perfusionist said:
If you understood anything thats been said or knew anything
about the characteristics of records in the various ways they
can be cut, and their intrinsic eccentricity you would realise
this is simply not possible.
A motor controlled tangential arm must have some form
of sensor that indicates any deviation from straight,
this implies that it has a horizontal pivot for the arm,
albeit only a small amount of horizontal freedom is needed.
(e.g. fixed LED/lens and receiver, mirror on the arm)
This signal is used by the motor control servo feedback loop
to control the position of the tangential arm base keeping the
arm straight.
Simple as that.
It copes automatically with eccentric records and transverse rates.
Why didn't you just ask how a motor controlled tangential arm works ?
(there are several other types of tangential arms)
🙂 sreten.
Hey, what's with the attitude? It is my understanding that you want to learn something, right?
FYI the groove pulls the arm laterally in the air-bearing designs and in the motorised designs a stepper motor moves the arm assembly when needed. From what i remember a B&O unit works like that: the tangential arm is free to rotate at a very small angle; once this angle is exceeded an optical system detects it and a motor moves the arm to a new position.
analog_sa said:
Think you'll find that the motor is usually DC, and the optical
system gives a continuous signal, think of a spot moving off a
sensor. The feedback tries to maximise this signal, rather than
an angle being exceeded, giving a continuous analogue DC
servo feedback loop, with much lower error than the above.
🙂 sreten.
Hi,
I can't recall a tangential arm that worked the way described here...
Analog_sa is correct in that in all motorised tangential trackers I know of the arms' lateral movement was allowed some runaway from perfect tangency and than corrected by a servo motor.
Usually this was done by a photoelectric sensor and a motor inside a servo-loop that corrected the tangency error by repositioning the arms' moving assembly as it approaches the innerside of the record.
Examples are the Goldmund T3, B&O, Revox and no doubt some others.
Their predecessor was the old Rabco which was a mechanical design, an evolution of that was the Souther which is still used by Clearaudio, for instance.
Cheers,😉
I can't recall a tangential arm that worked the way described here...
Analog_sa is correct in that in all motorised tangential trackers I know of the arms' lateral movement was allowed some runaway from perfect tangency and than corrected by a servo motor.
Usually this was done by a photoelectric sensor and a motor inside a servo-loop that corrected the tangency error by repositioning the arms' moving assembly as it approaches the innerside of the record.
Examples are the Goldmund T3, B&O, Revox and no doubt some others.
Their predecessor was the old Rabco which was a mechanical design, an evolution of that was the Souther which is still used by Clearaudio, for instance.
Cheers,😉
Strange - because we are describing the same thing !
The sensor and feedback loop don't care if its the arm that moves
or the base with the fixed sensor. The end result is the same thing.
My point is its the base that moves to keep the arm straight,
via the feedback loop, of course there must be some lateral
movement of the arm to keep the feedback working,
like the voltage difference between the input pins of
an op-amp due to the open loop gain not being infinite.
The same reasoning applies here.
🙂 sreten.
Hi,
Sorry, I suppose I just misread you then.
Either way, I don't think those motorised designs are still used or manufactured today.
There are still true air bearing tangential tonearms around though which I find more interesting as a design.
Cheers,😉
Sorry, I suppose I just misread you then.
Either way, I don't think those motorised designs are still used or manufactured today.
There are still true air bearing tangential tonearms around though which I find more interesting as a design.
Cheers,😉
"Now, an LP is played at a constant angular velocity. This means that the time it takes for a certain (fixed in relation to the LP) radial to pass under the needle is shorter as you get nearer the hole. SO: the tangential velocity as you call it is not constant! It becomes faster as you get nearer the hole and depends on each record.
So it is not possible to set a fixed tangential velocity for a linear tracking arm. This is probably what you wanted to know. The solution is to let the arm be dragged from rim to center by the force the groove on the stylus or use a feedback system."
Thank you very much, that's what I was getting at myself. I looked at many tangential tonearms, and they have gearing or motors moving them. I have been reading that website for some time now. I have one more question raised by what you said.
I have been told that the coefficient of friction for an average LP is .3 but it is varied by grooves and stylus shape. Asuming one is using a tangential tonearm, how much weight can a needle drag laterally while following the radial grooves? At what mass does the weight become a problem?
