Eminent Technology ET-2 Tonearm Owners

Hi Ct & Frogman!
"My place" is a historical house we run as a hostel, with a cultural program (which I organize with a few friends). The remaining time (not much) & energy (still enough) is spent on hifi and developing concepts & ideas. This has a company name and is actually more like a small personal counseling service to customers (who often are or became friends).
And I play in an amateur wind-instrument band.

And yes, Frogman, I read about and appreciated your similar experience with the screw torque. (And the sax (?) playing).

You're right with the several modes to mis-tune an ET2. It needs a bit of 3D operational understanding and then every screw (inside the brain too) falls into its place.
I leave the basic side-ways adjustment (the two posts) with very low torque, and keep the locking tilt screw fixed. I do lateral (and vertical) fine-tuning with the tripod screws in a balanced way: up one, down the other. The torque on the central pillar fixing screw stays the same this way.
I think it is really moving the VTA lever itself which dis-balances the arm a bit and the mechanism is not quite keeping the arm quite level. I pull the VTA lever with one finger using the bearing as a base, so there are (almost) no forces going to the plinth and the connection to it.
I "feel" the same about less interfaces is more (underneath the Tiptoes). But I got good results with small single steel pieces underneath the tiptoes. I think a whole steel or alu plate is less desirable. I will look to improve on this in my actual ad-hoc installation, which has a few things to improve.
BTW another tweak I use is trapezoid wedges of cardboard which I slide in (vertical) underneath both ends of the bearing housing (or damping trough) until they slightly lock the bearing to the plinth. This absorbs any horizontal elastic movement of the bearing on top of the pillar. They have to be applied with plyers, and should put symmetrical force to not unbalance the arm. I do this after having found my preferred settings (... :-)

all the leafs are brown... the fall is coming.. :-)
Yes Blue-Tak is something to have around, and I use it (I'm swiss :-) on my jewelers screw driver - when I remember...

No, I wasn't in LP mastering / cutting, but I have a profound interest on the electronic & physics side of that hobby. I make recordings of concerts at our place, and did some recordings already in the 70's with a semi-pro JVC portable cassette-deck from my brother.
And I did a few Magnepan modifications for customers, and made new low-diffraction panels, and calculated all new crossovers for MG3 and MG 3.3s.

I had to trade in the ET1 for the ET2, so I don't own it anymore, contrary to the ET 2 it was slightly crude in set-up and detail solutions - but it worked very well.

My main messages about the arm (which I always loved) are as follows:
- If the bass is off, it's probably the set-up (side level or maybe a too softly springed counterweight stud).
- If I only touch the VTA adjustment a bit on *my* ET2 - the side-leveling is severely off. If it's not, I was lucky. To check is better than to believe...
- The "critical theory" against air bearing arms, and specially the ET2, stands IMO on very shaky feet, as I tried to show above. If there is a structural problem it might be the slight elasticity in some of the joints.
- subchassis turntables have two main problems with an air bearing arm:
- the lever mechanism of the shifting mass offsets the subchassis (-> increases sideforces)
- The needle drag problem, ie. the varying friction with tracking, potentially sets up rotational forces in the subchassis, with every arm. This leads to a slight instability in the subchassis and pitch uncertainties. But it also kicks air bearing arms in a different, very undesirable and quite audible way: Side-ways. (Much more audible than a radial tonearm, where the kick goes much more along the stable axis needle to arm bearing).
So one gets both problems - the bass instability and the pitch instability.
But it's still feasible and it can sound very good.
- In my book of experiences, every screw that creates a touch more grip than needed to keep a joint or a connection from gliding, introduces quite severe grades of transistor-feedback sound. I understand this as a necessary minimal torque, but every further bit of torque is like tuning the screw up, in the sense of a steel string. Tune these down & out, and they ceae to resonate. It is less storage of resonant energy. So my ET 2.5 runs with minimal torque on all screw connections. It sounds less artificial and more integrated to my ears.

My ET 2.5 sits now on a partially spring suspended Technics SL1210 with some modification on the electronics. With a very old, original AT OC7, which sounds very, very good here. I know, it shouldn't :-)

I'm a (very) long time user of the ET2 arm, I owned the ET one before. I still find the arm superb!
I have a first comment, others might follow.
The question of resonance frequency is mostly re-cycled from some sources, that did a lot of calculations a lot of time back. That's OK, but there is one aspect missing IMO.
The main signal is cut laterally, one can safely assume that if there are very low frequencies on an LP it is cut *laterally*. The main disturbance laterallyis off-center records (0.55 Hz).
The off-phase info is cut vertically. The off-phae low frequency part is not doing anything helpful in normal rooms, so usually the LF signal below ca. 100Hz is blended to mono.
This leads to severely different optimal resonance frequencies horizontally and vertically. More will follow.
Let it be said that my most memorable analog experience in the 70's was a FR64 & FR7 combo - it had *decidely* "suboptimal" resoance frequency, ie. far below 10Hz. It sounded better than anything else I heard at that time, and I still the *bass" and ambience of that front end. There's a physical reason for it, I think.

I remember this combo clearly because it was visible that it had rel. low damping and an unusually low resonance frequency. I knew and was used to tonearm/cartridge combos with "correct" resonance frequency. This one was incorrect - and sounded better than anything I had heard (in the low-mid frequencies, in dynamics and resolution).
If "correct" resonance frequency would be the main reason for optimal or non-optimal sound, such suboptimal resonance frequency should have almost destroyed any potential of good sound. (that's what I thought then).

The normal warp & rumble zone is usually quoted between 2 to 5Hz - it leads basically to vertical accelerations. If one assumes that there are no important signals to be traced below 20 Hz or even below 50Hz (because the big excursions involved in low freq. creates problems for the cutting head *and* the cartridge. And because even in big listening rooms it is difficult to reproduce phase differences in LF) one places the optimal *vertical* resonance frequency between 5 and 20, or more radically even below 5 and 50Hz.
-> Vertical optimum is around SQR(5x20)=10 Hz or even SQR(5x50)=21 Hz
(so long so good, so often heard).

Horizontally there is a main disturbance at 0.55Hz (33.333rpm/60"), maybe 0.75 Hz (45rpm/60"). But there is a good reason to extend extraction of musical information like ELF reverb etc. down to at least 5Hz - if possible. Cutting and manufacturing LPs has several modes of creating vertical disturbances but only one or two horizontal issues: Off-center records (which actually could and should be corrected) and horizontal bumps on the cutting lathe - rare (but I have one such record in thousands of LPs).
So there is a moderate, but constant excitation/acceleration on 0.55Hz (basically correctable) and a useful extended LF range down to 5Hz which leads
to an
-> optimal horizontal res. frequency of SQR(0.75 * 5) = 2 Hz

I agree that low horizontal res. frequency is not without problems (with highly off-center records). But preventing this problem with much higher (ie. standard) horizontal res. frequency means loosing very worthwile ELF information (which, as said above, is cut mainly in the horizontal plane.)

So the *often" alluded main "achilles heel" of air bearing arms is actually an advantage, in an "objective", physical sense. About the same time, I drew these thoughts up on paper, I found the new Mörch arm which does *exactly* follow these reasonings, with superb result. It couples the cartridge to a *very* high horizontal mass. Experiments show that the rigid coupling of that horizontal mass vs. the cartridge is sonically important.

