Direct Drive vs. Idler Drive vs. Belt drive


I'd like to know your thoughts on the strengths and weaknesses of each drive system. I can see that direct drive is more in vogue over the last few years but is it superior to the other drive systems? I've had first-hand experiences with two out of the three drive systems but looking to learn more.
scar972

Showing 17 responses by atmasphere

Can you get parts for older lathes and keep them running at spec? Is a high quality modern turntable quieter than a lathe?
With my lathe its helpful knowing a machinist :)  No worries keeping it at spec. I think the Technics SP10 Mk3 is in the same ballpark of speed stability; most turntables are not. But the Technics motor was meant for LP mastering anyway.


With regards to surface noise I agree most of that is in the groove, not the turntable. The faults of turntables seem to be speed stability (heard as a shimmer in the soundstage; if you get to play reel to reel you'll understand what I mean right away) and resonance in the plinth, tonearm tube and platter. With regards to the platter, it has to be dead, and also control resonance in the LP itself- which is not a trivial matter! While the former seems to be understood in high end machines, the latter seems to get little notice, even though its just as important.


The surface noise is mostly an artifact of the pressing plant, although I do notice when the mastering engineer speeds up the cutter threads to cut the leadout grooves at the end of the LP. Otherwise though I simply don't get rumble despite the electronics having full bandwidth to 2 Hz.
By the way, most of the vintage lathes are worm drive.
Furthermore, DC motors have zero torque at constant speed.
@dover  These two statements are false. I own a Scully lathe which is typical of what was used to make many older recordings. Its not worm drive. It employs a 1/8th horsepower syncronous motor which drives an anti-vibration coupling which then drives a transmission (which allows for shifting from 33 to 45 rpm). The transmission employs bevel gears in an oil bath but no worm drive. Its output is a shaft which drives another vibration isolation coupling; that in turn drives the platter directly, which rests on a set of carefully machined bearings (which require frequent attention).


DC motors have torque even at constant speed, so long as there is a load. This link might be helpful:

https://www.motioncontroltips.com/torque-equation/
However its a bit of a misnomer to equate a servo controlled multi-pole motor with a simple DC motor, and its also worthy of note that AC motors have a similar property of lockup being the point of highest torque output (as I'm sure you know quite well). I'm guessing that what might have been conflated/misconstrued in your comment above is the fact that when a motor has no load, its **current draw** is its lowest. But I think you'll find that how much torque it has is another matter altogether! Certainly a motor at constant speed will be making torque...
Its one thing to 'scope the waveform that is driving the servo operation, its quite another to see how the system responds to that waveform. I very much doubt that the platter speed variation can be represented by a triangle wave!
I’m guessing that AR didn’t believe there was a market for a more expensive model, given that the AR turntable was all about outstanding value for money. In fact it would take decades before it could be beaten on that score.
I not sure I agree with this... budget turntables have always been problematic. Empire was making better machines back in 1961, but they might have cost a little more. Its certainly telling that an Empire goes for more money now than an AR.  I think the issue comes down to cost and that AR sort of 'got away with it' when they used a clock motor to run the turntable. Allied used to sell the AR; I had to deal with headshells with stripped cartridge mounting threads, stripped threads for mounting it to the arm tube, warped headshells, perished platter pads (the foam was terrible but the turntable couldn't support anything better); it was annoying that the overhang couldn't be set and the bearings in the arm were pretty cheesy. The platter bearing seemed alright though. If you really want to look at a budget machine that was better at ticking the boxes, perhaps the old Connoisseur BD2 would be a better choice. They didn't have any better bearings, but because they were at a 45 degree angle, the chatter was a bit more bearable if you'll pardon the expression.


