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.
21192e21 5d00 4ded bea1 72869e5cf35dscar972

Showing 28 responses by lewm

Mike, It is interesting to hear from you on this subject.  About 7-10 years ago, you had 3 turntables set up.  One was a tweaked Garrard 301, another was a Rockport Sirius, and the third was a Technics, as I recall, eventually replaced by a Dobbins Technics and then by Dobbins' The Beat.  Have I got that about right?  You were my guru on this subject. Based on your posts and those of some others, I got into experimenting with different drive systems myself.  Having never owned anything but belt-drives, I felt that I knew what they do, up to the price point that limits my selection.  So I acquired a Lenco and highly modified it, and eventually an SP10 Mk3 in slate+wood plinth, a Kenwood L07D, bone stock, a Denon DP80 in slate, and a Victor TT101 in a massy wood and aluminum damped plinth.  Again, I had to play in the cost area that is compatible with my level of affluence.  It has been a very educational experience.  I now run two turntables feeding one system that drives Beveridge 2SW speakers and three turntables feeding my main system that drives a pair of electrically modified Sound Lab 845PXs. It's wretched excess but fun to listen to the same LP interpreted several slightly different ways by the different turntable/tonearm/cartridge combinations.  I must say the highly tweaked Lenco (with input from Win Tinnon) holds its own quite nicely vs the 4 direct drives, and I firmly believe these turntables outperform belt-drives under $10,000 that I have heard.  However, a neighbor had a Dohmann Helix turntable with the built-in Minus K suspension.  That is a way more than $25,000 belt-drive that made me feel a bit of lust.  Or I should call it curiosity to try it in my own systems.  But that isn't going to happen because of cost.  Anyway, it's interesting that you decided to go back to having multiple turntables, after having been happy with the NVS for so long.  Carry on.
If the Brinkmann has a 20-lb platter, that would make it a heavyweight among direct-drive turntables.  Only the SP10 Mk3 and probably the SP10R platters would be similar in weight.  Therefore, the Brinkmann must have a relatively powerful motor, as well.  Since I know it is a coreless motor, that makes it even more appealing.  In my experience, I have come to favor slightly coreless motors (e.g., in my Kenwood L07D and my Victor TT101) vs iron core motors, even though on the other hand the SP10 Mk3 in my 80-lb plinth is the most neutral.  The Kenwood is wonderfully "musical" compared to any others.
On the notion of motor noise in direct drive, there is an ocean of ignorance.  If you hear noise, it is bearing noise, the turntable's bearing.  All turntables have such spindle bearings.  So you cannot attribute that sort of noise as a bugaboo of direct drive.  In essence, a DD motor cannot generate any noise except spindle bearing noise, because the moving parts (the rotor) make no physical contact with any stationary parts (the stator).  Moreover, the motor rotates at a snail's pace compared to the motor of a belt-drive, where the drive spindle is rotating at a much much faster rate.  DD motors CAN generate electrical noise, e.g., EMI.  They generally rely upon the platter and some shielding of the motor casing to prevent EMI from affecting the cartridge.  I do believe the SP10 Mk2, at least my two samples, generated some EMI which did affect playback, faintly.  The Denon DP80 is quieter which is why I've kept it over the Mk2.  And I really do not believe that the SP10R motor made an audible noise, unless it had been abused by hundreds of show-goers prior to being played with.  Which is possible.
When naming DD turntables with 20-lb platters and heavier, I neglected to include Mike's NVS and probably the Rockport Sirius, but I can no longer recall whether the Rockport was DD.  Also, any of Dobbins' variations on the SP10 Mk3 theme.  There may be others of course.
“Sexiness and tonality”. That’s what cordless motor driven DDs do better than iron core types, in my experience.
Mike, Platter of SP10 Mk3 is listed as "10kg" or "21 lbs", everywhere. Not near to 30 lbs. It's heavy enough as is.
Platter weight 9.8kg = ~21 lbs.  I see no conflict with what you previously wrote, if you're talking about the Bardo.
With any DD that uses any sort of servo feedback to maintain constant speed, the feedback circuit was designed for a particular platter mass.  If you change the platter mass dramatically, it is quite likely you will do bad things for speed constancy.  Likewise if you were to change to a much lighter platter compared to the OEM one.  Now, based on reports of others and my own experiences, there apparently IS some leeway within which some perturbation of platter mass will not do notable harm to speed constancy, but since every manufacturer had a different thought on the feedback loop and how to have it operate, there is no general rule about how far you can go.  I also don't know whether to credit some of the reports one can read on this forum, if you search the archives. Nearly everyone who replaces a 1kg platter mat on a DD with a 5kg platter mat, sometimes more than doubling the total mass of platter plus mat, says how wonderful it is. To me this says more about listener bias than it does about the physics.  And I also don't agree that there is a linear relationship or anything near to it, between platter mass and SQ. Belt drives with 200lb platters do not excite me.
