Sean...But it doesn't. This servo system was not designed by amateurs. |
El, it has to have some amount of cogging, otherwise there wouldn't be any need for the sensors. Every time it makes a correction it cogs and with an off center record it would be doing so continuously. Of course it is another matter whether this detracts from the sound.
I'm curious where the sensors are mounted in relation to the pivot point. Your spec of .05 degrees seems like an awfully low figure. With a 3 inch arm that would be about 70 microns at the stylus. I'm not saying it can't be done, but that does seem awfully low for a consumer table.
While I agree that it is an interesting exercise in engineering, one has to wonder why nobody builds such a system anymore. In the cost is no object world of high end audio things like stepper motors, hall effect sensors, and microprocessor controllers are dirt cheap. Perhaps it is a matter of perception given the amount of trashing they took early on. |
Herman...Your assumption that there is "cogging" really means that the control process exhibits overshoots. It is entirely possible to design a control process that does not overshoot at all.
Regarding off center records...remember that the arm pivots, and this pivoting takes care of following wobbly grooves. Remember that the arm servo is biased so as to move at the speed necessary for nominal groove spacing, so the servo does not need to follow the instantaneous arm angle. There is filtering, and probably a notch filter at the 33.3 Hz frequency. Again I say that the Sony engineers knew what they were doing.
I agree that the 0.05 degree tracking error is very very small. Probably smaller than it needs to be. Pivot to stylus measures 7 inches, so 0.05 degrees represents 0.006 inches. An angle of 0.05 degrees is 180 arcseconds. This is easily measured. In the inertial guidance systems that I used to work with we routinely implemented angular measurements on the order of 1/10 arcsecond. |
J.A. Michell recommends to use the Mobil 10W30 motor oil. I use it with my Rega P25 and I'm very happy with it. |
El, one of us is confused and I think it is you.
Consider a record where the spacing is wider than average. You lower the stylus into the groove, the arm is perpendcular to the rail, and begins moving toward the spindle. However, it is traveling too slowly for the wide groove spacing and soon finds itself lagging behind. The sensors see this and before the arm can get more than .05 degrees away from perpendicular the arm speeds up momentarily to recenter. It just cogged over. This will continue all the way across the surface.
Consider an off center record. From the point of view of the arm, the off center record presents a groove that is on average moving toward the spindle, but also constantly wandering back and forth. So to maintain the perfect angle the arm sometimes has to speed up when the grove is wandering toward the center and actually back up when it is going the other way. The pivot will allow the arm to move back and forth with the groove to a certain extent, but when the angle approaches .05 degrees it will cog.
The only way for your table to keep the arm at the perfect angle is to know in advance which way the groove is going. It doesn't. It senses when the arm is not centered and then reacts to this.
If the arm was tracking the groove perfectly then there would be no need for it to pivot and no need for sensors. The very fact that it does proves that it is not always at the perfect angle and cogs to correct anytime the arm approaches an angle of .05 degrees. I'm not saying it overshoots. I'm saying that Sony biases the arm to move at a rate that will track the average record, and that any record that is not average will require the arm to cog. |
Herman: your example of an off-center record is a bit extreme. While this does happen, even a pivoted arm is "wallowing around" in the grooves as it the disc shimmies and shakes in an oblong rotation. There's only one TT made that actually deals with this in a very thorough manner and it is a Nakamichi.
The ReVox TT's with the "brick-like" linear tracking arm uses some type of optical sensor that reads the groove spacing and adjusts accordingly. I don't know the specifics of this system, but you can very clearly see the light shining down on the disc from within the "cartridge carrier" ( for lack of a better term ). This was my first linear tracking table and quite honestly, i was not the only one impressed by this $150 Ebay purchase. Everyone that heard it after i set it up in one of my systems commented on how good vinyl sounded.
As a side note, Kavi Alexander of Water Lily records uses a ReVox ( as far as i know ). I remember that he was looking for one of these, so i dropped him a line and gave him some background as to the differences between the various models that ReVox made. He was so grateful that he offered me a free disc from his label, but i declined his gracious offer. After all, if we can't help each other out without expecting to get compensated for basic info, we are a doomed breed. Excuse me while i park my dinosaur : ) Sean
> |
Herman...Confusion is not the issue. It's just that you are citing the kinds of problems that a slipshod servo design might exhibit.
By the way, the reason that relatively inexpensive linear tracking systems did not become very popular is that:
1. For low end systems they add significant cost relative to a simple pivoting arm.
2. For high end systems, they eliminate all the exotic designs, precise setup,tweeks, and endless adjustments that are dear to the heart of audiophiles. Just plop the record down and push the button. How boring! |
El, eplain to me how your arm can keep from cogging over to catch up when the groove spacing is wider than average. You can't. If the arm is moving at a constant speed and the groove's relative motion toward the spindle is faster, the arm has to either make periodic corrections or keep getting further and further behind. Either we're talking about 2 different things or one of us is wrong.
These aren't the "problems of slipshod design." They are the issues that any arm has to deal with. I'm not saying that your system might not be the best solution to the problem, but to deny that it exists is ludicrous.
Sean, I don't think my example of an off center record is extreme. El tells us his sylus won't deviate from the center line more than .006 inches. That is less than 1/128 of an inch, about the width of the markings on my ruler. Not much. I just tried a number of random records and several of them were easily moving back and forth more than 1/128 of an inch.
