Arts and Crafts in the Lab
Feb 14th 2007Wallabynerd alert
One thing that may surprise people who don’t work in a lab is how much stuff is held together by glue.
Glue, or tape, or apiezon wax. And most of all by the graduate students and postdocs. (Very few professors at research universities actually work in a lab after they’ve gotten tenure. Rather, they hire students to do it for them, while they sit around working on grant applications to get money, and reports on what’s been done with the last pile of money. And going to meetings. Lots of meetings. And occasionally teaching.)
A big part of this is due to the “temporary” nature of most experimental setups. “I just need this to take one set of measurements so I can see if the idea works,” people will say. “If it does, I’ll talk him into coming up with the money for a better setup.” Of course, there’s only two options at this point — if the experiment works, then clearly it doesn’t need any changes; if it doesn’t work, then there’s no point spending the money, now is there? And if you *are* getting useful results out of it, then John over here has some samples that’d be interesting to try with your idea, and I’ve always wondered what would happen if we … And so the “temporary” setup gets a permanent position in the lab.
Not better equipment, of course. “Can we get a motorized travel stage so I don’t have to adjust this micrometer sixty two times every experiment?” “Motorize it? Nonsense! I did that experiment when I was a graduate student and I didn’t need a motor.” Nor a more stable arrangement — “it’s working now and there’s stuff we need it for, you can do that later.”
Of course, the other problem is that very few academic labs are fortunate enough to have a complete set of every possible bit of equipment, all designed from the start to fit together. For instance, my lab uses quite a lot of optical parts. Did you know that there are about 30 different ways of holding a lens? Did you know that there are half a dozen different types of screws and posts for holding the lenses on a table? Did you know that these are all slightly different sizes, and that an 8-32 and an M4 screw have very nearly the same diameter but go in very different things?
Now, my old advisor had the sense, at some point before I started working with him, to absolutely insist on standardizing on one set of parts. The entire lab (set of students) combed the entire lab (facilities) and checked that every screw that *looked* like a 1/4-20 really was 1/4-20 and really would screw into the table, that every screw that looked like 8-32 really was 8-32, that every kinematic mount really could be attached to every post and that every post could be mounted into every post holder. Painfully dull, I know (I heard about it from the survivors), but the result was that you really could just reach into a cabinet and pull out the type of mount and the size of post you needed and you could *get stuff done*.
Not so in my current lab. Some people order from Thor; some from Newport; some from Edmunds. I suspect some stuff has been imported from Mars. And the result is that you can spend an entire afternoon trying to pair together a lens holder and a post. *Even when you’ve worked here for years.*
3 Responses to “Arts and Crafts in the Lab”
Dianne Callender on 19 Feb 2007 at 11:18 pm #
Dad is replying right now as one scientist to another. As wife to one, I can assure you that most of his PhD experiment was held together with duct tape and apiezon wax. I watched him apply a lot of it. But I have an alternate source for your non-compatible screws and nuts and bolts and pieces: aren’t they leftovers from a previous project, such as repairing the lawn mower, or some lab device, and these are the parts that didn’t/wouldn’t go back into the engine?
ABC on 19 Feb 2007 at 11:28 pm #
How very true! Only it is broadly distributed: Also seen in government labs and in industrial; perhaps more so in the latter. Because it is a given, to the funding people, that, if you kludged it to work, that it will do “until we can get an outside source to pay for redesign.” And no outside source will do so, for they want to go on to the next thing. So you’re stuck with the kludge to make advances to the next thing. I once had a technician, who had, by his estimate, crawled 60 miles through the innards of a certain B-52 wiring it for instrumentation, working for me a few years, who very wisely said, “If you don’t have time or resources to do it right the first time, when will have enough time or resources to get it right?” I was impatient at that time, having to produce results *soon*, and my apparatus was justly described as “a large vacuum chamber obscured by apiezon wax on every fixture” — but it worked, and my successors on the project ever afterward sought unsuccessfully to replace the functional but cobbled together second-hand chamber and electron beam welder and other parts of the instrument.
Another possibility for the origin of the disparate parts of odd sizes is that they may be parts left over from fixing one’s car: GM uses both SAE and metric parts, not interchangeably, in the same vehicle. They may have been parts left over in reassembly, or off the old clutch, or out of the old fan, or …, and someone “Thought that looks like a useful size.”
But you have described the situation very well. Maybe someone could get adapters designed and built to allow different vendor’s parts to work together. Or, quietly toss the smallest set in the middle of the night. Or arrange a swap with another lab, so each could be closer to homogeneous in its own hardware.
Fresh Wallaby Juice » Blog Archive » When explaining isn’t helping on 02 Aug 2007 at 8:28 pm #
[...] (Yes, there’s ways to compensate for that. No, I don’t really want to set one of those up, even if we had the parts handy. Which we don’t, because it’d just be silly for a spectroscopy lab to have any extra optics lying around. Some days it’s actually hard to find a piece of plain old glass. Much cheaper to pay three postdocs to spend a day digging around in a closet to find something that will work one-tenth as well, then spending a week setting it up. But I’ve already done that rant.) [...]