Life as a Physicist

Last post I mentioned the LHC update that was given at a recent meeting at CERN. One cool thing Steve Myers’ showed during his talk was a discussion of the quality of the splices and how it might affect the LHC’s ability to run.

For a sample of the trade-off, check out this plot, stolen from page 46 in the talk.

Along the x axis is the measured resistance between two magnets (across the splice). The units there are nano-ohms – something only the most expensive multimeters can measure. If you remember your Physics 101 course, you remember P=I^2R (power is current squared times resistance). The units of P are Watts (!) – just like your light bulb. These are superconducting magnets, of course. The magnets are very powerful and so have 1000’s of amps of current flowing through them. So even small R’s mean decent heat sources. Heat warms up the magnets, and makes them no longer superconducting – and that can be a disaster (a few of these is not a problem – it happens every now and then – but a chain reaction is what caused the last September accident). So – the splices, which aren’t superconducting, need to be excellent and have almost no resistance. Like 10-15 nano-Ohms.

The Y axis is how much current you are pumping through the magnet. Current is proportional to the magnetic field, which is proportional to the energy we can run the LHC at. As you can see, if you can run at about 6700 amps you can run at 4 TeV. If you run at 8300 amps then you can run at 5 TeV.

The red and green lines are the keys to reading this plot – they are two different conditions for the state of the copper joints. The LHC machine folks always talk about the worst case scenario (the red line) – but I’m not 100% what the difference between the two is. Lets say you want to run at 5 TeV. Follow the 5 TeV line over from the left of the plot until it hits the red line. You see that it drops down to 58 nOhms. That means all splices have to be less than 58 nOhms in order to run at this energy. The machine is full of these splices. So this is a bunch of work checking these guys!! [listen to the video on the agenda page at about 30 minutes in]. So, one of the things the LHC engineers are doing is measuring all the splice resistances and then putting them up on that plot to see where they are.

BTW, nominal is 10-12 nOhms, and they need to be less than 25 to run at a full 7 TeV (two beams at 7 TeV gives you 14 TeV, the design of the LHC).

Sorry if this is old news…

CERN management recently had a council meeting. These meetings take place between the council and the CERN directory general. Big funding changes, new projects, major schedule changes, a new country wants to join CERN, etc., all have to be approved by this council. As you might imagine the recent council meetings have been dominated by the “schedule changes” (I don’t actually know as a function of time if that is true, but I would imagine).

What is nice about the current CERN DG is that he usually immediately sends a message out to the public and the the CERN folks. Much better than reading about an updated CERN LHC schedule in the Geneva newspaper. Even better, a presentation to all of CERN (and an open webcast) is schedule. The last one just happened (there is a video link and slides link at the top of the agenda, just under the main agenda title).

Everyone is eager for data. I’ve discussed what I think are some of the pressures on the accelerator division previously. This meeting is a continuing part of that conversation.

It is clear they are doing a huge amount of work. They have a lot done. A huge amount. From a physicist’s point of view, the most frustrating thing about Steve Myers’ talk was there is no date and no energy. It wasn’t clear to me what the plan was until someone asked a question at the very end of the talk. Basically – they will have measured the splices (electrical connections) in all of the LHC in early August. Those splices are what caused the disaster last September – so it is important that all of them be carefully measured. And once they have measured everything – then they can start a discussion with the experiments on start up schedule and energy.

Next time something on the trade offs…

Well… I wrote this post last night – before the press release that arrived in my email this morning at 10:50am (pacific time) – there will be no beam in the LHC until early spring… I’ll leave the rest of the post below, though I don’t know if it has any value at this point… But I’m too lazy to write a new one!

I’m not privy to inside information (if I was I wouldn’t write this post!). But you can do the following calculation yourself.

The LHC is going to be down for 2 months. That means the machine will be cold and ready to start power up in mid November. The current winter shutdown will probably commence at the end of November. That means only two weeks of time to power the machine up, put beam back in, try collisions (at 900 GeV), sync up the RF system and perhaps try acceleration. 900 GeV, by the way, is the beam injection energy: one doesn’t have to commission the whole acceleration sequence in order to run the LHC at 900 GeV.

Not a lot of time. From the accelerator point of view I’m not sure it makes sense. ATLAS and CMS, as well as others, however, are probably dying for any data they can get their hands on. They have had a taste, but even a week of collisions at 900 GeV would go a long way to helping with the commissioning. That isn’t to say that both experiments won’t have plenty to do during the winter shutdown even if the collisions don’t occur. I believe both will want to open up their detectors and get inside to the parts that are not easily accessible, for example.

Finally, there is politics. Closing out the year without collisions will be “a tough beat”. The newspapers are watching everything we do and they will surely report on this as a disappointment. Scientifically (given the information I have – which is the same that everyone else has and is undoubtedly not complete), it seems like the case for beam this year is on thin ice.

This whole post is based on the premise that the problem that occurred is as the LHC accelerator division expects: something that will take one or two days to fix. If a magnet needs to be swapped out with a spare or something similar to that, then I’d guess we will almost certainly not have beam. We should have the answer to that question as soon as they have a look inside the magnet, which will occur when it is warm in a little less than a month. As CERN has said:

“It’s too early to say whether we’ll still be having collisions this year,” said James Gillies, chief of communications for CERN, in an e-mail message.

The winter shutdown is due to the very high electricity costs in Geneva. I’m used to electricity use spiking in the summer as everyone turns on their A/C. Anyone know what causes the high cost in Geneva during the winter?