Since it pivots it can track deviations up to the .006 limit without any correction. After that it is either making corrections or it is deviating more than .006 inches. You can't have it both ways. |
Herman & El: You guys are discussing something in specific that i'm not really super fond of i.e. servo-driven linear tracker. Based on past experiences, i still believe that such a design has the potential to be better than most conventional pivoted designs.
Having said that, my thoughts are that the servo-driven mechanism is still going to be closer to the center of the groove more of the time than ANY pivoted arm over the duration of the LP. We already know that a pivoted arm can only be "centered" twice on an LP. We also know that this figure is based upon optimal cartridge installation / alignment.
With that in mind, how often do you think that the stylus is going to be centered on a pivoted arm as the off-center record yaws back and forth? On top of that, you have the inertial mass of the arm following that rotational irregularity, supported only by the cantilver of the cartridge up front and the bearing at the back of the arm. This not only causes the arm to follow that motion with NO form of correction, you're introducing more horizontal deflection into the stylus than should be there. My guess is that under these conditions, the stylus of a pivoted arm would be centered a big fat "ZERO" times on a record like this. That's because even when the arm gets to the point where the stylus should be centered, the stylus is probably going to be being pitched about sideways as the record wobbles around its' rotation.
Now how often do you think the linear tracker is going to be out of the center of the groove? Not only can the arm measure and correct for this ( to some extent ), but the arm starts off with the stylus following the grooves rather than pivoting across them? Before answering that, ask yourself how a record is originally cut i.e. with a pivoted arm or with a linear tracking cutting head? Now factor in how that cutting head is driven i.e. with a servo-driven motor.
If common sense has any input into your thought process, the law of averages will tell you that it is with the "inferior" servo-driven linear tracker, flaws and all. Pivoted designs came about because they were FAR more cost effective and easier to design and produce, not because they worked better. Sean
>
PS... Most of the early motorized linear trackers put the money into the mechanics of the table's arm, meaning that the rest of the table i.e. the attention to motor vibration and the platter suspension were highly compromised. The fact that many of the motors driving the platter were also driving the arm compounded the vibration problem, which was already poor to begin with. |
Guys...I guess you still don't "get" the concept behind the Sony arm. Perhaps my explanation was not the best.
The gist of the thing is that the Sony arm IS A PIVOTED ONE. What moves with the linear motion servo is the pivot point. Even if the servo were to "cog" (which it doesn't for reasons I have explained) this would only cause the arm angle (tracking angle) to vary by some tiny amount, at a frequency too low for any sonic effect to be audible. There are no sideways forces on the stylus from the servo.
Herman...perhaps it would help if I drew an analogy to driving a car in steady moving traffic on the interstate. You press the accelerator to a certain point which moves you along with the traffic. Sometimes you ease up a bit, or press down a bit so as to vary the speed according to local conditions. If you are a good driver (servo) these corrections are so gradual that your passengers never notice. In similar manner, the Sony arm smoothly varies the pivot point movement speed, but it never stops or reverses direction, which is what I think you mean by cogging. Because the arm is always moving in the same direction, at nearly constant speed, friction direction reversals, which can destabilize to fine control servo, do not happen.
Sean...The arm tracking is a completely separate servo. Actually, If I remember right the TT has five (count em!) microprocessors. Platter motor control, Fast arm movement, Fine arm movement (while playing the record), vertical arm control (biotracer), and one more to run the show. |
|
El: I know how the Sony arm works and understand how it is allowed to pivot at the rear i.e. it isn't "anchored" at the rear like the older HK / Rabco motor driven designs. Having said that, even the HK / Rabco has enough "slop" in the arm assembly to allow a certain amount of lee-way should sidewall thrust come into play.
As such, i was trying to keep the discussion somewhat "universal" in scope and not just concentrate on that one specific product. There were many other mechanically driven linear trackers out there and that is what i was talking about. After all, the comment pertaining to "why aren't these designs more common in the industry now if they were so good then" was previously used. I was trying to respond to that and at the same time, keep things current with the comments being made about your Sony. Hard to be both specific and general at the same time without some form of confusion arrising : )
Herman: Maybe the Bose servo-suspension is what you need in your car : ) Sean
> |
Sean...By the way, that Bose writeup on active suspension reads as if Bose invented it. Interesting reading, but typical Bose propaganda. Many people have worked on active suspension over at least four decades. GE was doing it when I went to work with them in 1961. |
Bose may not have invented it, but maybe they'll be the first to impliment it in a manner that is both effective and user friendly. After all, with all of the profit margin that they make on their plastic products using the lowest grade materials, they can surely afford to invest in quite a bit of R&D. It's obvious that they aren't re-investing the funds that they make off of audio into improving their audio products, so why not shovel it into another line of products all-together? Sean
> |
Ummm ... I use Wet Platinum. It's a silicone based "personal" lubricant. The stuff is hard to wash off even with hot water and soap ... and it's the most slippery stuff I've ever touched. Millenium ID also works well. And with what you've got left over after applying it to your spindle ....
Seriously, it's what I use. It decreased start-to-speed time, and it reduced the noise floor.
--Paul |
When I purchased my Oracle MK4, there was no bearing oil sent with it. I contacted Oracle, and the brother of the designer, who now runs the company, told me that Mobil 1, 5W30 synthetic, was the oil I should use. |
