Hacker News Reader

Comment by ImaCake | Hacker News Reader

I was briefly involved in Single Molecule Biophysics, where microscopes push the limits of what you can do with visible light in order to understand protein behaviour, where I heard the often repeated warning: "the microscope is an experiment itself". This was born out of the incredibly fancy and poorly understood microscope, which was experimental, and it was no guarantee that you were measuring the protein and not some weird light physics effect!

View on HN

> and it was no guarantee that you were measuring the protein and not some weird light physics effect!

My thesis was in fluorescent bio physics at the nano scale (STED). We literally had a PI effect. Normally, the 'PI effect' is that when the professor comes in, the experiment stops working, mostly due to Murphy's law. With our case, it was because the extra body heat caused the mirrors to expand and drift out of place, ruining the alignment. This also meant it took a while for the mirrors to come back into alignment after lunch and first thing in the morning. What a mess!

So yes, in microscopy, the apparatus very much is part of the experiment and it takes ages to get things controlled and imaging correctly.

I worked for a while on a team that designed an electron microscope from scratch. There was a phenomenal amount of work to take it from a bench top kluge that sometimes takes an image to a reliable product.

When designing sensitive instruments we had to use almost a paranoid attention to detail.

Thanks for your reply! I actually found out one of my current coworkers helped build a single molecule scope, he talked about the use of an air cushion beneath the giant breadboard of lasers to ensure they stayed nm accurate.

Most optics tables are floated on air pistons these days, they aren't super expensive. It helps with vibe issues and with shock absorption (there is a lot of slamming your head on the table when doing optics alignment :P ).

Stupid question: normally the labs I work in have excellent temperature control for this reason. In fact, some of the labs are actually rooms-within-rooms (with the outer room having AC, and the inner room having another AC). Was this for budget reasons, or were you physically placing your bodies extremely close to the optical equipment while running experiments?

The lab was in a bio lab, so the AC was set to STP. Well, mostly. Just having a warm person breathing in the tiny room was enough to throw the mirrors off. Since we were doing active tuning and development on the optics, we had to be in there, and it took a bit to get the room to a steady state with us in it. Fortunately, during the summer or winter, the only thing what would change was humidity, not the temperature, and that seemed to not affect things too much.

Also, not a stupid question. We thought of it too and it took us a week to collect enough data to rule out the AC system as a cause.

Not sure how to say it, but however brief your involvement was, I'm glad to read your comment here on Hacker News.

Reply on news.ycombinator.com