hulk SMASH

August 21, 2010

Shocking revelation for everyone: stuff breaks in science labs, all the freaking time.  I know, I know.  Stay with me here.

Anyway, the other day I was attempting to fix some broken stuff in my optics lab.  Well, maybe not “my” optics lab, but the one I work in every day.  Yeah, that one.  A number of little things had broken or at least were not working, and it all piled up into one big explosion of broken.

Okay, not THAT big an explosion of broken.

While a few broken things can almost always be ignored (I’ve met many a science experiment held together by tape), some of them really, really can’t.  One of those is anything that causes you to lose signal completely.  On Thursday, I had one of those on the optical path, as well as another one on the temperature sensor.  So here I was, hoping to measure optical signals as a function of temperature, without knowing the temperature or the optical signal.  Yeah.

Anyway, for the optical signal, I spent a good couple hours troubleshooting detectors and amplifiers, moving around optics, even checking the monochromator…no signal.  I quit in frustration and went and read comics on the internet for half an hour.  Grrrr, stupid experiment, grrrr.  Then I went back into the lab, pretty much prepared to just take a cursory second look and then turn things off for the day, when I found the problem.

I had a shutter closed.  A shutter that’s intended to completely block the signal when you want it to.

Yeah, I’m an idiot.

In my defense, I never use that shutter, and the fact that it was closed was probably a complete accident.  Nonetheless, I ended up largely dismantling a sensitive system in search of a problem that didn’t actually exist.  It took me another hour or more to put everything right.

After fixing the first nonexistent problem, I was feeling elated, although rather silly to be sure.  I moved on to repairing the temperature sensor.  Taking a first look at it, I was confronted by a huge tangle of wires, some of which were severed.  I retreated quickly into looking up a .pdf of the user’s manual for the temperature controller online.  It was largely unhelpful, but it did let me at least hazard a guess at what some of the tangle was for.

So, back into the fray I went.  After a bit of investigation, it was pretty easy to tell which wires did which, and since I’m a big smart scientist and all, I knew that the broken ones needed to be reconnected.  Yay!  Except they’re temperature sensitive, so soldering was out.  And oh yeah, one minor detail: they’re the tiniest wires I’ve seen.  I had to strip the coating off these mini-wires, and twist them back together, then protect the join.  Not hard really normally, except I could barely see this wires they were so thin, and I had to work in the dark (another optical experiment was running).

I felt like The Hulk trying to play the piccolo.

These hands were made for smashing, not for delicate music.

So that took a little longer that I might have hoped, and the wires got a little shorter on the way than originally planned.  Fun times.  But guess what?  After hours of repairs that weren’t always necessary, the experiment WORKS BITCHES.  HAH!

When you ask me why my PhD takes so long, now you know.



  1. I once spent weeks trying to track down a problem with a wonky board (which read data from a small drift chamber). This is literally weeks, except that only a small portion of the total time was spent on the issue, until the last week. Every day for five days I’m talking to people, looking at read-outs, blah, blah, blah. Well, I have no experience with electronics (or not enough to matter), and suddenly I’m at the point where the only people that know anything about this board are either a thousand miles away or are telling me things that I can’t understand (the latter being the engineer who built the board, but also built close to a hundred just like it). I can’t understand them because (as it turned out) they weren’t true!

    I spent 20 hours over two days in the LEACH (which is a small trailer-like building that is inside a much larger building which is really just a cover for a pit 6 stories into the ground and is just a 3-foot-thick concrete wall away from a very radiation-hazardous environment). Oh, joy. I spent them on the telephone, poring over the output of said boards (there were two), whilst on the phone with a professor in Colorado (I was at SLAC), when possible e-mailing (this was 1988, e-mail was practically unheard of outside of academia and a very few geeks) the output so we could go over it together along with the code which was to analyze the output of the board (beautifully spaghettied FORTRAN).

    Finally, after all that, I discovered that the boards were not, as the engineer claimed, identical, but had different parts (different FPGAs no less, in a twist that would not revisit my life until many corners had been turned) with different programming and so were producing different signals for the same input (but not, of course, the test input, because that would make everyone’s life too fucking simple!

    Am I bitter? Nooooooooo! It’s given me this story!

    May you be so lucky as to have only the story above.

  2. […] of my first crushes.  Her blog makes me laugh a lot.  To get you started I recommend this post on what working with fancy scientific instruments is really like and this one on how shitty being a grad student with a chronic illness can […]

  3. […] was paid out as an hourly wage for 50 40-hour work weeks a year, but we all know that’s not how many hours I worked), but it’s not enough to make me enjoy my job or to pay the bills […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: