Take two belts to a turntable?

Vinyl Frontiers: Stretching credibility and rubber

[Double trouble]

Further Horizons on Horizontal Hooke's Law

[Italian version here]

Product: Double pulley & double belts on Michell turntables
Manufacturer: OEM (historic) or aftermarket pattern pulley (Audiosilente)
Price: £18 at time of purchase via ebay, now £23 due to currency fluctuation, evidence that YMMV
Author: Mark Wheeler - TNT UK
Reviewed: Summer to Autumn 2020

How might two belts be better or worse than one?

Today we are exploring whether there are any advantages or disadvantages to using two belts on an already optmised motor-belt-platter system. John Michell, designer of the Michell Orbe and Michell Orbe SE no doubt optimised the relationship between the rotational effective mass of the Orbe platter and the torque of the motor via the modulus of elasticity of the belt, to minimise wow and flutter while maximising vibration isolation from the motor.

“Obviously two belts will transmit twice as much motor vibration as one belt,” state Plebs, stage left, “and the whole Michell Orbe project was about lowering the noise floor.”

Ignoring, for a moment, the unscientific ambiguity of a term like "noise floor", let us consider the theoretical implications of two round section belts on a turntable. In theory:

These latter two points are particularly pertinent to the current Michell Orbe but will not apply to earlier iterations of this venerable turntable, for example those with AC motors. As with the earlier test, the popular iPhone apps for measuring absolute speed and detecting wow and flutter were inconclusive, at best. Despite careful calibration and accurate centreing on the platter clamp, even without the dynamic load of stylus in groove, consistent results were impossible to obtain. The Pedersen Orbe should be less prone to effects like precession than a subchassis atop a trio of conical springs. However the results in each of the grooves of the aftermarket pulley varied from test to test, as had the results from the original pulley. Such variation envelope was equally evident when the two iPhone apps were tried on the Garrard 401, with neither belt nor subchassis. Ultimately, multiple tests were averaged for each condition. The averaging of the measured results undermines credibility but did indicate an improvement in wow & flutter under stylus load.

[Double trouble]

Two belt history

Michell did offer two belts for a period in the distant past. This was on the Gyrodec which has a lighter platter and simpler isolation system. The lighter platter might magnify any effect on primary wow frequency, of the reduced elasticity of a double belt arrangement. With less isolation between motor and main bearing, the Gyrodec belt might proportionately be less significant for motor noise isolation than on the Orbe. John Michell believed that the second belt was actually inaudible and told your Old Scribe that he'd never met anyone who could reliably tell the difference.

The second belt had been fitted in response to two reviews, published in different magazines from different publishers. The Flat Response generally described any source component (other than a Naim tuner) as inherently totally inferior to the Linn Sondek LP12. The Flat Earther's favourite comic criticised the Michell Gyrodec as lacking the Linn's PRaT potential. Meanwhile, the less political HiFi Choice more specifically identified a problem of dynamic wow on their sample (quite possibly the same abused review specimen heard by The Flat Response) of the Michell Gyrodec. Dynamic wow is the momentary brief slowing during heavily modulated passages caused by the extra stylus load presented by big groove wiggles. Dynamic wow is one of the reported errors attributed to poor suspension 'bounce'. In the days of Linn supremacy in the British audio comics, any aspiring high-end turntable had to compete with the tune-following hype of the Flat Earth followers.

HiFi Choice measured a single belt running 2% slow and 3% when loaded by a stylus playing a record. They also measured start up to stable speed taking 9.5S which they felt did not bode well for maintenance of pitch stability under load. HiFi Choice did also find unweighted rumble to be below 77dB and vibration breakthrough so low that the scale of the measuring equipment had to be raised by 10dB to detect any at all. The primary advantage of the Orbe over the Gyrodec is increased vibration isolation and the second is greater platter mass. The former extended the Gyrodec's low noise lead over many rivals while the heavier platter of the Orbe could well mitigate that dynamic wow issue by sheer force of inertia.

