Nice work.That is excellent. Whenever you have time and if possible a picture from top ?
Best Regards.
Rabbit test arm - first impressions
Even if it is just a bit more than a mock up, it works.
It tracks correctly (VTF 2.0) without cantilever binding (at least not visible), no evident variation of behaviour from external to inner grooves. Side force and stylus-drag seem to collaborate to motion, finally.
I could say that it is a typical weekend project (it took just a day, recovering some pieces from other arms) but it is not at all: mechanically it is very simple (the simplest possible?) but the construction is not trivial. The problems reported by Doug - rigorous alignment of the axes, perfect arm leveling - are crucial for a uniform rotation that does not introduce skating where we are trying to avoid.
The "Birch gadget" behaves well with just a little tension: glueing the belt to a point on the fixed pulley helps, but it is not essential. Some elasticity is also favorable: normally it does not act, but with marked eccentricity it helps absorbing the inertia of the lateral mass.
A functional arm, if I'll ever do it, will be quite different from this: you need to load the bearings in the best possible way, you need a precise but light vertical pivot (very short arm) and the horizontal mass must be limited maniacally. The main problem of this arm is the cantilevered structure itself, a mechanical nonsense.
Also important some reference marks for set up and control: if you inadvertently (forcing the arm improperly) move the belt, all the beautiful 2wice calculations become useless. Instead a very rigid transmission may be dangerous, with eccentric LPs. A lifter would be comfortable too, but where can be placed?
The next step will be listening (not so good probably, with this crude test arm) and then some measurements in various points: much more interesting to design a real arm
carlo
2twice - idler wheel: you're completely right, my arteriosclerosis progresses. Without it goes the right way, but faster than the crank! Now it's me waiting to see your clever solution playing -(Imho hard pulleys may bring to impossible loads to each other, even with minimal offset). Thanks again for calcs
Hiten - something better (attachment)
Even if it is just a bit more than a mock up, it works.
It tracks correctly (VTF 2.0) without cantilever binding (at least not visible), no evident variation of behaviour from external to inner grooves. Side force and stylus-drag seem to collaborate to motion, finally.
I could say that it is a typical weekend project (it took just a day, recovering some pieces from other arms) but it is not at all: mechanically it is very simple (the simplest possible?) but the construction is not trivial. The problems reported by Doug - rigorous alignment of the axes, perfect arm leveling - are crucial for a uniform rotation that does not introduce skating where we are trying to avoid.
The "Birch gadget" behaves well with just a little tension: glueing the belt to a point on the fixed pulley helps, but it is not essential. Some elasticity is also favorable: normally it does not act, but with marked eccentricity it helps absorbing the inertia of the lateral mass.
A functional arm, if I'll ever do it, will be quite different from this: you need to load the bearings in the best possible way, you need a precise but light vertical pivot (very short arm) and the horizontal mass must be limited maniacally. The main problem of this arm is the cantilevered structure itself, a mechanical nonsense.
Also important some reference marks for set up and control: if you inadvertently (forcing the arm improperly) move the belt, all the beautiful 2wice calculations become useless. Instead a very rigid transmission may be dangerous, with eccentric LPs. A lifter would be comfortable too, but where can be placed?
The next step will be listening (not so good probably, with this crude test arm) and then some measurements in various points: much more interesting to design a real arm
carlo
2twice - idler wheel: you're completely right, my arteriosclerosis progresses. Without it goes the right way, but faster than the crank! Now it's me waiting to see your clever solution playing -(Imho hard pulleys may bring to impossible loads to each other, even with minimal offset). Thanks again for calcs
Hiten - something better (attachment)
Attachments
Last edited:
Doug, thank to you for the advices and appreciation.
Of the cables I could imagine; in a well done arm the path should be carefully studied - what do you mean for "arrangement"?
It surprised me how sensitive it is to leveling, more than any bubble level: but yesterday's surprises were even greater
1- I have mounted a shorter belt (much more tension) in order to make precise geometry checks in the next days, and - incredibly - the friction is even smaller. Evidently the tension reduces the off center load of the bearings
2 - I tried to make the Diyrak String Test: advances on the geometry, without rotating! Maybe it's me that I do not know how to do (it's needed a parallel traction, and it's not simple) but it behaves much better than a normal one that turns immediately, and also of the syrinx that rotated just a little, but still rotated.
