advertisement


DIY Tonearm

Paul R said:
It's unclear to me what the damping fluid is damping.
Click to expand...

The arm tube is going to resonate driven by the stylus tracing a record. It is terminated at one end by the bearing point and the stylus at the other (which will be nodes). I suspect the theory was that you could dump energy into the damping fluid a bit like the strings on a guitar dump energy into the body of the instrument.


Paul R said:
Or why the parts of the bearing support would move independently.
Click to expand...

I doubt there is much movement, I suspect the bearing cup will rotate on its spindle if you twist it with your fingers, but in normal use it would be stationary as all the arm movement comes from the point sitting in the cup.
 
If it doesn't move, it doesn't damp. Which makes it a bit complex and not necessarily the only way to damp the underside of the unipivot. Perhaps just using a suitable plastic type material to mount the cup of the unipivot on would suffice? An interspersed stack of spring and nylon washers maybe.

If you put an ARO on its side, does the oil fall out?

Paul
 
TPA said:
No Mark, the Aro sKale uses O rings to locate it on the stub so it is free to rotate without altering position. I preferred this to the original metal on metal system.
Click to expand...


That's essentially decoupled twice then as the counterweight stub is already decoupled.
 
You don't want excess mass in the head shell. I think the design principle is always:
For a given effective mass, maximise stiffness.

Using more material, you can increase stiffness, at the expense of effective mass. If you hold effective mass constant in the design, you need to use light stiff materials, especially round the headshell area, as they contribute most to effective mass.

Aluminium or magnesium alloys are a better bet than stainless steel; what matters is the ratio of stiffness (Youngs modulus) to density. This turns out to be the same ratio that sets the speed of sound in the material - so a good materials for head shell and arm tube have high speed of sound.

For the counterweight, you want high density, so you can get the weight as close to the pivot as possible. It turns out to be better to use a big weight at short distance, this minimises the contribution to effective mass, and leaves more to invest in the rest of the structure.

Densities for counterweight materials:
Brass: 8.5
Bronze: 7.5 - 9 (depending on tin content)
Lead: 11.3
Steel: 7.5 - 8

And light materials:

Aluminium alloys: 2.6 -2.7
Magnesium Alloys: 1.7-1.9
Carbon fibre composite: 1.6 (ish)
 
PD - you missed out tungsten, density twice that of lead and coincidentally almost identical to that of gold, which is why some crooks sell gold-painted tungsten ingots to mugs.

(Looks up - density 19.3 g/ml)

I must say though - I wouldn't fancy machining tungsten. Rega/Origin's suppliers do it OK but I bet it's a lathe operator's nightmare. That and titanium would be a delightful morning's work, featuring a whole range of damaged tools and equipment.
 
winchman said:
Once you have a block of it its not difficult to grind it using the correct wheel, that's how the sharpen tungsten cutting tools
Click to expand...

Sure, and you can drill it using ultrasonic techniques. But these are expensive ways to fabricate a shape; I guess the question is if the saving in effective mass is worth the extra expense.

Here is a simple argument:

We start off with a symmetrical arm, with a 9" (or 12") counterweight stub, and a mass equal to cartridge and headshell at the end of it. This equals the effective mass contribution of the front side of the arm, and doubles the total. Now we halve the stub length, and double the mass at the end - this still balances. But because the effective mass goes like the square of the distance from the pivot, the doubling of counterweight contribution is more than overcome by halving the distance, so the back end is now only contributing half as much as the front. If we make the stub length (to the middle of the counterweight), a tenth of the arm length, then the backside contribution is only a tenth of the front, and we are into the law of diminishing returns. The counterweight needs to weigh about 10 times cartridge and headshell (plus a bit for the arm tube); this might be 100g or more centred at about 2 and a bit cm from the pivot. Using something like lead (or brass or steel), this is maybe 10cc of stuff (maybe 12cc for brass or steel), This could be a cylinder an inch in diameter and an inch long (roughly). This example could be pushed, with a bigger weight closer to the pivot; this make take you from a 10% effective mass contribution to 5%, leaving a bit more mass budget to use on rigidity in say the head shell area.

These numbers are just hand waving for examples; you need to do the calculations properly using integrals as the object sizes are comparable to the distance from the pivot.

Using a higher density material allows a bigger weight to be placed closer to the pivot but you can see it only makes a small difference to effective mass; you need to work the numbers properly for a detailed design.

I guess the other issue to bear in mind is that very heavy weights load the bearing hard; at some point this loading will cause trouble.
 
I think you can get close enough treating the headshell, cart, arm tube and counter-weight as point masses for balance and as similarly for the moment of inertia, which gives you the effective mass.

So roughly, if Mh is headshell mass, Mt is tube mass, Mc is counter-weight mass, L is the tube length and D is the distance from pivot to counterweight cg,

MhL + MtL/2 = McD (static balance)

and

Ia = MhL^2 + MtL^2/3 + McD^2 (inertia about the pivot)

effective mass of the arm is then

em = Ia/L^2

give or take. And factoring in a cartridge weight for the balance equation would be sensible.

