Modified mechanical anti-skate

Re: Modified mechanical anti-skate

Postby Golear » Fri Mar 15, 2019 10:19 pm

Unfortunately, Fermat's Last Theorem has already been proven, so I don't think I can suggest anything as your next project..... :-)
Golear
Senior Member
 
Posts: 240
Joined: Wed Nov 30, 2016 4:51 pm

Re: Modified mechanical anti-skate

Postby Votan » Sun Mar 17, 2019 4:15 am

Golear wrote:Some ideas I might explore in the future:
1. Cut the length of the upper arm, so that it's just enough for one O ring. Reducing the mass of the upper arm will move the Center of Gravity closer to the axis of the lower arm.
2. Replace the upper arm with a bolt which is screwed into the hub. It might be better to avoid the O rings entirely.

But there is the above idea of yours stated in Jul 05, 2018 in this topic, waiting to be explored! :)
So (theoretically) cutting 11 mm from the upper pole, leaving enough length only for 1 donut to fix fishing line, we have the modified torque (of truncated bare poles plus 1 rubber donut) accordingly: T1=14sinΦ+3,38cosΦ, all the rest remaining the same. Making the calculus for Φ from 0° up to 90° we have the Table II below.
Truncated upper pole of a-s mechanism.jpg
Truncated upper pole of a-s mechanism.jpg (65.51 KiB) Viewed 426 times

Table 1. Anti-skating in relation to angle Φ.jpg
Table 1. Anti-skating in relation to angle Φ.jpg (59.42 KiB) Viewed 426 times

Comparing Table II to Table I (where the upper pole isn’t truncated) we can see that now we have a much more “linear” increase of a/s toque (from VPI a/s mechanism ONLY) as stylus approach lead out grooves (especially if we start from 10° in the lead in grooves).
So Golear’s insight proved to be prophetic!
But taking into account that always exists as well the additional a/s of the lemo wire (even it’s in its more relaxed state), who knows exactly if this much more "linear" increase of a/s with truncated pole is better suited to the relevant increase of the skating force (with more or less unknown growth rate), compared to that of a not truncated a/s mechanism?
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.
User avatar
Votan
Member
 
Posts: 72
Joined: Sun Dec 02, 2018 4:59 pm
Location: Greece

Re: Modified mechanical anti-skate

Postby Golear » Tue Mar 19, 2019 8:11 pm

Yes, indeed!

For convenience, I used the rubber O ring to secure the thread on the upper arm. There might be another option: file a notch on the upper arm, close to the hub and secure the thread via the notch. This will further reduce the mass on the upper arm. The less mass there is on the upper arm, the better. And the closer the thread is to the hub, the better.

Thanks for all your work crunching the numbers. I didn't go by "insight", but by looking at the line along which we know center of gravity will lie, and then seeing how that changes as the anti-skate mechanism rotates. And then trying it out and having a listen. One can't argue with your numbers, though.

And you might want to investigate replacing the rubber O rings and going with a small metal block, as I did. I found a big difference there. But you'd have to use your ears for that.

And as Mr Putty suggested, there will no doubt be differences based on the type of thread.
Golear
Senior Member
 
Posts: 240
Joined: Wed Nov 30, 2016 4:51 pm

Re: Modified mechanical anti-skate

Postby Votan » Wed Mar 20, 2019 4:02 pm

Golear,
“Insight” was a poetic abbreviation of what you have observed by carefully watching (and hearing) the a/s vs. skating force counteracting through arm’s movement from lead in grooves to lead out ones.
Btw, how many grams is the weight of your small metal block, as well what is its CofG distance from the outer end of the pole?

Finally a clarification concerning the “crunched numbers” in the tables of my previous posts: Those are representing the torque T created from the a/s mechanism with reference to the pivot of its hub (for each Φ angle) and not the anti-skating torque Ta/s that is applied in the tonearm. For my 3DR-10’’ the Ta/s should be:
Ta/s=3,76xT, for the lead in groove, linearly increasing up to Ta/s=4,08xT, for the lead out groove, where T: the torque from the tables (known for each Φ).


{Calculations:
Ta/s=Fxd, where:
d: the each time distance of the arm’s pivot from the fishing line (perpendicular to the f.l.). In my 3DR-10’’ I measured it as d=23,5mm up to 25,5mm linearly changing from the lead in groove up to lead out groove respectively.
F: the force (for each Φ angle) that the a/s mechanism’s hub applies to the fishing line (supposing for simplicity that fishing line remains always into the a/s mechanism’s plane)
F=T/r, where T: the torque from the tables (known for each Φ and for each different loading of a/s mechanism), r: the hub’s radius, r=6,25mm
Finally, Ta/s=Tx(d/r)= Tx(23,5/6,25) =>Ta/s=3,76xT for lead in groove, Ta/s=Tx(25,5/6,25)=> Ta/s=4,08xT for lead out groove.}
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.
User avatar
Votan
Member
 
Posts: 72
Joined: Sun Dec 02, 2018 4:59 pm
Location: Greece

Re: Modified mechanical anti-skate

Postby Golear » Thu Mar 21, 2019 1:06 am

Sorry, I can't weigh the metal block. But it would be a tiny bit heavier than two rubber O rings and 3 brass washers. And you might have to look at the photos for the distance.

The difference was, I believe, in the way vibrations are handled by the mechanism. By screwing the metal weight securely on the lower arm, more details seemed to come through. But the O rings might be better in other situations, so this is something to try out.

(My turntable is packed away right now. Moving house.)
Golear
Senior Member
 
Posts: 240
Joined: Wed Nov 30, 2016 4:51 pm

Re: Modified mechanical anti-skate

Postby Johnny » Thu Mar 21, 2019 11:11 am

Can someone post in laymen terms what all these calculations prove with respect to the proper use of the mechanical antiskate.
Johnny
Senior Member
 
Posts: 533
Joined: Mon Sep 29, 2014 7:26 am

Re: Modified mechanical anti-skate

Postby Brf » Thu Mar 21, 2019 12:57 pm

Johnny wrote:Can someone post in laymen terms what all these calculations prove with respect to the proper use of the mechanical antiskate.

First my apologies to Votan, as I have not reviewed his calculations as he has obviously put in a lot of work and effort and thanks to Golear for advancing the subject. However, I think that everyone is over complicating the issue somewhat.

The anti-skate’s rotational torque is a result of the force of gravity acting on the anti-skate assembly’s center of mass in relation to the axis of rotation. The a/s assembly with two poles, which are 90 degrees opposed, will have a centre of mass located horizontally equal distance between the poles and vertically approximately and midway the length. If you let the a/s device swing frees around its axis, the assembly will look like an inverted “V” with the a/s center of mass directly under the rotational axis. With the centre of mass directly under the axis of rotation, no torque is produced.

When you add mass to “one” of the poles, all you are doing is shifting the A/S center of mass towards the direction of the added mass. Again, you can test very easily. Add 2 donuts to one pole and let the a/s swing freely around its axis of rotation. You will now notice that the 2 poles are no longer equal distance from an imaginary line drawn below the axis of rotation. An extreme example would be to add a lot of mass to one pole, the pole will then shift to the point in which it would be directly under the axis of rotation (6o’clock position with the other arm in the 9o’clock position).

When an external force is applied to the a/s assembly (pulling of the fishing line), it will be resisted by the rotational torque caused by the a/s centre of mass being shifted away from being perpendicular to the force of gravity (under the axis of rotation). If the external force is greater than the force of gravity on the a/s assembly’s center of mass the a/s will begin to rotate. As the a/s rotates, its center of mass moves farther away from the horizontal axis of rotation which results in more resistive torque. The maximum resistive a/s torque is achieve when the a/s center of mass is the farthest distance from the horizontal axis of rotation i.e. parallel and inline with the axis of rotation.

Conclusions:

The a/s center of mass is constant, the force of gravity is constant, and the only variable is the horizontal distance of the center of mass from the axis of rotation. Since the center of mass travels in an arc, the increase rotation torque increases linearly until the center of mass is parallel to the axis of rotation (6o’clock to the 9o’clock position) Beyond parallel, (9o’clock to the 12o’clock position) the rotational torque starts to decrease. After the 12o’clock position the a/s flips over.

Added or changing the mass on the poles only changes the assembly’s center of mass. As the center of mass travels around the pivot, the assembly’s center of mass determines the arc of travel. The larger the arc, the more resistive force.

The location of the fishing line to the upper pole only serves to provide a mechanical advantage of the external force on the assembly’s rotational torque. This is useful for fine-tuning.
User avatar
Brf
Site Admin
 
Posts: 3941
Joined: Tue Aug 26, 2014 9:23 am

Re: Modified mechanical anti-skate

Postby Votan » Thu Mar 21, 2019 1:38 pm

Brf thanks a lot, no need for apologies; as well it’s always a pleasure to peruse your posts.
All these calculations of mine (whose criticism is not only welcomed, but also desirable) just try to approach with tangible numbers in charts what you scientifically just explained. So in laymen terms, IMHO they prove:
1. That a/s force (torque) provided to the arm from the a/s mechanism as it is, is not continuously increasing during the movement of the lower pole from 06:00 o’clock up to 09:00; o’clock. I take 06:00 o'clock as a start (though a/s starts increasing earlier as Brf described) because, at list in my setup, a/s mechanism is restricted from the junction box to start earlier (say from 05:00' o'clock).
2. That depending on the loading of a/s mechanism as it is, (number of rubber O rings and/or brass washers, etc.) a/s force starts increasing from 06:00’ o’clock and tops somewhere between 07:30’ ~ 08:00’ (depending on the loading, the higher the loading, the higher the top) and then it goes degreasing.
3. That the above mentioned increase of the a/s force is not linear, but its gradient is decreasing until it tops in the points above mentioned.
4. That if you make your higher pole truncated (Golear’s idea) you can move the a/s curve’s toping up to 08:30 ~ 09:00’ o’clock of the lower pole.
5. That if you make your higher pole truncated, then you must load your lower pole a lot more in order to achieve the approximately similar a/s force to the not truncated one.
6.. That if a not truncated a/s mechanism is engaged such a way that its lower pole is in 06:00’ ~ 06:15’o’clock the moment that the arm tracks in the lead in grooves, you can give to the arm a continuously increasing a/s force up to approximately 07:30 ~ 08:30’ o’clock (of the lower pole) depending on its loading (the more loading, the higher the top). If you do that, then because most LP records have the lead out grooves before lower pole reach 07:30’ o’clock, you are in the increasing area of a/s all the time.
7. As well, to give an idea of the magnitude of the a/s force (in g.mm) depending on the lower pole’s angle and on its loading.
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.
User avatar
Votan
Member
 
Posts: 72
Joined: Sun Dec 02, 2018 4:59 pm
Location: Greece

Re: Modified mechanical anti-skate

Postby Brf » Thu Mar 21, 2019 3:13 pm

Hi Votan, a couple of comments:

#1 #2 and #4 – we might be saying the same thing. When I use a clock face reference, I am referencing the a/s assembly’s center of mass position, not the actual pole position. As an example, with a two unloaded poles 90 degrees opposed attached to a single axis of rotation, the center of mass is 45 degree from each pole (right in the middle). I this instance, maximum torque is achieved when the center of mass is at it farthest from the axis of rotation. That would position the two poles 45 and 135 degrees to achieve the center of mass at 90 degrees.

#3 The distance used to determine the torque is the distance from the pivot to the force acting on the center of mass but measured perpendicular to the direction of the force. With Fg constant acting on the center of mass and the distance increases linear, as defined by the arc, I am not us how torque cannot be anything but linear.

#5 Truncating the top pole only shifts the center of mass towards the lower pole in the same manner as adding mass to the lower pole.

#6 In my situation, I custom tie the fishing line so that the center of mass begins to move at the lead in groove and its center of mass rotates to increase torques towards the run out groove. A/S is always increasing during playing an album

#7 A/S force is determined by the center of mass of the entire a/s assembly in relations to the axis of rotation. Pole position, mass, angle are just components of the the center of mass.
User avatar
Brf
Site Admin
 
Posts: 3941
Joined: Tue Aug 26, 2014 9:23 am

Re: Modified mechanical anti-skate

Postby Votan » Thu Mar 21, 2019 7:18 pm

Brf wrote:Hi Votan, a couple of comments:

#1 #2 and #4 – we might be saying the same thing. When I use a clock face reference, I am referencing the a/s assembly’s center of mass position, not the actual pole position. As an example, with a two unloaded poles 90 degrees opposed attached to a single axis of rotation, the center of mass is 45 degree from each pole (right in the middle). I this instance, maximum torque is achieved when the center of mass is at it farthest from the axis of rotation. That would position the two poles 45 and 135 degrees to achieve the center of mass at 90 degrees.

#3 The distance used to determine the torque is the distance from the pivot to the force acting on the center of mass but measured perpendicular to the direction of the force. With Fg constant acting on the center of mass and the distance increases linear, as defined by the arc, I am not us how torque cannot be anything but linear.


I’ am afraid that we should go to calculus once more, something that could bother those who watch the topic. The non linearity of the a/s mechanism torque occurs from the calculations and relevant charts and sketches of my previous posts.
In any case, with respect to your comments, I agree that with the two unloaded poles 90° opposed, the center of mass (C of G) is 45° from each pole (right in the middle).
As well, I agree that the each time a/s torque is T=FgxD, where D is the each time distance (the lever arm) from the hub’s pivot to the force Fg, always measured perpendicular to the direction of the Fg (which is vertical).
According to what you mention, in a random position of the a/s mechanism where lower pole has an angle φ from vertical (where 0°<φ<90°), then the a/s mechanism’s C of G should be in an angle φ+45° (see sketch attached below).

A-s torque of bare poles.jpg
A-s torque of bare poles.jpg (38.62 KiB) Viewed 280 times


If R the (constant and known) radius from hub’s pivot to a/s C of G, then distance D=Rxsin(45°+φ), so the a/s mechanism’s torque should always be:
T=FgxD=FgxRxsin(45°+φ)
That means (Fg and R been constant and known) that a/s mechanisms torque T is not at all linear changing, as depended from sin(45°+φ). For example, with bare poles and for the lower pole in 3 discrete positions of 06:00’, 07:30’ and 09:00 o’clock we should have:
For φ=0° (06:00’), T=FgxRx0,71, For φ=45° (07:30’), T=FgxRx1,0, For φ=90° (09:00’), Τ=FgxRx0,71, i.e. not at all linear increasing, but a curve, exactly just as an extreme case from the calculations in my previous posts and charts. Indeed, in the above extreme example, from φ=45° and up, a/s torque starts degreasing (turning point).
Of course, adding load to the lower pole, the turning point of the a/s non linear curve goes higher from φ=45°. For example (looking at the charts of my previous posts) with 4 rubber donuts in the lower pole turning point is above 60°, adding 1 brass washer it goes above 75°. This is a conclusion in full agreement with what Golear noticed by ear in his setup, namely degreasing a/s torque in the inner grooves.
Truncating the higher pole, with 4 rubber donuts turning point goes above 75°, adding 1 brass washer, it goes up to 90° namely all the way increasing but not linear.
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.
User avatar
Votan
Member
 
Posts: 72
Joined: Sun Dec 02, 2018 4:59 pm
Location: Greece

PreviousNext

Return to Support Forum

Who is online

Users browsing this forum: No registered users and 2 guests

x