[xsb7244]

On Vinylengine I read the thread Some Thoughts on Turntable Design by jrlaudio.  He offers his facts and opinions on proper turntable design.
He may be the Consumer Reports of turntables.  I wish we still had Audio magazine to report about turntables. 

[neobop]

Thanks for posting the info for that thread.  Very interesting stuff.  I'd say it's more about design theory, but I've only read the first page so far.
neo

[xsb7244]

Feel free to disagree with John LeVasseur.  Not everybody agrees with Consumer Reports.  At least they do measurements  of some kind on every product and they offer their recommendations and rankings.  Audio magazine did objective testing but no recommendations or subjective  comments.  With no Audio magazine around there is no lab testing just subjective comments about turntables.  Is the turntable design flawed?
Is the turntable manufactured well?  What are the compromises?  Does the turntable measure well?  To save money did they use a 25 cent part instead of a better $1.00 part?

[neobop]

Consumer Reports?  I've only read his first long post and a number of members comments, but it hardly looks like Consumer Reports.
This guy has a couple of engineering degrees and an extensive history in recording industry and designing electronics.  In this first post he comments on such things as tonearm length, arm/cart resonance and damping, and speed variations.  I agree with everything he says, so far. 
Maybe I'll skip over the members comments and read the rest of his posts.  I saw this late last night and I was too tired to tackle a 13 page thread.

For those of us interested in these subjects, I think this might be a good basis for discussion.  Even if you have a problem with the management of VE, don't cut off your nose to spite your face. 
http://www.vinylengine.com/turntable_forum/viewtopic.php?f=18&t=58779

neo

[J-Pak]

I enjoyed reading his posts. I never considered that tonearm resonance could effect wow and flutter, but it makes sense now.

[watercourse]

Well, I can see how tonearm resonance can create distortion and frequency modulation, but I still can't get with lumping it into the W&F pot... and no reason is given for why it is being lumped even though I brought it up.

Despite the posting, and the fact that I agree with most of what jrlaudio says, tonearm resonance still seems to me to be a separate and different process than motor frequency drift - which you can measure directly and control - and when you do, you do not influence the modulation caused by the tonearm resonance. And vice versa.

Oh well, in the end it doesn't matter, but I'd like to be able to keep it separate even if it's only by definition.

[dlaloum]

It is the scrubbing effect, as the arm moves up and down (or side to side) due to the resonance, the styles is (relatively) moved forwards and back within the groove - effectively altering the speed at which the groove interacts with the stylus.

Given the way in which W&F is measured this phenomena is always lumped in.... if W&F is measured using an external mechanism (such as strobe /or laser) then this factor would not be included.
The traditional means of measurement using a test record with a tone, includes it as part of the measurement, although it is frequently "forgotten".... this aspect of performance was also discussed by Shure in their white paper documents published along with the launch of their damping brush... (ie: damping brush damps resonance, thereby reducing W&F)

[watercourse]

Thanks for the explanation David. If tonearm resonance effects have been lumped into W&F numbers all along, then this is news to me and I've learned something. I guess when I read specs for W&F for turntables, I assume it measures only motor speed fluctuations. If it does lump in tonearm resonance, this sure seems to be dependent on the cart used, compliance, weight of cart, the arm's characteristics itself. In other words, the specs should state what the reference is for the numbers published. For tables without arms, what do W&F numbers mean, just the motor specs, I presume?

Again, it's not that I can't imagine or that I disagree that arm resonance can and does cause frequency modulation, it's just that lumping these effects into an overall W&F number seems not the best way to help diagnose whether the correct course of action is to attack arm/cartridge compatibility, motor issues, and/or perhaps even bearing issues when one has audible W&F...

I can also see why overall W&F performance might be a good measurement of how well turntable components work together as a unit as well.

Just being nitpicky I guess.

[dlaloum]

no you are quite right - I was aware of the scrubbing effect, and of the way of measuring W&F, but until this thread I'd never mentally put the two together....

[watercourse]

It is important to know that arm resonance can affect W&F when as you say the "traditional" method of measuring is tracking a reference tone, and one other thing some of us have learned the hard way: the pressing itself can contribute to W&F when you have off-center pressings or if the wax is not ideally flat...

[neobop]

I think some modern separate tables might measure W&F externally.  It might give them better looking numbers.

