jump to navigation

A few pictures of the CMX muon system February 22, 2007

Posted by dorigo in personal, physics, science.

As more forgotten pictures emerge from my hard drive, I am brought to remember my years as a Harvard post-doc, when I worked at the commissioning of the upgrade of the Central Muon eXtension (CMX) for the CDF II detector, in 1999 and 2000.

The CMX is a system of detectors designed to detect muon tracks as they exit the central detector in the pseudorapidity region [-1.0,-0.6] and [0.6,1.0]. That just means muons with trajectories making an angle with the beam between 40 and 60 degrees, very roughly speaking.   They are made of eight layers of single-wire drift chambers measuring 6″x1″ (inches) in section and about two meters in length. Each layer of chambers is staggered with respect to the others, so that all muons hitting the detector are bound to leave a signal in at least four layers despite the uninstrumented regions between the chambers. The 8 layers are sandwiched by scintillator planes, read by phototubes at each end. The system has a nice redundancy, which allows for a high detection efficiency.

All of the 2135 CMX drift tubes existed well before I started working at it: already in Run I CDF had four arches of CMX sections, covering 120 degrees of azimuth each – 240 degrees covering the positive rapidity region [0.6,1.0] and 240 the negative rapidity region [-1.0,-0.6]. My job was to assemble, test, and install the 35% of chambers left behind (on the shelves in the Harvard HUHEPL building), which were necessary to cover one of the two 30 degrees gap at the top of the detector (with a “keystone” module) and two 90 degree gaps at the bottom (with two “miniskirts”).

Below you see the two modules making the “keystone”, assembled and installed on the ceiling of the CDF collision hall. A railing system allows them to be taken apart as they appear here.

Below you see one of the north 120 degree arches, set just a few inches away from its working position. At the top there is the juncture with the keystone. In the background, the yellow iron of the forward toroid, with a double ring of extrusion called “snout” made with 24″ thick steel, to stop stray particles from the beam from hitting the muon chambers.

About the snout, it was designed by none other than Melissa Franklin herself. I think I will take the time to discuss the problem here for a second, since it illustrates very nicely some issues with the design of particle detectors.

Muons are the only particles we can detect which traverse thick slabs of matter undeflected, so we place “muon chambers” outside the massive cores of our particle detectors to measure their ionization and reconstruct their momentum. Very few light hadrons are able to punch through our calorimeters, and their rate is manageable and they constitute a tolerable background to real muon signals. However, when in Run II the position of the forward toroids (of which you see one in the picture above) had to be moved closer to the central detector to satisfy the needs of the improved design, a problem arose: particles exiting the interaction region at small angle could hit the structure, bouncing off towards the muon chambers without traversing enough matter to stop them, and producing a large background rate, as pictured in the following graph:

Bringing the yellow structure closer means that the path of green tracks (background) to the CMX arches is much shorter than in Run I, and thus their arrival time is much more similar to that of real muons which had to cross the central detector. The two scintillators of which the CMX is endowed were able to reject the late signal of backgrounds in Run I by electronic modules which perform the average detection time of the two light pulses: this improves the timing resolution and allows to reject the “green tracks” of the picture above. Below you see a cartoon describing the issue: on the left you see the two scintillating planes (in black), and the two light signals (blue arrows) resulting from the passage of a track (in red). On the right, the arrival time of the signal (in red) and the arrival time of the background (in green) are separated by an electronic “gate”, a requirement that the signal is synchronous with a certain time window.

In Run II, the time difference is insufficient due to the shorter path of background tracks, and thus the need of the two “snouts” shown in black in the first graph above.

Now for the other pictures. The keystone is the first module I worked at, in the winter of 1999. After assembly and installment on the ceiling, it required electrical work. I remember hanging up there, 30 feet above ground on a flimsy cherry-picker… It was fun! below, you can see some detail of the wiring, the preamplifier boards, the gas system tubing (transparent), and the high voltage cables (in red). The thick black cables are those carrying out the amplified signals. Below and above the pack of drift chambers you can also see the scintillators and their curved light guides.

Bear in mind that these detectors had been designed in the early eighties, and produced in the late eighties for CDF Run I…. Nowadays the muon chambers look much fancier than that!

Finally, a picture of myself besides my beloved chambers. On my right, you can again see the steel of the “snout”.


1. Carl Brannen - February 22, 2007

Of all the manlifts available, the one I hate the most is the scissor lifts. They have small bases and shake badly from side to side at any height above 10 or 12 meters. And you can drive them around.

The worst thing for me is having to sit around in one waiting for someone on the ground to do something. While I sit there, I think about how often earthquakes come around. Other devices don’t bother me so much, I don’t know why. In fact, they’re all dangerous.

2. dorigo - February 23, 2007

Yeah, I have worked on those as well – scary!
I must say, though, that I am unaware of stories of injuries with those devices. Probably I did not dig enough in the matter🙂


Sorry comments are closed for this entry

%d bloggers like this: