This next coming week is ‘ATLAS week’. It is a big meeting held a few times a year where a large fraction of the ATLAS collaboration gets together (mostly at CERN but sometimes at other places) to talk and hear about the status of the detector. I have only been to a few ATLAS weeks so far and I have to admit that they seem very diplomatic. Too diplomatic. Let me explain.
In the city of ATLAS, your group, which is the Univ of Chicago group for me, is like your family. They are the people you share offices with, the people to whom you complain and confide. Next, is your sub-detector group, which in my case is the Tile Calorimeter or TileCal. They are your extended family. People you may or may not like but are in any case stuck with. The other sub-detector groups such as the SCT, TRT, the level-one trigger, etc are like the rest of village. You know who they are, you might be friends with many of them but they are not family. The other detectors such as CMS is like the next town over with whom you have a football rivalry. For most of the year you eye them with suspicion, during the week of the ‘Big Game’ with outright hostility.
ATLAS week is therefore like a town meeting. And each sub-detector group gives an overview talk on its status. Now I understand that each group doesn’t want to display all its dirty laundry to the rest of the collaboration. But it seems like every talk has the theme of “Everything is on schedule. Nothing to see here, just move along”. And actually I believe that. I have yet to hear in the rumor-mill of major problems. But every sub-detector has made minor mistakes. What is important to note though is not that mistakes were made but that they are minor. So, I think it would be refreshing to see a talk called, ‘All of the stupid things we have done since the last ATLAS week (but everything is on schedule)’. That would be a nice change.
Because we all have done stupid things. These are very big and complicated systems. It is unrealistic to expect that every possible failure mode was thought of and accounted for. That everything was assembled perfectly and to spec. We are not perfect and therefore neither is sub-detector that we work on.
But glossing over one’s mistakes is something done rather often. I remember a particular case of this when I was a graduate student on SNO and I had gone to do maintenance work on the detector (which was located on the 6800 ft level in an active nickel mine). It was during the mine’s annual shut down, which meant we had very limited access to the detector. On this particular day, I remember that I am the only electronics person underground, doing some repairs knowing that we won’t have access again to the detector for about two weeks.
In trying to diagnose some problem, I am probing using an oscilloscope at the back of one of the electronic crates. I have to stick my entire arm between a six inch space in the powered crate to reach the back and I am probing, blindly, on a bunch of very tightly packed pins. The other hand is changing the scope settings. But because the probe’s cord is not long enough, I have the scope propped on a chair. To balance the chair+scope, I am standing like a flamingo; one foot on the ground, the other hovering in the air keeping the scope from smashing into floor. To top it all off, I am on the phone with one of Penn’s electronic masters, Josh. So I have the phone wedged between my ear and shoulder.
I think it is pretty obvious where this is going.
At some point while fiddling with the scope and arguing with said electronics expert, a bright blue sparking at the crate becomes apparent in my peripheral vision.
Now any self-respecting high-energy-physics’ electronics system has a overabundance of LEDs. And these LEDs can have one of two meanings. Either:
They are on and they should be off. Meaning: you’ve got problems.
or
They are off and they should be on. Meaning: you’ve got problems.
In my particular case, all of the ‘off’ ones turned on, and the ‘on’ ones turn off. Meaning: I’ve really got problems. What had happened was that as I was probing on some very finely spaced pins, I shorted one the ECL signal lines to a +24 voltage line, thereby blowing every single ECL driver in the crate. Of which there are many.
Like many academic institutions, the Penn group is all about ‘Learning Experiences’. Such as
Monica: “uhhh… I think I just blew all the ECL drivers in the crate”
Josh: “Well, let this be a learning experience for you. The cage goes up to surface in four hours. That crate had better be working’.
And that is what I spent the next four hours frantically doing. Trying to get those ‘off’ LEDs off and those ‘on’ LEDs on.
Later on some call, when someone asked why certain repairs hadn’t gotten done that day, the response from Penn was, ‘we had some unexpected electronics problems.’ And even though at the time I was relieved to have been spared collaboration-wide humiliation, it would be been refreshing if the response was, ‘Man, you think your graduate students have messed up. Let me tell you what OUR graduate student did!’
So, to anyone preparing an ATLAS overview talk for next week, I say; We are all part of the same small town. If you tell us about your stupid mistakes, we’ll tell you ours. And we can all have a good laugh. And then move on.