Paul
 
PigletsDad said:
You don't want excess mass in the head shell. I think the design principle is always:
For a given effective mass, maximise stiffness.
Click to expand...

Does that make sense though? If you look at the forces a tonearm needs to deal with due to the movement of the stylus tracing the groove on the record, they are tiny, so why does everything need to be so stiff?

What I think really does matter in a tonearm design is how it resonates and that will be a function of more factors than just stiffness.
 
O/k haw about a sand filled tone are like the Well Tempered and a dangley counter weight like the Roksans?

Pete
 
Paul R said:
I think you can get close enough treating the headshell, cart, arm tube and counter-weight as point masses for balance and as similarly for the moment of inertia, which gives you the effective mass.
Click to expand...

Effective mass is a misleading name, as you correctly identify, it is the moment of inertia of the arm that matters. You also need to consider the vertical and horizontal planes - Pierre Lurne suggests (in "On hi-fi, neutrality and the 'pure mass'" I posted earlier) it is desirable to have the same M. of I. in both planes, though I am not that convinced.
 
John Channing said:
Does that make sense though? If you look at the forces a tonearm needs to deal with due to the movement of the stylus tracing the groove on the record, they are tiny, so why does everything need to be so stiff?

What I think really does matter in a tonearm design is how it resonates and that will be a function of more factors than just stiffness.
Click to expand...

Sure.

What you need to think about is the mechanical impedance that cartridge sees. Away from the frequency of the resonances, this is more or less given by the stiffness. At the resonances, it roughly the stiffness divided by the Q of the resonance, which can easily reduce it so it becomes comparable to the cartridge compliance, colouring the sound.

If you raise the rigidity of the structure as a whole, the impedance minima at the resonances are raised as well, reducing their effect.

Obviously, structural damping helps a lot. Carbon composite is quite good, while aluminium is fairly high Q.
 
Pete MB&D said:
O/k haw about a sand filled tone are like the Well Tempered and a dangley counter weight like the Roksans?
Click to expand...

Filling the arm tube with sand will change its resonant characteristics, I think it is hard to predict whether that will be good or bad. Others, like Audio Origami use foam filling. Funk Firm also using some damping material. If you look at the Aro design, the headshell and Arm cup both have sections of greater diameter tubing which surround the main arm tube. This will also have some (constrained layer?) damping effect.

The Roksan counter weight design means that there is a point contact between the weight and counter weight stub, rather than a large contact area. If you were modelling the resonant behaviour of the counter weight stub, I assume this would make it a bit easier as you have a well defined node. Naim uses a rubber bush to try to decouple the counter weight stub from bearing cup and therefore the main arm tube. It is hard to predict whether this has a positive effect, I assume they did it by trial and error.
 
Interesting as this all is guys are we in danger of over complicating things. The ARO was a well regarded design and we started off by thinking about cloning it and if possible addressing it's shortfalls. As far as I can tell from reviews the shortfalls were in deep bass and maybe high frequencies so maybe we should determine what we think contributes to these and then see how we might address them
 
PigletsDad said:
What you need to think about is the mechanical impedance that cartridge sees.
Click to expand...

As someone else pointed out earlier, the best way to do that is to ensure the headshell material is the same as the cartridge body. Every time you join bits of material together with different impedance you will get an impedance mismatch which may or may not be useful. E.g. an aluminium alloy headshell and a carbon fibre arm tube will have an impedance mismatch. This is why SME and Rega, for example, favour a one piece casting.


PigletsDad said:
Obviously, structural damping helps a lot. Carbon composite is quite good, while aluminium is fairly high Q.
Click to expand...

Agreed, not many really good carbon fibre arms out there though, I wonder why?
 
timH said:
Interesting as this all is guys are we in danger of over complicating things. The ARO was a well regarded design and we started off by thinking about cloning it and if possible addressing it's shortfalls. As far as I can tell from reviews the shortfalls were in deep bass and maybe high frequencies so maybe we should determine what we think contributes to these and then see how we might address them
Click to expand...

I think it is hard to address the short falls without a reasonable understanding of the mechanics.

My guess would be that the shortfall in the deep bass is due to the instability of the unipivot compared with a gimball arm. Even though the counterweight is hanging well below the pivot point, it can still swing like a pendulum. Some small movements when trying to track bass notes may explain the softness. If you look at the Graham Phantom arm, it has magnetic dampers specifically to prevent this.

As for the high frequencies, my guess is that is due to what is below the bearing cup - the pseudo-bearing which is oil damped. I reckon this might be what makes the arm sweet sounding compared with un-damped arms. Change that and I would bet you change the high frequency performance.

Just guesses at the moment though :)
 
"My guess would be that the shortfall in the deep bass is due to the instability of the unipivot compared with a gimball arm"
This ties in with a lot of the reviews of the ARO.

"As for the high frequencies, my guess is that is due to what is below the bearing cup" This makes a lot of sense but is the damped bearing trying to correct something inherent in the arm design or is it trying to isolate the arm from any crud coming from the motor-belt-platter-bearing-armboard path?
 
You must log in or register to reply here.


advertisement


Back
Top