The ET 2(.5) and other air bearing arms have the same high hor. mass advantage "built in for free".
The decoupled counterweight of the ET 2 gives simply more freedom with problematic pairings of cartridge, arm and off-center records.
Maybe, when it is not necessary, blocking the decoupling could have advantages sound-wise.

BTW in my experience it is extremely important to have both the platter and arm *extremely* well levelled in the horizontal plane. Side-forces on the cartridge cantilever lead to plummy bass. Probably because a variable pull with varying tracking friction on a laterally deplaced cantilever gives dynamically varying side forces - which might energize the horizontal resonance. It sure is very audible. In my setup it is important to check regularly, floating the arm with two equal blobs of Blue-Tak on both sides of the arm, one at the backside of the arm, the other on the counterweight.

@flemke: I agree! These are gradually more consequent and less reversible ways of what my cardboard wedges underneath the manifold/bearing do.
The wood option might create a more stiff but more vibration conducting, but less absorbing pathway than cardboard.
I might try a hardwood wedge someday. I like tweaks to be as reversible, low mass and "elegant" (not necessary in the visual aspect ;-) as possible.

A question to all:
Does anyone have tried grounding the air supply's static charge?
I think Richard mentioned that he connected the brass of the pressure gauge to ground (which ground, where?).
Has anybody else tried it?
There are - expensive - antistatic "hoses" available. Would be interesting to try, maybe even only on the short distance between the last air tank (preferably antistatic & grounded too) and the arm.
Other grounding ideas? Like putting a copper grid into the air tank and potting the grounding wire in hot glue?
BTW my simple air tank is a 20 gallon fuel "canister" (unused :-) with a long and a short aquarium tube entering the cover of the tank reaching the top and the bottom), the whole sealed by hot glue and padded with acrylic wool. Very simple , cheap and working well.

Interesting “facts discussion” about linear vs. radial tonearms, most have probably read it:
http://forum.audiogon.com/cgi-bin/fr.pl?eanlg&1266367593&openfrom&1&4&&st100
Sadly mostly boring (ie. repeated, but skewed arguments)… but some interesting points. A main point is the quotation of a High Performance Audio Review test in the 80's measuring much higher actual sideforces and deflection of the cantilever on a SAEC radial tonearm vs. an air bearing arm.
The reaction of the (radial) “experts”: Hilarious, ridiculous! The facts don't follow their invented orthodox belief, so the facts must be wrong! The reaction of other more pragmatic non scientifical audiophiles: This is only one situation, we must collect other… (measurements).
Experimental science means, one well set up experiment with a non expected measured outcome can expose any pre-existing contrary hypothesis as wrong. In this case the “excessive side-forces”. But what am I as a humble pragmatic thinker compared to the pope(s)?
Over most of the thread the effect of mass and bearing friction on a cantilever are confounded, and bearing stick vs. bearing friction isn't even mentioned. Neither are there attempts to define the magnitude of differences of (horizontal) friction of radial vs. air bearing arms. I think/suspect that even with the “lever advantage” the best radial arm may just quite cope with an air bearing. Horizontally that is. The rest is wrong thinking about the optimal horizontal mass aka. the desired horizontal resonance frequency – for best bass reproduction with lowest phase shift. The dynamic side forces on a cantilever are simply a result of friction (mostly non existant in an air bearing) and mass. Ie. they correlate with the horizontal resonance frequency – and are taking part in a compromise, where the sonic optimum lays much lower than audiophile orthodoxy.

Frogman: If I'd live in NY or the states I'd shurely ask you for lessons on my clarinet(s). It would be fun tweaking my fingers tangentially to the holes with your help and having them (the holes) optimized by tweaking the diameters... :-).

Dover: I found your posts very interesting and I like what I'm hearing with magnetic damping - in my setup, with a small neodym magnet close to the bearing tube placed at the "backside" of the bearing, direct wiring some cms away from the bearing tube. Not tried yet the "naked" alu arm tube yet.

Setup: IMO my original ET wiring was introducing too high lateral forces. I use twisted thin Audio Consulting silver wires outside of the arm, fixed via two or three small double sided foam ahesive pads. The wires go in a light curve, with a vertical radius, L/R some cms separated, to 7cm behind the arm tube, in the middle of the travel path.
With this, the arm is a *very* precise leveling tool. It shows the slight change of levelness introduced by the level forces of the tonearm on my battery-modded SL1210 with "trick-subchassis". So I have to look for a very slow movement toward the outside at the beginning and the end, but perfect level in the mid of the LP, or mid arm travel.
Every time I heard slightly tubby bass, it was caused by an involontary change in lateral balance. Unbalanced side forces affects bass - one already has it built in with too stiff wiring.
I'll try Dovers idea of a slight overhang, this is such a delicate tweak, that it will create much smaller forces than any wrong balance setup.

Decoupled counterweight & ET: I think it is a brilliant idea to reduce the effective lateral mass - in case of rather high compliance cartridges. It makes the ET2 much more widely compatible, than the ET one was (I owned it), or most other air bearing arms, specially those with stationary outer bearing, long bearing tubes and counterweight at the opposite end.
My axe to grind is that optimizing towards a higher lateral resonance frequency than say 4-5 Hz (at least) is simply neither necessary nor improving sound. Based on the same thoughts that stand behind the Moerch DP-8 arm.
Ie. with moderate to low compliance cartridges the counterweight decoupling introduces a new (sonic) compromise instead of solving a problem. But to really hear the advantage of a rigid counterweight, one has to do a bit more fiddling with the ET2 counterweight fixing than just using BluTak to bypass the spring (which I usually did).
For high compliance - I assume - it's ingenious.
BTW I use a pressurized air bottle (rental system in Switzerland) as air supply at the moment.

Addendum: The VTA adjustment mechanism of my ET2 introduces slight (but IMO much too much) lateral imbalances when changing VTA. I assume Ct0517 will tell me that my VTA mechanism has a problem - but it was never overtightened or any thing like that, it has equal spacing around and moderate to low "clamp".
- so, after changing VTA I would *always* adjust lateral balance... :-(
- I use very small vertical cardboard wedges between arm housing and plinth to *slightly* couple the bearing to the plinth. One "outside", one on the counterweight end, underneath the damping trough. This improves stability and focus and is audible (and reversible).

- BTW I measured/listened to the energy put into the arm, arm base and plinth while playing an LP, by using a structural feedback microphone. I heard a wide-band replica of the LP signals across the whole arm structure into and including the plinth. With no audible HF roll-off "after" the bearing, it just went lower in level across the plinth, a bit absorbed probably by inertia. So much for "air bearings are inherently less stiff than radial arm bearings".
- And I heard bearing noise... like white noise, created by air turbulence of the exiting air. This has tobe put in the backside of our heads when increasing pressure on the bearing, although it's relatively constant.

Hi Chris (?) the compressed air bottle is a 200 atmospheres 60 liter steel tank. It provides "air DC" (to my arm & head) :-)

Wires: the Audio Consulting wires are twisted single strand 0.15mm silver wires with thin isolation. I dress them carefully (two L/R separated looms in a circle of almost 245 degree with ca. > 15cm diameter) going up from the arm in 90 degree to ca. 4cm above & behind the arm, slowly turning > 200 degree and back to vertically down. I measured the forces applied by the elasticity and weight of a single winding, it's around 0.05 gm. This is audible - if one does not correct for it. But I tune the downforce by ear anyway. The forces are low, and less in the horizontal plane. The end effect is what the arm does: It still skates off very slowly outside the middle 5cm to the inner and outer grooves.
The wiring works actually very well (far better than the original wiring), and I prefer to keep the wires twisted for hum compensation. Some time I used 0.07mm single strand copper wires, the applied slightly less force but sounded IMO worse.

