Ambient music where no-one would dare listen
The third of three videos I made on Sunday 16 Nov 2025, and the second to feature this new Explorer, as I call it. Here it floats into a hidden world above the M25, north of London. It's an aqueduct approx 100m long, with a very low covering, making it utterly inhospitable to this particular sound artist, and, I suspect, almost anyone else. But I am happy to send my instruments into it, to capture video, and play/record some sounds in there. This was the result, making its way along the length of the watercourse in real time.
Heart Of Noise Cambridge #14
I gave a short talk on the Randomatones at Heart Of Noise Cambridge #14. As well as showing the usual red and blue standing units, I demonstrated a new prototype hanging setup where the speakers rotate on their support cables.
New hanging speaker prototype
A prototype arrangement that takes last year's Walthamstow Engine House installation up a step, with a motor at the top of each suspension wire to rotate the hanging speaker when a sound plays. This is unlike the normal rotation of the standing or floating units I have built so far, which is random. Here, the idea is that the rotation helps to indicate that a sound is playing from the speaker. There are already LEDs to do that (green for off, blue for quiet sound, red for loud sound), but I want something to make the installation more interesting to look at, especially as people (not least me) like to video. Rotating the speaker also makes the sound more interesting in a drier acoustic environment.
Building this prototype, I discovered couple of problems with the idea. At first, I had the motor down at the speaker end. This was so that I could put electronics on the speaker unit that would trigger rotation from the LED setup I already had. I got all the electronics working, but the motor didn't have enough power to turn the speaker. For no reason I can remember, I tried putting the motor at the top, and hanging the cable instead from it. It rotated fine. Vague memories of physics tell me this is to do with angular momentum. At the speaker end there is a lot, because it is somewhat heavy and bulky. Where the cable is attached, except for some inherent elasticity between it and the speaker end, which acts only within the very slim cable, there is almost none.
Another issue is that having been careful to get the rotation starting and stopping at the beginning and end of playing a sample, only a very light speaker will move that responsively. The horn speaker here is well over a kilogram, meaning the cable twists before the speaker starts turning. With this lag before the heavy horn speaker eventually moves, the rotation happens out of sync with the sound being played. It may as well have been random.
In a week I will test this setup live at Heart Of Noise Cambridge, and I'll see how it looks, then decide what to do about controlling the rotation. In this prototype I am driving movement from the Raspberry Pi, meaning I have some control over when it happens and for how long. In the Engine House I'll be using a different computer with none of the GPIO ports a Raspberry Pi has. Random movement is not that easy with ancient electronics.
Building this prototype, I discovered couple of problems with the idea. At first, I had the motor down at the speaker end. This was so that I could put electronics on the speaker unit that would trigger rotation from the LED setup I already had. I got all the electronics working, but the motor didn't have enough power to turn the speaker. For no reason I can remember, I tried putting the motor at the top, and hanging the cable instead from it. It rotated fine. Vague memories of physics tell me this is to do with angular momentum. At the speaker end there is a lot, because it is somewhat heavy and bulky. Where the cable is attached, except for some inherent elasticity between it and the speaker end, which acts only within the very slim cable, there is almost none.
Another issue is that having been careful to get the rotation starting and stopping at the beginning and end of playing a sample, only a very light speaker will move that responsively. The horn speaker here is well over a kilogram, meaning the cable twists before the speaker starts turning. With this lag before the heavy horn speaker eventually moves, the rotation happens out of sync with the sound being played. It may as well have been random.
In a week I will test this setup live at Heart Of Noise Cambridge, and I'll see how it looks, then decide what to do about controlling the rotation. In this prototype I am driving movement from the Raspberry Pi, meaning I have some control over when it happens and for how long. In the Engine House I'll be using a different computer with none of the GPIO ports a Raspberry Pi has. Random movement is not that easy with ancient electronics.
