Vacuum Leaks Lead To Another LHC Delay

suraj.sun tips this story at ZDNet about a new problem with the LHC. Quoting: "The restart of the Large Hadron Collider has been pushed back further, following the discovery of vacuum leaks in two sectors of the experiment. The world's largest particle collider is now unlikely to restart before mid-November, according to a CERN press statement. The project had been expected to start again in October. To repair the leaks, which are from the helium circuit into the insulating vacuum, sectors 8-1 and 2-3 will have to be warmed from 80K to room temperature. Adjacent sub-sectors will act as 'floats,' while the remainder of the surrounding sectors will be kept at 80K, CERN said in the statement. The repair work will not have an impact on the vacuum in the beam pipe. CERN has pushed back the restart a number of times, as repair work has continued. To begin with, scientists said the LHC experiment would restart in April 2009. In May, CERN [said] that the restarted experiment could run through the winter to make up some of the lost time."

Re:Is that first thing we need ?

[samkass]

Particle interactions with more energy than LHC can produce happen in the Earth's atmosphere every day. But outside of a carefully controlled environment with extensive sensor equipment, they can't be studied. The LHC is not about creating energies never before seen on Earth-- it won't do that. It's about doing so in an extremely controlled manner than can be measured and investigated.

Re:At least...

[Anonymous Coward]

don't worry about the pills, they aren't working

Re:puzzled by the headcrabs ref?

[PaintyThePirate]

More importantly, Gordon Freeman is apparently an engineer at CERN.

No worries though, he has a crowbar [reddit.com] now.

Nope.

[Anonymous Coward]

Nope. It's like Ultra-High-Vacuum applications -- the one single real-world technical application even more frustrating than programming.

(Disclaimer: I am a scientist working on UHV applications, and I am a programmer :-) And I enjoy both. And I'm doing both for a living... Hm. Hang on a minute, I'm just realizing my masochistic tendency... :-p )

No

[Werthless5]

The explosion happened last September, so it can't be a year behind the new schedule; it hasn't even been a year since the explosion! The schedule set after the explosion was to run again the following September, so it's now predicted to be 2 months behind that schedule

And it's really not too bad, since the SSC was far more overbudget than the LHC has ever been and was being footed solely by the US (whereas the LHC is international). And we're not really losing anything from even a one-year delay. Also, consider the fact that experimental particle physics is but a single aspect of physics, one side of a multifaceted subject.

As for cost, the total LHC cost after 10 years of running is expected to be less than $10 billion total, and that includes the full design phase (greater than 10 years). That means the cost/year is less than $500 million, a drop in the bucket for any modern nation and certainly no problem for CERN's 20 member states and six observer states.

Re:Worrisome

[Anonymous Coward]

What's worrisome is that these same scientists who can't seem to build this thing without some fatal flaw are the same scientists telling us there's nothing to worry about when they create a black hole.

Sorry if I'm missing intended humor in your post but that just doesn't make any sense.

These are construction flaws. The fact that the black holes they may be able to create are not a threat has nothing to do with any sort of special containment. It's simply that the size and level of energy is no where near enough to last even nanoseconds.

The ignorance about the dangers of particle accelerators is disconcerting.

By the way, if you want a good look at modern physics, read Brian Greene's "The Fabric of the Cosmos". Really good read.

Ok. From your posts, I gather that neither of you have ever worked with vacuum chambers. Let me tell you that having leaks is neither a design nor a construction flaw, it just *happens*. This is because, depending on the quality of the vacuum you want, you have to use only certain metals and glass and some very specific plastics. Seals are created by compressing a copper gasket between stainless steel flanges. Then, you will have to bake the entire chamber at about 150C for at *least* a week in order to get all the water out of the chamber. Some leaks are small enough that you will discover them only after bakeout. There are only limited possibilities to tell whether for example a copper seal is engaged completely tight. He leak checking is about the best you can do. So give those guys a break.

Re:Fermi has done it again!

[Anonymous Coward]

No. There's a huge community of folks here that very badly want the LHC to get going. Fermilab is a center for computing for the CMS experiment in the US.

Re:Worrisome

[Hurricane78]

Are you that badly informed, or just that unfunny?

There are giant amounts of particles with way higher speeds colliding with our atmosphere all the time, creating the same type of black holes.
The type that is apparently so unstable, that all those particles did not create one single black hole that are us all.
Go figure.

And try to not get your "knowledge" from the loudest and dumbest of all people.

Re:ZOMG,

[treeves]

Which choice of words? "Vacuum leak" is a common expression.

Re:Is that first thing we need ?

[Anonymous Coward]

Looking at the cosmic ray particle spectrum (google 'cosmic ray spectrum') one can see stuff at 10^20 eV, that is a lot higher energy than the couple of TeV these particle accelerators achieve (no mean feat). Here's a list of some observatories that look at cosmic rays:

Pierre Auger Observatory : http://www.auger.org/index.html [auger.org]
HESS : http://www.mpi-hd.mpg.de/hfm/HESS/ [mpi-hd.mpg.de]
MAGIC : http://magic.mppmu.mpg.de/ [mppmu.mpg.de]
Icecube http://icecube.wisc.edu/ [wisc.edu]

Re:Nope.

[thrawn_aj]

Masochistic is right. I just spent all of last week trying to find and fix a vacuum leak in my experiment. Luckily it was a (relatively) straightforward setup and after I got to the point where I wanted to strangle someone (or wreck the lab with a hammer - I'm not too fussy about my violent outbreaks :)) I just swore a terrible oath at the thing and machined a new one from scratch.

I'm not complaining about the work (it's sorta like having an irritating kid - no matter what, it's still your kid :)), but vacuum leaks can be seriously frustrating, especially the ones that show up with a delay so you have not the slightest farking clue where the damn thing is leaking.

Since I feel like venting (no pun intended :P), I'll let y'all in on just how one leak checks a vacuum chamber. The leak-checker is just a glorified pump that can pump down to really low pressures. The stuff coming out of the chamber is also directed into a mass spectrometer that is tuned to register and count (usually) helium atoms. You spray helium gas over the outside of the vacuum chamber and if there is a leak, some helium gets sucked in and registers on the spectrometer. The bigger the leak, the bigger the count and a simple calculation converts this into ccs of helium coming in per second with a pressure difference of 1 atmosphere across the leak (1 atm outside and ~0 inside) - that by the way is where the unit standard ccs per min (sccm) or standard cubic feet per min (scfm) comes from - you may have encountered these units in several diverse places (anywhere that gas flows are controlled through pipes). Of course, these days most of this is automated but we have this really cool leak-checker (Air Force surplus from the dawn of time :)) that is so freakin' awesome and not too automated. It's from the 60's and still works perfectly o.O

The bottom line is that finding a small leak* in a man-sized chamber is difficult to begin with. Imagine how insanely difficult it would be to do this in the frakking LHC! And there, since they deal with subatomic particles, they need even better vacua than I do. Gawd I'm glad I'm not the guy in charge of finding leaks - I'd probably start gibbering and running around in little circles if I had to deal with it :P.

____________________
*Here, small usually means somewhere around 1E-9 std.cc/s - at this rate, it would take more than 25 thousand years for a vacuum chamber the size of a beer stein to fill up due to air bleeding in from the outside. Much more than that actually since the rate would go down when the pressure difference decreases but this will do for now. And yet, such small leak rates can wreak havoc in delicate experiments (for instance, in a recent one where I was trying to measure the flow conductance of nanoholes - very tiny flows and leaks can screw things to hell).