Two Michell Belts Reviewed Before

Michell responded to the flat earth flat belt hegemony with a second belt for the Gyrodec and off it went to the comics for appraisal. HiFi Choice approved the change and backed this up with measurements. HiFi Choice found that with two belts the Gyrodec dynamic wow was halved.

Roy Gandy of Rega included a second belt on flagship models (P9 and latterly the £3k RP10), stating that “Two belts iron out the inconsistencies that exist with all rubber belts”. Gandy was also referring to thin round section belts like those on Michells. Another key advantage argued for the Rega is the increased surface area of two belts, especially on the small pulley which should reduce start up slippage and hence glazing of the belt, among other virtues.

The Audiosilente Double Pulley

Although Michell themselves no longer argue for a two belt solution, the Audiosilente double pulley for Michell turntables is widely available from Fleabay and other mail order sources. This Audiosilente double pulley was therefore purchased from their fleabay store. Audiosilente have no idea that this customer would be publishing a review, so the sample is representative of what you, dear reader, will receive. Audiosilente produce a remarkable range of those must have parts to repair and refurbish old Thorens, Garrard and Lenco turntables, as well as classic SME and Fidelity Research pick-up arms.

The Audiosilente double pulley is beautifully machined, looking remarkably similar to anything coming from Borehamwood. The Audiosilente pulley groove profile is slightly different from the Michell groove shape and the lower section is broader. Installation is simple and the double pulley fits the bearing shaft well, suggesting that pulley eccentricity (which would cause wow) is no more likely than the original Michell pulley.


Test Procedure

The Michell Orbe SE platter was thoroughly cleaned and degreased for the first test of new belt for old.

The belt swapping tests were referenced to the Garrard 401, SME 3012 II. The previous article includes a lengthy preamble about turntable set up and test procedure. This included fresh lubrication with Michell supplied spindle oil, after the bearing well was thoroughly cleaned. All the belts involved in this test were used in the same orientation every time, whether this makes any difference or not, to eliminate another variable. In the previous test we compared two brand new Michell turntable belts with each other and with a very old one and established the audible and measured consequences. Those new belt tests were conducted on the same double pulley as we are about to use with two belts, ensuring that any differences are not due to a different pulley.

“This sounds like unscientific hogwash,” Challenge sceptical Plebs, stage left, “Superstition or obsessive ritual,”

Audibility Tests

First the listening tests were undertaken and then the measurements. This is to minimise a priori expectation of difference and therefore reduce the potential for confirmation bias.

The speed differences had been obvious between the new single belts and the old single belt. There was also an obvious preference for the second permutation of the two belts arrangement, but not on grounds of speed as might be inferred afterwards from the measured results. Permutation 1 seemed lightweight with less discernible soundstage delineation and scale. Musically permutation 2 seemed more rhythmically coherent.

  1. Belt 1 top groove with belt 2 bottom groove
  2. Belt 2 top groove with belt 1 bottom groove

“This is yet more unscientific hogwash,” Challenge sceptical Plebs, stage left, “What a bunch of subjective phenomenological drivel!”

Two belts immediately gave the impression of sounding slightly slower than 1 new belt. The effect was of pace not pitch. However one permutation of two belts seemed superior to the other permutation. This was the second permutation tried, but the process was repeated back and forth several times. There were short excepts and whole album sides in each comparison. This could well be due to manufacturing tolerances adding up or cancelling, depending on permutation.

Two belts was consistently preferred on all types of material, over either new belt running solo.

With 2 belts the bass seemed stronger without being obviously louder. With two belts the nuances of Lou Reed's voice on Transformer seemed to ride more clearly over the bass, being more reminiscent of the Garrard 401 than the standard Orbe with 1 belt. It becomes more frequently and more obviously apparent that Lou Reed is modulating the sound of his voice by moving nearer and further from the microphone mid phrase. This is why we want to spend money and hours optimising our audio systems, to be able to experience more of the performance. This becomes especially obvious near the end of side 1, during Walk on the Wild Side in the line “Jackie is just speedin away...” so do have a listen when you've read this.

Returning to the old single belt left Reed with a lighter voice, as though trying to sing while lying down with someone sitting on his, instead of Andy's Chest. After Transformer, an album with the most bass complexity near the end of a side, Tapper Zukie's MPLA offers strong bass all the way. Bass impact & pitch are much improved by 2 belts on MPLA.

With two belts the pitch of vocals held much better over heavy bass modulation. This would fit the old 80s reviews' narrative of dynamic wow. The extensive Michell factory results of imperceptible difference between one and two belts may be at variance with these findings because of the Pedersen suspension modifications. The capacity of the subchassis to move equally but in the opposite direction to belt flex might explain why the sprung subchassis Gyrodec seemed to John Michell and friends to be unaffected by a second belt, in contrast to the fixed chassis experience of Roy Gandy at Rega and numerous Rega users. Most of the other multiple belt arrangements are on high mass turntables using thicker belts, whose application may therefore be too different to transfer to this context.

Two Belts and Solo Piano

Solo piano has been the de facto material for judging analogue time performance, whether it's tape or turntable. Keith Jarrett's Köln Concert is one of the finest pieces of piano capable of the repeated listing essential for accurate evaluation of audio gear. Anticipating a weightier left hand as a consequence of the 2 belts arrangement, confirmation bias was confounded when bass weight was identical. The most obvious changes were more percussive delivery and the performance even more engaging and addictive. Fundamental building blocks of music are pitch and PRaT. The 2 belt configuration, in combination with the Pedersen Michell modifications affect both these parameters. The improvements on piano are noticeable in pitch stability, especially above middle C. An increase in the percussive qualities of Jarrett's playing and instrument did not adversely affect pitch stability, indeed the reverse was true.

Hania Rani's Esja is another solo piano recording deserving repeated playing and full concentration. The differences between single belt and two belt configurations reinforced the impressions set by the Keith Jarrett Köln Concert.

Test Results: Measured

As noted in the earlier review, the lower groove seemed to run slightly faster. This could be due to machining variation, or alignment with the platter. Both variations of new two belt positions were tried,

  1. Belt 1 top groove with belt 2 bottom groove
  2. Belt 2 top groove with belt 1 bottom groove

Using the RPM iphone app on a selection of heavily modulated discs (including tracking test discs) and a low compliance Decca London cartridge dynamic wow was measured between 0.45% and 0.50%. With two belts this dropped to 0.29% to 0.32%. These results would tend to support the HiFi Choice measurements of the two belt Michell Gyrodec.

Some Reminders

In the test of two new belts against an old one, this was what was heard:

The audible difference made by new belts over old were predictably most explicit in the percussive qualities and pitch consistency which were most noticeable with solo piano recordings. Two belts simply built on the difference between one old belt and one new belt. Despite this being a cliché the solo piano really is for some reason the toughest test of pitch stability, hence it becoming the stereotypical dynamic wow test. The Keith Jarret Köln concert worked as a test for the single belt replacement and works equally effectively for the two belts test. Reviewers are prone to drone on about the Köln concert substitute piano being magically revealed by some new fancy bit of wire. The difference a second belt makes is not in such a magical domain. A dynamic bass slam cliché is Miles Davis' Tutu which did not disappoint in it's capacity to foreground the subtle difference between one and two belts.

Noise Levels

As described in the Pedersen modifications test motor noise could be less isolated by the Pedersen chassis arrangement than the standard sprung Orbe SE. The Michell Orbe SE is actually marginally better in this respect than the full fat Orbe. The effective solution is a motor pod sized circle of material like Deflex or similar Cured Polyurethane Polymer beneath the motor pod.

A doubling of the connecting mass between motor and belt does not seem to have any negative effects on noise or feedback. Both phenomena tested with "silent" grooves or single sine tones created no difference at all. The accuracy and precision demanded by the microgroove record could easily be upset by something as fundamental as a second belt. Fortunately this is not the case.


