Item Audio "Spoke" for the Linn LP12 anybody..?

[sq225917]

I have climbers hands, it wasn't hard to rotate the subchassis anticlockwise by a small amount.

@Sondek, yes the movement was limited and the rising rate was significant. I doubt it could be stimulated to make the same movement by the forces attempting to drive the platter.

I think it needs some tweaking still. it'll never bounce accurately until there is correspondence between the geometric centres of the spokes and suspension. I'm not sure if it would even be possible to create an alignment that works due to the requirement for near triangular positioning and limited real estate on the top plate.

 

[sonddek]

Bounce isn't necessarily relevant to performance. In an unspoken deck it might help resolve resonances into vertical motion, supposedly the least harmful, but in a spoken deck it shouldn't matter very much.

 

[Item Analogue]

No, it doesn't. A cartridge generates voltage output in proportion to groove modulation. It can't be "driven harder" unless you're playing a different record with 1812 on it.

Why can't a cartridge that is being supported, relative to the surface of a record, in a manner which is more stable than another, be able to resolve more information? It will be able make larger and more dynamic cantilever excursions than one that is held with less stability. I am sure that the cartridge output will be greater if the rate of change in flux induced by the soft iron core within the cantilever is correspondingly higher. Again, I am sure that this would be easy to measure.

Two Hi-Fi mags have expressed an interest in reviewing the Turntables in the near future, and the likes of Paul Messenger, Russ Andrews and David Price could all identify clear preferences for the Spoken LP12.

 

[Chris H]

I listened to the headphone demo that was set up at the NAS at Whittlebury. I approached, honestly, with an open mind. When I heard the huge difference in output, probably 1.5db, I was immediately sent into looking at what was wrong. As a DJ, (and an audiophile) I am very familiar with setting up two supposedly identical decks, and checking their output. The cartridges that were on both decks were not far off in quality compared with DJ carts, as compared with the low output MCs I am used to on my LP12 and other decks. The only time I ever hear such a significant difference in output is when there is a problem, such as a damaged stylus, dirty or loose connection, or other problem. A dirty tag/ headshell connection can easily cause this problem. As Peter swapped out the stylii in your dem, that was ruled out. At a quick glance I noticed that on your spoke deck you were using Eicmann Bullet RCAs, which are known to be very high quality, and on the other deck you had some other much lower quality RCAs. Also the cartridge tags on the spoke deck looked to be much higher quality than the non spoke deck. As to other differences, I did not look too closely. The switch used in your box could have potentially also caused a problem.

Why were both demo decks not identical? After all this is what is really necessary for an accurate, much more convincing comparison. I strongly suspect that there was a problem in possibly the cartridge, or some difference, or dirty contact somewhere in the chain. There should not have been a 1-1.5db difference between the 2 decks.

I was not looking for the spoke demo, I just happened on it. I had an open mind, at first....

As for the bounce issue, I wonder how you have found yourself so far into the LP12 world as to be developing a mod for it, but still going against long established common wisdom for this deck with regards to optimum suspension settings. Almost anyone who has ever seriously explored bounce on the LP12 suspension has found a combination of spring settings that basically negates any bounce whatsoever. That is how bad the suspension was set on the spoke dem deck. The other spoke demo that was present, the suspension/ top plate/ arm board only, had a much better bounce, as did the non spoke deck, though neither being great. Certainly if you set up LP12s, ensuring that the bounce is similar on both decks is a no- brainer.

Really I had an open mind, I just wanted a meaningful comparison.

It does make sense to me that the additional mechanical connection present in the spoke mod will allow more of a path for motor noise to be transmitted into the important bits during playback, and a well setup deck should tend to bounce in a vertical plane.

Additionally, standard folding tables are probably not the best support for LP12s, that flat table top surface is highly resonant. Ensuring a careful, identical, optimum setup next time will allow people to have a fair comparison..

Sending out a spoke kit to an objective, respected LP12 technician will yield a much more accurate opinion.