I hope you will experience and improve it yourself: knowing the weak points of these arms you can do a far better work than me. I remain with the desire to make an extensible Thales, unfortunately I am a mad dog that does not give the bone easily.
carlo
Of the cables I could imagine; in a well done arm the path should be carefully studied - what do you mean for "arrangement"?
It surprised me how sensitive it is to leveling, more than any bubble level: but yesterday's surprises were even greater
1- I have mounted a shorter belt (much more tension) in order to make precise geometry checks in the next days, and - incredibly - the friction is even smaller. Evidently the tension reduces the off center load of the bearings
2 - I tried to make the Diyrak String Test: advances on the geometry, without rotating! Maybe it's me that I do not know how to do (it's needed a parallel traction, and it's not simple) but it behaves much better than a normal one that turns immediately, and also of the syrinx that rotated just a little, but still rotated.
I hope you will experience and improve it yourself: knowing the weak points of these arms you can do a far better work than me. I remain with the desire to make an extensible Thales, unfortunately I am a mad dog that does not give the bone easily.
carlo
Last edited:
Carlo,
"arrangement" means where do the wires exit the wand at the pivot end and where do they go from there. Even very flexible wires can push or pull the arm.
The tighter belt may improved the bearing alignment by tilting the cantilevered arm up a bit and it may be lifting some of the weight. That would be an added benefit of your design.
The diyrayk string test result is interesting and a puzzle.
That's a good looking arm and better built than any of mine.
"arrangement" means where do the wires exit the wand at the pivot end and where do they go from there. Even very flexible wires can push or pull the arm.
The tighter belt may improved the bearing alignment by tilting the cantilevered arm up a bit and it may be lifting some of the weight. That would be an added benefit of your design.
The diyrayk string test result is interesting and a puzzle.
That's a good looking arm and better built than any of mine.
Last edited:
Rabbit arm measurements
geometry - see attachment
system friction
side resistance (90° push - measurements taken at the headshell - same method as in#1528)
C1- 1,2 mN C2 - 1,5 mN C3 - 1,8mN
stylus drag resistance (0° pull - measurements taken at the headshell)
C1- 1,5 mN C2 - 1,7 mN C3 - 2,2mN
carlo
String Test - really strange: of course, as the syrinx, this arm is forced to follow his geometry, but in syrinx there was a visible tendency to rotate, and here could not be seen. Consider too that the guiding system is much more elastic. The elasticity of the guide with respect to the eccentricity (but not to the small tracing forces) is the most important thing of this arm, imho. Someone will think that a flat belt is not Hi-Endish, I just know that efficiency is 95 - 98% - and this since the time of the Egyptians.
belt tension: exactly, a free bonus
geometry - see attachment
system friction
side resistance (90° push - measurements taken at the headshell - same method as in#1528)
C1- 1,2 mN C2 - 1,5 mN C3 - 1,8mN
stylus drag resistance (0° pull - measurements taken at the headshell)
C1- 1,5 mN C2 - 1,7 mN C3 - 2,2mN
carlo
String Test - really strange: of course, as the syrinx, this arm is forced to follow his geometry, but in syrinx there was a visible tendency to rotate, and here could not be seen. Consider too that the guiding system is much more elastic. The elasticity of the guide with respect to the eccentricity (but not to the small tracing forces) is the most important thing of this arm, imho. Someone will think that a flat belt is not Hi-Endish, I just know that efficiency is 95 - 98% - and this since the time of the Egyptians.
belt tension: exactly, a free bonus
Attachments
Last edited:
I was trying to explain that the driving mechanism (belt + pulleys) obviously forces the arm to follow the Thales circle and therefore it is difficult to understand if you are generating some skating because there is no visible misalignment.
Applying a traction aligned with the cantilever the arm maintains the alignment at point B and then 0 offset. With the syrinx it happened the same, but forcing the traction a certain rotation appears (perhaps elastically deforming the string of driving mechanism). Here the traction stretches the belt but drags the crank too, mantaining the alignment. I'm speaking of relevant traction, completely different from that of S. D.