Gawd- the original AR tt was awful. The motor was underpowered; the speed would oscillate back and forth just by setting the arm on the LP. The foam platter pad was a joke but the suspension and drive really needed it to be light. While the bearing location in the arm was an innovation, it was vastly overshadowed by the simple fact that the arm bearings had slop, making it chatter when anything complex showed up in the musical program. The headshell was plastic and over time wore out (from being installed and removed from the arm tube) or warped as it perished. It was a machine that was cleverly built to a price point, and didn't challenge machines like the Empire 208 that were around at the same time.  


Given how innovative AR was at the time, its a bit surprising that they chose to under-serve their concepts as presented in this machine. If they had been a bit more aggressive and upscaled it a bit, they would have built a classic.  
I think HP of all people nailed it when he noted that digital reproduction effects a high pass filter albeit at a very low frequency which seems to rob the music of its natural ambiance. The bass drops off a cliff, metaphorically speaking. You don’t hear it as lack of bass per se.
@lewm  In a nutshell, vinyl has more bandwidth than Redbook. On the bottom end its only limited by the mechanical resonance of the playback. On the top end it can go out well past 50KHz, although my cutter is bandwidth-limited at about 42KHz.
But I can't hear the sleeve bearings on my little Premotec either - not until it is connected to the platter by a belt, and then only through the playback amplification. So I think that sleeve bearings are a problem anywhere in the chain, and the closer to the platter the worse it is. What do you think?
When the belt is tensioning/loading the bearings from the side (as happens with a belt), they're going to make more noise. When the bearing and the motor are synonymous, then its the thrust bearing that is likely more important.
They had a naked motor available for admiration, and so I picked it up, put it next to my ear, and I could hear the sleeve bearings when I twisted the spindle.
Once the motor is prepped for operation it might not do that. I pulled the motor out of my Technics as I am taking it to a machinist so I can run a longer spindle. Its really quiet!
The problem is that next to something like a SOTA Cosmos with any decent arm on it those Technics tables sound like finger nails on a chalk board
A friend of mine (Warren Ghel, currently at ARC) developed the platter pad used on the Cosmos. SOTA got an exclusive contract to use it. I ran a Cosmos for quite some time- and then discovered that a mildly modified Empire 208 equipped with the same arm (at the time, an SME5) as the Cosmos sounded better. With more mods to the Empire (a plinth machined of solid aluminum and damping the platter) the Cosmos simply had to leave. Since then I've gone to using the Triplanar and I've working on a number of Technics SL1200Gs and GAEs; IMO they are a better turntable than our model 208 (the production version of the modified Empire). Based on this I have trouble taking the statement in your quote seriously! The Technics machine is excellent; equipped with the right arm and platter pad its a formidable machine; I don't see how a Cosmos would compare. 

Right Atmasphere, if on a "proper stand" which includes the floor and everything under it, a situation most of us have no control over. On any wood joist floor you will be in trouble. Those of us on concrete slabs are lucky. I personally do not think that level of speed stability is all that important. Like distortion in amps it becomes a numbers game. But, I do not have any prolonged experience with modern DD tables. My bias comes from the unfortunately distant past. You as an audio manufacturer get a lot more opportunity to play with this stuff than I do and I do believe I am jealous of that. My life just lead me in another direction.
What tonearm would you put on the 1200G?

@mijostyn I missed this earlier! I have my turntable on a custom Sound Anchors equipment stand, which is built to accommodate an UltraResolution Technology platform (n.l.a.) for both the preamp and the turntable, which of course sits on top. The stand in turn rests on a set of Aurios Pro bearings (unfortunately also n.l.a.). Despite wood floors and joists, no footfalls or any such like, even in my previous house which had considerably less sound flooring.


The Triplanar seems the obvious choice for tonearm as it is state of the art.


When the two are combined the soundstage takes on a spooky real quality much like you hear on tape. My surmise is since the speed is so stable, the arm oscillates less over the groove so the soundstage has less shimmer as the tracking pressure on each groove wall is more constant.