The various forums and social media groups devoted to turntables in general, and Michell turntables in particular, frequently debate the one belt vs two belts issue. These are hotly debated with the overarching binary politics of the day in English speaking communities. The same names appear taking the same positions and making the same points to support these positions. There is little evidence that many of the contributors have actually compared the alternatives. A few do own old two belt Gyrodecs and those who've compared seem to divide equally into three camps:

  1. One belt better
  2. Two belts better
  3. No discernible difference

Every other opinioneer seems to be writing from what they believe would be the difference or not of two or one round section belt. These writers appear to write as an article of faith, not experience. Either that or they simply fail to mention any actual experince. In response to this the following points need to be made before any comments respond to this review.



The two belt solution was an attempt, in this system context, to continue the progress made with the Pedersen subchassis and suspension in reducing the inherent weaknesses of the belt drive sprung subchassis approach to spinning vinyl discs. Suspended subchassis turntables have equally inherent advantages in feedback and other noise isolation. Having a Garrard 401 alongside a Linn, Thorens and Michell Orbe SE, as well as the occasional direct drive passing through, focuses attention on the differences fundamental to approaches as much as between individual design execution.

The two belts, in combination with the Pedersen modifications succeeds in bringing the Michell Orbe SE closer to the strengths of the Garrard 401 without the unmodified 401 weaknesses (by today's standards) in noise and focus. The potential noise transmission drawback of an extra belt and less bouncy subchassis isolation is mitigated by more isolation between motor pod and support. The collective whole is still a Michell Orbe in character and voicing but in some important respects now turned up to 11. The modified Orbe SE still offers a massive soundstage. The two belt Orbe bass control is even better, in conjunction with the Pedersen suspension. The modified Orbe mastertape type presentation, improved by the Pedersen subchassis, remains intact with a second belt.

Now, without losing its inherent strengths, this Two belt Michell-Pedersen Orbe SE gains some of the qualities usually associated with fundamentally different turntable drive systems. Subjectively the combination of tied suspension and two belts also seems to increase dynamics and transient attack. Dynamic wow is measurably and audibly reduced.

The perfect turntable does not yet exist at any price. This particular Michell Orbe SE is getting closer to the primary strengths of the best direct drive and idler drive turntables while maintaning its strengths of control, low colouration and immense soundstage. With the Pedersen modified suspension and subchassis and the Audiosilente double pulley these gains are made without losing the benefits of a belt drive subchassis. Fortunately these benefits are also made while building on the built-in virtues of the well established Michell Orbe SE.

DISCLAIMER. TNT-Audio is a 100% independent magazine that neither accepts advertising from companies nor requires readers to register or pay for subscriptions. After publication of reviews, the authors do not retain samples other than on long-term loan for further evaluation or comparison with later-received gear. Hence, all contents are written free of any “editorial” or “advertising” influence, and all reviews in this publication, positive or negative, reflect the independent opinions of their respective authors. TNT-Audio will publish all manufacturer responses, subject to the reviewer's right to reply in turn.

Music enjoyed while writing this review

Reference system

on vinyl of course

  • Lou Reed: Transformer
  • Tapper Zukie: MPLA
  • Keith Jarret: The Koln Concert
  • Hania Rani: Esja
  • Bob Marley: Exodus MFSL LP
  • Charles Mingus: Mingus-ah-um MFSL 2LP 45rpm
  • Miles Davis: Tutu
  • Frank Zappa Teenage Rockin Combo: Dumb All Over, 1981 concert recording

Equipment used in this review:

  • Turntable: Michell Orbe SE

  • Turntable Modification 1: Pedersen subchassis & suspension kit

  • Turntable Modification 2: Pedersen Orbe SE replacement 2 layer isolation

  • Turntable Modification 3: Pedersen Michell Armboard isolator

  • Turntable Modification 4: Polycrystal PolyCrystal Point Discs beneath Orbe feet and Deflex plain sheet beneath Orbe motor pod

  • Pick-up arm 1: HadcockGH242 SE with all silver wiring from cartridge to silver Eichmann Bullet plugs

  • Turntable shelf frame support: Origin Live Ultra wall support, chrome plated, nylon cord replaced by chains, ouriggers filled with kiln dried sand

  • Turntable shelf wall isolation: RCD Cups (larger surface area than Polycrystal Point Discs to reduce wall plaster deformation)

