Tuesday, February 08, 2005

Day 4 of the evening Shifts

The day started oddly: the power went out. I took my razor to the trailers to shave, arrived at the busiest time of the day: the parking lot was absolutely full, except for the slot by the front door that a fellow was just vacating.

Some monitors, such as the one above me that displays 11 Xterm windows for the Consumers, have had the same configuration so long that the monitor is suffering screen burn.

The accelerator suffered a quench earlier today. "Tev Plan: next shot setup around 18:00-19:00." In the meantime we're doing what's called "L2 torture testing," which shakes down our Level-2 trigger system. And we're doing more Pulsar testing.

Did I explain that the Pulsar board is a VME card that has room for several "mezzanine cards" (plug-ons) that give it more memory or provide different kinds of I/O protocols. It carries a pretty standard FPGA, which is like a program-it-yourself computer: You get to specify what happens with every bit at each clock cycle, how big the word size(s) are, and so on. They have pretty obvious application in triggering; where you want to rapidly process a haystack of hits to find a needle or a hundred of tracks, or to organize silicon data, which is read out from the silicon system with an eye to speed rather than ease of use. You can learn more than you ever wanted to at this site.

There may be a ground fault in the accelerator. They may find it soon, or may need to turn off the main injector and go look (which takes quite a while).

I understand the Superbowl was played last night. I wonder who won. No, we didn't tune any TV to the game. That's a problem with cable and TV monitors. Years ago somebody unplugged one of the ubiquitous accelerator info TV's and watched fuzzy images of some game or another, but that's a bit harder when the TV is just a monitor. I'll live, I think.

I gather that the network decided not to stage a simulated sexual assault this year.

I'm calculating and typing and from time to time checking the Wall Hadronic Calorimeter histograms. One two channels are hot, but their hotness comes and goes. What am I typing? Stuff like "m^{\prime\prime} e^m + {m^{\prime}}^2 e^m" Why type this sort of gibberish (LaTeX)? Because it is way easier to do than work with Microsoft Word or OpenOffice. OpenOffice even borrowed the LaTeX syntax for creating equations, and it is still more awkward than using straight LaTeX. MS Word is a joke.

Duh. One of the hot channels is due to a TDC problem. We're running cosmics, so TDC problems dominate. In a real run, stray energy forms the background.

TrigMon errors filled up the disk. About a dozen worthless files, from 1 to 3 GB each. Hmm. Deleted one too many: one looked like it might have been useful. Actually, now that I think about it, I could have whipped up an awk script to parse out the crud. The expert we called suggested we call a different expert next time.

The Ace (a new one) is about to start some calibrations. I've not done verification before. . .

Before he could finish the calibrations, the job ran out of space. Some InhibitGUIxxx file was growing beyond all bounds. It is gone now: Ah, Bill killed it.

Slowly figured out how to use DBANA for the calibrations. The documentation is not very uniform. Some things were simple, others tricky.

No beam tonight. They're still looking for their ground fault...

No comments: