overdue for an update
Hmmm, what have I been up to? So I've written up a quick review of the SCUBA diving that Christine and I did in St Lucia: it's here. I'll try to write about topside activities at some point.
Been rather busy at work running experiments on the sample that I prepared. In fact, I think it deserves a rant.
So, most NMR experiments have a large number of parameters whose values need to be properly calibrated, or else the experiment will work like a flaming pile of crap. Some of these parameters must be recalibrated for each new sample that is put in the machine. Other parameters remain almost completely constant, and only need to be recalibrated for a change in hardware. It's the responsibility of one of the facility members to publish these calibrations. (see where this is going?) So I get on the spectrometer, get my sample all set up and calibrated. As I'm setting up the experiment, I realize that some of the permanent calibrations that I need were never done on this machine. Grrrrrr. I was able to determine approximately what the calibrations should have been, and tried running the experiment. I think it worked.
Fast forward to Friday of last week. I've moved to a different spectrometer that appears to have free time. Then I realize why it has so much free time. Turns out that this special calibration program was run on the machine, to make the amplifiers respond in a very predictable way, so that careful calibration would no longer be necessary. However, this tossed all of the previous calibrations out the window. And no one bothered to check whether or not this program actually worked. Mmmmmm, fun. After asking around, I eventually found a few calibrations that allowed me to run a couple of experiments.
It's really infuriating how it seems like (at least here), people are interested in physical infrastructure (new magnets, faster computers, the like), but they seem unwilling to invest in informational infrastructure to help people use the physical infrastructure in the best way possible. Sigh... can't trust anything anymore - have to do it myself.
Been rather busy at work running experiments on the sample that I prepared. In fact, I think it deserves a rant.
So, most NMR experiments have a large number of parameters whose values need to be properly calibrated, or else the experiment will work like a flaming pile of crap. Some of these parameters must be recalibrated for each new sample that is put in the machine. Other parameters remain almost completely constant, and only need to be recalibrated for a change in hardware. It's the responsibility of one of the facility members to publish these calibrations. (see where this is going?) So I get on the spectrometer, get my sample all set up and calibrated. As I'm setting up the experiment, I realize that some of the permanent calibrations that I need were never done on this machine. Grrrrrr. I was able to determine approximately what the calibrations should have been, and tried running the experiment. I think it worked.
Fast forward to Friday of last week. I've moved to a different spectrometer that appears to have free time. Then I realize why it has so much free time. Turns out that this special calibration program was run on the machine, to make the amplifiers respond in a very predictable way, so that careful calibration would no longer be necessary. However, this tossed all of the previous calibrations out the window. And no one bothered to check whether or not this program actually worked. Mmmmmm, fun. After asking around, I eventually found a few calibrations that allowed me to run a couple of experiments.
It's really infuriating how it seems like (at least here), people are interested in physical infrastructure (new magnets, faster computers, the like), but they seem unwilling to invest in informational infrastructure to help people use the physical infrastructure in the best way possible. Sigh... can't trust anything anymore - have to do it myself.