cleeds, I took examples of 1kg vs 5kg out of thin air. Most OEM platter mats weigh less than 1kg (2.2 lbs). I picked 5kg as the upper extreme, just because it is 5 times what might be the weight of a heavy OEM mat. Furthermore, a lot of the vintage DDs have rather lightweight platters.  So the total platter mass (platter plus mat) might be under 5kg.  When you add a 5kg mat to such a platter, you are about doubling its mass.  Can that be wise? Now we know you can bring any platter to a dead stop with palm pressure, so we know there is SOME platter weight that is too much. Just because a platter continues to rotate and the music is not obviously off-pitch on the low side,to indicate gross slowing of the speed, does not mean that one has done no harm to the operation of the table with a very heavy mat. Not too many people have the objectivity to be able to hear that (listener bias), and fewer people still will be able to make the relevant measurements that would reveal a problem. I certainly do not know that ALL turntables will be functionally harmed by a 5kg mat; I just used that as an example. Even among DD turntables, I think it’s impossible to generalize, because there is quite a bit of variability in motor torque and the "tightness" of the feedback loop (how much error is tolerated before a correction is initiated) also varies quite a bit from one design to another. For example, Technics seems to have favored very tight control. Whereas the Kenwood L07D uses a looser feedback loop. So perhaps I added to the confusion by seizing on two particular exact numbers in my first statement. I hope this is more clear. Heavier is not always better.
Nottingham Hyperspace has or had a very thick mat that appears to be graphite in photos, about one inch thick.
I bought a very early version of the Hyperspace, from one of their dealers in Florida with whom I never met.  When it arrived, I was chagrined to note the absence of that nice graphite-like mat.  I complained to the dealer, and his response was that I got an early model built before they went into production of that mat.  Instead there was a flimsy foam rubber mat on mine.  The problem was never put right, maybe because I am too lazy to make demands upon the dealer, or because the table sounded so good as it was.  That whole incident preceded my fascination with DD turntables and the Lenco.  I think I got tired of the Hyperspace because I hated having to give the platter a shove to get it started.
Bob, how could the changes you describe have affected bearing noise?
Ralph, One of the virtues of a coreless motor as originally designed by Dual and used by Pioneer, Yamaha, and Kenwood, and even now Brinkmann, is that the field it generates is in the horizontal plane, mainly. The engineers who designed for these companies and those who designed for the iron core motors in the Denons and Technics turntables, were not ignorant of this potential problem.  They took care to provide shielding.  In the case of my Kenwood L07D, in addition to the motor being encased in a metal that adds shielding (the motor in its case looks like a metal discus, sealed all around top and bottom), the platter uses a thick heavy sheet of stainless steel for a "mat".  Kenwood call it a "platter sheet".  Nevertheless, L07D owners are known to add even more shielding by various methods.  In my case, I had a machinist (Colby Lamb in Oregon) make me a new platter sheet out of pure copper.  It mimics the original in shape and weight to within a few centigrams.  I've been meaning to buy a cheap field strength meter to see whether I can detect a difference in radiated EMI, comparing the original stainless steel sheet to the new copper sheet, but I have not done it yet.  Although I like to think my copper platter sheet makes a big difference, because I paid Colby $700 to make it, I really cannot say the difference is huge, because I heard no problem before, and I hear no problem after, maybe a very very subtle difference.  Anyway, I don't care because the copper is beautiful to behold.  With my SP10 Mk3, the platter is simply enormous and thick, made of brass and aluminum.  It's dead silent.  With my SP10 Mk2, when I owned one, there I may have heard a faint grayish coloration to the sound, only when comparing it to the former two turntables named above.  Belt-drivers like to claim that the faint coloration they say they hear with the Mk2 is due to the servo "hunting" to maintain constant speed.  I think it's more likely due to EMI which could easily be ameliorated by either using a metal mat (which many do use with the M2 and others) or by simply adding shielding under the platter.  Perhaps the reason people like metal mats on some DDs has to do more with blocking EMI than with the inherent goodness of metal as a mat.  One reason I sold the M2 and kept my Denon DP80 is because the DP80 lacks that faint coloration I thought I heard from the Mk2.  (This is before I was lucky enough to find my Mk3 and the L07D.)
Mijo, blah-blah-blah.
what could be cheaper to make than a typical modern belt drive in the under $3K price range? They multiply like rabbits. THAT is why belts became dominant. You can build one in your basement. ( I have respect for some expensive belts, starting with Dohmann Helix.) And most suspended turntables are not properly suspended. Or do you still want to ignore the placement of the motor, on or off the suspension. Either way is a compromise. If you’re obsessed with suspensions, Minus K or Herzan or possibly Vibraplane are ways to go. And each of those can be used with any drive system.
This discussion hasn't "boiled down" to anything at all.  The subject at hand is always a function of who feels like posting at any given time of day or night.  My personal predilections are against excess complexity per se.  I don't want a compressor in my room OR in the next room OR in my system anywhere, not because an air bearing is not excellent but because I don't  want the complexity, the hoses, the moisture problems, the filters, etc.  And probably also because the turntables that employ air bearings are beyond my upper limits in what I am willing to spend.  There IS a point where cost is laughably high.
Thuchan, I do take your point re complexity. I admire industrial art for art’s sake. For that we also have cars and cameras, along with vintage turntables, vacuum tubes, and oddball speakers like my Beveridges. Just no compressor in my audio system. I had to draw the line somewhere.
