For the beach bum http://www.americanradiohistory.com/Archive-Wireless-World/50s/Wireless-World-1957-05.pdf Page 228 Barlow : Limiting factors in gramophone reproduction.
part 2 http://www.americanradiohistory.com/Archive-Wireless-World/50s/Wireless-World-1957-06.pdf page 290
part 2 http://www.americanradiohistory.com/Archive-Wireless-World/50s/Wireless-World-1957-06.pdf page 290
Last edited:
Thanks Bill, the rest of the publication is interesting from a historical perspective.
But, as to Barlow's article, it considers only static indentation, properly done IMO. And acknowledges "As the record moves under the stylus the system is not the same as the static indentation case so far considered", but does not go on to discuss it.
IMO it nicely illustrates a classic historic error: to consider static deformation applies to the dynamic situation in vinyl playback. And all discussion that follows, based on that misassumption, might be well-thought-through and self-consistent, but isn't relevant because static deformation isn't at issue, and I venture can't happen for reasons which have apparently been overlooked........
LD
It would seem that if a stylus was moving relative to a record groove, for there to be no vinyl deformation at all, the stylus would have to be moving faster than the speed of sound in vinyl? Otherwise, its hard to see how there could be no deformation (although any deformation could be small in a dynamic case, relative to the static case).
I only scanned the papers and need to read them properly, but one thing stood out. Barlow considers plastic deformation as real based on his static measurements then goes on to surmise that, once deformed the damage is done and will neither recover nor get better. So the first play does the harm and after that subsequent plays make no difference. This of course was the days when men were men and VTF was 8g. But this should be testable, at least if you could borrow an ELP to do contactless scans before and after.
That was a ways back, but it would be useful to know what magnitude of effect you are looking for. I have played the same cut twice or more numerous times and never heard a difference. For that matter isn't this standard practice in many situations, equipment comparisons using LP's would be rendered useless.
The test is made more difficult in that no two consecutive plays can be perfectly aligned.
Actually I have a precise understanding of where this heresy or it's denial will lead to. It's SFA.
We aren't in a position to change the vinyl properties in a record. I see the discussions here on fat tube cantilevers etc and am willing to bet thiis is unlikely to result in better cartridges ... or better vinyl sound, which is of course the true aim of the exercise 🙂
Err.rh! The whole idea of an EXPERIMENT is that the RESULTS are used to modify your view of the world. It's called EVIDENCE
Let's get back to my proposed experiment. It's simply to play & record the Frequency Sweep track of a Test Record on a 'modern' cartridge ... and then repeat this immediately afterwards.
We are simply looking at the Frequency Response. One of the most obvious outcomes is an either/or result. (There are other outcomes related to THD and if you have a SEM but we have the technology to look at FR conveniently)
- the response is unchanged on the 2nd sweep
- the response changes on the 2nd sweep
If the result is [2], there are several possible explanations. The 2 most likely are
- the first playing has excited the lucky particles in the vinyl such that though it is physically unchanged, the lucky emanations react with the diamond stylus differently the second time round
- the vinyl had been physically deformed, ie it has compliance which is plastic in the short term and hence also has a damping coefficient
Then this deformation must have taken place under the dynamic conditions of playback and the compliance & damping of the vinyl has some effect on the stylus, cantilever, cartridge bla bla. ... and the stylus, cantilever, cartridge bla bla has some effect on the vinyl.
Note that if the result is [1], it does not preclude the significant effect of vinyl compliance ... it just suggests the deformation is elastic.
But if the result is [2], I really can't see how you can preclude the effect of significant vinyl compliance, damping bla bla ... unless you believe in lucky particles 😀
Actually, Don, like a true scientist, changed his mind as new EVIDENCE is gathered. In 1978/9, IIRC, with new cartridges, he said the deformation would recover. What I can't remember is whether he was referring to his specially cut lacquers or our CBS and B&K Test Records.billshurv said:
Decca's John/Jean Walton showed, with some of the earliest SEM pics of record playback in the early 70s, that with a tip mass of <0.5mgm, the groove is actually polished on first play.
And don't for a moment think the old gurus had no idea of cantilever resonances & bending modes. The Decca FFRR cartidges used a clamped steel cantilever and its modes provided practically all the mechanical properties of the cartridge. These & the effect on response, trackability bla bla were agonised over by those versed in the art in those days.
Last edited:
mea culpa, mea culp, mea maxima culpa.
I should clarify .. my sparring with lucky over denial of vinyl compliance is simply "my d*ck is bigger than yours" sh*t stirring. The practical significance and what this will lead to is SFA 😱
But I am sad that lucky ... one of the few people who could carry out my experiment .. prefers the sh*t stirring to actually trying something (which might disprove one of his pet beliefs 😀 ).
I'm not really interested in a SFA outcome.
_______________________
What I really hope to get out of the EXPERIMENT is guidelines on how to measure a cartridge 'accurately' with our present resources.
I hope to use this measured response to modify Guru Wurcer's DIGITAL RIAA EQ, both his EVIL FIRs and our IIRs, to incorporate correction.
The desired outcome from this is better vinyl sound.
_______________________
And thanks Bill for digging up that wonderful site and those issues. Besides Don Barlow's words of wisdom, those issues have several other iconic articles including Great Guru Baxandall's stuff.
I should clarify .. my sparring with lucky over denial of vinyl compliance is simply "my d*ck is bigger than yours" sh*t stirring. The practical significance and what this will lead to is SFA 😱
But I am sad that lucky ... one of the few people who could carry out my experiment .. prefers the sh*t stirring to actually trying something (which might disprove one of his pet beliefs 😀 ).
I'm not really interested in a SFA outcome.
_______________________
What I really hope to get out of the EXPERIMENT is guidelines on how to measure a cartridge 'accurately' with our present resources.
I hope to use this measured response to modify Guru Wurcer's DIGITAL RIAA EQ, both his EVIL FIRs and our IIRs, to incorporate correction.
The desired outcome from this is better vinyl sound.
_______________________
And thanks Bill for digging up that wonderful site and those issues. Besides Don Barlow's words of wisdom, those issues have several other iconic articles including Great Guru Baxandall's stuff.
Last edited:
Well, deformation can't be absolutely zero, you are right, but rather be so small as to be negligible that it might as well be, IMO.
What determines dynamic indentation is stylus profile, especially how sharp the contact region is in the direction of record movement, and the relative velocity of the vinyl material to that dimension. And downforce, of course.
If one started with static indentation and a still record, then very slowly sped up the record, albeit to a fraction of normal speed, initially some sort of 'bow wave' would form ahead of the stylus. This bow wave would propagate around the stylus, but not at the speed of sound within vinyl: rather at a wave velocity determined by surface factors and material properties, much as a ship's bow wave does not propagate at the speed of sound in water.
If the record speed is slowly increased, the bow wave would become progressively less deep and more aligned with the groove, until the depth effectively became zero and the bow wave effectively disappeared. The stylus is then sliding along the surface of the vinyl, effectively without deformation, and IMO this happens well below normal playback speeds.
There's always a stress pattern associated with stylus contact, and this moves with the record, of course. From the record's point of view, changes to the pattern propagates within it at the speed of sound, but the pattern moves at the speed of rotation of the record. Stress doesn't necessary mean yield though, however strictly the yield can't ever be absolutely zero. Small enough to be negligible, and certainly so relative to flex yield of a cantilever, IMO.
LD
Last edited:
Okay, thank you. That sounds pretty sensible.
So, even though the physical process in phono playback is actually somewhat complex, you seen quite sure it must be negligible?
I guess what has me wondering is that some people can sometimes hear and find objectionable very small distortion in, say, amplifiers, even though they are listening through speakers with much more distortion. Apparently, in that case the sound of different kinds of distortions involved can somehow be picked apart and distinguished by poorly-understood brain processing.
Given that observation, how do you know that small vinyl-deformation-related distortion products can't be heard in a way somewhat analogous to amplifier distortion, that is, despite whatever other higher level distortion is present?
EDIT: By the way, when I mentioned the speed of sound in a medium, I was thinking about some kind of limit with which molecules can effectively bump into each other and propagate a mechanical wave front. Even if the bulk behavior, as enough molecules are brought into motion, is to form a bow wave over a slightly longer time frame.
So, even though the physical process in phono playback is actually somewhat complex, you seen quite sure it must be negligible?
I guess what has me wondering is that some people can sometimes hear and find objectionable very small distortion in, say, amplifiers, even though they are listening through speakers with much more distortion. Apparently, in that case the sound of different kinds of distortions involved can somehow be picked apart and distinguished by poorly-understood brain processing.
Given that observation, how do you know that small vinyl-deformation-related distortion products can't be heard in a way somewhat analogous to amplifier distortion, that is, despite whatever other higher level distortion is present?
EDIT: By the way, when I mentioned the speed of sound in a medium, I was thinking about some kind of limit with which molecules can effectively bump into each other and propagate a mechanical wave front. Even if the bulk behavior, as enough molecules are brought into motion, is to form a bow wave over a slightly longer time frame.
