Peachtree Nova

There's a low threshold to get enthusiastic about products such as this Peachtree Audio Nova. First of all, it has a great deal of features. It's a 80 watt integrated amplifier. It's a switchable Class A tube or transistor preamp. It's an outboard SPDIF & USB DAC. Equipped with Sonos ZP-90 (digitally connected via coax/optical input) it becomes a high-resolution wireless multiroom appliancy. And so on. No radio, no CD player tough.

Second, it looks good. It looks good despite the fact that such a retro wooden chassis (if it's real wood, I'm not sure) has been IN in the audio land for some years already, and if the matter is investigated more deeply, never really vanished from the market.

Soundwise I found the Nova be not perfect but musically largely satisfying. I expecially liked its DAC part. The DA chips are the same ESS 9006 Sabre DACs that can be found in some McIntosh top CD players. The Nova's DAC is said to sport high quality capacitors, 11 regulated power supplies, transformer coupled digital inputs: 4 x 24bit/96kHz and one USB (16bit/44kHz). The DAC reclocks the signal before 24bit/192kHz upsampling procedure to minimize the amount of jitter and to better the SN ratio (120 dB). On the rear panel there's a switch for choosing one of the two digital filters: Sharp Slope and Soft Slope. The former measures better, the latter may sound better. It's up to you. Three analogue inputs, headphone out et cetera.

With the Sony CD player and Dunlavy Athena speakers, the Nova sounded first a little crisp, light, pale, and bright generating an impression of an analytical and slightly nervous reproducer, clean and pure but at the expense of warmth and softness. Both frequency ends were less well represented than with the Accuphase. Rhythmically a nice performance though. The locus of the otherwise fairly 3D soundstage was higher up than with the Accuphase, - I don't know why.

As a DAC only, Nova sounded more mature. Somehow the whole presentation became more balanced, more normal, and almost intrusively dynamic. Male voice sounded wholesome and tonally correct, no artificiality whatsoever. Compared to Firestone Fubar IV DAC (through the Nova), the Nova DAC was a step more effective, airier and more analytical. More info. No huge difference but music amenated from the speakers differently now, less muffled.

Switching between the preamp's transitor circuitry and the Russian made Electro-Harmonix 6922EH operating in Class A didn't produce - apart from some guitar and organ samples - a major difference, tonally or otherwise. If anything, the tube provided the sound with further spaciousness.

The power amp of the Nova is said to feature a Class AB Mos-Fet circuitry. This was something that I started to doubt during my listening sessions. Not only because the unit is so light that it cannot sport a massive power supply with a heavy AC transformer typically found in similarly powered AB power stages. However, my main suspicion came from the Nova's subjective bass reproduction, which was minor in the amount, but first and foremost inconclusive, unsystematic, unpredictable. The bass notes popped out bouncing, as if in the form of indefinite bolls. Sometimes there was the note as expected, sometimes not. Very strange. That was the kind of phenomenon that I've come across with small power Class D amps.

So althought I quite liked the Nova's music making capabilities, especially as a DAC, I was slightly embarrased by its bass performance. Untile I read this from the latest issue of the Swedish Hi-Fi magazine Hifi och Musik:

"When the TDA7293 (the output IC in the Nova) is used with a high rail voltage and in 1 chip per channel, the SOA (safe operating area) can be exceeded with reactive imoedance's lower than 4 ohms. This ID will fail when this happens. This is inherent to the IC itself.

The TD7293 can safely drive lower than 4 ohms by paralleling them, which would double the cost of the Nova's power amp section. This would be the only protection that would save the amp under this condition. However, with less than 1 per cent failures in the field we chose not to go with the added expense.

The Nova does have protection to shut down if the IC faults ad sends excessive DC to the speakers, however.

Also the iNova, iDecco and Decco2 use the same IC and have added thermal protection and additional circuitry to help with colateral damage caused if te IC should self-destruckt."

It was written by the manufacturer in reponse to HiFi&Musik's query on why the Nova behaved like it did in their power cube test (collapsed into 2 ohms, 60 degrees inductive load). Now, I'm not sure to what extent the reply would be helpful in explaining my listening impressions, and the Nova's bass performance especially. Perhaps to no extent. Dunlavy Athena is a 4 ohm speaker, but as far as I know, no major drops in the impedance response. So we're far from the extremes conditions of the cube test. And yet, when I read the manufacturer's answer, I nodded: this must be it! Something in the speaker/amplifier interaction didn't work as it should have.

Maybe Athenas were a borderline case for the Nova. I did try other speakers as well (eg. 2-way Dynaudio bookshelf model, Amphion Ion), and although I was happier with the outcome, my doubts were not entirely removed. I'm telling this basically in order to stress the importance of choosing right speakers for the Nova. Or so it seems to me. As said, there were many good things about the Nova's sound, so it deserves speakers that adjust to its preferences. And of course, if the power amp section still doesn't satisfy completely, from its preout -connection the Nova can be used as a DAC preamp only, with quite good and promising results.