The materials and construction used to make equipment furniture is by no means the end of the story.
Each piece of equipment needs feet and those feet play an important part in the energy 'coupling' process. Fitted with the usual rubber feet (felt and cork are also used) the equipment acts as a reservoir storing the feedback energy - and stored energy is just what we don't want! We need to dump the energy into the floor.
Luckily, energy - like water running downhill - will 'flow' into a higher mass if it can. Rubber feet or 'high tech' squidgy feet prevent this happening. They are 'decouplers' that form a higher resistance to the flow energy, trapping it in the equipment and the feet themselves.
It is often claimed that these 'decoupling rubbers' turn the energy into heat. This is a fallacy. It is extremely difficult to turn energy from one form into another and the process is very inefficient. Just how difficult it is and how little of the input energy is turned into heat is quite surprising. A few years ago I asked one of the biggest manufacturers of 'decoupling' rubbers if they could tell me exactly how much energy was turned into heat in their products. After a short silence they admitted that it was so low that they hadn't been able to measure it! That is just what I expected to hear.
Consider this for a moment: the kind of rubber we are talking about is used in squash balls. These balls are designed to bounce very little (and there are different grades or bounce rates) and to be hit very hard into a hard surface. If they convert energy into heat at any appreciable efficiency at all the heat generated by hitting the court wall would cause them to melt and run down rather than bounce off!
So what does happen to all the energy you put into the ball with the racket? In simple terms the energy has just changed down in frequency to the 'resonant frequency' of the rubber. Every material has a 'resonant frequency' the level of which is determined by its mass. Every object or substance converts energy into its resonant frequency and so stores vast amounts of energy. The low bounce rate of the squash ball is an expression of the low resonant frequency of the rubber- you can hear the 'thud'. A tennis ball, on the other hand, bounces better and you can hear the higher frequency in the 'dong' ring sound.
If you use this sort of material as equipment feet (and some people use squash balls cut in half) the feedback energy is converted down to the resonant frequency of the rubber and it just sits there and 'wobbles'. I have to tell you that I think this is a very bad idea! The destructive effect on sound quality is enormous. It increases the 'time smear' and changes the balance of the sound to a slow, boomy bass heaviness.
The Solution for Equipment
A hard material with a highish resonant frequency will perform better at 'coupling' equipment, so that energy flows quickly and easily from one thing into another and finds its way to the floor (the highest mass around) as fast as possible. The floor, of course, 'couples' the energy directly to the speaker.
A number of obvious hard materials could be (and are) used for equipment feet: steel, aluminium, titanium, stainless steel, brass, ceramics, carbon fibre, even wood. All of these materials sound different because the sonic signature of the material has changed the feedback frequency response. I have tested all of these materials and found that wood (hardwood) distorted the sound the least and its sonic character was the most 'natural'.
From my research, I have found that the cone shape is the most effective for dumping energy out of the equipment, and that the larger the diameter of the cone the better it works.
The Solution for Racks and Cabinets
If wooden cones are the best feet for 'coupling' equipment to shelves and coupling shelves to racks or cabinets, what do you use between the cabinet or rack and the floor? The answer depends on what kind of floor you have and whether or not you have a carpet.
For bare wooden, tiled or concrete floors then wooden cones are ideal. The bigger the better. If you have fitted carpet with underlay then you need feet that will penetrate the carpet to stand solidly on the concrete or wood. Machined steel spikes are the best solution, despite the fact that steel is not an ideal material in sonic terms.
If the floor under the carpet is concrete, you need nothing else, but if you have
a wooden floor you need some way of stopping the spikes from sinking into the wood. Crosshead screws fixed into the floor are an excellent solution - the spikes can then be located into the screw heads (see instructions below).
Fitting spikes to a wooden floor covered with carpet.
1. Place the rack, cabinet or speaker stand exactly where you want it and mark each spike hole in the carpet. (This can be done by placing Sellotape under the spikes and pressing down so that the spikes puncture the tape.)
2. Insert a screw (use 1" (25mm) No. 8 crosshead Philips or Posidrive countersunk screws) into each hole and screw it down. The screws will disappear into the pile of the carpet and become invisible (I suggest doing this while your partner is out!). Each spike will now sit exactly on a screwhead and the result will be a great improvement in bass definition and clarity.
Why I recommend three feet
Stability is a very important aspect of coupling and one that, through misunderstanding, provokes argument. One of the biggest problems with most loudspeakers is the way in which they wobble about on their floor stands or wheels. As the speaker cone moves it rocks the cabinet resulting in a loss of information and energy.
I am often asked why I recommend using three feet. The answer is simple: stability is vital and whilst four feet may look more solid they allow micro-rocking. Even if the rocking action is very small, in this situation (dumping energy) it is very significant. By using three feet (two placed at the front, one at the back) you create a stable triangle and prevent information loss. Unconvinced? The improvement brought by using three points rather than four can easily be shown by doing a simple listening test.
