2008 RMAF – – – all things analog.

Back in 2002 Audio Asylum correspondent Klaus discovered a letter in a 1967 issue of the Journal of the Audio Engineering Society where RCA reported some measurements they had made of speed variations caused by stylus drag. Using a metal master, the RCA engineers measured a 0.078% speed change when playing a silent groove and a 0.079% speed change with a modulated groove. The speed change was even lower on a vinyl record, 70% of that measured with the metal disc. Since the difference between playing a modulated and a silent groove on a metal master was 0.001% (0.079% - 0.078%), and a vinyl record reduced that difference to 70%, the real speed change caused by a modulated groove would be .7 x .001% or 0.0007%. That's 2/10000 of an RPM for a 33 1/3 RPM LP. I really doubt that anyone can hear the difference between 33.3333 RPM and 33.3331 RPM. In any case that speed change caused by stylus drag would seem to be buried beneath the wow and flutter of even the most stable turntables.

Of course those measurements are over 40 years old now and I suppose that today's much better stylii might have even lower drag and speed change. It would be nice to have contemporary measurements of the phenomenon but until someone repeats the RCA experiments stylus drag would seem to be inconsequential.

Robdoorack,

Interesting stuff, but not necessarily conclusive of anything. Read the posts by Teres and Thom Mackris just above (well, some of Thom's anyway!). Depending on the time span over which "speed" is measured, the effects of stylus drag may produce no measured difference at all, yet may still be audible.

Consider this analogy:

A. You decide to time me running laps around a track, but the seat you chose to observe from can only see the start/finish line. It has no view of what's in between.

B. You observe that I pass the start/finish line precisely once each minute, so you conclude (correctly, from your perspective) that I'm travelling at a steady 4 minutes/mile clip (assuming a 1/4 mile track).

C. What you don't know, because your chosen vantage point doesn't let you see it, is that half the circumference of the track is actually a foot deep pool of water. This slows me down to 8 minutes/mile speed, but I make it up by blazing through the dry half of the circuit at 2 minutes/mile speed.

D. Your limited resolution of measurement (you can only see and measure in whole laps) leads you to the false conclusion that I'm running at a steady pace. The reality is that my speed is varying all over the place. Only the long term average is steady.

E. A turntable with a motor/drive system that allows deceleration due to drag, but then re-accelerates to faster than average speed when the drag is removed, could easily maintain a perfect AVERAGE speed of 33 1/3 rpm, while producing audible or even horrible sonic speed changes that a once-per-revolution measurement would never detect. A longer period of measurement would be proportionately less likely to detect them.

So, the experiment quoted was vaguely interesting, but proves virtually nothing. The human ear is vastly more capable of detecting short term speed changes than the crude experiment you described.

Doug

P.S. If anyone ever observes me running 4 minute miles, please let me know!

Robdoorack, Thanks for the interesting data. The measurable deviation cited was more than I would have expected. As Doug points out the measurement time period is very significant. Most people are not able to detect relatively large errors in average speed. For the majority of us the threshold is more than 0.1% and nobody can detect a 0.0007% error. However short term deviations are a different matter. It is well documented that digital jitter of 10 picoseconds is audible. That's a short term deviation of 0.000000001%, one billionth of a percent! So it should not surprise us that a short term analog speed deviation 10,000 times greater would be audible.

Wow some interesting reading. The speed issue has been with us from the beginning of vinyl playback right. We all have accepted the fact cause were here. I am in constant adjustment between lps. Observation done with a KAB strobe on top of my center weight as I am able to check when playing a lp. I am tuned to hear pitch changes and thats the reason for the constant checking and tweaking. All of the reasons are known for this yet there is no drive system that will remove our drag problem that i am aware of.

In the end though the sound of vinyl play back is so far superior to digital that i can live with the drag in my hearing. IMO

No pun intended ha.

yet there is no drive system that will remove our drag problem that i am aware of.

hummmmmm.

as far as stylus drag/groove modulation.....i think that the Rockport Sirius III's drive system may solve this issue. you have an pure induction direct drive motor with zero torque ripple, a 55 pound platter, an air bearing, and a servo system (coaxially mounted optical encoder) which samples the speed every .000001 of a second.

the only support i have for my opinion besides the design description is the unique way that the Rockport handles musical peaks to my ears. until one hears it on a familiar Lp it's easy to dismiss the signficance of it.

a side note on the servo; when it is engaged a green LED flashes to red. the servo is never engaged unless the 'floating 250 pound plinth' is bumped or if someone jumps up and down on the floor in front of the tt. once the platter reaches speed it stays there. there is no hunting occuring. stylus drag does not cause the servo to engage on the Sirius III that i have observed. maybe the servo engages without the LED flashing, but i don't think it works that way.

i have a Technics SP-10 Mk2 sitting next to the Rockport which does a very good job on this issue.....but not at the Sirius III level of naturalness and continuousness. of course; there are more differences between the SP-10 and the Sirius besides stylus drag to cause these differences.

there may be other tt's which are able to do this same (stylus drag/continuousness) thing equal or better but i've not heard that.

anyway; stylus drag can be solved.....but it's not easy to do.