Resonant frequency and the springiness of sprung turntables

[Seanm]

I'd always thought that cart/arm matching was just a matter of compliance and effective mass. There's an interesting thread running at the moment started by mark-hants about his Thorens 150, on which Torstoi has related the advice of a Thorens expert, suggesting that the springs are also a factor. Any thoughts?

 

[YNWOAN]

Well two things spring (compliance joke) to mind; 1/ the suspension should be acting outside of the arm and cart loop and, 2/ the TT suspension should be at a frequency sufficiently low as to not interact with the arm mass/cart compliance.

 

[Darth Vader]

Takes me back to Uni and coupled oscillators. So I guess in TT design you need to pay attention to decoupling each oscillator. My own TT does not have suspension springs so thas one way of removing the coupling.

Cheers,

DV

 

[YNWOAN]

In an absolute sense some compliance exists irrespective.

 

[Seanm]

So I guess the problem with putting Linn springs on a Thorens (the proposal being discussed in the original thread) would be that the lighter mass of the Thorens platters would cause the suspension to resonate in such a way that it interacts with the other oscillator? Or am I mixing up the two issues mentioned by YNOAN and DV?

 

[YNWOAN]

I've not seen the origonal post you refer t but you cannot simply juxtapose springs simply because they fit. The resonant frequency of the suspension is actually a combination of the Spring rate/frequency and the mass it/they are supporting. Having said that, the arm/cart compliance should exist outside of that of the suspension.

 

[mark-hants]

The technical parts of this are above my understanding atm though I am learning plenty.

The platter on the 150 is very heavy. Is the Linn one really heavier?

I suspect that had I removed the platter while the deck was in storage the springs would have fared better.

 

[martin clark]

you cannot simply juxtapose springs simply because they fit. The resonant frequency of the suspension is actually a combination of the Spring rate/frequency and the mass it/they are supporting.

True, but there's an interesting outcome that's not widely appreciated - the key is displacement under load: regardless of load or springs, the total displacement from unloaded to loaded is the direct measure of frequency.

If you run the numbers, you find that a given total deflection of [a spring loaded with a given mass] is just another way of expressing frequency: it's a ratiometric relationship invariant with total amount of mass. Counterintuitive perhaps, but true. Equally then - if spring type A and spring type B used under a given deck result in about the same total deflection under load - the resonant frequency and the spring rates are pretty much identical.

For small differences in deflections between spring 'types' (let's say, 3mm or so, out of perhaps 10-25mm total travel) - then the effect on F[sub]resonance[/sub] is very, very small - certainly at the low frequencies typical in this use. 2-3mm difference in height by simply swapping springs will only affect things by a few tenths of 1Hz. Strange but true.

And yes, considering coupled oscillators analytically quickly becomes a mess ..!


[I don't have a turntable, this is simple an argument from basic mass-on-springs physics ]

 

[Michael 1]

In principle there are 4 frequencies which have to correspond to each other:
- 16 Hz lowest recorded frequ. cut on vinyl
- 12 Hz (my recommendation) tonearm resonant frequency depending on compliance of the cartridge
- 5-8 Hz sub-chassis swinging frequency, as stiffer the springs are, as higher the frequency
- 2-3 Hz foot fall sound
In this way you avoid oscillation.

The sub-chassis swing frequency you can simply count per second, with the watch in you hand. (Or more precisly with a video)

The main problem of sub chassis are uneven loaded springs. e.g. a heavier tonearm will squeese the sping on the right side more. Adjusting the other springs with the screw nut is not a solution as it changes the spring rate to a higher frequency.
In such a case the platter will never move up and down like a cylinder - it will wobble, moving also the arm and cartridge sidewards.

The optimum is reached when all 3 springs have exactly the same load and the screw nut is at the same position for all 3 springs - while the platter is even without any adjustment. Typically this is never exactly the case, also with the original weighted tonearm.

A lengthy procedure helps:
- Dismount the subchassis
- Put the subchassis including tonearm, platter vinyl record (all the weight) on 3 springs/mounting points and measure the pressure of each - I used a flat digital kitchen scale
- Add weight (e.g. glueing aluminium blocks) in a way that all 3 springs will have the same load - I makes sense to move the arm 1/3 over the platter, in order to simulate the force under operation.

After reassembling, the platter will swing up and down without sideward movements.
As mentioned above, the swing frequency should not as low as foot fall sound and not as high as the tonearm resonant frequency.

When the springs are stiffer, the frequency may be too high. However, it could again be reduced by adding weight...

In case you do not trust the values provided for effective tonearm mass and cartridge compliance, you can always easily check the real arm resonant frequency: Move a speaker close to the turntable and run a sinwave-generator (free software) via the output of you PC. At the resonant frequency the tonearm starts "jumping".
(Also expensive test records exist)

 

[torstoi]

The problem, or other way round the task he is faced to solve is,
that he has a more heavy arm installed than the original one, but (probably) still has the original springs.

Further he thought of using an MC cart, which would obviously require or suggest the use of a heavier headshell, ie 10 or 12,5 gr to meet the compliance required by most such carts
& it was mentioned that this could possibly make necessary the use of a heavier counterwheight. (not sure it's necessary, but it's a possibility)

So all these changes in the upper store to get some MC cart working well with the arm - as an isolated system first off-
will have effects to the lower store by the wheight applied
& the suspension should of course work perfectly to not disturb overall performance.

As I has probs dealing the set up of suspension of my LP12 with the original and intended arm, the Ekos, I'm surely no expert on setting turntable suspension right.

Contrary to my initial thoughts, the expert who has worked with Thorens stated that the Linn springs have certain problems in use, but these problems occur in a frequency range that doesn't come into play in this particular case.

It is beyond me a bit, but in short that means , obviously Linn springs should be possible to face the added wheight in this application.

It's obvious though, should he aim to go the MC route incl. all necessary measures, a generic set of springs from a Thorens supplier won't do

& also he will need some expert getting his spring settings perfect
or he's going the full distance and learn how to do it himself.

I was not suggesting the springs were playing into the arm-cart compliance harmony,
which I see as an isolated 'system'.
edit: rather said, by principle they could, it's you that has to take care they don't..so the cart-arm compliance remains within it's intended range w/o being disturbed by other frequency influence as far as possible.

Thus, as DV pointed out, an unsuspended deck has it's advantages here, but the Thorens as such is a very nice and classic deck, I'd keep it.

 

[Seanm]

Ah, sorry Torstoi, it seems I misunderstood the advice of the Thorens expert that you kindly translated and related on the other thread (which is here, by the way, in case others are confused/interested: http://www.pinkfishmedia.net/forum/showthread.php?t=181508).

Specifically this bit:

Also the subchassis resonance of about 4 Hz is set higher, which makes the choice of carts a harder.
But as with a SME 3009 due it's low inertia-mass only very soft cartridges are recommended, this is less of a problem.

That's a bit clearer now thanks to Michael's post above.

I've learned a lot from this thread though, thanks all for sharing your expertise.