Hi @BRCSD, good question. Others may want to respond with their own classroom Shake placement experience, but I want to add mine too. I have no knowledge of the building you’re in, what floor you’re on, etc., but basically there are a couple of guidelines, in order of importance, that should point you in the right direction:
The Shake should be situated as near as possible to a structural element of the building. For example: a load-bearing wall, joist, or girder directly tied to the ground (ideally, bedrock).
This should have the dual effect of slightly reducing mechanical and footstep noise. This effect can also be achieved by, for example, putting it in a basement or ground floor, but then you don’t get the same tactile feedback of being able to stomp around right next to it…
As far as possible from footsteps and mechanical noise sources like furnaces, pumps, and fans.
Depending on the building this may be impossible to really achieve, and since you can get a somewhat similar damping effect by putting it near a structural element, it’s sort of secondary. As you’ve noticed, noise levels increase during the day for various reasons anyway (heating systems, people walking around, cars driving by outside, etc.)
The upside of mechanical noise, however, is that it generally occupies a very narrow band of the frequency spectrum. So if you’re looking at a spectrogram in SWARM, for example, it’s really easy to tell that there’s something spinning, somewhere nearby, that’s causing noise to appear in the data.
As far as I know, there’s no way to filter a helicorder in SWARM to only show specific frequencies of data. I think someone has asked the USGS about adding this in the past, but as of now they have not implemented this feature. So it’s probably not your SWARM settings that’s causing noise to appear.