Did you really check for absolute levelness after changing the VTA position, finding *no* change in side balance?
The imprecision is shurely on a very low absolute level, but not low enough in my case. I found out the hard way, by ear that there was a problem.

Hi Chris, BTW the pressurized air bottle is true. :-)
I forgot to answer an earlier question of yours: I prefer to use less counterweights farther out on the I-beam. I prefer it because this way the center of gravity of the arm is more centered along the bearing, and because I have less lever forces on my semi-springed subchassis. It helps to float the arm better across it's way.
It also reduces the maximum lateral mass, effective at "DC", below the resonance frequencies of arm/bearing and I-beam/counterweight.
Less counterweights farther out in effect tunes the I-beam / leaf spring resonance lower down. In the end, this is a more important aspect than the leaf or counterweight looked at separately I think.
I went through many such tests during the 80's and beginning of nineties, taking some of them up in the last years.
I didn't like the single leaf springs with my medium to low compliance cartridges, I preferred the double spring ones - I don't think I have a triple leaf I-beam.
There is another aspect of the I-beam, which is critical / sub-optimal: The weights are not centered around the I-beam, ie. it exerts an offset torsional force on the leaf. This means that any vertical movement of the arm activates a hidden torsional resonance, which slightly modulates tracking force. The vertical axis is the one axis, where you want *absolute* "true inertia", and as few resonant modes as possible, including arm resonances, as it affects the critical downforce.
My feeling is that the stiffer beams control this aspect better than the single leaf beam.
I always preferred the I-beam leaf with a bit more damping than originally provided, using BluTak.

For balancing I move two Blue-Tak "saddles" along the I-beam, and the arm lever extension, to keep lateral forces equilibrated.

My ET2, now with ET 2.5 bearing is an early one. I ordered the bearing for around 0.8 bar (ca. 11 psi) and it still runs OK with ca. 3 psi. Though below ca. 6 psi things go audibly downward, the magic disappearing somehow. In fact the ET 2.5 has "simply" more surface and is therefore already stiffer, but the price is a higher lateral mass, probably 10g more.
Sending the arm from Switzerland to the US - and parting from it - would probably affect my sleep... not in the best way. :-) It works good enough... I'm Just checking lateral balance after adjusting the VTA.

Slaw, could you tell me which of my post you find offending and maybe why? I have no problem with disagreement. But I know sometimes my tiredness of repeated misinformation pushes me for words that fit my feelings. :-/

Regarding your photograph, Frogman...
A real audiophile is one who looks first at the turntable and tonearm in the background. I - almost - did.
:-)

Hi ye all! Did anybody ever try a Decca in the ET2.5 (or ET2 arm)?
I once did with an ET one and was not blown away - it didn't quite fit, but this was with a rel. light subchassis turntable, which acerbates several problematic aspects: High horizontal, quite low vertical compliance (one third!). Subchassis resonance above the horizontal resonance frequency of the arm/cartridge system.
I assume, but don't know, that using the decoupled counterweight could help here really. The question being:
Which stiffness, which weight/distance.
My feeling with decoupling and the rel. high compliance in the horizontal plane goes into the direction of
far out weight, with medium to low stiffness spring.
But I'd like to hear of real experiences, not virtual ones, or concepts. Of the latter, I have enough myself :-)

Thanks a lot Frogman! The wiring is an issue for sure, but the EM-field situation is actually the cause, and the noise is the symptom. And the ET2 / wiring / Decca system seems to be very sensitive. What turntable do you use, and how close are other power supplies?

Thanks Ct0517 & BdP! I am wondering that the conceptually very similar London air bearing arm (based on the Cartridge Mans's Conductor arm), with similar wiring, and at least these four "naked" unscreened (or five!) wires without screen for about 15cms, is the recommended arm for the London cartridges. It want hum, I'm shure (not the cartridge :-).
I suspect it's something to do with a missing ground link from the arm to the preamp. Does the original ET2 wiring has a 5th wire? Mine certainly does, but I have a different wiring & routing to the original one.

Yes, this makes sense, and in an additional view-angle it's also like looking at ground loops like HF currents, overlaying each other with more or less phase differences, because "tuning" the screws (down) also tunes (down) the speed of each transmission path individually, and tunes (down) the Q of the resonances individually.
This BTW is also an aspect in different ways of mechanical equipment grounding. Depending on how you ground vibrations, more or less energy travels at more or less speed across the room, ie. across the floor, eg. to ones "ground receptors", the feet.
I mainly think in terms of rigidity as: "how much is *really* necessary". This & not more. One really has to check by ear, with a "natural" and integral way of listening. The "low rigidity way" keeps the musical range much more together, sounds more fluent. But it may sound less "impressive", less bassy, less "brillant" etc. Instead one hears more of the structure of sound, and listens into the quality of instruments *and playing* quite a lot easier.

Yes frogman & CT0517 on a John Deere, what nice illustration of Americana seen the other way :-) It shurely tunes bass sensitivity, although only in a limited interval range...
Regarding "spit", isn't what most lay people (ie. the non-blowing family members too) think it is. It's mainly distilled vapour from the lungs, with maybe a bit of sulfur molecules here and nitrogen mol. there. But not much spit, I think. At least not anymore after one played two or three years.
I call my baritone sax my personal distillery. But the bowed neck (?) of a bass clarinet works already nicely as such.
Regarding trombone: This is a really great instrument! (And - ha! Ray Anderson will play november 11 in our concert program! :-) The instrument is extremely powerful in a seductive way, a lot of colours and expressive. Must be fun to play with this huge tool-box.
It's funny you mention this aspect, of first getting the fundament right, the rest then falling in place: If I'm asked about what defines the quality of a well set up ET2 (or even more a ET2.5) it is the "full-range" sound from the *lows* up, surrounded by a lot of air in the bass (first): Bass ambience. Yes, it's also sublime upper ranges, but it starts with a lucid, airy bass range with a high resolution of bass timbre & pitch.
It sounds "correct" by itself while not sounding tight-assed NFB correct.

Hi Dover, thanks for your clarifying comments!
Regarding the pressure compensation, it might help to use the analogy of a stiff power supply (air supply & reserve connected to manifold) with a few local supplies (local pressure zones around air capillary openings to the bearing) coupled with high series resistance (capillaries). As current (air flow) drops on one of the sub-supplies, the voltage (air pressure) rises - kind of a passive feedback!
Regarding the usually alluded inherent "stiffness" of mechanical bearings including unipivots, it's worth to consider the following thought:
No material is stiff, everything is more or less elastic. (With some unique properties subsummized in the poetic word "character", importantly damping, and including speed of transmission). Reduce pressure area, and elasticity increases. This affects resonance frequency inherent in any elasticity / mass combo. *Point* coupling as in a unipivot or a spike point, looked at on an "atomic level", is in no way making the coupling stiff, it's the opposite. The surfaces meet in kind of a balanced force & elasticity state, a bit like a jelly ball swimming in water, to put it to the extreme. You don't get steel more elastic than with a perfectly pointed unipivot interface. Then think "it" as an elongated point and you see something like a short subminiature "string" at the end of the point - quite elastic, like a very small piece of microscopic harpsichord string. "Flatter" points like balls have much less of this, and make stiffer bearings - that depend more on extremely complex polishing processes. Some arms use the tip of a roller pen, quite clever!
The whole "argument" (rather a mythical marketing image?) of the "mechanical diode" is moot. Point coupling shurely does "something" (as everything we do does) but shurely it is not "stiff coupling" or magic diode processes. It might eliminate eg. multi-point rattling by a multitude of low pressure indefinite points, eliminating noises of "buzzing paper on a comb"-effects, tingling in metal-to-metal sonority.
And... air bearings are at the total other end of the scale!
That's what Bruce tells us since a long time.