Cheers!

 

[James]

Why can't a cartridge that is being supported, relative to the surface of a record, in a manner which is more stable than another, be able to resolve more information?

More information, yes. More amplitude, no.

It will be able make larger and more dynamic cantilever excursions than one that is held with less stability.

For this to happen, the whole cartridge would need to deflect in sympathy relative to the groove to reduce absolute drive to the coils. Speed stability of the platter has nothing to do with cartridge output unless it is stopped altogether.

I am sure that the cartridge output will be greater if the rate of change in flux induced by the soft iron core within the cantilever is correspondingly higher. Again, I am sure that this would be easy to measure.

Seeing that you have both spoken and unspoken decks, you're in a prime position to do so. Let's see some data.

James

 

[sq225917]

@Item, you still seem to be missing some basic understanding of how turntables and cartridges work. Let's not forget that the arm is fixed to the sub-chassis which is fixed to the bearing and platter so the spoke makes no difference to this- though your choice of a bolted arm board on one of the demo decks would, as opposed to the wood screws on the standard arm board fitment.

Suspension is there to filter out environmental noises, inc motor noise from reaching the deck. Bypassing this juts couples the noise to the cartridge more directly, there's no free lunch here, though coupling at certain frequencies could be more beneficial than others. certainly less suspension = more speed stability, that isn't under question.

But if you want to claim that the Spoked deck is louder then you have to accept that this is all from coupled noise overlaid above the cartridge output.

 

[MTrei]

Just stumbled on this fascinating thread after seeing a kit on Epay. I have been thinking about this issue for decades, but long ago switched to a Roksan which specifically tackles this exact problem. However, I do set up turntables on a professional basis, including literally hundreds of LP12s.

Dressing the arm cable always felt like an inadequate solution to the problem, but it has always been critical factor in performance. When the RB300 came out, people said it didn't work well on a Sondek. I felt that the original skinny RB300 cable was to blame, so in my setups I would add on a thicker piece of cable alongside the arm cable from the arm base to the p clip as a stiffener. This worked quite well, but I always felt a more radical solution would be better. This looks like that more radical solution.

As other have noted, I am somewhat concerned about noise transfer from the top plate to the subchassis. In my solutions (in my head) instead of a stiff rod I was thinking about a tether that could be carefully adjusted for zero slack. This would restrict clockwise rotation while remaining quite compliant vertically.

Hopefully I can try this out for myself someday. All I need is an LP12 owner who's willing to have six holes drilled in his top plate, and three in his subchassis!

 

[sonddek]

Just stumbled on this fascinating thread after seeing a kit on Epay. I have been thinking about this issue for decades, but long ago switched to a Roksan which specifically tackles this exact problem. However, I do set up turntables on a professional basis, including literally hundreds of LP12s.

Dressing the arm cable always felt like an inadequate solution to the problem, but it has always been critical factor in performance. When the RB300 came out, people said it didn't work well on a Sondek. I felt that the original skinny RB300 cable was to blame, so in my setups I would add on a thicker piece of cable alongside the arm cable from the arm base to the p clip as a stiffener. This worked quite well, but I always felt a more radical solution would be better. This looks like that more radical solution.

As other have noted, I am somewhat concerned about noise transfer from the top plate to the subchassis. In my solutions (in my head) instead of a stiff rod I was thinking about a tether that could be carefully adjusted for zero slack. This would restrict clockwise rotation while remaining quite compliant vertically.

Hopefully I can try this out for myself someday. All I need is an LP12 owner who's willing to have six holes drilled in his top plate, and three in his subchassis!

Two points:

1. As far as I can work out, a Xerxes stores motor energy in the spring that limits motor rotation. This scheme is quite similar to how a Sondek stores motor energy in the suspension. It inserts a resonant buffer between the motor's power and where it's needed, the platter.

