A little while ago I blogged about acoustic conduction here and introduced the idea of acoustic conduction in solid materials. And earlier I wrote a bit about supports in a blog here, and discussed the problem of microphony (the actual mechanism of how vibration pollutes our signals).
The next things to think about are the principles of coupling and isolation – and it's actually quite a big topic. This is a materials and acoustic transmission thing and, in essence, is the way one physical object is connected to another in terms of acoustic propagation through that connection. And there’s a direct parallel here between acoustic transmission through material junctions and signals passing through electrical junctions – and it's about impedance matching. To me this is the key to understanding it all.
When we need to ensure the effective transmission of critical electrical signals through a connection, or between one section of a system and another, we sometimes talk about impedance matching. And in hifi there are many specific examples of this. Most speakers are somewhere in the range of 4-8 Ohm impedance and usually when employing a valve amplifier (who's natural output impedance would be several kOhms), we need to use an output transformer to create a fairly close impedance match. A digital transmission circuit should be impedance matched at either end (often with a small transformer) and the hookup cable matched too. RCA digital for example is normally configured to 75Ohm impedance. If we don't match these impedances we get poor signal transfer and lots of reflections and standing waves (particularly with the digital stuff).
Well its the same with physical signals (vibration) in materials. Hard materials (metals, ceramics, glass, hard woods and hard plastics) conduct acoustics at high speed and with very low losses, and soft materials (soft woods, rubber, soft plastics) are by comparison slow and lossy. And yes, there is science here if you're interested:
The unit of specific acoustic impedance is the pascal second per metre, often called the rayl, after Lord Rayleigh. The unit of acoustic impedance is the pascal second per cubic metre, called an acoustic ohm, by analogy to electrical impedance.
For comparison, the acoustic impedance of steel is around 45 Rayls, whereas the acoustic impedance of spruce is around 25 Rayls. Soft plastics and rubber are typically in the range of 1.5 - 4 Rayls. Note that the values of acoustic impedance strangely seem to go in the opposite sense to electrical impedance. A high Rayl figure represents high signal transmission.
So I hope you can now get a sense of where the debate and considerations for vibrations in materials is going when considering hifi equipment and it's setup. A typical component chassis is made up of materials with a high Rayl (steel or aluminium). It may be fitted with hard feet or relatively soft feet, we may consider using steel spikes or sorbothane, and it may be stood on a glass and steel rack, or a wooden rack. But do we ever give real consideration to the impedances of the various materials involved and how they may be affecting levels of microphony within our circuits?
So the term coupling and isolation can now start to take on a more scientific meaning in our minds. Coupling is when we impedance match through a material's interface, for the purpose of high vibration conduction (steel spikes on the bottom of a steel speaker stand) and isolation is when we impedance mismatch for the purpose of low vibration conduction (such as silicone isolation pads under turntables).
So when it comes to listening optimisation, these principles open up a huge range of considerations and options. Whether it's racks or flooring, plastic feet with felt on or steel spikes, damping layers (which might be causing problems as well as benefits) and so-on, they all have an effect on the levels and character of microphonic pollution of our musical signals.
Okay, I'd better stop there for this blog. I hope this information sows some seeds. I'll do the next blog in a few weeks and start focussing in on how we can do some tests and experiments with different supports to start discovering how particular components in our system respond.
Comments