thesis analysis, I think you have a point. However, to assume that the delays of the LHC are driving students and post-docs to different fields is a bit of a stretch in my opinion. Some American students have switched to Tevatron experiments recently (as they’re nearing the end of their degree and simply can’t wait any longer for data to come in, and unlike they’re European counterparts, aren’t allowed to graduate on a Monte Carlo analysis), but I’ve yet to hear of a single person moving entirely away from collider experiments, simply because they’re discouraged by the delays.
student, if such delays continue then obviously you will see more people
migrating away from collider/accelerator physics. as I mentioned WMAP
papers have more HEP citations than any particle physics experiment.
The migration to astrophysics seems unlikely by students already in the pipeline; however, we might have seen an exodus to the Tevatron had the real extent of the damage been made public earlier. Recall that the “He leak” was to be fixed, and LHC ready for beam in “early spring”. Now we see a steady dribble of two week schedule slippages.
To compound the misery, the detectors have been closed and are going through periods of 24 hour shifts to be ready for beam in summer ….fall … winter …..
We all know this is a difficult undertaking, but the management of information is starting to have a real effect on the participants lives.
Yes Peter, I’ll email you what I’ve got when I have it. But only if you stop erasing my rants about the LHC. 🙂 That is the deal with Satan. I had such high hopes for it and the europeans just ruined everything.
Do grad students and postdocs really have the ability to so completely shift their research focus within the timeframe that the final burst of LHC delays has covered (1-3 years)?
…and if some people did shift their focus as you describe… would that even be a bad thing? There’s some really important stuff happening in astrophysics and cosmology.
Don’t forget the paper that argues that the LHC will never be operated (at close to full luminosity) due to a time travel cancellation effect from the future.
http://www.fqxi.org/community/forum/topic/230
Every time I read about new delays and new problems I wonder if the idea might just be crazy enough.
Are you unable to log on from China due to internet filtering? Hope you start blogging again soon.
nigel,
I did have internet access periodically in China, no posting just because I was on vacation… Internet filtering did keep me from following Lubos’s blog.