Thanks Ct0517 & BdP! I am wondering that the conceptually very similar London air bearing arm was "built for London" cartridges... (based on the Cartridge Mans's Conductor arm) - it must have functionally similar unscreened wiring, at least these four "naked" unscreened (or five!) wires without screen for about 15cms.
Could it be that it has something to do with a missing ground link from the arm to the preamp - does the original ET2 wiring have a 5th wire? Mine certainly does, but I use a different wiring & routing to the original one.

Richard
Yes i have two twisted silver wires running left and right of the tonearm. They are mounted to the tube via three very small pieces of double sided adhesive foam strips, to hold the wires and prevent them from touching the metal arm tube. They "exit" the arm connection bending a bit up and then down in a wide radius to the plinth, where they are supported again with an adhesive foam pad. And there is a separate ground wire to one of the arm screws.
With careful wiring dressing the arm is completely free in movement, and very sensitive to the smallest lateral imbalances, as it should be.

Hi all - interesting bits & pieces!
Ct0517: interesting to hear more from internal design parameters of Bruces bearings. Some time I'll make photographs of my wiring etc.
Dover: The wire experiences are very interesting, as they are real findings. I'm not convinced however that the reason for the RF issue is solid core or silver, it could be the insulation environment of the wires affecting HF reception too, or the smaller wire diameters. As always the observations are true, the causes sometimes can be more complex than expected.
Were the silk litz wires individually isolated (HF litz) or just the whole bunch? (Besides HF detection / diode-property could also be a cold soldering joint, but I don't suppose that).
My turntable still has the original alu plinth (not too long anymore), but the motor is fed by batteries and the tt as such has no connection with the mains when running. grounding a "raw" alu plinth surface is a PITA...
The Deccas by definition have each channel ground connected ("matrix" summed through the vertical coil). So the ground loop is there for the small EM-field buggers to sneek behind and find. Luckily they oversee the possibilty under certain circumstances - but which? Hum is a dark asian martial art, teached under monumentally wound trees and learned by the nasty little buggers. There is some hope though in the light swords of master Yoda...
Your check list will be very helpful for many in trouble! Like a friend who messed up the ground connection on his new, outstanding phono stage, complaining about everything except his wiring arts.
- Generally, I find that one considerable advantage of low impedance MCs is that they don't need screening through bypassing / dumping external EM-fields through their low impedance.
It's certainly a drawback that exactly the type of cartridge that is more sensitive to capacitive loading - and therefore the sound of the dielectric too - needs screened interconnects. IMO & IME screening doesn't help the sound quality except for eliminating potential hum.

- ct0517: It's interesting what you quote about the wire loom. While radial tonearms have the advantage of leverage on the point where the wire exits the tonearm tube and enters the base (tube), the distance is short, and the wires are somehow uncontrollable in their movement, so they are inclined to rub / stick / jump - and this "out of sight". The latter problem is nonexistent in tangential arms, if the wires are properly arranged. Thales' Micha Huber prefers *very* thin wires (ca. 0.07mm), probably to have the arm movement as free as possible. If this is necessary "even" with radial arms, not even a trace of a problem will remain regarding freedom of movement will remain with such wires for tangential arms either. IMO even two twisted 0.15mm wires, carefully arranged, do not create a problem.
- Dover: The silk wires are interesting, as these have much more electrostatic "field" surface covered with natural insulation than the compared solid core wire, this might (on sinister or twisted paths) affect RF sensitivity...
Somehow a "better" insulation with thinner litz wires should be "capable" of better radio reception :-/ ;-) In this case: The opposite!
Isn't reality from time to time baffling us?
Actually it's more of a surprise if we actually and really *know* something for sure.
- Decca: I had a first try yesterday with a artisanally rebuilt Decca, in the Garrot tradition. Yes, it's still a bit of a hassle concerning hum and screening. My 1.25m unscreened solid core wires are too long, but the hum with one phono stage we compared yesterday was quite acceptable even in my suboptimal situation (the Lehmann Audio Black Cube).
The first results are extremely impressive - in a truly musical way. This cartridges liquidity, high resolution, ambience retrieval (not 1:1 = big spaces) and grainfree transparency is *hugely* impressive!
Yes, there is a slight upper mid forwardness (but still in the range of "realistic") and yes, there is a certain upper limit in tracing (now, in my present setup), which is slightly modulated by the very low horizontal LF resonance (not yet damped enough). I will experiment with different spring / counterweight settings.
Still my rebuilt Koetsu Black is - as yet - more general purpose, and extremely natural too, and as yet it traces better.
- Phono stages, in short: The Black Cube is good enough by far to show off the superb qualities of the Decca. It seems to sound better on MM than MC and it's good. The Lehmann was a trace gray on MC, but well organised, generally transparent, dynamically stable and transparent. I had some issues with the Koetsu on high level historical horns (on fff).
. The same issue on blasting horns did arise with an austrian phono stage from pure dynamics. This one was very good, transparent and open, with very good stage in all dimensions. The bass was very good too, and it was fuss free in setup. (It costs around or below 1000$)
. The real stunner was an older LFD MM0, same circuit as the LFD Mistral and later basic LFD phono stages (but with a much beefier power supply). While I had a big issue with some intermittent RF signals ( ;-), probably radar, it was truly superb in timbre, agility, space, musicality and dynamics with a very natural, open bass. My feeling is that the LFDs phono stages might be considerably underrated in the press.
. My Audio Synthesis Passion phono had probably an issue with the supply connections, resulting in a not typical slightly opaque and undedamped quality in the bass, and a slight haze. The Passion Phono was still very transparent and true to instrument characteristics with wide and high stage, depth was a bit shortened probably by the mentioned issue. Generally and over long years this is a superb Phono stage.

The real fun in research is the reality. It has a certain from time to time recurring character trait of disrespecting the holy truth of any hypothesis. Without that, life would be pontifical s***. :-)

Regarding RF: It was only a problem with one phono stage, that might have some issue in the power supply (while still sounding very impressive). The power supply is one of the places where rectification "happens" (yes, it does), the others are nonlinearities in active devices or shaky contacts. No problem even with ca. 1.5m of twisted non screened solid core wires. That's my experience in the context of my system.
Very important: I compared the phonostages with a superb insulation transformer that does a lot (by sounding like less devices in the signal path).
I placed a Stadler air cleaner / ionizer in my listening room, that ran for very few hours (1-2) but was unplugged while listening. After experimenting ca. three weeks with "high quality" water in a humidifier which improved the "atmosphere" or energy but made the sound somewhat dull. The results with the ionizer are amazing: Remarkably more full and transparent - and no charged LPs... !!
We suspected already during an earlier listening sessions, that the slowly increasing amount of dry air exiting the tonearm creates - a lot! - of static charge on the vinyl surfaces. And it had a considerable effect sonically. The ionizer, building up a certain reserve of air ions (smells also a bit like Ozone), seems to eliminate this problem. Hmmm...