This is not just me but but all the audio big wigs in Miami circa 1980. The drive behind these tables was that they were cheaper, easier to assemble, much lighter (less expensive to ship) and could be pumped out in large numbers which they were. Since the Japs are capable of turning out an extremely polished device and the marketing hype was good they sold in droves until the digitally mandated vinyl crash. There are some newer DD tables I find intriguing and would like to hear but given my own experience I would never buy one sight unheard.
One other problem is that I have never seen a DD table properly suspended. A drive be it belt, DD or Idler is entirely dependent on it's plinth for isolation. Cartridges being the very sensitive devices they are will pick up any vibration transferred from the environment to the drive. This is best and most reliably done (IMHO) by having a very inert sub chassis suspended by a system with a resonant frequency around 3 Hz.
Since that time Technics did quite a lot on their top-end machines to deal with those issues. The latest round seems even better- The SL1200G is now one of the most speed-stable machines made. A Sutherland Time-Line sorts that out quickly enough.

There are a good number of belt drive machines that lack in the way of suspension, but IME, I've found that if you really want a turntable to sound right, its got to perch on a proper isolation platform and in turn on a proper stand, whether suspended or not. I found my SOTA Cosmos to be dramatically affected by this practice; when I went to our Atma-Sphere 208 it was too and I found the lack of suspension on it to be of no consequence whatsoever, perhaps because our goal was to make the machine otherwise quite dead. FWIW, the Technics SL-1200G uses 6 different means of killing vibration (including a very dead plinth) so despite no suspension it seems to work just fine if on a proper platform and stand.


Our model 208 is as speed stable and neutral as any belt drive, but IMO the Technics is better. That's why we came up with an armboard for it so a tonearm that does it justice can be installed.

The problem with Direct Drives is a big oscillating magnetic device directly under a very sensitive magnetic device.
@mijostyn   What is the result of this? We've been making and playing belt drive machines for about 20 years but when playing the Technics machines I don't hear anything in the way of noise floor that would suggest this is an issue.
WRT to platter mats, the concept that materials cannot absorb/attenuate vibration is not correct.
@antinn I suspect you misread my post, as you and I are on the same page w.r.t this statement; it is the same point I was making (or at least was attempting...)
Thanks for your thoughts, Ralph. About air bearings, I think that a very heavy platter on an amorphous carbon thrust bushing is virtually immovable - at least, mine is. You need a lot of force to raise it - and more to lower it - 87N per micron, plus the inertia. Compare that to a fragile cantilever. For radial stiffness, see above. The New Way figure is 34N/u.
@terry9 If the platter is unable to move up and down or in any other plane when subjected to vibration, such that whatever vibration might be present in the plinth is the same as that at the base of the arm and the surface of the platter, then the needed mechanical engineering principle is satisfied and the pickup will be immune to that vibration.

my low flow-low pressure air bearing CS Port LFT1 turntable and linear tracker give me more detail than any conventional bearing turntable and conventional bearing tonearm i have heard. the bass is otherworldly. magnificent.
If you didn't use the exact same platter pad, cartridge and arm on both machines, this might only be anecdotal wrt the platter bearing type.
The ET2 with its large surface "stationary" bearing is more rigid at audio frequencies than any metal bearings. Bruce Thigpen has done the analysis , documented it
I doubt this very much, on account of one of the metal bearings suppliers in the US requires a security clearance to obtain their bearings (several grades harder than commercially available bearings so they are used in aerospace applications). Now if you can tell me that Bruce has that security clearance (as apparently Triplanar does) then I'm more likely to give this some credence.
all air bearings are not created equal. painting with a broad brush is just not informed. there are general characteristics I do agree. but exceptions too.
More correctly, painting with a broad brush is not so much 'not informed' as it is less likely to be completely accurate.

