subnet, Austria, September 2020

'Closed System' (working title)

Week Two (23rd - 29th September 2020)

To visit Week One click HERE and to see Week Three HERE

The cooling system has been attached to the supporting baseboard structure, with measures in place to reduce transmission of the sounds of the motors.

Last week I already made some recordings, using Arduino to control/power the pump motors as they would not function with recorded sound. I will return to this approach to see if I can make it work this way, but random Arduino programming is causing some nice erratic rhythms right now and accentuating the movement of bubbles in the system.

The short video shows four aspects of the system with some recording methods - the ones you don't see are shown in the images below.

At times I am using contact microphones which have a wooden toothpick affixed. This is to soften the sound somewhat and also to prevent grounding issues when attaching brass piezo discs (the basis of the contact microphone) to metal objects, but it also works well with the radiator part of the cooling system as I can jam the toothpick into the small internal fins to pick up the sound inside.

You will also notice an Aquarian hydrophone being used as a contact mic. This microphone picks up the lower bass frequencies of the system that are not apparent when using piezo disc contact mics. In the last section of the video you will hear the effect of using this mic, but please listen with good headphones or speakers!

I am already feeling the focus of this project is slipping away from me as I am unable to power the pump motors with sound energy, and also the sound of the motors is more present than I has hoped, but this is of course because the sound is being transmitted through every single part of the system. Therefore I have to work with it, or find ways to lessen the effects...

Attempts at this have been made to amplify the sounds through various means acoustically. I have an old cassette walkman, donated by local arts and music community Data which I have modified in a manner demonstrated to me by Daniel Luna of Un Loquer, Medellín, Colombia back in 2013. It turned out that the cassette head no longer functioned so I replaced it with a miniplug socket. Now it works relatively well as a preamplifier for contact mics.

Connecting this to the system and adding a small amplifier circuit to power various speakers, I can now filter some of the sounds. One way is shown in the video to the left, where I have used a transducer inside a metal bowl, but placed it onto a computer fan so the lower frequencies are dampened and feedback is also lessened to a degree, although can happen if the volume is at a certain level because of all the elements being in contact with each other and the table.

I have included an annotated photograph to show the elements used in this stage.


I am writing this section during the final day of this second week of the project, as I am looking after the exhibition I am part of in Marseille currently. I have decided to concentrate on certain methods of extracting or amplifying the various sounds discovered while the cooling system is running, but utilise some methods I have never used before with a live sound feed.

I will position up to six microphones on the cooling system, choosing positions that sound particularly strong and/or different from the others, but also that hopefully don't pick up the motor sound too strongly. I will feed the signal from each microphone into my computer and use a programme called Ableton Live, a Digital Audio Workstation, where I can then automate several parameters of the sound, for example volume level, equalisation, clarity and even add some effects such as distortion or reverberation (the latter is very unlikely as I am not fond of this effect, preferring actual reverberation in a space).

So for this stage I have repositioned all the equipment needed to make this possible during the next week. I have added photographs to illustrate a little what I mean but for some readers this will not be clear until I make some video documentation.


In an attempt to explain to those not familiar with this type of programme, the upper photo shows the 'envelopes' of the parameters affecting the sound in a graphic manner. So if a line rises towards the right, this parameters will increase, and therefore if one declines, the parameter decreases.

The result I am aiming for is the sound heard directly from the cooling system will slowly alter and evolve as the signal of one microphone changes and then fades as we hear another. The whole process will run without further input and be fully automated.