The science of anti-skating.

Artie

Super Member
As I've been looking at some modern, (budget), tables, I see more and more eliminating the anti-skating device, or making it dubiously "automatic". This got me to thinking of what it might take to design one from the ground up. Problem is, everything I "Google" either talks about what it is, or how to set it up on a particular table. Nothing about its engineering. The two main questions I have are:

1. Should the anti-skating force stay uniform across the full sweep of the record, or be stronger at the outside groove relative to the inside groove? Or vice-versa.

2. Does the speed, (33/45/78), affect the amount of force?

Just curious if anyone has any insight to this. Designing one's own anti-skate mechanism could be a fun project.
Artie
 
Anti skate is a complex problem. As you may have surmised, it will be stronger at the outer grooves than at the inner ones (higher groove velocity) and higher RPM calls for higher antiskating. It is a function of stylus geometry (conical stylus requires the least, line contact the highest. And it is modulation dependent. I imagine it may vary slightly with different vinyl formulations. And it probably is frequency dependent too.. IMHO the best way to set it would be with a test record with high level modulation levels. And a scope to look for mistracking. All that being said, my own practice is to keep it on the low side and to make sure tracking force is high enough so no mistracking occurs. Good luck with your research.
 
Thanks Steve. I had an old Thorens once that used the opposing force of a couple of magnets to create a non-contact, progressive, (regressive?), anti-skating that was quite ingenious. Wish I had that table back. Can't remember the model number. TD-160 maybe?
 
and to really screw the pooch. study the moment of motion (friction), moment of inertia (torque is adequate here). The sudden loss of "mass" at MOM has to be exactly offset by a sudden inertia ( the universe hates a void ) Lower frequencies have a higher friction coefficient, higher frequencies have a lower coefficient. The swinging of energy forces between friction and inertias cause all kinds of "understanding issues" and the pictorial of "skating effect" showing force vectors and the like, they even add a non existent force vector line to solve their own riddle.
 
Hmmm. I'm thinking that step 1 would be to acquire a really, really good test record. :idea:

(And maybe one of those "blank" records.)
 
This is why I use a linear tracking tone arm. It makes all the difference in performance and guarantees great performance on the inner groves.
 
There a lot of complex interactions involved in playing a LP. Most of your better tonearms do supply a varying amount of antiskate via the eccentric action of the antiskate adjustment bar, the monofilament and weight.

The HiFinews test record offers a number of obstacle type tracks for dialing in antiskate as well as three tracks at the second to highest velocity at three different areas of the record. I have posted in the McIntosh forum, there is a connection in this forum, to screen shots of the misstracking and correct tracking of three of these tracks.
 
I'm trying to understand this a little better as well.

The engineering of different arms manages the forces involved in different ways. My SME 3009 S2 gives me only limited ability to adjust "anti-skate" and I have ended up adjusting the vertical tracking force between two different weights to optimise anti-skating. With this arm the differential in force required between inner and outer grooves is regulated by increasing and decreasing lever length. The changes in force required to compensate for higher velocity on the outer grooves are adjusted through a cam-type action on my Formula 4 but with this arm the actual mass of the anti-skate is adjusted through pellets.

My favourite system is that on a top of the line Kenwood table from the late 70's a L07d, which has very fine adjustments available across a big range. I don't own this table but it's superior nature is really clear.

I use a hifi news test record to optimise anti-skate. It requires a lot of patience because settings are often not continuously variable. I've found the best approach for me is to try a new setting and then make a series of adjustments to accommodate that change before testing it. If it's not an improvement I return to the earlier configuration and leave it be for a while before working in the opposite direction.

For me it seems that the longer I work at an issue continuously with a turntable the more likely it is I make a mistake or make things worse rather than better.

Ironbark
 
The Shure Audio Obstacle Course record has increasing levels of bells that are good for setting anti-skate. Adjust until mistracking occurs on both left and right a the same level. Mistracking damages grooves, so don't go any further than necessary and as little as possible. IMO, experience with anti-skate is hard-won and there's little design info freely available. Fortunately it's an approximation at best, so many imperfect schemes work well enough to get by.
 
Some do argue with this method, but for me it takes the guessing out of the situation, and sets a great perfect balance. it takes the grooves out of the record to show the pull direction, and lets you adjust how the needle is behaving when its on a record groove. Put it on , watch the arm move to one direction, adjust, try it on the outside, center and inside , get a good compromise , finished. this one cost $18.
 

Attachments

  • 20170510_102457.jpg
    20170510_102457.jpg
    39.7 KB · Views: 63
  • 20170510_102731.jpg
    20170510_102731.jpg
    24 KB · Views: 61
Hi,

I have an SME M2-9R Arm fitted to a Thorens TD150 MkII. The anti skate is done the same way as a 3009. The slot for the Grams weight at the tip is picked and then the pulley for the weight is positioned at 90 deg to the rod on the top of the arm when the arm is at the outside edge of the platter.. As the arm progresses to the centre of the record, the cord for the anti skate then pulls at an angle so the force is reduced. Things are still a compromise but I have a 12" single which has a complete blank B Side. No matter where I place the arm on the record it stays stationary. I think that is good enough for me!!! As long as the cord is set at 90 deg at the outside edge of the platter I don't think there can be much improvement as there are too many variables.

Cheers

GPS16
 
It seems to me that this should be a basically simple, 2-stage mechanism. Part one would be some method of determining absolute tonearm position at any given moment. Could be magnet with Hall-affect device, optical sensor with stepped vanes, or maybe a synchro-resolver. Part two could simply be a small electro-magnet coil, playing against a fixed magnet, so that a slight force could be varied, linearly, with any given tonearm position. A simple little Picaxe controller could be the "brains" of the whole thing.

The "trick" is, knowing how much force, and which direction, when. :idea:
 
... As you may have surmised, it will be stronger at the outer grooves than at the inner ones (higher groove velocity)...

No, you have it completely and utterly back to front.

The skating (hence the antiskating force required to compensate) force increases towards the centre of the record. The stylus is always pulled towards the centre and the antiskate spring/magnet compensation pushes the opposite way.

The main contributors to the entire skating force is the disparity in the groove wall curve/length on each side of the stylus and the tonearm pivot point causing the stylus to 'pull' towards the spindle. All other forces are insignificant. Think, if you reversed the rotation, the arm would push out and off the edge of the record wouldn't it?

The spring systems apply little if any force for the first third of a record and rapidly increase the lateral outer force as the arm gets closer to the centre.

Have you ever balanced a tone-arm? Neutral balance the arm horizontally, zero the antiskate. Move the arm half way across the record, wind on a bit of antiskate and tell me which way the arm moves...
 
Last edited:
Have you ever balanced a tone-arm? Neutral balance the arm horizontally, zero the antiskate. Move the arm half way across the record, wind on a bit of antiskate and tell me which way the arm moves...

That's a good point. I remember doing that back in my "tech" days. It moves outward.
 
In answer to the OPs questions:

1: The antiskating force is not uniform across the record. It increases towards the centre.
2: The faster speeds create more inward force (bump the arm with a 45rpm single vs a 33rpm and see the difference with A/S off)
 
It's definitely not linear.

One could measure a few TTs using floating horizontal neutral balance and a jeweller's scale standing up on the platter mat with the headshell pushing on the weight stage at various points across the arm travel. Take down those points and plot a curve.

Trouble is, TT's anti-skate adjustments are pretty inaccurate and vary considerably in their quality and implementation. And many vintage ones are gummed up with grease, foam damping in the springs has festered and some springs are stretched. The graduated stepped bar/string and weight A/S as used on Empire arms is one of the few that holds up over the years I would say.
 
I would think a completely electronic anti-skate could be designed. But would you really want to?

Along the lines of the magnetic Thorens style, except using adjustable electromagnetic force antiskate system programmed and derived from a 'force map' -essentially multipoint look-up tables specific to the particular arm/TT combination.

The values could only be derived by experimentation with the arm/turntable combination you actually used. It would need a test record with an entire side of test tone, perfectly pressed and one would play the record into the capture mode. The system would then analyse the increases in groove wall distortion and create a specific elctromagnetic anti-skate force map to completely offset that distortion and/or cancel it, from the outer edge to the centre.

The system would need capture modes for each speed I would think 33/45/78. You have to re-map each time you changed a cartridge or changed your VTF. It'd be a nightmare IMO.
 
Back
Top Bottom