Angling for 90° - tangential pivot tonearms

Surely off topic doubts, and with no direct experience; I was interested in it because Studer was one of the best European constructors, and that geometry is the closest to the radial guide used in #1957 video. However on those models there are not many positive reports: mechanical complexity (resonances, failures), compliance and azimut pitching problems, difficult setup etc.
Just to say that if Studer itself had problems, it is certainly not an easy task.

carlo
 
Last edited:
wooden arm in action

Just saw this video that features a tangential pivot tonearm in action. Sorry for the dizzying camerawork. It looks similar to Doug's work.

pivot-linear-tonearm.jpg
 
The KLaudio arm is not a pure "pivot" tonearm in the spirit of the OP. Yes, it is pivoted, but it also incorporates a linear ball bearing slide.

In his comments, Mike Fremer considers all of us who value the concept of linear tracking to be "obsessed" and "misguided". Why should we care what he says in his comments on linear arms? Aside from commenting on workmanship, he really said nothing of interest.

Ray K
 
I don’t agree with everything Micheal Fremer said especially about linear tracking arms. I am a firm believer of linear tracking arm. However, he does offer some valid points sometimes.

I don’t want to generalize about the view on pivoted tangential tracking arms but want to just refer to Klaudio’s arm. In early part of this thread, I said I don’t like Klaudio's arm because it has too many parts and the construction is too complicated. It seems to me that Micheal Fremer’s listening experience supports my concerns.
 
I've refined this geometry a little more.
1st groove @ R=146.05mm
Null groove @ R=129.94mm
Last groove @ R=60.33mm

Tangency error @ 1st groove = +0.15°
Tangency error @ last groove = -0.07°
Most of the play area is spent between NULL and last groove where error goes from 0° to -0.07°, not quite tangent but pretty close.

The ratio of the 2 "gears" or cams is reverse 1:1.4667.
Center to center distance 50mm.
 

Attachments

  • geo1.PNG
    geo1.PNG
    120.9 KB · Views: 329
  • geo2.PNG
    geo2.PNG
    90.8 KB · Views: 299
  • geo3.PNG
    geo3.PNG
    139.3 KB · Views: 145
  • geo4.PNG
    geo4.PNG
    136.2 KB · Views: 140
  • geo5.PNG
    geo5.PNG
    120.8 KB · Views: 142
Glad to ear you again 2wice.
This new data are nicely surprising for me, so near to those used to build the Rabbit (see the design in post #1719). Graphically I found much bigger errors, if instead are so small it's really a good new.

Rabbit Tonearm data
crank rotates cw by 62.6° on main pivot P -
HP moving pivot rotates ccw by 42.6°
Pulley sizes - ratio = 1:1,470
P main pivot = Ø16,0 mm
HP moving pivot = Ø24,0 mm
flat belt thickness 1mm = ratio 1:1,470 (Ø25/Ø17)
moving mass = under 100 g

Geometry:
pulleys axis distance = 50,0 mm
distance - main pivot P - platter spindle A = 218mm (> 225 mm with eccentric base)
distance - moving pivot HP - C2 = 167mm (eff. length)
0 offset - tangency error < + - 0,5°
0 overhang - error < + - 1mm


Now I do not understand if you're using gears or the previous cams.
And just two, without the idler wheel?

Merry Xmas - carlo
 
Glad to ear you again 2wice.
This new data are nicely surprising for me, so near to those used to build the Rabbit (see the design in post #1719). Graphically I found much bigger errors, if instead are so small it's really a good new.

Now I do not understand if you're using gears or the previous cams.
And just two, without the idler wheel?

Merry Xmas - carlo

Merry xmas Carlo and all.

Not new geometry, just an optimization of #1651

The idlers are still there in the model, I've just removed them for simplicity.
If you look closely, the 2 circles in the drawing do not touch. And the sum of their radii doesn't add up to the length of the carrier arm.

I'm using gear relations in the simulation to approximate cams as the cam relations tend to slip.

I've found a new way to optimize the geometry, using a spreadsheet and a solver.

There are 5 major variables + 1 minor, it iterates those variables to approach the target output tangency.

I think I can tweak the algorithm to get it even closer but it might mean losing the nice rounded dimensions it has now.
 
Clear, thanks.

Playing with Linkage simulator i found something a bit better than the original, used for * 3d Rabbit #1950. Same ratio, but the 2 pivots distance increased to 54,5 mm. You may find measures on that attachment.

ciao carlo
*...some day i'll print it. Have you found how to get those cams?
 
Sorry, I was curious how to make them fabricate, with those really minimum tolerances. For the "rolamite like" ribbon, what are you going to use?
c
starting torque: consider that both Rabbit and 3d Toy (and Doug's Bunny too) have no such problems, despite multiple bearings used. Side force + stylus drag aids each other efficiently. Imho 2 idler could work too, more when using jewells as intended
 
Sorry, I was curious how to make them fabricate, with those really minimum tolerances.

The 2 idler version can be turned to +/-50 micron very easy on a lathe as they are round pulleys really.

The 1 idler makes use of non-round cams to gets to absolute tangency but has to be cut with a wire-EDM machine that can get to 5 microns.

For the "rolamite like" ribbon, what are you going to use?

Ultimately I would like to use Beryllium-Copper foil as the strip but it is not available in South Africa, so I'll practice with SS foil.

starting torque: consider that both Rabbit and 3d Toy (and Doug's Bunny too) have no such problems, despite multiple bearings used. Side force + stylus drag aids each other efficiently. Imho 2 idler could work too, more when using jewells as intended

That's good to know. There are only 2 jewel bearings that just keep the main pulleys apart, so not part of the load. Tension in the foil in the 2 cam version is what worries me, but not for the 1 cam version as the foil never makes a complete loop.
 
Beyond cams tolerances (seems all feasible today) what concerns me more is the non extensible metal strip, that can raise the tension from zero to too much in a moment. How it's controlled, with the idler wheel?
+- 50 micron are comfortable even for a junk Chinese lathe and my old eyes: i like to press fit the TA parts to get mechanical continuity, and this needs much better coupling.

c