That's why I found this Diyrayk test surprising, I expected a rotation that I did not see. You know a lot more than me, maybe you could analyze the behaviour of this arm and tell us what happens.
ciao Carlo
Applying a traction aligned with the cantilever the arm maintains the alignment at point B and then 0 offset. With the syrinx it happened the same, but forcing the traction a certain rotation appears (perhaps elastically deforming the string of driving mechanism). Here the traction stretches the belt but drags the crank too, mantaining the alignment. I'm speaking of relevant traction, completely different from that of S. D.
That's why I found this Diyrayk test surprising, I expected a rotation that I did not see. You know a lot more than me, maybe you could analyze the behaviour of this arm and tell us what happens.
ciao Carlo
I do not think I know more than you.
If you apply a force F aligned with the cantilever to the tonearm and as a consequence the arm moves toward the spindle, that is by definition skating. The magnitude of the turning moment is F X (the distance of the line cx-B from the center of pulley P).
If you apply a force F aligned with the cantilever to the tonearm and as a consequence the arm moves toward the spindle, that is by definition skating. The magnitude of the turning moment is F X (the distance of the line cx-B from the center of pulley P).
Rabbit arm now playing.
Sounds good (pleasently surprising as you said, Doug), good with my jurassic amp+ speakers (technics suv 6 + allison 3): no issues with just 1,5 VTF.
Ready to be heard on a much less forgiving one: triode phono - single ended 300b mono - fostex fe208S BLH: all diyed, of course.
some details more
as said from the first post
Skating (the magic word of this thread): equal to every Birch no one excluded, even those with stellar co$$$$ts.
I do not think it has no skating. I've tried to make an extensible Thales just for how the P Birch point is located. I just said that the string test made by me does not show it.
Pulleys: reading carefully the 2wice calculations you will find a ratio slightly different from 3/2 = 1,470, to get the smallest possible error. With pulleys r 10 + r 15 you have to add 1/2 belt thickness (0,5mm) that gives a ratio of 1,476. that is very very near. Effective lenght (from HP ) is 167mm (average value between HP>C1-C3) getting 2 nullpoints.
So, buying the pulleys, the arm can be made by anyone with a pillar drill, on the famous kitchen table: do not forget to add a lot of precision, to get the right flavor.
carlo
Sounds good (pleasently surprising as you said, Doug), good with my jurassic amp+ speakers (technics suv 6 + allison 3): no issues with just 1,5 VTF.
Ready to be heard on a much less forgiving one: triode phono - single ended 300b mono - fostex fe208S BLH: all diyed, of course.
some details more
as said from the first post
Skating (the magic word of this thread): equal to every Birch no one excluded, even those with stellar co$$$$ts.
I do not think it has no skating. I've tried to make an extensible Thales just for how the P Birch point is located. I just said that the string test made by me does not show it.
Pulleys: reading carefully the 2wice calculations you will find a ratio slightly different from 3/2 = 1,470, to get the smallest possible error. With pulleys r 10 + r 15 you have to add 1/2 belt thickness (0,5mm) that gives a ratio of 1,476. that is very very near. Effective lenght (from HP ) is 167mm (average value between HP>C1-C3) getting 2 nullpoints.
So, buying the pulleys, the arm can be made by anyone with a pillar drill, on the famous kitchen table: do not forget to add a lot of precision, to get the right flavor.
carlo
Attachments
Last edited:
Hi Carlo,
Nice work, it is good to see it working, not only on paper. Do I read it correctly that it is not tangential at every radius, but has two zero-error null points? What is the maximum error between nulls?
Are you using special bearings? How do you make sure the arm is vertically stable? Bearing have a minimum "play" off axis.
Nice work, it is good to see it working, not only on paper. Do I read it correctly that it is not tangential at every radius, but has two zero-error null points? What is the maximum error between nulls?
Are you using special bearings? How do you make sure the arm is vertically stable? Bearing have a minimum "play" off axis.
Niffy impressive achievement for a days work...
