Torsional vibration; Gimballed vs Unipivot

[IslandPink]

OK Folks,

I had a moment of clarity the other night while listening to some music . This is not the normal state of mind I seek at this time, but you never know when it's going to hit you I suppose .

I was musing about the transmission of vibration through a tonearm to its support, wondering about why this interface is still important musically .

I think I now understand why there is STILL vibration that has to be resisted by the arm-pillar, for a gimballed arm, and why unipivots differ ( crucially ) in this respect .

From what I can see , the gimballed arm is decoupled for two important motions ONLY - the side to side pull caused by lateral stylus ( music ) motion, and the up and down vibration caused by vertical music modulation and record warps . These motions are then mainly resisted by the counterweight, which is the preferrable state of affairs as far as I can see - a short and relatively 'clean' mechanical interaction . There is at least one important motion that is NOT effectively decoupled by the bearings . This is the torsional ( twisting ) motion - caused by lateral stylus motion and the moment of the stylus around the arm-tube centre axis . This is mentioned in
Mark Baker's woffle on his Origin Live website and promo material, with reference to submarines attacked by depth-charges and suchlike .

There are other effects , maybe significant or not - a push-pull effect along the length of the arm, for instance . I don't know how big these effects are, and I don't think they differ from gimballed to unipivot arms .

In any case, if you think about the torsional mode , this one ( high-frequency, as usual, due to RIAA ) goes down the arm and may cause chatter in one or both of the bearings , but even if it doesn't , it ends up as
a bending motion on the arm pillar or on the arm mounting flange and on any interface between arm pillar/flange and the plinth . Any difference in hardness here then causes a reflection of energy back ; any looseness causes 'noise'
to be generated which also feeds back . In either case, this junk signal has no chance of coming back in phase with any musical signal and must cause distortion or clouding of the original music .

What's very interesting to me , is that this mode of vibration, and the problems that can be caused, do NOT happen with a unipivot - this torsional mode is decoupled at the unipivot , and is resisted solely by the counterweight and any outriggers , together with the basic stiffness of the arm tube . This seems preferable .

I can already feel JamesD chuckling to himself, thinking how each new bunch of audiophiles has to rediscover these audio truths for themselves and present them as 'new insight' ... !
Hope this is of interest .

Mark

[Clive]

Mark,

I wish you'd stop trying to get me to spend more money on a Hadcock/Morsiani etc.

Your arguement seems logical and quite obvious....now that you've made the point. Wish I'd had the clarity of thought.

Clive

[Black Stuart]

Well thought out,
and as someone pointed out on another thread - why did the belt drive take over so completely from the idler - cheaper/quicker and simpler to produce - ergo more profit.

Black Stuart

[James D]

Mark,



Very nicely thought through. Now are you and Clive going to make your own unipivots? It really is the only way to go...

James

[IslandPink]

Hmmm,
Yes, usually have some form of 'clarity of thought' after I've spent a lot of money . 'Welcome to the next level' as JC Morrison said , this time in analogue gear .

Now then James, where were those DIY links - there was a nice-looking acrylic one I saw somewhere .

You must understand this project has to take its place after the SE amp ( end of this week ?! ) and the phono mods , then the subwoofer ...

Mark

[bornin50]

Hi All,

If this is crazy let me know.

This decoupling of the unipivot idea, do you think it also applies to knife-edge designs, a bit? And could similar effects explain the positive user feedback for the Rega vtaff, which decouples the arm from the fixing?


Regards

Colin