This review page is supported in part by the sponsor whose ad is displayed above
Reviewer: Srajan Ebaen
Source: Zanden Audio Model 2000P/5000S; Opera Audio Reference 2.2 Linear; Raysonic CD128, Ancient Audio Lektor Prime [on review]
CD Recorder: Olive Symphony with Red Wine Audio battery mod
Preamp/Integrated: ModWright SWL 9.0SE; Music First Audio Passive Magnetic; Bel Canto Design PRe3; Wyetech Labs Jade; Supratek Cabernet Dual [on loan from owner]; Melody I2A3 [on extended loan]
EQ: Rane PEQ55 active merely below 40Hz
Amp: 2 x Audiosector Patek SE; Yamamoto A-08S; FirstWatt F3 & F1; Bel Canto e.One S300; Eastern Electric M-520
Headphones: AKG K-1000 w. hardwired Stefan AudioArt harness; audio-technica W-1000
Speakers: Zu Cable Definition Pro in custom lacquer; Anthony Gallo Acoustics Ref 3.1
Cables: Zanden Audio proprietary I²S cable, Zu Cable Varial, Gede, Libtech and Ibis; Stealth Audio Cable Indra, MetaCarbon & NanoFiber [on loan]; SilverFi interconnects; Crystal Cable Reference power cords; double cryo'd Acrolink with Furutech UK plug between wall and transformer; Crystal Cable Ultra loom [on extended loan]
Stands: 2 x Grand Prix Audio Monaco Modular 4-tier
Powerline conditioning: 2 x Walker Audio Velocitor S fed from custom AudioSector 1.5KV Plitron step-down transformer with balanced power output option
Sundry accessories: GPA Formula Carbon/Kevlar shelf for transport; GPA Apex footers underneath stand, DAC and amp; Walker Audio Extreme SST on all connections; Walker Audio Vivid CD cleaner; Walker Audio Reference HDLs; Furutech RD-2 CD demagnetizer
Room size: 16' w x 21' d x 9' h in short-wall setup, with openly adjoining 15' x 35' living room
Review Component Retail: $2,695 - $2,995
TVCs. The moniker has stuck. Transformer volume controls. These three little letters distinguish passive preamps using multi-tapped transformers or autoformers from those that rely on the far more ubiquitous resistive plastic and carbon wipers or ladder types for signal attenuation.
Because of a transformer's air gap, a TVC, broadly speaking*, can interrupt ground loops (not true for autoformers). Because of the transformer's secondaries wired to relays or switch poles, TVCs never sport smooth but always stepped controls. Those using switches rather than relays nearly always lack remote control.
There seems to be little good reason for the latter except that superior attenuation transformers are significantly more expensive than pots. TVCs are thus intrinsically less cost-effective than regular resistive passives of equal chassis and connector parts.
|And perhaps it is because the superior Seiden or Shallco switches traditionally favored as the manual control to which the transformers are wired would take quite a stepper motor or belt drive to move. That'd be noisy or somewhat clunky in operation and require an AC or DC power supply. TVCs thus nearly exclusively eschew remotes. To sell that as an asset, one could try to cop the standard high-end audio retort whenever convenience is lacking - better sonics. Greater purity. Or simply call a spade a spade and make it hair shirt audio. It's certainly nothing to brag about if this mostly obscure breed of preamps wished to attract a greater market share.
|Bent Audio's John Chapman to the rescue. When Thorsten Loesch's collaboration with UK transformer firm Stevens & Billington first netted the requisite commercial TVC parts way back when -- by now S&B is in its third generation -- Bent Audio became the original source for kit and assembled TVCs in the US market. DIY HiFi|
|Supply in Hong Kong with whom Loesch was collaborating on modifications for discontinued Opera Audio models then marketed the Django on the other side of the globe. To give credit where due, Swiss tweak king Audio Consulting predates all consumer TVCs by having resurrected this vintage method of signal reduction first. While makers like Audio Zone, DIY Paradise, Music First (S&B's subsidiary), Promitheus, SAC Thailand, Sonic Euphoria and others have since joined the small TVC parade, John Chapman once again leads it, by adding full remote functionality and scalability to the base TVC recipe with his latest Tap.
His? Yes, it's his design thru and thru. Yet a win/win arrangement with Music First Audio has altered the badging from Bent Audio to Music First since Chapman first introduced the Tap. Under this agreement, Bent Audio covers Tap sales for the US/Canadian markets while Music First handles the global remainder. Both companies use the same S&B trannies. Both have collaborated from the very beginning to define the raw part and its performance. Competing rather than working together now would be rather counterproductive. Instead of developing their own remote-control version of the accoladed Passive Magnetic preamp, Music First has quietly adopted the Bent Audio Tap platform as its own.
Readers whose upstairs calculators jammed upon hearing that will envision ghastly sticker shock now. After all, the demurely clad Passive Magnetic already clocked in at a rather stout $2,495. That's when it was first launched in America. US importer Fidelis AV's introductory special at the time had always been scheduled to increase to at least $2,995. Cite painful conversion from British pound to US dollar as the primary culprit for that eventual necessity. How much more would the far tricker Tap have to be? Double? As it turns out, barely nothing at all. It starts at $2,695, just $200 more than the plain Jane Magnetic. Bent Audio handles American sales direct to make it so. Dealers are eliminated. How this welcome pricing solution affects Music First's existing Passive Magnetic remains anyone's guess. While sonically spectacular and there fully competitive with other $3K preamps, it was bloody expensive for being a basically empty and modest box filled with a rat's nest of leads but zero convenience features for emotional balance.
About that rat's nest, the Tap completely cleans house. It terminates its transformers with Arlon circuit boards from whence neat ribbon cables for the control signals connect to the 34-level 2dB-step attenuation matrix. Music signal wiring between i/o ports and transformers is OCC copper hard-wired to the connection points. The 6dB step-up gain previously switched manually now engages automatically whenever volume exceeds unity gain. The infrared commander controls level, balance, inputs, phase, mute and display. The front panel duplicates these functions. Misplacing the remote won't ever interfere with operating the Tap. Any input/s can be configured as HT thru-put/s with the push of a back panel button. For outputs, there's twin pairs of RCAs and XLRs. That's a total of four. The transformer wiring for the Tap is custom UP-OCC copper sourced from Neotech. (On the subject of why metallurgy in attenuation transformers can affect sonics, Jack Elliano of Electra-Print offers that "silver is more sensitive to flux variations than copper to where it may reveal very high-order harmonics".)
The modular Tap comes in two trim levels, the $2,695 one-input (RCA or XLR, both installed) and $2,995 6-input version. Expansion modules connect via fiber optics to eliminate any possibility of ground loops and are star-grounded to the back-panel screw terminal. All volume switching is performed via optical encoder and sealed Coto relays spec'd for millions of trigger actions. The entire control system enters sleep mode post commands to avoid clock noise. Even the ground connections for each input are isolated - only the chosen one is connected. Phase inversion is accomplished with added relays per input to avoid extra switching in the signal path. An external ground lift switch allows further customization.
The display shows 1 thru 34 for the possible 2dB attenuation steps and also confirms balance offsets. 6 LEDs to the right correspond to the 6 inputs. Green indicates active input, orange phase invert and red an input that was programmed as a Home Theater thru-put. Three LEDs and associated control buttons on the left are for display on/off, tape-out on/off and system mute. (S&B connoisseurs may question the 34 since the TX-02 part traditionally had 24 steps. 34 is the correct number. Bent Audio commissioned the transformer maker for additional taps. Incidentally, the auto gain feature proceeds in smooth 2dB increments across the zero line of unity gain so there isn't a sudden jump of 6dB but +2, +4 and +6.) Lastly, the transformers in the Tap aren't potted: "The mount is a pair of aluminum brackets that sandwich the core laminations and bolt to the bottom plate. This bracket does a nice job of closing gaps in the core and I used aluminum rather than stainless steel since it is a bit more inert from a vibration point of view. Also, it had to be a non-magnetic material. It makes for a package very resistant to vibrations once it's all put together."
ln summary, John Chapman's Tap -- shared as it is with his Music First chaps in the UK -- is the current epitome of the full-function TVC built to virtually bullet-proof reliability standards. It shares this top distinction only with the far dearer AVTAC Pasiphae at $6,585. Below the Tap sits the new Le Club HiFi Django. It commands $1,775 and offers S&B parts and remote control but less complex connectivity sans XLRs and neither display nor HT and balance functionality. For the exact mix of features and pricing on - um tap, today's unit does seem to occupy a singular position in the market.
Readers hoping for further technical background on the genre should visit my previous review of the Passive Magnetic, Stevens & Billington's own tech page or -- for serious tech -- this PDF from Jensen transformer maestro Bill Whitlock. Should the question nag what powers the Tap's display and relay switching, yes there's an external 9V power supply. It operates entirely outside the signal path to substitute for coach potato calories. There's no reason to believe the Tap will break new sonic ground over the Music First Passive Magnetic. However, wiring, connectors and enclosures as the final variables can make a small difference when a circuit gets as minimal as this one. Should this be the case, the Tap's construction is of a higher level, its signal paths are shorter and its switching hardware is far more advanced. Sonics could thus be marginally superior. That subject is nearly mute though. To gain all of this active functionality on top of the Passive Magnetic's comprehensively reviewed sonics -- for essentially the same coin -- is already cause for champagne and fire works. What else could one possibly want on this subject? Even the cosmetics are rakish.
In general, TVCs excel with ultimate transparency, non-existent noise floors and superior low-level detail even at subdued playback volumes. (To be complete, there's yet another form of transformer-coupled passive preamp that uses a 1:1 input transformer with a resistive attenuator plus fixed resistor across the secondary to create an idealized constant rather than varying load for the transformer and insure linear impedance matching regardless of attenuation. MagneQuest's forthcoming Ingot designed in collaboration with Steve Eddy is one such device and will eventually be offered with optional valve or battery-powered JFET buffer.)
Let's briefly revisit the argument about TVC's superiority to common resistive potentiometers - because the latter throw excess gain away as heat while the former convert voltage into current. What does this imply? With line-level interfaces between CD player and amplifier, the job at hand isn't to transfer maximum power but maximum voltage. For maximum power transfer, we'd want identical source/load impedances. But in the case of low-level signal, the amplifier wants to see a low output impedance from the TVC to complement its own high input impedance. If the TVC can lower the effective output impedance of the source component connected to it, there's a more optimal impedance transfer to the amp, hence superior sonics.
That's because the source -- most likely the CD player in our case -- has to supply less current to drive the amplifer's input impedance to the same voltage level. This equals more efficient use of the voltage presented at the transformer's input. The same effect is used in electrical power transmission. High voltage power lines are reduced in voltage via transformers. "Imagine global warming if we turned that unused voltage into heat!", quips John Chapman.
With a TVC, the transformer turns ratios limit how far its output impedance can be lowered.** Mathematically, the transformer's impedance is reflected as the square of the turns ratio while the signal itself is attenuated merely by the turns ratio. For our purposes, this math works out neatly. If the signal is attenuated by a factor of two, the output impedance drops by a factor of four. If the signal is attenuated by a factor of four, the output impedance drops by a factor of 16. Consecutively lowering the output level on a TVC thus logarithmically drops its output impedance. [This is essentially the same benefit as accorded by the recently reviewed Burson Audio Buffer - though the latter's is a fixed rather than variable lower impedance - Ed.] With respect to impedances, a TVC is naturally at its worst at zero attenuation, i.e. unity gain. That's exactly where the resistive attenuator performs its best. You thus don't exploit a TVC's special benefits at or near unity gain.
Regarding other technical differences between a TVC and potentiometer, a TVC can provide very high levels of common-mode noise rejection even from an unbalanced source which a pot cannot. With the TX102, a very high but existing HF limit "is useful to eliminate high-frequency hash riding on the source signal".*** A TVC has a magnetic core with hysteresis. Its nonlinearities can be significant depending on signal level and frequency whereas a pot is essentially free of nonlinear distortions. A TVC presents a varying load impedance and a varying source impedance depending on position whereas a pot presents a constant load impedance and a varying source impedance.
The only technical argument sensible to the user in the end is that passive preamps are measurably constrained or liberated by ancillary choices and chosen settings. Try before you buy thus becomes more vital than usual. At least discuss the relevant impedance values of your partnering gear, interconnect lengths, the most likely volume control window and your system's overall gain structure via the Bent Audio forum. You'll be advised fairly whether the Tap would be electrically copasetic or not.
Adds John Chapman: "I also build resistor passives so don't think I'm against them. A TVC system done properly is simply very expensive. For many, a resistor passive is far more cost-effective. It's just that from a straight technical perspective, the TVC does a better job in a passive environment than a resistor-based passive. Despite the retro vibe transformers give off, the way they technically behave attracts me. I measure all gear on my Audio Precision System 2 dual-domain analyzer. It continues to amaze me how well these TVCs measure. This does not mean that a TVC will always sound better than a traditional resistor passive. Systems vary far too much to generalize.
"I came rather slowly to the TVC camp. I was making resistor passives long before that. After hearing the Swiss Silver Rock, a local friend asked me years ago what I though of using transformers for volume control. I though it was a crazy idea. It was not until I looked at the way transformers treat impedances that I could really appreciate their advantage in this application. Still, I tend to slow folks down who plan on using TVCs in active DIY preamp projects. They do work very well but I think that for the money spent when you have full control over the downstream impedances like you do in that active case, you are not using much of what you really pay for in the TVC. For those situations, a nice clean resistor attenuation circuit can work very well indeed."
Depending on where you stand on the passive preamp issue, these devices either don't suffer any additive errors at all or instead evidence an audible lack of signal conditioning such as active preamps will provide. It's fair to say that passives of any persuasion are specialty products. TVCs -- or magnetic/inductive volume controls as they are also called -- are a further subset of this distinct minority. Passives aren't universal solutions by any stretch then. Still, I've had excellent results especially between tubed sources and valve amplifiers. In the right circumstances, TVCs can be the hammer. Of the sledge variety no less...
* The interruption of ground loops is a function of a two-winding transformer and that the primary and secondary be magnetically coupled, not electrically. A dual-winding transformer like the S&B unit thus will interrupt ground loops but an autoformer-based volume control cannot provide this electrical isolation between its source and load. This does not consider the more involved analysis of capacitive coupling between a transformer's primary and secondary windings. To distinguish between transformer and autoformer attenuation, we're informally referring to them as TVCs and AVCs, respectively. [This from transformer guru Mike LeFevre.]
** Theoretically speaking, DCR could be argued to act as a quasi voltage divider whereby the load impedance itself attenuates a portion of the signal. Yet John Chapman feels that a well-designed attenuation transformer's DCR effects will be marginal at best. The amount of spurious attenuation implied by that argument depends on the transformer's winding resistances, turns ratio and the connected load impedance. With a superior transformer design driving an appropriate load, this type of parasitic attenuation is kept very small compared to the keyed-in attenuation provided by the TVC's setting. "It is the impedance translation of the transformer, not the DCR that matters. Transformers operate in a fundamentally different way than a voltage divider. This does not make them better than resistor dividers but it does make it hard to draw exact comparisons between their operation."
*** A TVC also has a high pass filter function which limits ultimate low frequency response (whereas a pot will pass DC). A transformer driven from a reasonably low source impedance will usually have an LF -3dB cutoff. With the TX102, this occurs in the single-digit Hertz range to only be of academic concern.