The Decca continues to amaze - fantastic! And this already with the Lehmann Black Cube... :-)

Regarding EI (there are several other "square" cores BTW) vs. toroidal transformers: Things are not that simple. While a toroidal has the advantage of a more simple "flowing" magnetic circuit, it's main "advantage" is the absence of an air gap - an economic advantage. This makes for considerably less iron for a toroidal, and more inductivity with less windings. This makes for a low (ohmic !) loss transformer, with low impedance windings.
However this creates disadvantages: The saturation of a toroidal is very sharp, and with the industry practice (and the practice of almost all audiophile transformer builders) of driving the transformers too close to the magnetic saturation point of the cores ("Efficient" ! Bling!) they are *the* achilles heel of excessive sonic mains sensitivity.
And within these issues DC on the mains (a strong disturbance of high variability) drives a toroidal transformers very easily into saturation, a massive disadvantage!
The last related disadvantage of a toroidal is the inherently lower ohmic loss: It produces (much) higher charging current spikes, which on their own magnetize the (hard limited) toroidal core much harder, and again pull away reserves from the iron core. Low loss is not per se better, on magnetic circuits it creates very audible problems.
A correct air gap of a well made EI transformer (not the usual cheap chinese ones) makes it much more linear, and it absorbs DC on the mains with much less hickups. And yes, it usually sounds better.
As does a well designed "artisanal" insulation transformer with EI core.
One example more of reality vs. hype and hypothesis.
And a toroidal "out of reserve" creates very strong hum fields BTW.

In my system, there is a low count of transformers (and an increasing count of batteries). The LFD, with some hum and interference problem in our test, has toroidal transformers ;-) Close to my phono front end (1m) there is only one transformer plus the phono stage transformer, if there is one.

My LFD MM0 looks like it has the same circuit board as the LFD Mistral, but has a double mono power supply fed by a very generous toroidal transformer. It's switchable between MM and MC, it seems to sound very good on both inputs. I saw that one was sold on ebay for 275£, and another one advertised in spain vor ca. 310€. These are extremely attractive prices. Sonically, to my ears it plays well into the 2000$ phono stage league.

Transformer saturation is one of two main loss factors with transformers, magnetic and ohmic losses. The doubling of the power in high power amps with low impedances highlights the problem of ohmic losses. There are two questions about reducing transformer losses: Does it correlate with better sound, and at which output power levels? And which design optimisation has to take places with which compromises to reach the intended goal. Think also about the sonic qualities of single ended amps, some tube amps too with extremely good bass, but no doubling of the power to lower impedances. The optimisation of audio quality in the major signal range is important - even with considerably compressed music the main energy range is at -10 dBU rel. level, which means at one tenth the peak power. With better material it's rather around -20dB and more. In a normal listening situation one is inclined to reduce the replay level. Generally 90dB medium level, except peaks is very loud. My speakers need less than one Watt average for this. Everything that improves sound quality from zero up to this power range has a lot to do with sound quality. Doubling the power into half of the impedance would be desirable, if there were no inherent sonic compromises. I know this flies in the face of orthodox engineering, but there is some empiric and well thought through truth to it.
Reducing ohmic losses leads to an extremely high current path for the capcitor / rectifier path, with considerable problems, exacerbated by "stable, high energy" designs with huge load capacitance.
Interestingly one of the better sounding Goldmund transistor amps as well as some Cello designs did search for an *optimal*, not maximal load capacitance, with very good results.
One of the reasons is reducing the stress factor inherent in increasing charge current peaks with increased capacitance. The same design compromise is involved with low loss transformers.
There is a way of thinking (shared by some famous names) that a "transformer only provides voltage & current and has no influence on the sound whatsoever". If one starts to experiment with different (brand) transformers, one finds the opposite is true.
This has - in empirical evidence - strongly to do with the magnetic circuit, and wire diameters (and rectifier and power supply design). This is a wide - empiric - research field of patient work, which very few people have done from both sides, the producing and listening side. Most any "normal" transformer is built following general design rules, strongly economical ones, which overlook some key factors, which are relevant "only" to audio. Optics are sometimes also optimised.
It's BTW a question too, how desirable a wide bandwidth is for a power transformer, which is an inherent property of the general winding technique of toroidals.

- Ct 0517: The funny thing with ESLs, Quad etc. is that they do much better with something closer to a current source, this means they deliver much more voltage on high impedance than a low one. Doubling the power in each halving of impedance is the definition of a perfect voltage source. A voltage source gets tricked out on the voltage increase needed for the bass impedances (except if one uses insanely powerful amps). And it delivers the current in the highs (doubling, doubling, doubling the power) like a preussian soldier by executing it's order - but there is an intrinsic stress in that, of which thermal stress is not the least.
There were some highly regarded german SS amps that worked as a current source. It's a pity I never heard one or could try one.
So you listen on Quad ESL 57?

A few comments about the offered arms:
- the ET one is a good choice for a low to medium compliance cartridge, usually a MC. I think the bearing was almost  (?) as high quality as the "low pressure" ET2 bearing. The bearing (and the headshell / arm / bearing connection) is less stiff than the ET 2.5 though.
The main advantage of the ET 2 and ET 2.5 is that they are more adaptable (than most any arm on the market except the Mörch).
- Surge tank: My Airtech tanks end caps broke during a long time of being stored away. My replacement was a standard HDPE gasoline tank, 10l (ca. 2 gallons) with two 5mm holes drilled in the screw cap, with silicone tubing entering each hole, one with a short piece entering the tank, one with a long piece touching the "floor". The tank is filled with polyacryl wool for damping. The seam of screw cap / tubing then is "glued" with hot glue. This stays air tight since 10-15 years, and i's cost were almost nothing.
- My original ET pump with ET 2 arm gave  ca. 0.25 to 0.3 Bar. With the improved tightness of the ET2.5 air bearing I'd expect 0.4 Bar. This is at least enough for a (well) "working" bearing. My high pressure manifold still runs at 0.15 Bar - much better than the original "low pressure" ET 2 bearing!

Regarding levelness of tangential vs. radial arms:
The side-force on the stylus in a (non-servo) linear tracking arm is proportional to the off-levelness and the lateral DC mass, ie. lateral weight: From 75g up.
With a radial arm this lateral mass/weight force is almost cancelled in case of non-balanced arms with tracking force / "tracking mass" remaining, 1.5 to 3g. With dynamically balanced arms it's totally cancelled, so it is (almost) not critical with radial arms.
The sideways pull on the cantilever created by lateral bearing off-levelness and/or off-level platter and/or off-level record is visible at the moment the stylus hit's the record, but is not easy to see.
To have the arm helping leveling, the arm wiring is critical and needs to be thin and very elastic, ideally left/right separated with optimal wire looms. The radial arms have the levering advantage...
Optimising & eliminating sideways forces unleash the bass and dynamics of the ET2 linear trackers. It is *very* critical.
I doubt that in the 80's all too many ET2s were set up correctly, thus leading to the mythos that straight tracker have a "problem in the bass". 