I would be interested in hearing about an air bearing that has less play than a high quality bearing of conventional design.
I'll try again, since I'd be very interested to find out about this.
I did respond to this, in my post just prior to this.
Rock music will soar with direct drive.
@mikelavigne  Just FWIW, its not possible to build any sort of audio playback that favors a certain musical genre.
I must confess that I don’t really get this, as I don’t see how a mat can simultaneously be an effective and non-reflective absorber of energy and have the same hardness as the vinyl record.
If you are transferring vibration, to absorb that vibration at all frequencies the material to which the vibration is moving to has to be the same hardness as that which imparts it. In this way all the molecules move together. But at the same time, there is no material whatsoever that can receive vibration and not absorb some of it. This is easy enough to google, here's an example:
https://www.quora.com/What-is-best-known-material-that-can-transmit-vibrations-without-absorbing-vib...


So the trick to to create a platter pad that at its surface has the same durometer as vinyl, but internally is better prepared to absorb vibration- so this does not mean that the material is amorphous. Dissimilar materials are well-known to absorb vibration from each other (we use this principle to damp our preamp chassis, but of course any extensional damping compound is doing exactly that) so a platter pad composed in this manner would be quite effective.
At the risk of disagreeing with those who know more,
- agree with the late legend Tom Fletcher, who thought that powerful motors generated powerful vibration hence audible distortion;
- agree with the air bearing crowd who think that all conventional bearings cause vibration in the platter, which is audible;
If the motor has good bearings and the design of the 'table is correct, any vibration the motor has is quite minimal. To give you an idea of this, the motor in my mastering lathe makes 1/8hp but because the shafts which run between the motor and transmission have isomeric isolation, vibration from the motor is not picked up during the mastering process. The same thing can be done in playback; the motor for the Empire machines is isomerically mounted and simply makes no noise in playback; on top of that the Pabst motor used is notoriously silent despite its amazing amount of torque. Tom simply made too broad a generalization!

Air bearings of any sort provide a different problem! For a turntable to work properly **without coloration**, there must be no play between the platter and the plinth and the plinth must be completely dead while rigidly coupled to the base of the arm; in this way the arm and surface of the platter are only able to vibrate in exactly the same plane. The pickup is thus unable to interpret vibration as a coloration or noise of any sort. To this end of course there can be no play in the bearings of the arm and the arm tube must not be able to 'talk back' (resonant) to any vibration being picked up by the stylus; in essence the cartridge is held rigidly in locus. Air bearings of course violate this basic engineering principle. To understand this better, think about the steering of a car because its exactly the same mechanical engineering principle. The wheel has to stay on the road but has to be guided by the driver. If there is play in this scheme, the car will be dangerous to drive and quite scary. Imagine installing an air bearing in the linkage of the steering of a car!



The issue is you need a powerful drive. Direct drive easily has the best speed stability (Technics makes the most speed stable machines in the world and their drives are pretty powerful). A powerful idler drive can easily dust a weak direct drive or belt drive- a powerful belt drive can easily dust a weak idler drive or weak direct drive.

This is why you see a following around certain vintage turntables- all of them have powerful drives- the Technics SP10, the Garrard 301, Empire, Lenco and the Thorens TD124.


The problems with the vintage machines is usually a poor plinth design which does not take into account the effect vibration can have on the sound of the machine, and crude tonearm performance (solved by installing a modern arm). If the designer has done their homework, the plinth (and platter) of the ’table will be quite dead.


Another serious variable is the platter pad- it can color the sound since there is resonance in the LP itself when the stylus tracks it. If this is going on the LP can ’talk back’ to the cartridge. A good platter pad can make all the difference and this is a left out variable in the discussion of which drive is better!! A good platter pad will simultaneously damp the platter and silence the vibration in the LP; this has to be done with a platter pad that has the same hardness (durometer) as the vinyl. In that way energy from the LP is absorbed and not reflected back the the LP. If this is all correct, it will be very hard to hear the stylus tracking the groove with the volume off.

None of the vintage machines had anything other than a joke for a platter pad. For that matter most modern platter pads are no better. IMO this is the least understood aspect of LP playback.