So it is not possible to set a fixed tangential velocity for a linear tracking arm. This is probably what you wanted to know. The solution is to let the arm be dragged from rim to center by the force the groove on the stylus or use a feedback system."
Thank you very much, that's what I was getting at myself. I looked at many tangential tonearms, and they have gearing or motors moving them. I have been reading that website for some time now. I have one more question raised by what you said.
I have been told that the coefficient of friction for an average LP is .3 but it is varied by grooves and stylus shape. Asuming one is using a tangential tonearm, how much weight can a needle drag laterally while following the radial grooves? At what mass does the weight become a problem?
Hi,
??? To achieve what exactly?
Really? Anything else I should know?What website? The one with the German language you asked for translation?
I don't mind but be prepared, you have to know what your talking about....From what I read you don't have the foggiest.
I learn everyday, but I don't barge in with an attitude on something I'm asking questions about...
It can be approached differently, you know..............
Cheers, 😉
??? To achieve what exactly?
Really? Anything else I should know?What website? The one with the German language you asked for translation?
I don't mind but be prepared, you have to know what your talking about....From what I read you don't have the foggiest.
I learn everyday, but I don't barge in with an attitude on something I'm asking questions about...
It can be approached differently, you know..............
Cheers, 😉
analog_sa said:
If an air bearing arm could be light enough and parrallel to the groove angle, could the groove move the whole arm without servo-motor assitance?
perfusionist said:
Certainly can (assuming the bearing is also very near perfectly level).
Theoretically equates to a lateral displacement of the stylus and
the friction of the bearing, higher compliance cartridges having more
lateral displacement.
Practically the real issue is eccentricity, in records.The arm will be
moving side to side at (for 33rpm) at ~ 0.5Hz. (0.75Hz for 45's).
If the arm friction is low and its very nearly level then this will
dominate the lateral displacement of the stylus.
The lateral arm effective mass and the stylus compliance usually
(assuming friction is low) form a classic underdamped 2nd order
high pass filter.
In this case the higher the compliance/mass resonant frequency the
lower will be the the effective displacement of the stylus from 'true'.
(Assuming the resonant frequency is well above 1Hz)
If the arm is not level it will add a constant DC displacement, the
other enemy is stiction, inconsistent friction, as pure friction to
an extent can be used for a modicum of damping.
Arm/cartridge resonance below around 7-8 Hz is a disaster in
suspended subchassis turntables, and given the levelling
requirements for air bearing arms they do not mix well.
The main advantage of air bearing arms is not the tangentiality -
that goes with the territory - but the theorectically very high
torsional stiffness of the bearing arrangements.
I should also add without the the benefit of any lateral pivot the
lateral effective mass of an air bearing arm is its total moving mass.
Vertical effective mass due to the vertical pivot will always be lower.
🙂 sreten.
fdegrove said:
If anyone barges in with an attitude, its you Frank! You make nasty remarks and then have a thin skin when people respond. I am ignorant of the turntable physics, which doesnt allow you the right to insult me.
Your snide little remarks are all over DIYaudio. It's the rare thread where you contribute. Why is your psychology so abnormal?
Why dont you go back to trolling on the DIY TT thread? Maybe becuase the moderator reamed you out for poisining the atmosphere? I guess I'm havent learned your snobby audiophile aloofness. I"m not so knowledgeabel that I can tell people they are wrong, but that " I'm so smart I'm not going to tell them the answer, I'm just going to ridicule you".
You talk so much, yet you say so little. Your caustic comments would be welcome if you could add something constructive, but you post to stroke your own ego. The DIY TT moderator was very tactful in his treatise, but I dont have time for you wasting mine.
I'm sure I could benefit from your knowledge of audio, but you dont care to share, so then just go away.
I cannot find any positive comments you have made, except one... when you thanked someone for a complement! I guess having your ego stroked is all you are looking for, you pathetic little troll.
Calm down a bit. Even if you think that way (and you are entitled to, why not), then there is no reason to get that excited. I understand your anger, but you also have to understand that we tried to get it over to you that what you want -a constant tangential velocity- is not possible. Please try to understand that part also.
Just as a visible indication: http://www.cs.huji.ac.il/~springer/ look at the pictures....
Just as a visible indication: http://www.cs.huji.ac.il/~springer/ look at the pictures....
Havoc,
In general, you guys have helped me tremendously. If you think I'm acting out of the ordinary, you should look at some of the threads fde has been trolling on. This is his regular mo, and I'm just not in the mood when so many other people are helping me so much.
I understand now that the stylus is constantly accelerating tangentially, but also that its angular velocity is constant. What I realize now Is I'm looking for the angular velocity, which is constant, and a vector quantity. At first, I wanted to make a 10Hz arm which would be pulled by friction alone. Physically, I cant make it light enough to work on a high compliance cartridge.
If I've made a mistake on my calculations, let me know.
I'm wondering if its possible to make a motor whose controler is based on angular velocity, in radians/second, which would speed up proportionately with the tangential velocity. Tangential velocity is proportional to the distansce from the axis of rotation. This would keep the arm constantly accelerating. You see, if tangential velocity is constant, then angular velocity =2*pi*freq (rev/second). Of course, tangential velocity is never constant, but if we make a differential equation using dv/dt as it approches maximum velocity, we can get the constant to set our motor to, which might be very close to what is required for constant acceleration. This it what I'm wondering. If partialdv/partialdt=2pi*freq will work. Angular velocity is constant, no matter where on the radius of a spining mass you are. So if we can use a formula to set the speed of the motor based on that, it should stay tangent without servo-ing.
Of course the speed of the cartridge side is determined by the program material and the coefficient of friction, but if the speed of the back of the arm is well within the boundry of speed changes due to friction reached in a few radians, it might be a simple way instead of all these hysteresis servo systems.
The other problem I had is I'm aiming for 10Hz resonance. As has been pointed out, in tangential arms, the effective mass = actual total mass of arm+screws+wire+counterweight. Using Hz/1000 = 1/2 pi (effective mass x compliance), I solved for the mass of my arm, which could be no more than 3.8-4.2 grams to get 10 Hz. And that's including the counterweight! Even if I could make an arm that skinny, the cartridge (shure v15vx) is 6.3 grams, so the counterweight would have to be way on the other side of thepivot point to lighten that cartridge. I'd go over weight limits before I could cantelever out far enough. I think this is why many tangential arms suspend the cartridge over the surface with a cantilever, to keep the lateral/effective mass down low enough. angular
In general, you guys have helped me tremendously. If you think I'm acting out of the ordinary, you should look at some of the threads fde has been trolling on. This is his regular mo, and I'm just not in the mood when so many other people are helping me so much.
I understand now that the stylus is constantly accelerating tangentially, but also that its angular velocity is constant. What I realize now Is I'm looking for the angular velocity, which is constant, and a vector quantity. At first, I wanted to make a 10Hz arm which would be pulled by friction alone. Physically, I cant make it light enough to work on a high compliance cartridge.
If I've made a mistake on my calculations, let me know.
I'm wondering if its possible to make a motor whose controler is based on angular velocity, in radians/second, which would speed up proportionately with the tangential velocity. Tangential velocity is proportional to the distansce from the axis of rotation. This would keep the arm constantly accelerating. You see, if tangential velocity is constant, then angular velocity =2*pi*freq (rev/second). Of course, tangential velocity is never constant, but if we make a differential equation using dv/dt as it approches maximum velocity, we can get the constant to set our motor to, which might be very close to what is required for constant acceleration. This it what I'm wondering. If partialdv/partialdt=2pi*freq will work. Angular velocity is constant, no matter where on the radius of a spining mass you are. So if we can use a formula to set the speed of the motor based on that, it should stay tangent without servo-ing.
Of course the speed of the cartridge side is determined by the program material and the coefficient of friction, but if the speed of the back of the arm is well within the boundry of speed changes due to friction reached in a few radians, it might be a simple way instead of all these hysteresis servo systems.
