Tonearms, separation, soundstage and imaging

[Oldpinkman]

Apologies if this has been covered before, but I couldn't find anything with a quick search. My mate Arthur is snowed under (as I am really) and not available for humouring a mate talking tt design. But Mrs S has asked the question after listening to the FX3 with me, and we have this Professor Higgins / Eliza Doolittle thing where she just assumes I have the expert answer, and I hate to rock the domestic balance of power by disappointing her.

Why does a tonearm affect imaging and separation? It clearly does. And from experiences with cartridges I am pretty certain that "imaging" and "stereo depth" are down to the channel balance and separation characteristics of the various components. This is evident with the FX3 - the imaging is so much better than the F5 which in turn had more detail than the 774. And this correlates with evident separation - the soundstage goes wider as well as deeper, and there is quite a lot of "just left speaker" or "just right speaker" sound on recordings mastered that way. The technics U205 had great imaging and depth, and measured well on separation (well as in significantly better than "narrow" cartridges)

But how does a tonearm affect this? I have a pet theory, but I am not technical, and so this could be complete b****ks. Clearly one of the factors in arm design is tube beam resonance. I can see how a big spike at say 800hz will colour the sound. And, presumably, regardless of the channel supplying the energy to excite that resonance, its impact will affect each channel broadly equally, thereby reducing any separation effect. But that would only affect the separation at or around that frequency spike. How's this for an idea? Whilst avoiding a spike avoids colouration, reducing the level of total resonance in an arm must reduce that "blurring" across the frequency spectrum. If that blurring from interfering resonance, regardless of the channel originating it, affects both channels equally, that in effect puts a signal into both channels and reduces the difference between them, or the purity of their individuality.

Does that work?:scratch:
If not, how does an arm affect stereo imaging and separation? 'Cos it clearly does.

[Reffc]

Hi Richard

Tonearms per-se do not add anything, but it's how they react with the cartridge that makes all the difference and this is a matter of design.

If you read the bumff supplied with your cartridge, it will have figures for stereo separation listed. Where it achieves those figures is partly determined by the tonearm it's fitted to. Cartridges produce energy and in the case of moving coil cartridges, relatively speaking, a lot of energy. If a tonearm is not a great match in terms of mass, this will affect the sound. Too heavy and the cart can become over compliant and dull sounding with loss of things like stereo separation. Too light, and the opposite happens...you may get insufficient overall compliance and a lot of the energy from the stylus is lost to resonance in the arm tube, again affecting sound stage amongst other things. The bearings also play a major role as does the tonearm's ability to handle resonance. It should be as resonant free as possible by design, or at least shift its natural resonant frequency above the audible spectrum (resulting in a very stiff construction). The bearings should be as stiction-free as possible so that they are free to move with as little force applied as possible and without binding, even slightly, or without allowing the arm to move off-axis about it's pivot point.

All quite technical when you start to look at the detail, but the long and short of it is that a tonearm/cart combination should be assembled with the combined resonant frequency points within specific upper and lower points, the arm should be as stiff as possible, and the bearing quality must be high. There's more too it than that, but it offers a simplistic view. The differences between well engineered tonearms can be very little indeed with the same cartridge used providing you are comparing like with like on mass and compliance. That's all before you look at differences between wiring!

[Oldpinkman]

Hi Paul

Thanks for the suggestions. I am familiar with setting the main mass compliance resonance above the record warp, and below the audible settings (9Hz to 14Hz) and that is not this issue. You are right about resonance affecting the sound. Most armtubes have a spike ( a major resonance) and that affects that frequency. What I was trying to get to grips with is why does a good arm give such a good stereo soundstage and imaging. Or rather, since I think that is to do with separation, what arm characteristics affect separation. It is a mechanical thing. It is not the main cartridge compliance / arm effective mass thing - since that is easy to tailor and adjust. It is something else. Bearings again need to be good enough to transmit energy away to the armboard / subchassis evenly across the frequency spectrum and without chattering, but most modern half-decent arms have bearings that good.

And no arm on the market I have seen shifts their natural main resonant frequency above 20Khz. Most have a spike between 500 and 1200 Hz and several have more than one (often the counterweight too). Most have spikes exceeding 20db, some exceed 30db, and no other arm I have seen a graph for spikes less than 10db (and I believe that includes the Dynavector Martin). Fx3 doesn't really have a spike, more a small hump of 4db or less, which accounts for its lack of colouration, but has pretty low levels of energy at all frequencies - which I was speculating accounts for the separation thing, in the manner I described before. I wondered whether there was any knowledge on that point.

Yes wiring makes a difference - but it can only protect and preserve with accuracy the signal produced by the generators in the cartridge. What they produce is a function of the stylus moving relative to the cartridge body. If the stylus is moved (or the cartridge body moved) by energy other than the groove displacement, then something is added or something is taken away. So I guess it has to be mechanical energy that is changing the characteristics of how an arm sounds.

The question is which mechanical energy causes that imaging effect (separation) and how? Since I had forgotten the FX3 is quiet, I suspect it is a resonant tube which adds signal - and adds it pretty much evenly to both channels, which spoils separation where a particular frequency is supposed to be only on one channel. The lack of resonance in the FX3 reduces the unwanted added signal, which improves the separation and makes the output lower overall. I am struggling to remember whether other arms I thought performed well (Helius Orion and SME 5 stand out) were also "quiet" arms. Or whether there is some other explanation, since both those arms had fairly high energy levels and main resonances compared with FX3.

And then there's the conundrum of the missing bass, that isn't missing at all. I need that boy Arthur to get over his production crisis, although from his evasion I suspect he doesn't know for sure.

[Reffc]

You've almost answered your own question Richard ;-)

I think that the differences can be partly explained by the resonant behaviour of the arm (and also the effective damping). Some arms are much better than others by design and as you say, an arm which doesn't transmit loads of energy into the bearings is preferable. The ideal is an arm that is infinitely stiff but has no resonance of its own and of the correct match to the cartridge, but also one that is able to damp energy sufficiently well to allow the stylus to just get on and do its thing. Separation if you think about it, is the ability of the cartridge to accurately track the vertical record grooves without mis-tracking, jumping or being forced off balance by incorrect bias or VTF. In that respect, an arm must mechanically have excellent bearings which minimise the impacts of the inevitable friction such that it remains a small fraction of the force exerted on the stylus by the grooves, that it allows the stylus to track accurately across the record sruface, and that the arm has sufficient internal damping characteristics not to react alike a spring (or conversely like a soggy sponge) to influence the compliance of the cantilever.

When you start to think about how this all works, it's always a source of amazement that with so many variables that can give rise to distortions, how good a well set up LP system is, even something of very modest origins.