Angling for 90° - tangential pivot tonearms

I think it can be reasonably applied to the stylus a drag of 2-3mN and 5- 6mN of side force.

5-6mN is way to much side force. With a typical cartridge this level of side force will result in a lateral tracking error of about a degree. It will not be just the stylus/groove that will be misaligned but also the coils/magnets of the cartridges generator. If you are accepting this level of side force your lateral tracking error is likely to be as bad or even worse than a conventional pivoted arm. It really does appear that side force/bearing friction is the biggest problem with pivoted tangential arms not the geometry.

Niffy
 
Completely right* Niffy, but at this stage I just want to understand if this mess can move or not, using what I called the "skipping force", the maximum available.
If it does there will be the problem of using as little as possible with the most appropriate techniques, for now it is only a matter of vectors and after the Syrinx better understand everything before than after.
Speaking of side force, how much do you need on a good linear on bearings?

Carlo
*far, far worse than a good old Baerwald
 
Hi Carlo,

The side force required to overcome bearing friction with a linear tracking arm is dependent upon the type of bearing used.
With an air bearing the fiction is very close to zero so only a miniscule side force is required, measured in uN rather than mN.
With mechanical linear tracking arms the amount of friction is dependent upon the coefficient of friction of the bearings and the mass of the carriage. An arm using ball race bearings running on borosilicate glass will have a coefficient of friction between 0.006 and 0.01 dependent upon the quality of the bearings. An arm like the clearaudio which, if I remember correctly, has a mass of about 75g would require around 4.5mN side force to overcome bearing friction. (i am assuming that clearaudio uses very high grade bearings) So better than a conventional pivoted arm when looking at lateral tracking error but not by a huge margin.
With my arm using jewelled pin bearings running tungsten carbide wheels on a tungsten carbide rail the coefficient of friction is 0.0017, much lower. With my carriage mass of 55g the side force required to overcome friction is about 0.9mN.

Niffy
 
Last edited:
very, very low: must be really excellent bearings.
The data you quote are taken at the cart point or at the stylus? if I remember correctly the side force to a cart with friction applies according to the cosine. I think it is also essential to have a very short arm like yours, in spite of azimuth problems on warps. Do air bearings have laminar flows? because if not could be tricky with resonances.
Lots of questions better to talk on your forum, someday
Carlo
 
Hi Twice.
I've done an overlay on the dimensions and layout that you sent me to work on it: you managed to reduce errors in an incredible way. It would be nice to use standard diameter pulleys that can be bought, to allow a construction without a lathe
Even if the solution with runout could be feasible (I tried just for fun) now seems really unnecessary: + - 0.5mm of overhang is completely inaudible. And then, since i'm stupid, i like to K.eep I.t S.imple.
On this basis I think it's possible to make a mock-up, but since to be useful will almost be like making the working arm, before I'll have to re-study a bit of Static, to see what kind of forces can be found on P (and HP).
carlo
 
Not even if the tonearm moving against the moving pivot rotation will work.

Yes, there is something strange that I can not understand, but does not convince me: that knee joint that must open for 40 ° e.g. I tried to break down the forces and the torque on the pulley HP seems to act against, but should be impossible trough the bearing. Before doing even the mock up better to have clearer ideas on what happens.

Your solution seems to have less problems and is easier, if the pulleys can be kept in contact in the right way

Carlo
 
Carlo,

Some construction thoughts: I assume you are using ceramic hybrid bearings. Try those made for yo-yos.

An angular contact or a deep groove bearing is a good idea for the top bearing in the support column because it carries so much of the vertical load. The others can be radial.

Check out preloading: Bearing Preload Facts and Information - NMB

Consider alighiszem's binding force link at #1525 if you are going to use the cantilever design in Birch B #1605.

Side and stylus forces aren't the only ones. Gravity is a trouble maker for PLTs. The axes for horizontal movement have to be as vertical as possible or the wand motion becomes erratic.

Horizontal effective mass builds up quickly.

You may have to add half the belt thickness to the pivot diameters to be accurate.

I haven't quite figured it out, but it may be that skating can be our friend.

The parking problem I mentioned is about Birch designs with a bearing at point B. The headshell is very close to the platter at rest. Your design and the two arm design solve this.

I'm going to be traveling for about ten days. I try to stay in touch.

I agree with 2wice at #1628
 
thanks for the advices, Doug - always coming from an uncommon experience

Bearings - I'm using SKFs as usual, hoping they are worth the cost.
I know something about pre load: in my arms I have always used double bearings loaded by spring washers. Making the first gimbal 10 years ago I discovered that two bearings have less friction than one badly loaded

# 1605 - The B hypothesis was drawn to place the arm at a standard distance (225mm). The A convinces me a lot more (I think I have resumed your solution) because two well spaced bearings are stressed much better by this load, and maybe can be made very light.

Weight is definitely a problem - I'm thinking of fiberglass (pcb) for the "cranks" and thin aluminum tube instead of solid bar, but above all a spring instead of the counterweight.

Belt: the accuracy of ratio is crucial, and even some elasticity - troubles guaranteed. The Twice solution must be considered seriously: i've rotated two gears leaning on table and works surprisingly well

Frictions: I believe, however, that I will never reach the data reported by Niffy for linears (surprising even to be measured!). What I did in the second arm was much better than many commercial types lent from a friend who likes to build 3D printers, but still far from 0.9 mN on the stylus. That means just a small fraction able to push the cart.
I am comforted by the fact that your prototypes worked even with low VTF: and I believe that the guides, or the second crank were not friction less.

For now, I'm trying to understand something with faded memories of high school. Difficult that I could succed, because the system is really complex: two related rotational movements + one rolling (the pulley on the belt).
I decided to make a crude arm, reduced to the bone, instead of a mock-up. I will lose some more time but I will not risk surprises.

ciao - Carlo
 
Last edited:
You can reduce the error at the outer groove to 0.55°, middle -0.55° & inner groove to 0.55° with 2 null points at 1/3 and 2/3 of the record by changing
the length.
 

Attachments

  • Birch Gadget CAD5.PNG
    Birch Gadget CAD5.PNG
    65.5 KB · Views: 202
Hi 2wice (sorry, sometimes I've called you Twice). The more I think of your solution and the more I think is the right one, simpler and with better application of the forces (bigger lever arm),
To reduce hor. mass can be used only the necessary sector of the pulleys (especially for the mobile one) and for contact I thought like you, but the opposite, an O - ring on the fixed one. Something must be invented to keep them in touch with just the slightest possible pressure.
I tried to trace the curve with two gears and a pencil and I discovered that the curve described is a beautiful spiral. May you model one for the belt? I'm curious to see if it's the same

Alighiszem - hope not, even with a similar feeling. To do so it would mean that the resistance on the belt (which is fixed) is less than on HP bearing (wich is "frictionless"): seems improbable.
But what learned from Syrinx is that we can't waste even a bit of the minimal forces available. Even a beautiful geometry it's not a working arm. In this phase are needed calculations, graphs, not feelings

Hiten - Rolamite: no, please, I beg you... you're putting too much stuff in the soup
Rolamite is for linear displacement, and nobody uses since his brilliant invention
carlo
 
Last edited: