Jeff Rowland Design Group Model 2 power amplifier Bob Deutsch w/BPS-2 July 1997
Like many audiophiles, I've been impressed with the effectiveness of dedicated AC lines and line conditioners. Although power supplies in audio equipment are supposed to provide isolation from the various sorts of interference (EMI and RFI) that may be present on AC, cleaner AC power usually translates to cleaner sound. If reducing EMI/RFI from the power source is worthwhile, it follows that eliminating it should be even better (footnote 1). This is the promise of battery power: by having a source of pure DC, you are, in effect, rolling your own. No EMI/RFI coming in through the power line, and you don't have to worry about the line-voltage variations that are common in most areas of North America.
But wait, there's more! According to Jeff Rowland, conventional AC-based power supplies in amplifiers suffer from impedance fluctuations that are a byproduct of the conversion from AC to DC. This, in turn, results in ripples in the DC supply, particularly evident at low frequencies. With a battery supply, the AC/DC conversion is performed by the charger, so the battery may be thought of as a capacitor with high current capability, low DC resistance, and no AC ripple.
Of course, battery power is not without its downside. First, there's the cost: batteries that produce sufficient voltage and current to run a power amplifier are likely to be expensive. The service life of a lead-acid battery can be reduced by too much or too rapid charging, and by allowing the battery to discharge completely before charging. (NiCads, on the other hand, must be completely discharged before recharging.) The battery's power reserve has to be sufficient to last at least the average listening session, so we're not just talking about using a couple of AAs. Audio circuits are sensitive to voltage drop, so to maintain sound quality the batteries' output must be monitored, and provision made for switching from DC to AC if the need arises (assuming that the amplifier has an AC as well as a DC power supply).
Over the years, a number of high-end designers have introduced battery-powered products, but to my knowledge none has taken the uncompromising approach of Jeff Rowland. The BPS-2 weighs over 50 lbs and contains six 6V, 12 ampere-hour, maintenance-free Sealed Lead Calcium (SLC) batteries, as well as a microprocessor-controlled circuit that monitors electrical parameters and provides charging to keep the batteries in peak operating condition. Charging is ripple-filtered, which smooths out the charging pulses that would otherwise result in mini-charge/discharge cycles, thus shortening service life. Unlike NiCads, SLCs don't suffer from the "memory" phenomenon; and, unlike lead-acid automobile batteries, they're not damaged by prolonged discharge. The batteries' expected lifetime is six to eight years of normal use, with the replacement cost currently about $400.
Although the Jeff Rowland's Model 2 amplifier was initially available only with AC power, the battery power supply is clearly not an afterthought. From the time they were first manufactured, all Model 2s have been equipped with a large connector in the back that allows connection to the BPS-2 via a thick cable (made by Cardas). As with all Jeff Rowland Design Group products, the BPS-2's chassis is made of machined aluminum alloy, its size and general appearance an exact match for the Model 2, including the ¾"-thick front panel. The units are meant to be stacked, the BPS-2 providing a high-mass base for the Model 2, and have the usual impeccable Rowland fit'n'finish.
Ergonomically, the Model 2/BPS-2 combination was a dream to operate, with no operational glitches. Pressing a button on the BPS-2's front panel toggles between AC and battery (DC) supply, with a small light indicating whether the BPS-2 has been sufficiently charged. If the BPS-2 isn't ready, then pressing the button has no effect, the Model 2 continuing to operate from AC. The changeover from AC to DC power was accomplished without switching noise or any interruption of the sound.
When connected to a fully charged BPS-2, the Model 2 will operate for several hours without being connected to AC (playing music at moderate level, I got over four hours before the red warning light came on), although Rowland recommends that it be left plugged in. This was the way I used it most of the time. (But see below.)
Upon receiving the Model 2/BPS-2 combination (and after a suitable break-in period), my first task was to get my ears accustomed again to the Model 2 in its normal AC-powered mode. Much of the associated equipment in my system—see Sidebar—has changed since the original review, but my opinion of the Model 2 hasn't: I still think it's a superb amplifier, refined yet dynamic.
On changing over to battery power, initial impressions were equivocal. Switching back and forth, replaying the same segments of music (with the preamp gain control untouched, volume remains the same), I first had a hard time detecting any change. Eventually I was able to get a fix on some differences, but they were generally minor. In the battery-powered mode, the bass was a bit leaner and tighter, although not more extended. (The Model 2 has very good overall bass performance, but it's no match for a bass champ like the Bryston 7B-ST.) The upper midrange/treble was slightly cleaner, more delicate, with transients ringing more freely, but this was apparent only in the most critical comparisons. The magnitude of the differences was less than I've heard among high-quality interconnects or speaker cables.
The effect on dynamics was complex. At very high levels—louder than I normally play my system—I thought the battery mode was somewhat reticent, with not as much get-up-and-go as AC operation. At more sensible levels, however, the dynamic life of the music was enhanced by the battery mode, with a better sense of rhythm and pace. Once again, the differences were subtle, evident only on repeated comparisons.
I made these comparisons with a Chang Lightspeed CLS 9600 power-line conditioner and a dedicated AC line. While this sort of setup is not uncharacteristic of the kind of high-end system in which the Model 2 is likely to be found, it's arguably not the fairest set of control conditions under which to assess the benefits of battery power. In any case, it would be useful to know how battery power compares with "raw" AC.
To do this, I repeated the AC/DC comparisons with the Model 2 plugged directly into the wall—without going through the Chang power line conditioner, but still using the dedicated AC line. Results: still pretty close, but the differences were a bit more clearly in the favor of battery power, the AC condition sounding somewhat fuzzy and subdued. Finally, I plugged the Model 2 into a nondedicated AC line. This time it was no contest: In the battery mode, the sound had greater openness and transparency, with a greater "suddenness" of transients. The percussion instruments on track 3 of the familiar Chesky Jazz Sampler & Audiophile Test CD (JD37) actually seemed louder, the music having a more dynamic feel, and the soundstage was more precisely defined. By ordinary standards, the sound in the AC mode was quite acceptable; but, having heard the difference, I was eager to switch back to battery operation.
Oh, just one more thing...
Although the BPS-2 is designed to be left plugged in all the time, I would not have been true to the obsessive-compulsiveness that characterizes all audiophiles had I not bothered to check whether the sound improved when the BPS-2 was unplugged from AC.
The part of me that's not obsessive-compulsive would dearly like to report that this makes no difference, but I must say that it does. Not night-and-day, not chalk-and-cheese, but another subtle improvement that moves the sound away from the electronic and toward the musical. Having to unplug the BPS-2 is a pain, and remembering to re-plug it at the end of the listening session is even more so, but a man's gotta do what a man's gotta do.
So, is battery power worth it?
At $5800, the Jeff Rowland Model 2 is an expensive amplifier, and the additional $2600 for the BPS-2 is not exactly small change. Is this a cost-effective way to improve the sound of a system? Tough question. It depends (my favorite way to start answering difficult questions) on a number of factors, especially the nature of the system itself.
There's no doubt in my mind that the BPS-2 "works"—under certain conditions, it allows the Model 2 to operate in a manner that results in a more musically satisfying listening experience. However, the effect is greatest when the amplifier is not given ancillary support through use of a power-line conditioner and a dedicated AC line. Chances are, if you can afford $5800 for an amplifier, you're likely to have a listening room supplied with dedicated AC lines—and if you don't, you should. The cost of a good power-line conditioner and the installation of dedicated lines is considerably less than the price of the BPS-2, and will benefit the entire system. Using the appropriate AC power-line treatment gets the sound pretty close—albeit not all the way—to that obtainable through battery power. Unless everything else in the system is beyond reproach, I'd be inclined to spend the $2600 on a component that offers a greater potential for improvement. For example, the $500-$750 upgrade of the Balanced Audio Technology VK-5 to VK-5i makes more of a difference than adding the BPS-2 to the Model 2.
If everything else in the system is beyond reproach and/or cost is not a factor, then by all means go for the BPS-2. The improvement may be subtle, but it is an improvement, and at the Model 2's level of performance, subtle improvements are all you're likely to get.—Robert Deutsch
Footnote 1: Audio equipment can also be affected by the EMI/RFI radiated through the air. Neither battery power supplies nor power-line conditioners can help with this.