The waste bin of a diyer is always fullfilled; for this test arm the biggest work is making 2 holes in a piece of plastic e 2 in an aluminum bar: 10 minutes for a machinist, a day for a diyer.
Icsaszar: read 2wice work - was 2mm error on my drawing, reduced to 0,5 mm by 2wice: i think that is impossible to make something better.
Hiten: i'm designing something to overcome various problems, counting on Doug advices. But first i have to understand better the physics of this system: some reverse engineering starting from the meauseres.
No need of thin belt . 1mm is the right measure to get that ratio with standard pulleys.
Even the best bearings of the world (not 0 friction high speed please, 0 play) must be loaded properly - as I said, my test arm is instead a mechanical nonsense.
carlo
The waste bin of a diyer is always fullfilled; for this test arm the biggest work is making 2 holes in a piece of plastic e 2 in an aluminum bar: 10 minutes for a machinist, a day for a diyer.
Icsaszar: read 2wice work - was 2mm error on my drawing, reduced to 0,5 mm by 2wice: i think that is impossible to make something better.
Hiten: i'm designing something to overcome various problems, counting on Doug advices. But first i have to understand better the physics of this system: some reverse engineering starting from the meauseres.
No need of thin belt . 1mm is the right measure to get that ratio with standard pulleys.
Even the best bearings of the world (not 0 friction high speed please, 0 play) must be loaded properly - as I said, my test arm is instead a mechanical nonsense.
carlo
Last edited:
Hi all,
from what I've heard the rabbit cub deserves to grow.
Here some thoughts about building (attachment):
rotating "crank" supported in a better way (limited off center loads)
belt aligned with the bearings (correct loads, limited chattering)
adjustable leveling base (indispensible for set up)
adjustable distance from spindle (idem)
limited moving masses
a lifter will help, because two moving axes makes handling problematic (how to?)
advices welcome
carlo
from what I've heard the rabbit cub deserves to grow.
Here some thoughts about building (attachment):
rotating "crank" supported in a better way (limited off center loads)
belt aligned with the bearings (correct loads, limited chattering)
adjustable leveling base (indispensible for set up)
adjustable distance from spindle (idem)
limited moving masses
a lifter will help, because two moving axes makes handling problematic (how to?)
advices welcome
carlo
Attachments
I really hope some potential builders are looking at the thread now because I'd like to see some arms being built and posted on the thread. I think Carlo's designs, especially the Bunny are very DIY friendly.
Carlo: I had some suggestions, but you have already included them in your designs. Did you turn your pulleys or buy them? And where did you get the belt? Have you had a chance to listen to the test arm on your 300B amp? To me, LTs, both parallel and pivoting, compared to conventional pivoting arms is like an SE amp compared to PP. All of those can be very good, a pleasure to listen to, but LTs and SE bring something special.
I did some experiments about a month ago to try to understand Niffy and carlo's numbers. The results were only approximate but close enough to convince me of their results. I've come to accept the idea of PLT skating, but still think, because of the unique Birch geometry, that it is very difficult to separate skating from the movement that is built into Birch arms. For instance, the eff. length increases and the "offset angle" decreases. I think that means the skate should decrease across the LP, but I'm not sure. If one or two of the skating advocates would help with an analysis, that would be an excellent contribution to the thread.
Carlo: I had some suggestions, but you have already included them in your designs. Did you turn your pulleys or buy them? And where did you get the belt? Have you had a chance to listen to the test arm on your 300B amp? To me, LTs, both parallel and pivoting, compared to conventional pivoting arms is like an SE amp compared to PP. All of those can be very good, a pleasure to listen to, but LTs and SE bring something special.
I did some experiments about a month ago to try to understand Niffy and carlo's numbers. The results were only approximate but close enough to convince me of their results. I've come to accept the idea of PLT skating, but still think, because of the unique Birch geometry, that it is very difficult to separate skating from the movement that is built into Birch arms. For instance, the eff. length increases and the "offset angle" decreases. I think that means the skate should decrease across the LP, but I'm not sure. If one or two of the skating advocates would help with an analysis, that would be an excellent contribution to the thread.