Hi Chris! I tried making photographs of the wiring - but the spot is too dark, and the flashlight messes things up – but I'll try again.
- It's funny you mention that (even) increasing vertical mass by minimizing & shifting the counterweight out on the I-beam improves bass. Ie. moving the vertical resonance down from a frequency above 12 Hz in many cases. Because "too low" resonance is much more a problem with vertical resonance, not lateral, as LP rumble is mainly a vertical problem – and you hear even there that lower is better. I don't doubt this!
How much less of a problem is a really high *lateral* mass within a certain safety range (not below 2-3Hz resonance)!
(Except that down there is a problem with magnetic flux with some MM and/or MI? cartridges I suspect and found - not with MC.)
- I think the quote of "lateral forces are .1 gm. compared to .2 gm on a conventional arm" refers to the forces exerted by wiring stiffness and skating forces, not to off-levelness of the arm.
If one doesn't want to exceed 0.2g lateral force one needs to set up the arm to a level precision of less than 0.2mm over a 100mm travel! It's no problem to level the arm as such - but the wire may exert a "correcting" force for an off-level arm, and changes this sideways force at a different place than where it was level. A bit of a case of luck with the original cabling, or wandering setup of the turntable base.
- The increased FM modulation with radial tracking goes back to the dynamically varying friction which dynamically varies skating force with modulation, different radius, different pressing / vinyl surface, and different aerosol deposits on the vinyl. With the straightline trackers are always tangential except LF resonance, off-levelness or other reasons for a slightly off-tangential cantilever. But this error angle is always much lower than with conventional radial arms with strong offset angles. 
- This absolute offset angle is really much lower with the Thales arms, specially with the newest, the Thales Easy.

Hi Chris! The magnet damping doesnt't work the way you describe it. What you describe is the magnet "levitating" of the Verdier (and others) by opposing magnets of same polarity.
The proposed damping works by eddy current, the same as eddy current brakes. Basically you put a magnetic circuit very close to a electrically conducting surface. The magnetic flux crossing the surface induces a current in the conductor / surface, which creates a counteracting magnetic field.
I placed a neodymium kitchen magnet (or two) on a blob of BluTak underneath and very close to the spindle. (I like my setups to be a bit messy - but reversible :-).
My wires exit the arm relatively "straight" at the end of the tonearm proper, the magnet is on the opposite side of the bearing.
The small round kitchen magnet has a magnetically conductive (steel?) "guiding" sleeve around the magnet that guides the back side pole to the front, so you have opposite poles in the center of the magnet vs. a circular opposite pole around.
In my case it works "enough", so that the arm has no high Q lateral resonance. It's probably still slightly underdamped though. Which is IMO desirable as a trade-off.
I don't have an alternative anymore, as my oil trough is leaking.

a) regarding wiring: I have to try again to photograph - the lighting makes pictures difficult... patience :-)
b) Fridge magnets & damping: Drop a circular magnet into a vertical copper tube. An enlightening sight - the magnet centers itself and *creeps* down very slowly instead of passing quickly!
Magnetic attraction is needed for static force only or for DC so to say.
With movement or "AC" the magnet induces a current into a close conductor (for which alu is moderately good), which creates a counter field, ie. a damping action. OK :-)

Chris

So just a question.
Bass waves are slower than all the others. So why do people put subs X feet behind their mid and tweeter drivers ? Just asking.

If I was you Harry I would be very tempted just "temporarily", to set up in near field for one hour - just to hear what was possible with one sub.

I’m not exactly shure what this would mean? More upfront placement = faster bass?
Placement is a question of "allowable" phase angles between drivers caused by offset drivers (specially in distance to the listener). As long as this offset phase angle is not far off, there is no "real" problem.
Bass waves travel at the same speed as any acoustic wave in air, but have longer wavelengths, so offset distances in the low bass can be considerable without being audible. Usually it’s said to be desirable to be below 90 degree - I’d keep it lower, to be on the safe side in the cutoff region where the subwoofer is still emitting considerable upper bass levels.
At 40 Hertz xo this would still be 67cm for a quite tight max. 30 degree difference. This results in an allowable 120 degree two octaves above, at 160Hz, at eg. -24dB (for 12 dB/oct.)
But I agree, that close-field placement is the preferable "error", because a) the excursion / level of the subwoofer can be considerably lower, b) the decrease of room reverberation level relative to the direct sound level should reduce audible room resonances & bass colorations = c) the modification of the subs frequency response by room resonances drops an order of magnitude.
But... it works only if there is no spurious midrange energy exiting reflex tubes or coming from the driver. Still - the lower sub levels reduce distortion and "box talk" & "cone cry", both help.

Re/ VTA & gooseneck range not high enough: Having a BSc in audiophilia nervosa including a MSc in empirical material testing... :-)
I'd propose a surprisingly crude idea, which might work better than it should. Have a carpenter (or your right hand :-) make a wedge formed piece of wood - good sounding wood like solid fir or similar, or maybe also birch plywood. The wedge is placed between the ETs headshell (doesn't need to protrude from the wand) and the cartridge, covering not more than the interface between your cartridge and the arm wand. The wedge corrects for the back tilt angle of the arm if the manifold and gooseneck are in middle & correct positions. Maybe the wedge is 2mm front and 4.5mm back - make a drawing and calculate the height correction.
You could glue the wedge to the cartridge with a bit of white glue / elmers glue which doesn't hold to metal all too well. This makes an easily breakable connection. The other side (either to the arm or to the cartridge) uses screws - or maybe even thin double stick tape. *Not* having a totally rigid connection with natural materials in the "mechanical loop" often sounds more natural.

Regarding arm boards and optimal materials, it's very interesting to watch the diverse experience in idler drive plinths. A mixture of natural and slightly lively wood layers, combined with *some* more deadened materials, like *a bit* of MDF and maybe one layer of stone seems to yield very good results. Pure MDF not, neither Corian, synthetic stones, marble or metal is desirable. I think a considerable part of this experience should be considered when thinking about armboards. Solid wood and *maybe* a bit of acryl or alu smewhere should work well too.
Personally I'd seek other materials than overdamped polymers like Sorbothane - it's sounds like s*** as a platter mat, and it's audible wherever one uses it. Even natural rubber feet on speakers (or under analogue decks) sound IME sluggish. It's easy to overdamp surfaces and overtighten connections (not only because of eg. making cracks in the ET structure :-). Both result in kind of a deadened, "negative feedback" kind of unnatural sound.
Solid maple? A friend uses solid oak to very good effect for the adanalog arm.
For ET arms maybe solid sBruce is the optimal wood...?
An eminent discovery.

