Tuesday, April 05, 2011

Designing a lab

Designing a lab is not trivial, particularly if you have no experience in doing it before.  My new lab (day 2 of the move....) was perhaps the ideal circumstance: a new building is being constructed, and you have a very free hand in determining the layout, the facilities, and so forth.  In any realistic process you never get everything you want (e.g., this building does not have a building-wide deionized water system; I can't have unlimited space; there are restrictions based on cost and feasibility).  The challenge is to end up with functional space - laid out intelligently, so that work flows well and you don't find yourself fighting with the building or yourselves.  Sometimes this is not simple.  In my original lab space, for example, that floor of the building was never designed with vibration-sensitive work in mind.  The need to position certain pieces of equipment on the vibrationally quiet parts of the floor strongly influenced lab layout, rather than basic experimental logic.

Lab design ranges from the Big Picture (e.g., I have a couple of optics tables, so I should probably have a separate area with independently controlled lighting; I want isolation transformers to keep my sensitive measurement electronics off the power lines used for my big pumps.) to a zillion little details (e.g., where should every single electrical outlet and ethernet port be positioned?  What about emergency power?  Gas lines?  What fittings are going to be on the chilled water lines?).  Nothing is ever perfect, and there are always minor glitches (e.g., mislabeled circuit breakers).  You also want to design for the future.  If you think you're eventually going to need a gizmo that requires chilled water or a certain amount of 480V current, it's better to plan ahead, cost permitting....  The situation is definitely more constrained if you're moving into pre-existing space, particularly in an older building.  Like many aspects of being a professor, this is something that no one ever sits down and teaches you.  Rather, you're left to figure it out, hopefully with the help of a professional.


Heumpje said...

I have just finished moving our lab. We had to move an STM and an ARPES setup. The nicest bit is to have it finished :). We have moved into a completely new building and we had some input during the construction process. From the experimental point of view things have gone very smoothly: the ARPES setup was taken apart and put together again (by me) in only 3 weeks. The worst bit was connecting wires. This has taken the best part of three months. The setup in the old situation had organically grown over a period of several years and was a complete tangle of wires (measuring and power cables all tangled up). The new setup has this all sorted out, well labelled cables, measurement wires separated from power etc. It is a large effort, but if done properly one can benefit from the process for years to come.

Joel Kelly said...

We are moving our lab today. Bleccch. There's been a few fun surprises, like realizing the fumehoods won't be functional for another two months.

I guess I may be a wee bitter since I'll be graduating in a few months without getting to spend much time in the shiny new building. Which is quite a silly thing to be bitter about, since I'll actually be graduating!

Craig said...

We're just coming out of major lab renovations, trying to keep the existing equipment (UHV systems) online so we could keep working for the ~3 month construction period. The process has taught me:
1) That if a contractor tells you a time to meet you or a date for something to happen, they're lying
2) If you don't want a contractor working on something without your supervision (for a hypothetical example, ripping down an interior wall of your laboratory) you pretty much need to keep that something in your direct line of sight 24/7...

vibration consultant said...

It's stories like this that almost make me wish I'd stayed in materials science. There's just something about building a lab from scratch -- with nearly no money, of course -- that makes you feel like you know more about the world than just that narrow slice of science you're working on!

Anonymous said...

Doug, based on your experience would you recommend a serious lab building project or equipment building project for junior faculty (or a postdoc)? It seems like bad decisions in projects of this magnitude can be survived only by senior faculty.

Douglas Natelson said...

Heumpje, the kind of organic, very complex setup you describe seems to be very typical of AMO experiments as well. Hopefully our setup will be a bit more plug-and-play.

Craig, I've always said that general contractor is my fallback career, since apparently you don't even need to be very good at it to make money.

Anon., I'm not sure how you could ever really educate someone about this without putting them through the full exercise. As for doing this as a postdoc, while it would be a very valuable skill, I don't think it's really the right use of a postdoc's time (especially if the postdoc is just 2 years). Postdocs need to get research done on a rapid timescale, whether they want to be in industry or academia. Spending months designing or building a lab is not really conducive to that.

Heumpje said...

Unfortunately, the ARPES setup is not so much plug and play. It took little time to put the vacuum chambers back together, but a lot more to reconnect and realign things. I was the main responsible person on our side and did a lot of the preparation. During the move I was the one steering outside contractors, identifying and making sure that problems were solved. I did all the work on the system myself since there really is nobody else whom you want to trust with your 250K€ detector. It was a learning experience, but i don't think it matters if you are a senior scientist or not. As a physcisist you are trained to think
pragmatically and that is the only training you need.
Incidentally, I am a postdoc in my second year. I am working on a three year project and have some hope to get tenure here. Rebuilding the lab was not obligatory, but it seemed like a good test and quite some tinkering fun. Our lab setup has a net worth of 2 million euros, i managed to keep damages to the system at less then 1000 euros.

I also do not agree that it has to take all the time away from real science. I scored a PRL (actually my first as first author) and two others written in the period while we were moving. I have to admit that I am not 100% reliant on our lab setup since we also do experiments at the synchrotron.