Last edited:
My opinion is based on original calculations and modelling from a few years back now. Can't find them in my archive right now..... IIRC examining bulk elastic and loss moduli for vinyl, self-mass, impedance and losses arising, and available energy and force.
This is only original because, AFAIK, a dynamic deformation model has never been published: in the past it seems to have been consistently assumed that static models apply, which IMO is easy to show can't be true.
However, most of my stuff on this was reductio ad absurdum, aimed at showing contradictions if the static deformation model applied. But it also showed that dynamic deformation must be negligibly small.
If I can't find it, I'll do it again, if there's time and interest?
It's the moduli you're looking for, Markw4. Which arise from molecular behaviour, but no need to go there, fortunately !
LD
Last night whilst waiting for my platter bearing to descend (takes 20 minutes, but given hadn't cleaned and lubed it since 1999 could wait) I started wondering about dynamic (ie. modulation driven) deformation. As the static state is in equilibrium then as soon as you start tracing the groove then its just compliance at play, so we only need uN to move the stylus. I need to run the numbers but these appears to be an order of magnitude less than the VTF so it cannot add significantly to the pressure on the vinyl from the deflection.
Inertia would seem to be a bigger problem (especially warps).
Inertia would seem to be a bigger problem (especially warps).
I use the terms 'static' and 'dynamic' to mean the record is stationary, or moving, respectively.
Hope this hasn't caused confusion.
In an unmodulated groove, the only forces on a stylus are downforce and sliding friction, which might be considered 'steady state' or static, but I would dispute that friction is constant, rather there is a significant random element to it. So still not without controversy, but we can all get the gist I think. The stylus is not meant to deviate from the straight and narrow.
In a modulated groove, the stylus is compelled to move by the locus of groove walls. The cartridge in total provides an impedance to moving the stylus, ergo there is an applied force via the cantilever whose reaction is applied to the stylus via the groove walls, which are at 45 degrees, so have an upforce component. Then the net downforce is always depleted by modulation, and if depletion exceeds downforce the stylus loses contact with one or both groove walls: mistracking.
The nature of the cartridge mechanical impedance depends on the motion involved. High acceleration is dominated by inertia, high velocity by damping, and high displacement by compliance.
LD
The reason I though the warp case was the extreme was because, even in high acceleration, you are only accelerating the cartridge moving parts and the arm just looks like a mass. Over a warp you have to move this mass and particularly on the down stroke you would expect VTF to have a large peak. Here is where you would look for deformation?
Another thing that I didn't see in the archives that had survived the cull.. DJs talk about vinyl burn (and accuse ortofon of being worse than Shure for that). If your sliding model is correct then the burn could be caused by too slow motion, not too fast?
Another thing that I didn't see in the archives that had survived the cull.. DJs talk about vinyl burn (and accuse ortofon of being worse than Shure for that). If your sliding model is correct then the burn could be caused by too slow motion, not too fast?
I wouldn't take money off a beach bum anyway😛. I am just willing to invest a small amount to be able to test the theory. If it turns out to be wrong very little lost in the grand scheme to get some data. In fact a win win.
If we actually find out that its all fluff and the Shure M44-7 is as good as it gets cos you need DSP to actually fix things then it's still all good.
But we can change the properties of a cantilever; either by selection, or by Frankencartridge DIY......
And, IF rigid cantilevers are what we're after, then fat tubes present the lowest mass way of getting there, in a given material. That's a fact. A hollow fat, thin walled Al tube would beat most current and historic exotics for rigidity/mass........so IF the exotics genuinely outperform fat Al tubes, we need another explanation as to why, beyond rigidity/mass ratio........
LD
Last edited:
Yes, warp increases downforce, sometimes significantly. The record rising is the issue, and sometimes on warped records from the days of 10g+ nominal tracking force, there can be an audible burn from previous plays on the rising warp.
But in practice, downforce can't often go much above x2 VTF, because, on the way down from the warp, there needs to be positive downforce or the record will skip. So these days, warps don't cause burns unless VTF is insanely high, such as when the headshell is held by a DJ perhaps?
Warp forces aren't so different in magnitude from other extremes of normal programme material induced high stylus force though, just that old warp burns are long enough and characteristic enough to notice, methinks.
As to burn at slow speeds from startup, AFAIK it doesn't happen? Perhaps the transition to sliding is graceful, rather than ripping.....perhaps the transition happens at very slow speeds indeed.....AFAIK resting a stylus in a silent groove doesn't cause a click either......
PS: Just ran a rough test. Silent lock groove on never played record. 3g VTF, S-120 stylus lowered on to record, then rotated very slowly by hand over 30s for 1/4 revolution. Then lifted off. Then played normally, and listened to for signs of burn. result: normal silent lock groove, no audible sign of burn or changes at all.
LD
Last edited:
LD, thanks again for your cogent explanations.
Although, I don't listen to records often, what has been discussed here so far is kind of fascinating from a scientific perspective. The argument that static models are inaccurate seems plausible to me.
It might also not be much of a stretch, IMO, to wonder to what extent simple dynamic models are likely to be accurate. To make an electrical analogy, the concept of dielectric constant if fine for many purposes, but it turns out to be not all that constant in some situations and for some purposes, such as at high frequencies, with temperature variations, applied voltage, etc.
In the case of vinyl, and for claims there is some kind of semi-elastic/semi-plastic deformation that mostly recovers over time, there would be a question as to the existence of some recovery time constant, assuming it would be reasonably model-able as a constant in the case of playing a record, given factors such as the forces involved and the structural geometry of vinyl supporting the contact area. In other words, if there were to be slowly recoverable deformation, how long would recovery be expected to take, and to what extent would it be dependent on groove shape, etc. Under what circumstances would it be expected to most audible and/or most measureable?
Although, I don't listen to records often, what has been discussed here so far is kind of fascinating from a scientific perspective. The argument that static models are inaccurate seems plausible to me.
It might also not be much of a stretch, IMO, to wonder to what extent simple dynamic models are likely to be accurate. To make an electrical analogy, the concept of dielectric constant if fine for many purposes, but it turns out to be not all that constant in some situations and for some purposes, such as at high frequencies, with temperature variations, applied voltage, etc.
In the case of vinyl, and for claims there is some kind of semi-elastic/semi-plastic deformation that mostly recovers over time, there would be a question as to the existence of some recovery time constant, assuming it would be reasonably model-able as a constant in the case of playing a record, given factors such as the forces involved and the structural geometry of vinyl supporting the contact area. In other words, if there were to be slowly recoverable deformation, how long would recovery be expected to take, and to what extent would it be dependent on groove shape, etc. Under what circumstances would it be expected to most audible and/or most measureable?
Last edited:
Don Barlow circa 1978 said about 24hrs and it would be measurable on a Frequency Response sweep. He wouldn't have said this if he hadn't tested it. He wasn't just a polymath & scientist but also a most careful researcher.
But as the report is a Jurassic memory from a beach bum on the other side of the world, it would be wise to check this out. 🙂 Much easier to do today compared to the late 70s
Would you be willing to invest in a small amount to persuade George to do my Experiment on one of his Frequency Sweep records .. perhaps with your M97X? Maybe the cost of a Test Record if the 3rd measurement 24 hrs later shows permanent deformation?
I can't promise a win win as Result [2] of from this Experiment would kybosh da Deniers of Significant Vinyl Compliance/Damping/Deformation. 😱
But it would get you closer to a fully integrated Digital RIAA EQ. Presently only you & George seem to be interested in this.
Actually this is similar to the situation for loudspeaker diaphragms.
I've done a LOT of work on this including Laser Holography, the earliest Scanning Laser Doppler Velocimeter studies & FEA for both treble units & cones.
Aluminium domes have a distinct advantage over Titanium, Diamond-like Ceramics and even Beryllium. All BTW from practical tests on real experimental samples.
Flexural Stiffness is to the 3rd power of thickness which gives a big advantage to lighter materials. A hollow cantilever favours the 4th power of diameter but its more complicated cos you need to consider wall thickness bla bla .. Anyway, its all in Structural Engineering 101
BUT the End Conditions are VITAL .. Vinyl Compliance being the important one for cartridge cantilevers.
For cone materials, the winner isn't carbon, kevlar etc. It's Unique Fibrous Material (aka paper) .. its only disadvantage being a certain lack of consistency.
I specified & had made an Engineered Plastic to try & replicate the properties of UFM, which we used in our cones for many years. It worked well up to a sensitivity of about 90dB/W@1m. Above that loudness, UFM still reigns supreme.
Yes please lucky.
I don't disbelieve your mythical purged calculations.
I just don't believe SHURE didn't check out fat tapered Al cantilevers in their modelling. It's a pretty obvious route ... especially if you start trying to model this stuff properly.
Last edited:
The M97 is George's as well. I haven't seem him around in the last week, but sure he will appear like the shopkeeper (sorry for those who never saw 'Mr. Ben).
The part that is confusing me is the order of magnitude we are looking for without a reference to his later research. we talking less than 1dB or lots of dB and at what frequency does it start.
The part that is confusing me is the order of magnitude we are looking for without a reference to his later research. we talking less than 1dB or lots of dB and at what frequency does it start.