Hi Chris! I accidentally grew up with a Lenco B55 (practically same technology as L70/75/78) and a Thorens TD 124. These were the first victims of my brothers and my audiophile interests - we dismounted and reassembled them. I was more impressed by the ingenuity and elegance of the Lenco - *and* I still remember the sound as very upbeat and vivid, memories reaching almost 55 years back! Much more so than my slightly lukewarm memories the TD124, I'm sorry to say...
The Lencos plinths empty rooms are not for controlling resonances - they are for ventilation, as the motor, a shaded pole motor, has very low efficiency and consumes *35W*...! I rather think that in the end this is more of a weak point than the idler wheel. My impression is that the vertical low mass idler wheel is (within the context of idler wheels) a very good idea. The weak point there is a) the steering of the wheel with kind of a rod with limited stiffness and b) the exact positioning relative to the conical axe of the motor. But...
Rehearing the L75 in a friend's (superb) system with modded Avantgarde Trio etc. etc., after years of listening to a Well Tempered Signature, then a Funk tt with Funk arm was a revelation - already with a simple plinth. Of all "real existing" turntable platforms it is in musical, timbral, dynamical and coherence terms almost in another league - and it is a "stone age" concept. It really was a jaw-dropping experience. ca. 1980 I *never* considered re-using a Lenco or a SL1210). I was a *real* audiophile, who had to use a belt-drive... I bought a Thorens TD 126 - one boring lifeless turnatble. Then came the Rega 2, then a Walker CJ55. Both really much better in musical & colour-terms, then the WTT Signature in my system too.
Funny how  I moved from "scientific" musical medicine back to "evidence based" or empirical musical medicine.
Today I use a much modded SL1210, which has a slight advantage in purity / stability of upper midrange timbres.
In Europe Pet Reinders is the Lenco "guru". I can do without gurus, though he has a good solution for the basic motor / bearing platform.

those nasty motor wheel - rumblings.

I was actually amazed about how they disappeared behind the not inconsiderable LP pressing and cutting lathe rumble (the latter sometimes audible too), and behind the superb musical information & drive. I expected problems, but...

armboard holes are helping to provide tonearm isolation

I assume Jean Nantais has pretty good reasons and convictions why he does things the way he does.
But - looking at physics - how could these holes provide isolation from either acoustically transferred noise (which, if they'd change at all, would rather increase) or structure borne noise.
Besides: How does it sound?

Hi frogman, I share with you the admiration for Paul Desmond. I play the alto myself and he has +/- my ideal sound on the instrument (within the limits of my abilities, and the recorded sound as you point out).
Loading MCs: Your line of thoughts reminds me of my experiments  with the Magnepan MG3 speakers with their superbly transparent, but often a bit glassy sounding tweeter. I started with the usual recommendation to put a "normal" resistor in series, but this was loosing much too much transparency (a noiseless back-ground with kind of a plentyful sea of lively information). Then I used parallelled  ERO Resista metal film resistors, but the effect was not much better. Then I replaced the HF-Litz feeding the ribbon with a thin solid core wire and had success with more transparency and less "noise" and brightness. Ultimately I developped an independent, three-wired new crossover (with soldering to the ribbons... :-) which somehow integrated the superior mid/high transparency of the two-way MG 2.5 (and MG 2.6) into the three-way-MG 3.
Before these changes I had the feeling that my Koetsu Black / VdH 1 needle (Gyger upgrade) was sounding "bright".
Removing the transparency obstacles, which created stress and brightness, allowed to retrieve the VdH1 stylus information without glassyness.
I am skeptical of loading as a cure for problems created elsewhere, it's not really working in that context, it's too much of a compensation game (for me). I loose too much, gain not much.
Besides: The direction of the resistor is as audible as that of cables, as is the brand and type of resistor (and the preparation of the wires and the soldering including the solder).
To attain an experimentally "clean lab table" for these kinds of comparisons, within the context of the ultimately almost always slightly sloppy way people generally do their comparisons, seems pretty optimistic to me.

Frogman, I can only second what you wrote! To hear the elimination of contacts or to compare different connectors under controlled listening tests has been my most educating experience in audio. I think the effect of contacts on sound quality are highly underrated. I guess that from the omnipresence of brass based "professional gold" contacts in a wide field of the audiophile market. Listen once, never return back.I was an anti-WBT low-budget "snob" :-) but the copper based WBT RCA connectors and the Clearaudio MPC were hugely better than any silver- or gold-plated brass connector back when I compared these (with ca. 7 identically constructed & oriented silver interconnects). Yo don't get what you haven't paid for - except for soldering directly!

Regarding pumps and pressure:
a) Back in the 80's I found that the WISA 300 "high pressure" model (of which I bought several, directly from the company) was a single membrane pump, not a double one like the original HiBlow. My high pressure WISA model dropped in actual measured pressure level at the arm below the HiBlow, and had a high variation of pressure (mains frequency) - and the needle vibrated. This resulted in a "homeopathic 50Hz drone" that subjectively gave more, but a slihtly tubbier bass - this was *with* a ca. 7.5l tank.
b) I then chained two HiBlow pumps, ie. I hot-glued a piece of silicon tube to the inlet of one pump and to the outlet of the other pump. This resulted in (if I remember correctly) ca. 0.45 Bar (6.5 psi?). Three pumps were around 0.6 Bar, all this sounded very good, and was more silent  than a WISA pump. The HiBlow has a surprisingly steady flow and it's limitation can be surmounted by chaining two or three. On the ET 2's "normal" low pressure bearing the two HiBlows were a remarkable improvement.

This is indeed the way I tested this. Stylus in groove, platter stationary, pump running, phono gain turned up. In this test the HiBlow was audibly quieter than *my* WISA 300 high pressure.

Hi John, no, your WISA pumps work the same. But connecting them parallel and out of phase makes them work the same way as a double diaphragm pump, like the original HiBlow: Single ended vs. push-pull... The air pressure is much more constant, "stabilized" pressure being about the same as the pressure peaks of a single WISA.
How high is the pressure? I assume it’s still not much higher than the original pump, and the worse noise/vibration is remaining. Besides I was not able to connect WISAs in series (nor will such a parallel combo), contrary to the HiBlow pumps with their air-sealed housing, which allows operating them above normal atmospheric pressure.

A few thoughts regarding longer i-beam and leverage:
- While the captive air-bearing has advantages in stiffness, the ET design with stationary bearing has only one point with centered forces within the bearing along the arm travel. This "allows" some maximal leverage at the beginning and end of the travel.
- Having a short i-beam compensated with more weight on the i-beam side increases (lateral) leverage.
  Eliminating the leverage on the air bearing by using a moving air bearing has some advantages, but a drawback too:
  Either one needs an additional flexible air-tube, which increases vibration coupling between arm and tt chassis, and also vibrations within the air feed. 
  Or one needs an open bearing with less bearing stiffness, like the (very good) Adanalog MG 1 arm, or the many Ladegaard variations incl. the Trans-fi arm. (see below)
- It is desirable to keep varying leverage at a minimum, a)  on the air bearing as well as b) on a sprung subchassis. The latter will lead to varying lateral desiquilibrium, visible as increased "slipping" from a balanced middle position to outwards gliding at the start and end positions of arm travel.

- So a longer i-beam reduces the leverage on the bearing, and on sprung subchassis. It improves balance of the center of gravity along the bearing travel.  And it increases vertical mass, which often improves the sound too.

- Regarding "optimal" vertical resonance I did put up the idea, that (maybe) there is not much useful out of phase bass information below 100Hz - this might not be the case... as increasing the (out of phase) bass bandwith below eg. 15Hz, by lowering vertical resonance seemingly still is audible.
- However, there is an end to that, probably below 10Hz because of subsonic vertical garbage information with a maxiumum around 5Hz. This contrary to lateral resonance where the main problem frequency (correctable) is not higher than 0.75Hz.