I found the other subjects equally interesting - arm length vs alignment distortion vs eff mass.  That one was a doozie, especially for some of the guys around here.  I looked over the second installment - plinth design, vibration transmission and mounting schemes.  Interesting stuff.
neo

[neobop]

We were talking about the assertion that tonearm resonance effects or can affect wow and flutter.  Also the "myth" of dramatically reduced distortion with longer arms.  I meant to get back to this discussion earlier, but it's never too late.  This quote is from the first page of the link above:

"Case in point. Twelve inch tonearms. It's a fad right now. Mostly perpetrated by arm manufacturers who have lost a big share of the market. They needed something to stimulate sales. So began an effort to taut the lower distortion from a twelve inch geometry. If you take this in an empirical sense based in mathematics the distortion figures are indeed lower. They are also more constant (linear) over the playing area of a an LP. These are facts. But the devil is in details.

Many people have changed over to the twelve inch arms and many hear the difference, to a greater or lesser degree. However, most end-users do not have access to thing like Bruel & Kajaer test equipment. Nor do they fully understand the physics involved with all the parameters of changing a tonearm. Some do, but don't draw the correct conclusion from their observations.

Here is a case where a reviewer drew some poor conclusions about the "miracles of 12 inch tonearms", as he put it. I will not mention the author or its publication, not the manufacturers of the products involved. That wouldn't be fair, and could have a liability as well. However, the facts and details are what are more relevant for this discussion. The author wanted to see what improvement a 12 inch made made on his reference system. He carefully picked a 12 inch model for the same tonearm. He talked about how he carefully mounted the arm and the same cartridge he had used on his 9 inch.

Now, he reported several miracles. Of course he noticed a change in the distortion. However he also reported on several other things, like lower noise better bass extension, less smearing of the image. So I took a look at what he did and just ran ALL the numbers. Hear is the real difference.

First, the cartridge he was using was a very low compliance cartridge. The new 12" arm had substantially higher effective mass. There is no mention, in this very detailed article, about him using anything to change the mass in the cartridge mounting. When I looked at the numbers the fundamental tonearm resonance for the 9" version and this cart, this put the figure at about 13Hz. A bit high. (I was amazed this cart/tonearm choice was made in the first place). The 12" arm put this figure at about 8Hz, nearly ideal. So what does this mean from an audible perspective? For one thing, the wow & flutter and drift figures change dramatically. You may be saying, "What?" right now.

Yes, wow & flutter are greatly affected by the fundamental tonearm resonance. It is not solely a measurement of motor noise or drift. It is a description of modulation of higher frequencies by sub-audible ones of which the motor is only one contributor. This is why a wow & flutter specification on a turntable without specifying the arm & cartridge used, are meaningless. As it turns out wow & flutter and the modulation of upper frequencies is something the human ear is very sensitive. H.Saki in his seminal 1970 paper Perceptibility of Wow & Flutter measured the ears ability to perceive as little as .06% wow & flutter on a complex 5kHz tone with a modulation frequency of 3Hz. Most modern tonearm wands are damped. Therefore any harmonics of the fundamental tonearm resonance are effectively deadened. However this does not change the potential for modulation artifacts.

W&F manifest themselves musically in different ways. It can have a negative affect on imaging, causing image instability and shift. Obviously it has an affect on tonal factors, since harmonic distortion is produced. Actually it has a component of inter-modulation distortion as well. This affects tonality as well. Slight changes in W&F are not minor subtle changes, they can be very noticeable and pronounced.

Fundamental Tonearm Resonance also plays a big part in what B&K once referred to as BIM (Bass Intermodulation). A non term, but an appropriate one. It is a the affect caused by high level sub-audible signals boosted by an undamped tonearm resonance. These manifest themselves in the most obvious way as poor bass extension.

Now to tracking error differences between 9" and 12" arms. Even the theoretical change in tracking distortion is so incredibly small as to be almost non-existent from a musical perspective. Rarely is even this theoretical figure achieved BTW. However this the primary advantage taunted for 12" use. The theoretical figure is .09% max difference between the two. Less in practical reality. IT is simply not enough change to explain the huge audible differences reported in this article and by others.

I propose that what we have here is people believing something based on an erroneous conclusion. They have been told to expect a certain outcome and believe what they have been told. In the case of the article mentioned here the author contends it was the wonders of 12" geometry that solely contributed to lower distortion, better imaging, deeper extension and such. He even used the phrase "since all things being equal". When in fact it appears the physics (mostly the eff. mass) of the arm just suited his particular cartridge better. All things were not equal and unaccounted for in his conclusions. The improved tracking error from the arm being 12" was only a small contributor to the reduced distortion and most like had no other effect on the other factors mentioned. And that's by the numbers. Furthermore based on tests I conducted taking measurements on the same turntable/arm and cart combos using B&K test gear.

Sidenote: When doing the tests I also attached an appropriate compliance similar type cartridge to the 9" arm in question. The comparison, both measured and audibly showed very little difference in performance between the 9' and 12" arms. However, when using using the authors setups, the differences where quite readily noticeable between the two arms. So it wasn't the arms, it was the all crucial matching."