2. You can preview sub-chassis stabilisation by using sellotape and thread at the corners of the armboard:

Try different tensions. It makes an enormous difference to how pianos sound, for example. They begin to stay in tune. It does seem to have a deleterious effect on high frequency detail though.

 

[sq225917]

As will anything that couples the motor to the cart.

 

[MTrei]

Two points:

1. As far as I can work out, a Xerxes stores motor energy in the spring that limits motor rotation. This scheme is quite similar to how a Sondek stores motor energy in the suspension. It inserts a resonant buffer between the motor's power and where it's needed, the platter.

2. You can preview sub-chassis stabilisation by using sellotape and thread at the corners of the armboard:

Your analysis on the Roksan bearing mounted motor is not correct. It's there to ensure that the tension in the belt on both sides of the motor pulley remains constant, and this stops the motor from being pulled out of the center of its synchronous "pulse", reducing motor noise, and the slingshot effect you get when the belt is tighter on one side of the motor than the other. The Roksan motor is much lower torque than the LP12, but the main bearing is vastly lower in friction than the Linn while the rotational inertia of the platter is far higher. Almost all of the motor's power goes to overcome stylus friction rather than main bearing friction.

Also, I don't use a Xerxes, I have a TMS which isolates any remaining motor noise far more effectively than the simpler Xerxes arrangement.

The problem I see with the way you have attached the thread for your test, is that it short circuits the suspension in an unbalanced way. I would suggest trying it with a much longer piece of thread, connected at the right front corner of the armboard, and at the right rear corner of the plinth, with most of the length suspended over the gap. The way you have it tied in the picture is almost at right angles to the direction of the force you are trying to restrain, which goes from the arm bearing through the stylus. You'll still have the problem that the subchassis will want to rotate around the mounting point of the tether, while the three bar setup stops lateral motion evenly in all directions.

 

[darrylfunk]

Your analysis on the Roksan bearing mounted motor is not correct. It's there to ensure that the tension in the belt on both sides of the motor pulley remains constant, and this stops the motor from being pulled out of the center of its synchronous "pulse", reducing motor noise, and the slingshot effect you get when the belt is tighter on one side of the motor than the other. The Roksan motor is much lower torque than the LP12, but the main bearing is vastly lower in friction than the Linn while the rotational inertia of the platter is far higher. Almost all of the motor's power goes to overcome stylus friction rather than main bearing friction.

Also, I don't use a Xerxes, I have a TMS which isolates any remaining motor noise far more effectively than the simpler Xerxes arrangement.

The problem I see with the way you have attached the thread for your test, is that it short circuits the suspension in an unbalanced way. I would suggest trying it with a much longer piece of thread, connected at the right front corner of the armboard, and at the right rear corner of the plinth, with most of the length suspended over the gap. The way you have it tied in the picture is almost at right angles to the direction of the force you are trying to restrain, which goes from the arm bearing through the stylus. You'll still have the problem that the subchassis will want to rotate around the mounting point of the tether, while the three bar setup stops lateral motion evenly in all directions.

hi mike,
unfortunately the vast majority on here deny that the xerxes works like this and even though i asked them to ask touraj directly this hasn't seemingly been followed up.

plus the discussion will probably be best served by sticking purely to the lp12 and not the xerxes or technics or the well tempered stuff.....

hope you are well to.
is sound by singer still around?

 

[sonddek]

I did indeed ask Touraj. I did not think the reply in any way contradicted what I have said above. If we accept that stylus drag varies, then constant tension in the belt is not capable of maintaining constant speed. The tension in the belt needs to vary rapidly in direct proportion to variable drag. Hence constant tension in the Roksan belt means that any variations in stylus drag are being stored in the motor spring and released resonantly. This is A-level physics. The Roksan motor spring has the effect of shifting higher frequency speed fluctuations down to the subsonic domain, assuming that is the natural frequency of the spring system. Consequently, just like a Sondek, it has the effect of sacrificing pace and tune for timbre.

