LHC is Turning on FAST September 19, 2008
Posted by gordonwatts in Fermilab, LHC.trackback
During Frank’s talk he said that he was a bit "depressed that it took 25 years to build machine, but just a few weeks to commission" – implying it wasn’t enough of a challenge.
I am, frankly, amazed at how fast this thing has turned on. None of my previous experience had prepared me for how fast they have turned on. The engineering job is incredible. Here they turned on 10 September and they are talking about first collisions (at 900 GeV — really small compared to the planned 14 TeV final operation) this coming weekend!
Given the number of moving parts, electronics, etc., in this machine… well, this is a breath-taking to have gotten this far. As a long-time member of the US community I’m also feeling a bit embarrassed. Our last start up – Fermilab Run 2 – was not nearly as smooth. One key difference between the LHC and Fermilab’s startup, btw, is that all the bits are present in the accelerator from the get-go. Fermilab has continuously been adding bits to the accelerator to improve it (many bits were delayed due to the funding profile).
If we in the USA get a chance to build another accelerator we are going to have to make sure we step up and match (or come close to matching) CERN’s accelerator division!
UPDATE: Yeah Yeah. I know. 2 months of downtime now. 
What can I say? The accident was probably caused by me posting this. I don’t know if you saw the CERN press release – but note how it said that in a non-super conducting machine this fix would take a few days? The 2 months is because they have to warm the machine up and then cool it down again. What is a magnet replacement at Fermilab? 8 days? I wonder what the difference is in the cryo systems?
I am not going to be holding my breath for collisions this weekend or even this year after today.
Ironic, eh?! I wrote this the night before last… and the day it shows up… BOOM! Big quench (check this out: http://user.web.cern.ch/user/Welcome.asp for those of you not up on the latest problems from the machine…
But, JustAnotherStudent, how do you know that it will take all year to bring it back?
Real time to do anything in physics = Estimated time * Pi
Or, estimated time * i
I think at Tevatron they can warm up a single magnet, while at LHC they need to warm up a sector. As far as I know that would be the biggest difference. After a quench, as far as I remember, at Tevatron they need about 3 weeks to change a magnet …. that includes warming up, colling down (each 3-4 days I think) and then baking the beampipe takesthe longest. But yes, if there is no need for opening the vacuum I think a fix takes around one week.
way to jinx it Gordon
The message is that you should never write postings “Oh it’s so fast” right before a big quench.
No kidding, Lubos!!!!
I wonder if the LHC is slower just because of the cooling power required for the whole string rather than the single magnet. I should try and find out if there is anything extra going on.
Dear Gordon, I don’t want to discourage you but I find it obvious that the whole sector 34 will have to be warmed up, the magnets fixed, and the sector 34 will be cooled down.
It’s probably not just one broken magnet but the proton pipe may have lost the vacuum, too. I am not 100% sure but it is conceivable that in my text – click my name – there are new facts about it.
Lubos – no, I meant that I wonder why it takes so long for the LHC to warm up and cool down. The Tevatron would take 6 days for the same operation. I’m asking someone I known in the accelerator division to see if he knows why.
The official thing says we wait until monday. Whatever, it sounds a lot like they will have to break vacuum if it hasn’t happened already.