Eminent Technology ET-2 Tonearm Owners

Frogman et al,
There are many reasons why different tubes will affect the sound.
Depending upon the velocity, the air is likely to flow chaotically, forming vortexes and eddies as it moves down the pipe, it is not necessarily laminar.
The geometry of the tube, the roughness of the tube walls, the elasticity of the tube and the purity of the air all affect flow and chaotic behaviour.
Furthermore, air is compressible, which affects flow. And finally as air flows, compresses or decompresses, it can change temperature which affects pressure and therefore flow.
Keeping the tube as straight as possible will reduce turbulence, just like an optical cable with digital - oh the irony.
It is quite complex.

As an aside I would have thought it would be advantageous to have a regulator at the end of the long tube, just prior to the arm, as the length of tube after the regulator will, if my rusty memory on fluid dynamics is working, drop the flow rate going into the bearing.

In other words if 2 of you had the same regulator and 40ft of tube, and one places the regulator at the start of the tube, and the other places the regulator at the end of the tube, then I would expect that you would end up with different flow rates into the bearing. I would at the very least expect a significant impact on the "sound" from where you position the final regulator in relation to the arm bearing.

Suggested music for AB testing - JS Bach Air on the G String : Zoltan Rozsnyai/Philharmonia Hungarica on M&K Realtime for the audiophiles, perhaps Malcolm Sargent/New Symphony Orchestra for the purists on HMV ( 78rpm of course ).

Very interesting, Dover. Thanks for the explanations and music recs.

Regards.

The motivation for putting a soft tube between the two regulators was the finding that a small amount of bleed off between the regulators had a positive effect. This implied that the air pressure was not constant between the regulators. Or for that matter after the last regulator. More pressure equals more flow into the arm since it is not a compensated load and since a needle valve is not a compensated load either it will pass more or less air as the pressure increases or decreases. So the needle valve tends to smooth these pressure changes hence smooth out the air flow.

The soft walled tube acts slightly like a balloon, it tends to absorb small pressure perturbations by expanding with increased pressure. The surprising finding was that this positive effect was present even when the pump had turned off. This points towards the regulators themselves creating tiny pressure changes which the softer tube helped to reduce. Acting in a way like a surge tank.

If this is correct then it may be better to have the regulator remote from the arm. A long length of soft hose to the arm then helping to smooth out the regulator pressure irregularities. Obviously more testing is required. One test would be to fit a surge tank after the regulator in a rig with a regulator close to the arm. Or simply add another long length of soft tube which would be so easy to do....Chris. A small increase in regulator pressure may be required if this was done to compensate for the line losses.

It is fascinating stuff.

Richardkrebs,
From your explanation I conclude that you are putting an argument forward that the more components that are in the process that are inherently unstable, then based on mathematical probability you are suggesting that this may increase the efficacy of the air flow. You might like to research chaos theory. Another solution may be to use a rigid tube, but put lots of pins through the tube - if you do the calculations and modelling, it would be possible to have a totally chaotic air flow rather than patterns of eddies, which may be preferable.

Last year you said that the ET2 air bearing is rigid. I assume since you now claim that you can hear a difference with different pumps and air supply configurations, that you would now concede that the air bearing is not rigid.

I have studied engineering including fluid dynamics and the way I view the ET2 is that the air bearing is not rigid, and what you are doing with all this experimenting on air pumps and tubing is in fact minimising instability within the bearing. That is why you hear a difference.

I note that you are now running 17psi whereas previously you said that 12psi was optimum ( and that your arm was perfectly rigid ). Can you explain why you have changed your view. Have you removed the lead and put your arm back to standard with the decoupled counterweight put back in now ?

Dover
Below is a copy of a mail I received from Bruce T many months ago. You might find it enlightening reading.

"Richard,

The resonance of the air cavity is over 500Khz and does not manifest itself on the surface of the bearing, it is a well damped liquid bearing.
A statement - " the air gap allows movement at audio frequencies" - shows a lack of understanding regarding how a tonearm works.
In two out of three degrees of freedom (x,y, & z axis) a cartridge is completely free to move in any tonearm. How can it not move in the X & Y axis but magically pull and push in the constrained Z axis? The record is encoded in and the forcing functions greatest in the X (vertical) and Y (horizontal) axis.
A tonearm works because of mass present in all three axis with the forcing functions above the systems natural frequency. The tonearm components headshell, arm wand, are thousands of times heavier than the cantilever and stylus, so by a ratio of masses, they sit still.
To put this in perspective go run back and forth and jump up and down on a several hundred thousand pound untethered barge and watch the displacement of the barge, while it will move, its motion will be extremely small relative to yours and proportional to the ratio of masses.
I used three measurement methods when developing the tonearm, accelerometers, strain gages, and the simplest and most effective was the use of a second tonearm to play parts of the tonearm under test while playing a record. Measurements at the air bearing are more than 60dB (1 million times) below signal levels, lower than the pivoted tonearms I used for comparison. In reality the ET-2 has its highest inertia in the Z axis and appears rigid to the cartridge.
If a tonearm moved at audio frequencies it would reveal itself as dips in frequency response.
brucet"

The arm IS rigid at audio frequencies, however it is sensitive to supply pressure irregularities. These are two completely different phenomena.

re pressures. In my rig, I found 12 PSI to be optimum without an oil trough, 17 psi with one. There is no inconsistency.