17-Apr-2016, 18:38
A bit of background about me.
I always wanted to work in motor racing and wrote a computer programme to optimise suspension geometry in 1971, which got my foot in the door, but the total number of graduate engineers in the whole of Formula 1 back then, on the car as opposed to engine side, was probably 25 so full time jobs didn't crop up often.
At the end of my student apprenticeship in 1972 I ended up doing noise and vibration research and continued until a full time job at Hesketh Racing was offered in 1976.
So my experience was a long time ago but it was, at the time, pretty cutting edge, pioneering computer modelling at a time when programmes were on punched cards.
I had done my apprenticeship for David Brown central staff and got my first full time appointment at David Brown Gear Industries in R&D.
My work consisted of predicting and measuring vibration and trouble shooting vibration or noise problems.
I moved to Garrard in 1975 and spent my time on record players, particularly trying to make them accurate by clever engineering rather than throwing money at them.
So my experience is a long time ago but the physics hasn't changed since, though the sensors and analytical tools are massively cheaper today.
One of the first things I had to do at Garrard was acclimatise myself with the measuring kit and the strengths and weaknesses of record players. My boss was one for letting people learn by experience, so my first job was to measure the rumble of a record player, a Garrard 401 with contemporary SME arm in a solid plinth. We measured the output of a fitted cartridge via a RIAA stage since it was sensibly reasoned that that was what mattered. If one of the plinth resonances doesn't cause the cartridge output to be changed, for example, it is reasonable to assume it isn't important.
Anyway my first efforts to get rumble measurements gave inconsistent results. The record player was on a lab oak bench on the fourth floor of a building and the employee carpark was between the building and the road but when I looked for the inconsistency I found I was measuring the traffic flow on the road outside. This was a surprise...
In fact, in order to measure rumble, record players were always put on a concrete block suspended by angled springs calculated to isolate in all 6 degrees of freedom from about 10Hz. That way we could get consistent results.
We had suspension of most decks, but relied on installers for the pro 401. None of the reasonably priced suspensions isolated in all 6 degrees of freedom though but worked quite well vertically and horizontally.
Anyway, years later we moved to a bigger house. By then my record player was a Technics SP10 with a SME arm in a self built solid plinth I decided that I had space to move my record player and electronics into my study with just the speakers in the lounge. This made the sound much less atmospheric and more closed in with smaller stereo image!
I realised that a lot of the characteristic sound of a record player was because it picks up mechanical and airborne vibrations which act like a bit of extra reverb.
When it comes to electronics I know that both Naim and Goldmund include expensive vibration control into their most expensive kit, presumably for good reason.
I found my Goldmund Mimesis 36+ CD transport sounded better when "mechanically earthed" as recommended by the manufacturer rather than on an equipment rack.
Other than that I haven't tried electronics for myself.
Certainly many valves are microphonic, so valve electronics is likely to add a little something if it is in the room with the speakers.
I have heard that crystal clocks can be vibration sensitive too.
Anyway, the only person I know of in the hifi business who talks sense, from an understanding of the physics pov, is Max Townshend.
I haven't time to calculate an accurate figure but as a rule of thumb if you wish an isolation system to isolate from 20Hz up it will be flexible enough to deflect about 20mm when the weight of the component is put on it. If it deflects less it will isolate at some frequencies, transmit at others, and have a resonance in the audible range.
If the "isolator" is like a spike, or something, it is actually a coupling device rather than an isolator.
I always wanted to work in motor racing and wrote a computer programme to optimise suspension geometry in 1971, which got my foot in the door, but the total number of graduate engineers in the whole of Formula 1 back then, on the car as opposed to engine side, was probably 25 so full time jobs didn't crop up often.
At the end of my student apprenticeship in 1972 I ended up doing noise and vibration research and continued until a full time job at Hesketh Racing was offered in 1976.
So my experience was a long time ago but it was, at the time, pretty cutting edge, pioneering computer modelling at a time when programmes were on punched cards.
I had done my apprenticeship for David Brown central staff and got my first full time appointment at David Brown Gear Industries in R&D.
My work consisted of predicting and measuring vibration and trouble shooting vibration or noise problems.
I moved to Garrard in 1975 and spent my time on record players, particularly trying to make them accurate by clever engineering rather than throwing money at them.
So my experience is a long time ago but the physics hasn't changed since, though the sensors and analytical tools are massively cheaper today.
One of the first things I had to do at Garrard was acclimatise myself with the measuring kit and the strengths and weaknesses of record players. My boss was one for letting people learn by experience, so my first job was to measure the rumble of a record player, a Garrard 401 with contemporary SME arm in a solid plinth. We measured the output of a fitted cartridge via a RIAA stage since it was sensibly reasoned that that was what mattered. If one of the plinth resonances doesn't cause the cartridge output to be changed, for example, it is reasonable to assume it isn't important.
Anyway my first efforts to get rumble measurements gave inconsistent results. The record player was on a lab oak bench on the fourth floor of a building and the employee carpark was between the building and the road but when I looked for the inconsistency I found I was measuring the traffic flow on the road outside. This was a surprise...
In fact, in order to measure rumble, record players were always put on a concrete block suspended by angled springs calculated to isolate in all 6 degrees of freedom from about 10Hz. That way we could get consistent results.
We had suspension of most decks, but relied on installers for the pro 401. None of the reasonably priced suspensions isolated in all 6 degrees of freedom though but worked quite well vertically and horizontally.
Anyway, years later we moved to a bigger house. By then my record player was a Technics SP10 with a SME arm in a self built solid plinth I decided that I had space to move my record player and electronics into my study with just the speakers in the lounge. This made the sound much less atmospheric and more closed in with smaller stereo image!
I realised that a lot of the characteristic sound of a record player was because it picks up mechanical and airborne vibrations which act like a bit of extra reverb.
When it comes to electronics I know that both Naim and Goldmund include expensive vibration control into their most expensive kit, presumably for good reason.
I found my Goldmund Mimesis 36+ CD transport sounded better when "mechanically earthed" as recommended by the manufacturer rather than on an equipment rack.
Other than that I haven't tried electronics for myself.
Certainly many valves are microphonic, so valve electronics is likely to add a little something if it is in the room with the speakers.
I have heard that crystal clocks can be vibration sensitive too.
Anyway, the only person I know of in the hifi business who talks sense, from an understanding of the physics pov, is Max Townshend.
I haven't time to calculate an accurate figure but as a rule of thumb if you wish an isolation system to isolate from 20Hz up it will be flexible enough to deflect about 20mm when the weight of the component is put on it. If it deflects less it will isolate at some frequencies, transmit at others, and have a resonance in the audible range.
If the "isolator" is like a spike, or something, it is actually a coupling device rather than an isolator.
Devialet Original d'Atelier 44 Core, Job Pre/225, Goldmund PH2, Goldmund Reference/T3f /Ortofon A90, Goldmund Mimesis 36+ & Chord Blu, iMac/Air, Lynx Theta, Tune Audio Anima, Goldmund Epilog 1&2, REL Studio. Dialog, Silver Phantoms, Branch stands, copper cables (mainly).
Oxfordshire