The other problem I had is I'm aiming for 10Hz resonance. As has been pointed out, in tangential arms, the effective mass = actual total mass of arm+screws+wire+counterweight. Using Hz/1000 = 1/2 pi (effective mass x compliance), I solved for the mass of my arm, which could be no more than 3.8-4.2 grams to get 10 Hz. And that's including the counterweight! Even if I could make an arm that skinny, the cartridge (shure v15vx) is 6.3 grams, so the counterweight would have to be way on the other side of thepivot point to lighten that cartridge. I'd go over weight limits before I could cantelever out far enough. I think this is why many tangential arms suspend the cartridge over the surface with a cantilever, to keep the lateral/effective mass down low enough. angular
Hi,
You sure got a nerve, mister..Trolling, he?
It isn't and it's blatantly obvious why it isn't, unless you have a device watching what's going on. Which is where your controller comes in, which puts you back to square one: feedback controlled devices controlling and correcting errors.
I hope that one helps?
Your little theories are so full of loopholes that I wouldn't know which end to put an handle on them first so just keep on trying: it won't work, that much I can guarantee you.
In fact I've never seen anything like it...Never really looked at a record I suppose?
Cheers,😉
You sure got a nerve, mister..Trolling, he?
It isn't and it's blatantly obvious why it isn't, unless you have a device watching what's going on. Which is where your controller comes in, which puts you back to square one: feedback controlled devices controlling and correcting errors.
I hope that one helps?
Your little theories are so full of loopholes that I wouldn't know which end to put an handle on them first so just keep on trying: it won't work, that much I can guarantee you.
In fact I've never seen anything like it...Never really looked at a record I suppose?
Cheers,😉
No, my idea is NOT to servo, but to have a constantly accelerating motor. Although the variation in grooves changes the speed of the cartridge, if we make a motor finely accelerate so that it cannot overtake the fastest record, yet not undertake the slowest, then there is no need for servo. That is why I asked for coefficient of friction. Frank, I would suggest that instead of always pointing out how much you know and wont share, why dont you show me how smart you are and explain a better way. I challenge you to show my how smart you are. Instead of telling me the many ways my ideas are wrong, why do you not share your ideas, or have you any. Is your only purpose to snipe upon others and not share your vast knowledge?
perfusionist said:
IMO Frank is perfectly entitled to tell you that it won't work.
(And despair at the utterly tortured pseudo logic of the above,
which if analysed does not make the slightest sense whatsoever)
You don't seem to have taken in the information given to you,
simply put again - the problem is the eccentricity of records.
You keep asking questions that seem to blantantly ignore
this fundamental fact, and insist on trying to use concepts that
anyone with an understanding of the way records are cut would
know are far too variable and thus essentially meaningless.
Designing for a 10Hz lateral resonance frequency for a V15 in
an air bearing arm - you've discovered is just not possible.
That doesn't mean it won't work. Your design goal is unrealistic.
As stated before the resonant frequency must be clear of
the 1Hz region. So 2.5Hz to 5Hz is a practical proposition.
🙂 sreten.
Take a break, and then re-read everything said in this thread (cutting out the non technical bits).
NO! It is not. That was my mistake, corrected by sreten. Tangential velocity is constantly changing! Both accelerating and decelerating!
Yes, that is about the only thing you can be sure of. (come to think of it if you can. Sretens remarks about eccentricity might trow a spanner into that also)
No. Since angular velocity is constant, it does not give you any information. And it is linear velocity that is proportional to the distance from the axis.
Tangential velocity depends only on:
- program material
- how much running time the engineer wanted on the disc
- mastering proces used
- very probable: label
- very probable: lathe used
So those are the only variables you can use for setting your speed. An arm dragged along by the groove uses those variables. One with feedback just makes sure the variation of tangentiality is within some designed in limits.
NO! It is not. That was my mistake, corrected by sreten. Tangential velocity is constantly changing! Both accelerating and decelerating!
Yes, that is about the only thing you can be sure of. (come to think of it if you can. Sretens remarks about eccentricity might trow a spanner into that also)
No. Since angular velocity is constant, it does not give you any information. And it is linear velocity that is proportional to the distance from the axis.
Tangential velocity depends only on:
- program material
- how much running time the engineer wanted on the disc
- mastering proces used
- very probable: label
- very probable: lathe used
So those are the only variables you can use for setting your speed. An arm dragged along by the groove uses those variables. One with feedback just makes sure the variation of tangentiality is within some designed in limits.