The photograph of attached thread is of the near right corner of the armboard, just as you have suggested. I have also experimented with the rear right corner, as you have suggested.

A longer span of thread would be stretchier, more elastic and more resonant. The purpose of the thread is to stop horizontal movement of the sub-chassis, and it does this best when it is short and taut.

The primary delta of motor force is along the part of the belt before it reaches the motor. Therefore the first force that any thread needs to obstruct is motion towards the motor by a point on the subchassis to the left of the bearing. There are two suspension springs, arm cable and the majority of mass to the East of that point, so a clockwise yaw of the whole subchassis and anterior-posterior motion of the West suspension spring would be indicated as the largest motions. An ideal inelastic thread would attach the sub-chassis straight to the plinth from a point between the bearing and the West suspension column. Unfortunately, casual efforts imply use of elastic thread attached to the armboard. It's still an interesting experiment, so give it a try.

 

[darrylfunk]

If we accept that stylus drag varies, then constant tension in the belt is not capable of maintaining constant speed. The tension in the belt needs to vary rapidly in direct proportion to variable drag. Hence constant tension in the Roksan belt means that any variations in stylus drag are being stored in the motor spring and released resonantly. This is A-level physics.

it depends on what you believe is going on in the roksan system compared to the linn system.
i feel you are misunderstanding the roksan system and yes it is basic physics.

 

[sq225917]

I like the idea of something that is 'vastly lower friction than the LP12 bearing' exactly how have you measured that if the platters are grossly different weights? Surely the Roksan would spin for much longer accordingly. The largest factor on the bearing of either deck is the viscosity of the oil and the platter weight. Both bearings are around -70db which would imply a similar contact noise and therefore precision.

There has to be uneven tension on the belt either side of the platter simply because without it the system would be in equilibrium and the platter would never start up nor turn. It probably does a great job at reducing the effects of cogging and belt stiffness irregularities however.

 

[sonddek]

it depends on what you believe is going on in the roksan system compared to the linn system.
i feel you are misunderstanding the roksan system and yes it is basic physics.

I'm getting deja vu here. We've been over this. The Roksan motor spring stores energy supplied by the motor. It's a damper. Scraping your nail on the platter causes the motor to rotate until the force in the spring balances the increased drag. When the drag is removed the spring returns the stored energy. The spring is there to allow the platter to accelerate and decelerate gracefully. There are no voodoo elves inside Roksan decks.

 

[darrylfunk]

is the constant frictional drag of the system playing a record a larger force than that needed to slow the motor by the 'varying drag' that you mention?

 

[sonddek]

is the constant frictional drag of the system playing a record a larger force than that needed to slow the motor by the 'varying drag' that you mention?

I have no idea. It doesn't matter. The job of the motor is to render variations in speed, whatever their origin, so small that they are inaudible. The oscillatory effect of oscillating stylus drag may be underestimated. Placing resonant springy reservoirs between motor power and its intended destination serves to lower the frequency of speed oscillations, but it mightn't make them less musically destructive. If critically damped, power damping springs may sound preferable to hunting or prolonged excursion from correct speed, but that's no substitute for rapid recovery of correct speed (e.g. Technics). A motor unit ought to be powerful enough to ride rough-shod over the stylus's efforts to change record speed. Feeble motors with damped energy transfer aren't a great way of avoiding low frequency wow, even if they are good for suppression of motor noise.

 

[darrylfunk]

please don't confuse platter rotation and motor rotation.....you seem to believe that the roksan motor moves backwards and forwards about it's bearing when playing a record but it doesn't.....never mind if it moves when you poke the platter with a finger to slow down the platter.

i think you are getting steady state and constants and variables muddled up.

but back to the lp12......

 

[sq225917]

Is the Roksan really anything more than a light motor heavy flywheel deck with an interesting physical clutch arrangement for the motor?

 

[darrylfunk]

yes, if you analyse the design.

i think it's pretty elegant really.