On other threads we've discussed longer arms vs alignment error and even seen BS correlations with distortion.  Is it possible that another type of mismatch that favors a shorter arm could be perceived as better results on a longer one?  I think so and the opposite proposition could also be true. 

The assertion of wow and flutter being affected by arm resonance is based on intermodulation. 

neo

[dlaloum]

Great posting Neo - you are on fire today....

The arm motion / w&f relationship has another interesting corollary...

the more limited the vertical motion, the lower the W&F... or rather, the lower/lesser the angular change of the cartridge, the lower the w&f

A longer arm has a reduced angular change with any rises and falls, regardless of whether they are caused by resonance or by warps.

A shorter arm has increased angular change with any rise and fall.

If we keep all things equal (including the parameters relevant to resonance, ie the stylus compliance would need to be varied to ensure resonance was the same in the two differing mass configurations, or the two arms masses would need to be matched by adding mass to the lighter configuration - perhaps by using varied headshells) - or rather as many things as possible equal (there is some weasel space there I know!) - I wonder the degree to which the mere additional length makes a difference due to the reduced vertical angular change....

And whether that is a fundamental flaw in my Revox Linatrack arm (ultra ultra short 2" arm).

Measuring w&f as well as IMD are things I have never played with - possibly another exercise for a weekend when I am feeling bored...

bye for now

David

[watercourse]

I'm beginning to like this ultra-condensed thread of the VE thread better than the original...

[neobop]

This is the stuff germane to our previous discussion.  Like some others, I never really considered intermodulation to be related to wow and flutter, but if you think about it, it fits.  Think about it some more and why is 8 to 12Hz more desirable resonant frequency than 6Hz or 14Hz?  Resonance anywhere down there could have intermodulation consequences.  The answer we're usually given is that resonance above 12 or 13Hz is close to the audible band and is more likely to have consequences in that band.  Likewise, resonance below 8Hz is more likely to be exacerbated by warps. 

Everyone is pretty much in agreement about resonance approaching the audible band.  This is a situation encountered when you use a low cu cart on a light arm.  Not everyone agrees that a low resonance is undesirable.  Some high cu carts are exceptional trackers.  I guess results depend on the amplitude of resonance.  Arm damping tends to reduce the amplitude and spread resonance over a wider band of frequency.  The problem I hear with most hi cu carts on a heavy arm is sluggish response. 
Another aspect is lateral compliance/resonance.  All these calculations are for vertical dynamic compliance.  Most carts have similar vert and horiz cu, but many are somewhat different.  When you consider that eff mass is the same as moment of inertia (MOI), it effects the cart/arm being able to respond to change in a timely manner.  It seems like a contradiction in terms, but the arm has to be a stable platform for the cart, but at the same time respond to the tip/cantilever constantly dragging it across the record groove.

Arm length must necessarily effect warp behavior, which is described as warp wow.  This is the sound from the change in SRA as the cart goes up/down a warp.  From geometry, if you think of the arm as the hypotenuse of a right triangle, and the record surface and arm pillar as the other 2 sides, it's not hard to visualize.  As you increase the length of the arm any displacement from the record surface should have less affect than on a short arm.  This would make an interesting problem to calculate, but it's not as simple as it looks.  An imaginary line from the pivot to the needle is the actual hypotenuse.  The top of the cart is suspended above and it's that angular displacement of the cart that's the culprit. Still, you'd think SRA would be changed less with a longer arm.

I've never done any related math, but for some reason this seems to be less of a problem than you might think.  Back in my day I sometimes worked with Souther linear arms.  This is the arm that CA bought and became the prototype for their arms.  It has a very short arm tube.  Record warps didn't seem to be an additional problem, depending on the cart.  Moderate warps seemed no different than on other arms.  I didn't own one of these arms and I suspect there was a little difference and some of that could be in favor of a shorter, more responsive arm that can recover much faster. Long arms tend to have much higher MOI because it's also a function of length. 
neo

   

   

[*Scotty*]

The linear arm on my Maplenoll Ariadne has a length of 7.25 inches and combined with my AT 440ML seems to have very little trouble with moderate record warps. I'm not sure what defines a severe warp, but I have one classical record that is somewhere near an 1/8th inch in vertical displacement. I think one of the things that might characterize a severe warp could be the abruptness of the vertical displacement along with how much area is under the curve.
 Linear arms may tolerate warps a little better than a pivoted arm due to the relative separation between motion in the vertical and horizontal planes when compared to a pivoted arm, there is no motion around a pivot point describing an arc in the horizontal plane.
Scotty 

[neobop]

I don't understand calculations for linear arms.  I think there is a way to convert MOI to parallel motion, but how does it work?

On another forum I got into a discussion with someone who has a Terminator arm.  This is an air bearing linear tracker.  They say the weight of the arm including the sled is the lateral effective mass. It gets stranger from there.  Check this out, toward the bottom of the page:
http://www.trans-fi.com/terminatortonearm.htm

"Using pivoted arm criteria to measure an air bearing in a parallel tracker is about as useful as road test data for cars being used for hovercraft. So, the below is an attempt at an explanation:

The inertias of the vertical component is different from the horizontal component. Horizontally, because there is no pivot, the inertia is equivalent to its mass, in this case ~ 80g. The vertical component, by design, is approx 25x less, ie ~3g. Horizontal resonance is typically between 12-16hz, vertical 5-7Hz.

Vertical inertia must be as low as possible, & the horizontal is already high as per Ladegaards recommendations. To a certain extent, the air pressure being used can influence all of the above."

If the vertical eff mass is 3g, how can vert resonance be typically 5 - 7Hz?  The air pressure must dramatically increase vert eff mass.  This is a terrible explanation.  Instead of telling us how calculations work they make it look like magic or voodoo. 

Is 12 - 16Hz horizontal resonance desirable?  None of this makes sense to me and the link they provide doesn't work.  Instead, they fill up the page with "beauty" shots of this erector set looking arm.  Why in the world would I want horiz resonance 16Hz?     

neo

[dlaloum]

That also does not cover all the linear arms that are effectively a servo driven pivot arm with the pivot shifting on a linear track..... which I think is most of the 80's LT's (the good as well as the bad ones)

[*Scotty*]

They were correct in stating that there are different effective mass numbers. The effective mass of the arm in the horizontal plane will most likely be different from its effective mass
In the vertical plane. How much they differ will depend on the arm design.
 I personally don't see how a virtually frictionless air bearing is going modify the arms resonant behavior in either plane. Ideally the horizontal resonance would quite far away from the frequcies associated with off-center holes and still comfortably outside lowest frequency seen on the record. The vertical resonance should lie at about 8Hz just like a pivoted arm. It would have been nice if they would have shown how they mathematically calculated the two different resonant frequencies. They didn't "show their work" and that's a problem.
 Their arm, like the ClearAudio arm, has a low vertical polar moment of inertia and possibly low vertical effective mass as well which makes it much easier for the cartridge to change direction in the vertical plane in response to a record warp without unduly compressing the cartridges suspension which can result a rebound which tosses the stylus out of the groove. The cartridge can perform like a sports car on the hills rather than a semi-tractor  trailer rig.
Scotty

[neobop]

They were correct in stating that there are different effective mass numbers. The effective mass of the arm in the horizontal plane will most likely be different from its effective mass
In the vertical plane. How much they differ will depend on the arm design.
 I personally don't see how a virtually frictionless air bearing is going modify the arms resonant behavior in either plane. Ideally the horizontal resonance would quite far away from the frequcies associated with off-center holes and still comfortably outside lowest frequency seen on the record. The vertical resonance should lie at about 8Hz just like a pivoted arm. It would have been nice if they would have shown how they mathematically calculated the two different resonant frequencies. They didn't "show their work" and that's a problem.
 Their arm, like the ClearAudio arm, has a low vertical polar moment of inertia and possibly low vertical effective mass as well which makes it much easier for the cartridge to change direction in the vertical plane in response to a record warp without unduly compressing the cartridges suspension which can result a rebound which tosses the stylus out of the groove. The cartridge can perform like a sports car on the hills rather than a semi-tractor  trailer rig.
Scotty

Not so fast Scotty,
If Terminator vert eff mass is low, why is typical vert resonant freq 5 - 7Hz?  I think they used to recommend hi cu carts. With a pivoted arm this res would tend to be with a med/high mass arm.  Vert arm mass was just reduced with new wand to 3g!!  One of 2 things is happening here.  Either air pressure greatly effects vert mass or we don't have a clue about mass/res with a linear air bearing.  ET has a longer arm if I remember correctly, and air pressure is lower, 30 PSI (methinks) opposed to 45PSI?  ET works great with lower cu carts.  In the vert plane PSI must effect MOI or resistance to change or this makes even less sense than before.

The guy on another forum was playing with arm mass getting rid of buzz and sibilance with problem carts on Terminator.  I'll see if I can locate exactly what he said.  Looking at physics sites, parallel axis of MOI doesn't seem to apply.  With low mass mechanical linear arms, VTF and trackability seems to trump res considerations.  David, have you measured res freq on the Revox?
neo