  • Turntable shelf frame support: Rawlbolts embedded in hard plaster

  • Turntable shelf 1: 10mm laminated glass

  • Comparison Turntable: Garrard 401 in plywood & concrete plinth on Polycrystal Isolators

  • Pickup arm on turntable 3: SME 3012 II with FD200 damper

  • Turntable support 2: Wilko AV shelf frame, filled with kiln dried sand, RDC cone feet

  • Cartridges: Dynavector DV XX-2 MkII on Michell; Decca London (John Wright serviced 2020) on Garrard;

  • Phono pre-amplifier: Gold Note PH-10 supremely adjustable phono-preamplifier, on Polycrystal isolator cones

  • Line Pre-amplifier: Audio Research Corporation Reference 3 with steel top cover removed, casework damped by BrightStar Little Rock 5, tube rolled and vibration further controlled internally by silocone O-rings and Pearl valve cooler/dampers, on Something Solid end grain balsa plank, on kevlar sling supported by separate sand-filled steel frame support

  • Crossover: Behringer CX2400 analogue variable Linkwitz Reilly (4th order) 3-way active crossover

  • Treble Amplifier: de Paravicini (HiFi World) designed SET6080 in RATA Torlyte case and outboard mains transformer, and passive components to match mid and bass

  • Midrange Amplifier: Assemblage SET300B Signature variously tube-rolled and components to integrate with treble and bass

  • Bass amplifier: Breeze Audio Nelson Pass Zen clone SE class A FET 10W

  • Tuning planks beneath treble and midrange amplifiers: ERaudio Large SpaceHarmoniser on ERaudio steel cones

  • Vibration isolation under crossover and bass amplifiers.

  • Active Loudspeakers: 3-way active loudspeaker drivers:

  • Active Loudspeakers Bass drivers: Focal Audiom 12 VX with dual layer Cerwin Vega 2 layer cone roll surround, in 3 layer 25mm birch-plywood braced, hardwood \96 chipboard \96 fibreboard - birch-ply reflex cabinets, with Deflex subwoofer sheets & and with Deflex wedges and centre supported long fibre wool.

  • Active Loudspeakers: mid-range drivers: Focal Audium 7K with rubber solution surface damping removed. Mounted in separate enclosures with sculpted & radiused fiddleback sycamore baffle, 25mm birch-ply bracing, Deflex standard sheets. Rattletrap Extreme damped, decommissioned missile warhead, aluminium tipped copper mesh reinforced resin outer shells - literally.

  • Active Loudspeakers high frequencies: Focal TC120TDX with felt focus ring.

  • Active loudspeaker bass cabinet feet: Three tapped RDC cones seated in three Yamamoto Loudspeaker Spike Receiver.

  • Sand filled column stands under mid/top enclosure, with lossy polymer sheet interface and 3 Polycrystal Isolator feet


Extensive and ever evolving acoustic treatment including corner bass absorption, high frequency (above 2kHz) absorption at primary tweeter reflection points, high frequency diffusers at other critical points, sloping ceiling with absorber >2kHz. Solid walls and argon filled triple glazing and no radiators. Hanging rugs and thick textured floor rug between listeners and loudspeakers, solid slate floor.

Some wire is used to join these components together. No interconnects cost more than 10% of the device at each end, much of it made by the Old Scribe from high quality components without Pixie Dust. Old Scribe amplifier-to-loudspeaker wire (full range, mid-range, tweeter) is ultra-low impedance Black Rhodium S900, a low-Z variation (3x3mm^2 csa) on the Black Rhodium S300 & S600 cable that came out well in Ben Duncan's objective and subjective correlation tests, selected primarily to match the OPT/driver damping factor, not for any magical qualities. Bass only loudspeaker cable Naim NACA 5, which remains rarely challenged below 300Hz. Mains is supplied by an audio only ring main with Radex earth (ground) non-inductive connections and a technical earth. Crossover and power amplifiers fed by a minimum connections hydra. Sources and pre-amp from terminal blocks within the audio only ring.

Facebook logo

Copyright © 2020 Mark Wheeler - mark@tnt-audio.com - www.tnt-audio.com