- Trans-fi arm:
I trust several keen and independent ears, that this is a superb design, technically by design and subjectively by constant tweaking and improvimg.
- the stability/stiffness of it's "open" bearing is still very good by it's v-shape which forces centering of the air bearing by help of gravity.
- the vertical knife edge bearing is a very good and elegant design with no bearing chatter, and with selfcentering by gravity too.
- "optimal geometric design practice" followed by ET keeps the vertical bearing axis on the same height as the tracing / platter level, This reduces FM wow induced by vinyl warp,
-  However, this is not without a price: Lifting this tracing level has it's advantages too, leading to improved bass: Trans-fi and RS labs ar examples. This is, because putting the vertical axis in line with the cartridge cantilever ((ca. 20-30 vertical degrees)) eliminates vertical pull on the cantilever bearing, reducing vertical resonance of the arm activated by dynamic friction changes by tracking vinyl, and by variable friction...
No free lunch therefore for ET... and the above mechanism is rarely looked at. 

thoughts on two or three open questions & i-beam
@richardkrebs:
- I'm looking forward seeing pictures of your bearing tower!
- I did not yet try the elevated bearing position with the original ET bearing tower, as there will be drawbacks stiffness-wise because of the (practical but) rel. narrow tripod feet of the design.
- I think a 20 degree angled piece of good wood (with a slit / hole for the cartridge nuts) would do the major part of adaptation.
- arm lift adaptation could be tricky.

- magnetic damping: "My" magnets are kitchen magnets. They are round and have a simple collar-shaped pole piece which focusses and intensifies the flux around it. This will make a considerable difference in efficiency regarding damping compared to a normal "open" magnetic structure.
The magnets need to be placed really close, not more than 1mm on the closest place. The flat magnet shape vs round bearing tube  is obviously not ideal. My feeling is that it kind of cuts the peak of the resonance. It sounds and feels a step more stable - i don't like the sound of too much damping anyway.
It's difficult to extrapolate Bruces experiences & comment to different implementations without exact descriptions.

- i-beam: I checked the rotation pendulum formulas to be correct about this: If one doubles the length of the i-beam one halfs the needed weight, because of double the leverage. But the inertia grows with a square factor: 0.5 of the mass x 2^2.This results in double the inertia and 0.7 of the resonance frequency. 
The whole double mass double spring system of the ET is very ingenious but also quite complex. It is a 4th order resonant system instead of the usual 2nd order one. With one short attempt :-) i did not yet successfully find the correct CLCL model for simulation.In the most simple view there is an i-beam resonance and a bearing tube/cartridge resonance, with a zone where the i-beams spring and the cartridges compliance work in series (the i-beam springs damping can control the cartridge/bearing resonance, if the i-beam  resonance is well chosen... ;-).Problem is seeing / knowing what each one does. Simulation? Empirism?... :-)
My techno-intuitive thought on this is:
- The lower the resonance of the i-beam, the wider the frequency range over which the bearing/cartridge resonance can be controlled by it.
- the higher the resonance, the closer together both resonances and the more resonant interaction instead of control.

Does Bruce have the model? Or should I ask my more MATLAB-experienced son?

Thanks for the captivating infos!
- Although I agree on the idea of eliminating turbulence - isn't damping it not also introducing an additional flow resistance which will upset the  "balance" of implicite air flow resistances from air feed via calibrated air beaing openings to bearings air film? Personally I'd be very (self-) critical when changing this, I like to be double on the safe side about changing such crucial areas (at least with my limited understanding of air bearings).
In fact I'd contact Bruce about his thoughts regarding keeping the correct pressure on the distributed points/air openings when introducing additional flow resistance.

- raising the arm bearing: I meant lifting the back of the arm, and correcting with a 20 degree wood wedge between arm and cartridge (the wedge pointing to the front of the arm),

I don't know your degree, so I don't tell you mine... ;-)Except that you will find the formulas if you feel the need.It's a rather elevated compliance in these days, not Shure-high but probably, relating to ct517 experiences, it better fits a ET 2 and not a ET 2.5, as the mass of the ET 2 is lower. However, I think it's mainly a question of the arm wand, ie. using the original alu wand.
With this cartridge it's more about experimenting with vertical inertia, starting at the lower side, starting with more counterweights close to the bearing.
You always can increase vertical inertia by trying less counterweights in the middle and then a low count at the outer end.
I suspect that still the long i-beam with counterweights out and starting with a single spring might be best, as the high compliance / low resonance frequency of the cartridge might be better controlled by an I-beam that is set to the lowest most resonance.

Last time I did an extended cartridge optimization session for a customer (SME arm), we extensively tweaked the tracking force and bias (.. ;-) by ear.
Start at the lower end of the tracking force, and prepare small balls of Blue-Tak at 0.1 gram, measuring them with a precision scale 0f 0.01 gram resolution.
By adding them (side by side and not pushing them flat) you’ll find a spot where the sound starts to feel slightly compressed, slower and too creamy (= too high tracking force / too low VTA), instead of dynamic, fast and a bit dry.
Go back to one mini-blob less and do the same with mini-blobs of ca. 0.01 gram.
It’s surprising how audible this is. It’s worthwile to start with finding the best VTA, do the weight optimization, go back to VTA and do the weight thing again. The weight adjustment and VTA interrelate *somewhat*, but they are not the same.
*Check the lateral balance=calm free floating in the middle, when adjusting VTA with the ET adjustment!*
Otherwise you introduce a major variable, while tuning a subtle, important factor.
It’s not a real hassle. But it’s ear opening.

Hey Brad,
Also check for obstructions in the airline and nipple feeding the body of the manifold housing.
Dave

It’s always preferable to eliminate obstacles between nipples and the mani folds of our bodies...

Fumbling around (...) with air bearing arms, also of other makes makes things appear simpler than they seem:
It’s simply a question of turning the arm around the mounting hole to get the exact desired travel, as defined by the lift bar. It may look slightly twisted though, but that’s the correct way. (You might correct a bit with pulling the end travel "hood" pushed over the bar... a bit)
But basically the travel is set by turning the arm around the single mounting screw and finding the correct position.
Tangentiality then is set by adjusting the length of the arm, ie. (non) "overhang".
Actually all is very simple.

What’s less obvious is that it is important to have
c) the surface of recordings in average level,
b) the platter and platter bearing level, (maybe with a smidgen of defined tilt toward the motor, both for optimized friction or dynamic behaviour with belt drives) and
a) the arm level. With highly increasing importance:
It’s a) where the setup sensitivity of passive tangential arms differs widely from pivoting arms.
The whole mass of the arm - dynamic and static - pulls sideways with the slightest off-level.
A normal pivoting arm with tilted setup compensates gravity "error" almost completely (with only the tracking force / weight remaining). With dynamically balanced arms it cancels to 100%.

Fortunately the air bearing arm itself is maybe the best measuring device on earth for off-levelness.

Hi frogman, yes it seems so. But think about a "normal" mount on a rectangular base:
The arm is mounted "precisely" and travels exactly parallel to two sides of the plinth. Correct tracking then runs on a line exactly parallel to the sides, crossing exactly the platter bearing. This is achieved by the correct "overhang".
Assuming now the arm is slightly twisted (as seen from above) and runs at an angle to the sides. The tracking line then will be twisted the same, but if it shifted by correcting the "overhang" to cross the bearing again, it tracks as correct as before,
Twisted clockwise the overhang will be a bit less, twisted anti-clockwise it is a bit more than with "correct" parallel mounting.