I guess I will use it to inflate a balloon, so I can float over to Germany to hear Eckard's systems.
Tom, I liked your list of 16 yes or no alternative preferences. I calculate that makes for 65536 choices in turntables. (2E16)
Mijostyn, If you have a recording of fingernails on a chalk board, any good Technics table will make it sound like fingernails on a chalk board.  But why do you want to listen to that?
Thanks, Ralph.  That's a very key point about direct drive (the absence of a force in the horizontal plane that pulls the spindle shaft up against the bearing wall) that the belt-ers choose to ignore.
I built my modified Lenco based on Win Tinnon's recommendations.  Slate plinth, PTP top plate, platter painted and damped to reduce resonance, huge aftermarket bearing with under-platter clamp, from Jeremy in the UK, Phoenix Engineering Motor Controller.  It performs up there with anything else I own.  The total investment there is under $2K, partly I guess because I subcontracted the making of the slate plinth. Peter Reinders (source for the PTP) supplied a pdf file to program the waterjet; the actual cutting was done by a company in York, PA.  Slate came from the same quarry in PA used by OMA.  I was fortunate to source an NOS Lenco as the basis for that project, for only $500.
Mijostyn, have you had direct communication with techdas and clearaudio? I was wondering how you would know the basis for their marketing decisions. I don’t know about techdas but in my opinion clearaudio is a company that does not really make anything much. It seems they collaborate with other anonymous companies who make the products that are then marketed under the clearaudio name, in at least many cases.Before anyone attacks me, this is not to say that some of their products are not very good. I know that they have a large following. And in the current atmosphere, there are not too many companies outfitted to make direct drive turntables. Whereas as I have said before it is quite a simple matter to make a belt drive turntable and then market it under several different guises with increasingly elaborate bling  that can be priced at different levels from low to high.
Problems with two or three motors are: (1) They have to be perfectly in synch with one another, and (2) you now have two (or three) potential sources of noise that could be transmitted via the belt to the platter.  No free lunch.  I also think that we anal audiophiles worry perhaps too much about issues that are in reality way below background, if the product is well made and engineered.  So 2 or 3 or 4 motors (how about that?) might be fine.
Thuchan, I was only joking when I mentioned using 3 or 4 motors. However I think there actually was or is at least one commercial product with 3 motors as an optional “upgrade”. In these discussions, it is my tendency to favor direct drive and idler drive over belt drive, but I certainly do recognize that there are very fine belt drive turntables that have sounded wonderful to my ears. As I noted earlier in this thread, I think, the Doehmann helix turntable is one of those latter. But you have to spend what is for most of us big bucks in order to get to that level of performance. And lesser levels of belt drive turntables just don’t cut it for me.

Richard, I was surprised to read your comment about the difficulty of designing and building a direct drive turntable. With respect to vibrations. Because to me a saving grace of direct drive is that there is no physical contact between the motor and the platter, the platter itself is the rotor of the motor. It is driven by the magnetic forces generated at the stator which never touches the rotor. So the only issues are accurate speed and speed stability despite stylus drag, bearing friction, etc.This is not to say that designing and building the world’s best direct drive is not a challenge, but I would have thought that one of the challenges is not motor noise of the mechanical type.
So, although I was joking about 4 motors and never dreamed there actually IS such a turntable, Burmeister have built it.  I'll probably never get to hear it, so I leave it to you to consider. But the more motors you have driving the platter, the less of the circumference of the platter can be contacted by the belt, which means a smaller "contact patch" and greater energy losses, more opportunity for belt slip, unless each motor or each pair of motors drives/drive the platter via separate belts.  I'd guess that is the case with the Burmeister.
Burmester, not "Burmeister".  I just looked at a photo.  One cannot see how the belts and motors are implemented but from the text, I am not sure.
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. HP described it as a reduction in downward dynamic range. I hear it with red book CD and with SACD. I don’t know if it is still the case for current Hi-Rez streaming  types of reproduction. But a friend of mine who is into that sort of thing, and who also has high end vinyl equipment, does still hear a difference in favor of vinyl. 
I’ve got a couple of LP box sets of iconic recordings, by the Beatles or by Miles Davis, or by etc. Just because there is only so much time in the day, several individual LPs in these various sets have never been played. In recent weeks, because we have so little else to do, I have played some of these LPs, particularly the Miles Davis recordings on Prestige that were re-issued by Classic Records maybe 10 years ago. To my dismay some of those Classic sides are abominable. I compared the bad sounding LPs to originals on Prestige, where I own duplicates, for example “Workin’”. Old and worn though my original Prestige copy may be it kills the Classic reissue. In cases like this I far prefer digital. Too late to return that Classic box set but beware.
One solution to the issue of the side force applied by the belt, when you use a single motor, is to add a capstan or passive pulley 180 degrees opposite from the motor pulley.  This will equalize the side forces on the bearing at the expense of reducing the size of the contact patch between belt and platter.  I think the Kuzma Reference is designed in this fashion.  It also avoids the complications of synchronizing two or more motors, two or more belts, etc.