This blog post was written for Elin Darelius & team’s blog (link). Go check out that blog, more cool stuff being added there every day! :-)

—

Remember this little office on top of the tank? Exciting stuff going on there: Testing of the cameras that were just being installed in the last post! (Of course it will get even more exciting once we start rotating! :-))

The office — co-rotating with the tank. Important decisions being taken down there!

Of course we need to know which part of the tank the camera captures exactly, and what any given length on the pictures relates to in real life. That’s why we measured the topography in the last blog post.

Below, you see that office again and additionally get a glimpse one floor down into the tank, and people moving a large grid there:

The office — co-rotating with the tank. More important discussions!

This grid, with exactly known dimensions and positioned in different ways above our topography, lets us calculate the focal length of the camera, which will ultimately give us the possibility to calculate back from pictures taken from far above to actual lengths in the tank. This is necessary to calculate the currents inside of the tank.

The calibration grid on our topography. This will help us connect the pictures we are taking to actual length scales inside the tank.

Easy as this sounds, I can assure you that it is not. So many considerations need to go into this now, for example where do we want to put the origin of the coordinate system when we interpret the data? And how should the axes be oriented? We are in the Southern Hemisphere, so should we use this to determine the direction of the x-axis? Or should the direction of the jet that we will be introducing into the tank give us our x-axis orientation? The source (where we will have the inflow of the jet later) seems like an obvious origin of a coordinate system, but we will move it around on the topography, so it really is not.

Decisions made now will make it a lot more convenient (or inconvenient) to work with the data for years to come, so lots of different considerations going on right now…

First critical looks at the calibration. This is such an important step!

And there are two cameras that need to play well together, or at least the data coming from them needs to be connected seamlessly to each other…