Golear wrote:The dealer set up the mechanical anti-skate according to VPI guidelines. However, I found that the anti-skate force was less in the inner grooves than in the outer grooves. I could hear it, and see it in the way the loop deformed under tension. That led me to look for a better solution and that led to me starting this discussion. This is not to say that users' set-ups may be wrong. It's possible that the mechanical anti-skate works with some residual anti-skate from the wire and gives good results in the end. But I think in my case, I ran into issues because the dealer did such a good job removing the twist!

I agree nothing is 100%, and the wire may be adding/taking away something. There is the Wally Skater, which can measure the anti-skate across the whole path. But in my case, I can see just by looking at the loop that there's minimal anti-skate in the outer grooves, which rises via a cosine function to a max near the innermost grooves, then drops slightly from max at the very end of the music (90%, I think). It's not perfect, but don't hear any ill effects.

Golear finally is right in his findings! Namely:

In order to ad mechanical a/s in my JMW 3DR-10’’ tonearm, I decided to ponder a bit its behavior, namely doing the math to calculate the way it increases the a/s torque during arm’s moving inwards, comparing 2 alternative weighting of the lower pole, one with 4 rubber donuts, the other with the 4 rubber donuts and 1 brass washer. The whole calculus procedure is listed in the end of the post (in between {}), not to bother those who are interested exclusively for the results, listed below in the Table 1, with reference to the sketch below, as well as to the following symbolisms and explanations:

- VPI Anti-skating mechanism.jpg (42.72 KiB) Viewed 263 times

- Calculations do not take into account (it’s not possible) the extra a/s torque created from lemo wire.

- No Dual Pivot for the time being

Φ: The each time angle of the lower pole relative to the vertical.

T1: The total torque created from both (upper and lower) bare poles (without any donuts on them).

T2: The additional (to T1) torque created from all 4 donuts placed in such a way that the center of gravity of this quadruple is suspended in point F of the lower pole, supposed to be approx. in a 3/4L distance from the pivot.

T’2: The alternative additional (to T1) torque created from all 4 donuts plus 1 brass washer placed as above mentioned.

Ttot: The total Torque creating from the whole VPI mechanical a/s (wearing only 4 donuts) in relation to each time angle Φ.

T’tot: The total Torque creating from the whole VPI mechanical a/s (wearing 4 donuts plus 1 brass washer) in relation to each time angle Φ.

- Table 1. Anti-skating in relation to angle Φ.jpg (59.42 KiB) Viewed 263 times

Conclusions (of mine): Table 1 tells us that with 4 rubber donuts in the lower pole, torque is increasing from Φ=0° up to 60° plus (~63°), but from this point up it starts to degrease! Is this too bad? Maybe not, because if we take into account the lemo wire’s additional a/s torque as well, which supposedly is continuously increasing, then the total sum could be finally (but not so linear) increasing up to 75°. This is adequate I think, because if we start with the lower pole at Φ=5° in the lead in grooves, then most records have their lead out grooves way before lower pole reaches Φ=75°. But if you start with the lower pole at (say) Φ=15° in the lead in grooves, then there should be the problems that Golear detected in his setup, which isn't a good thing at all.

If in the 4 rubber donuts we add 1 brass washer, we can go up to Φ=75° increasingly, so here normaly no problem. But making preliminary tests for a/s in my 3DR-10’’, I found (via “By Eye” and “By Ear” techniques) this 4 rubber donuts plus the 1 brass washer version a bit too much for a/s.

{Calculations:

Let it be (see the above attached sketch):

L: the total length of each one pole, reaching till the pivot of a/s hub.

B1: the weight of each one pole, suspended in L/2 distance of the pivot (middle of the L length where the center of its gravity is), in point D in the upper pole, in point E for the lower pole.

B2: The total weight of (say) 4 rubber donuts (because those are the total that VPI provides), the center of gravity of this quadruple suspended in point F of the lower pole, supposed to be approx. in a 3/4L distance from the pivot.

B’2: As B2 but adding 1 brass washer to the already existing 4 donuts.

b and

a: The distance of points D and E respectively from the vertical passing through the pivot (This is the lever arms of the torque creating respectively from the upper and the lower arms).

c: The distance of point F from the vertical passing through the pivot (This is the lever arm of the torque creating from all 4 donuts in the lower arm).

According to the attached sketch there should be:

DO=L/2, EO=L/2, FO=3L/4, a=(EO)xsinΦ= sinΦxL/2, b=(DO)xcosΦ=cosΦxL/2

T1=B1xb+B1xa=B1x(b+a), =>T1=B1x(cosΦ+sinΦ)x(L/2),

T2= B2xc, =>T2=B2xsinΦx(3L/4)

T’2=B’2xc, =>T’2=B’2xsinΦx(3L/4)

For each cylindrical pole we measure: L=20mm (till the pivot), d=3,44 mm, so Vpole=(πd*2/4)xL=185,8 mm*3

Density of each steel pole is Dpole=7.530 Kg/m*3=7,53x10*(-3) gr/mm*3

Thus B1=VpolexDpole= 185,8x7,53x10-3=>

B1=1,4grThe weight of each rubber donut is Bdonut=0,12gr, so B2=4xBdonut=>

B2=0,48grThe weight of a brass washer is Bwasher=0,82gr, So B’2=B2+Bwasher=>

B’2=1,30gr

Thus, T1=1,4x(20/2)x(cosΦ+sinΦ)=>T1=14x(cosΦ+sinΦ), [gr.mm]

T2=0,48x(3x20/4)xsinΦ =>T2=7,2xsinΦ, [gr.mm]

T’2=1,30x(3x20/4)xsinΦ =>T’2=19,5xsinΦ, [gr.mm]

So:

Ttot= 14x(cosΦ+sinΦ)+7,20xsinΦ, [gr.mm], as well as

T’tot=14x(cosΦ+sinΦ)+ 19,5xsinΦ, [gr.mm]

}
Prime Signature+10”3DR in Finite Element(FE) platform, ADS, Lyra Delos, WW Silver Eclipse 8 Phono Cable, Groove Plus-SRX Phono Stage, Moon 740 Pre, Moon 820 PSU, ATC 100 Active Speakers, AQ Niagara IC, WW Silver Electra 7 A/C, FE Ref Rack, RPG Treatment.