Belt stretch

Jean,

You were doing okay until you said, "a rubber wheel grips without deformation".

Pardon me, but that's nonsense. Rubber is an elastic material. Elastic materials ALWAYS deform under load, then rebound back toward their original shape when the load is reduced - that's what "elastic" means.

Another aspect of this is that energies stored in an elastic material by compressive deformation are released with a time delay when the material is allowed to rebound. Depending on the TT design, this energy normally feeds back into the platter at a fairly low frequency. The cartridge picks this up as "mud" - nothing you can specifically hear, but a lack of clarity and some opaqueness to low level detail. Very TT specific, but a real engineering problem nonetheless.

If you believe a rubber idler wheel (or a rubber anything) is not deforming, you're fooling yourself. The amplitude and frequency of deformations may be adjustable by choosing rubber of different durometers, but they can never be eliminated.

That said, we're in agreement that Rccc should listen for himself. He'll hear what he hears and can make informed TT choices based on his own sonic sensitivities and musical priorities.

Geeze, I cant believe there's not one other belt slip skeptic out there. So you dont think that the idler sound may be attributable to something other than speed like say damping or ? I still havent heard anyone actually talk about a measured deviation in the belt system or if the speed is spot on wouldnt the differences in sound be due to something else or what resolution is audible (inaudible). What Im proposing is that the difference in sound may be due to some other forces. As I said before I could hear a difference in sound with different belt tensions while the speed remained stable. Although I follow the logic, bass extension and dynamics could be enhanced by other forces. Where the hell is HW when you need him?

Again...

You aren't hearing from other skeptics because your argument is with pure science, not conjecture. Belt creep is a matter of sheer physics; it isn't a theory. That said, there are some very ingenious workarounds for the problem. Frank Schroeder has one that addresses belt creep and implements noise cancellation simultaneously. I heard his turntable at some length, and it works. There are others who offset or minimize the issue in their designs, too. Whatever the drive system, it is the implementation of it that separates the men from the boys.

Doug is also correct in his assertion that rubber deforms. The key to success here is in choosing an optimum footprint, density, pressure, physical configuration, and mass of the rubber, so that it does its job in the least invasive way. This means that the rest of the turntable has to be designed in keeping with that aspect. It can be done, however.

The bottomline, I suppose, is that the type of drive isn't quite as important as the makeup of it, and if that makeup includes slip, there's your first obstacle to overcome.

Mosin, Im not arguing that belt slip exists or that rubber deforms or that there is some theoretical friction from stylus drag. Im questioning the audibility of these conditions and other than "I can hear it" and attributing "it" to the afore mentioned conditions no one yet has demonstrated that this so far unmeasurable speed variation is responsible for the effects they are hearing. If the speed measures stable with a strobe then what pitch related anomaly can be heard? When John hears better bass extension and dynamics with his idler compared to belt but both systems are showing no deviation in speed perhaps its not speed shift that he is hearing.

If the speed measures stable with a strobe then what pitch related anomaly can be heard?

1. Any pitch-related anomaly whose time duration from delay through recovery is shorter than the time that passes between two marks on the strobe disc. This would be a greater risk with low resolution strobes (ie, fewer marks per circumference).

2. Any pitch-related anomaly which slows the platter by precisely one (or any other whole number) strobe mark in 1/60th of a second (1/50 in some other countries). This would be a greater risk with high resolution strobes (ie, more marks per circumference).

A strobe isn't perfect. It's a an unnumbered, circular yardstick with markings giving it one very specific resolution. Changes that are completed below that resolution won't be detected. Changes that cause a whole number shift in mark position won't be detected.

You haven't responded to Mosin's or my suggestion: listen to a good quality belt drive table properly set up with a variety of belts. Hear the differences. Supply some explanation other than speed variation for the differences you'll certainly hear.