Solid State Preamp Reviews

Introduced at the 1988 Summer CES, this preamplifier from San Francisco-based Parasound costs $395 and is manufactured in Taiwan. It does away with mechanical switching for source select and tape functions, replacing it with CMOS integrated-circuit switches similar to those used in the British Linn LK1 and Quad 34 and 44 models. Construction is to a good standard and the circuit is carried on two main pcbs and three small ones. Following a signal from the phono inputs, the MM-only RIAA amplifier is based on discrete FETs, its output joining the line-level signals at the switching ICs, these controlled by DC voltages controlled by front-panel pushbuttons. The manual states that to get the best sound quality, the tape-out selector can be deactivated. It appeared that this could be done by selecting sources from left to right, then pressing Tape Dub 2-1 at the far left (though this didn't always work).

The selected signal is taken first to a small printed circuit board carrying a rotary "loudness" control, this off when fully clockwise, then to the volume control, this stepped. The only departure from a rational layout is seen here, as a relatively long piece of cable connects the volume control pcb to the stereo/mono switch, this on the other side of the front-panel bank of pushbuttons. The output from the volume control wiper is taken to the balance control, then to the tone-control amplifier, this again based on discrete FETs, as is the output circuit. Separate bass and treble controls are provided, as well as a tone defeat button. As far as I could make out without a schematic, the only capacitors in the signal path are a pair of 3.3µF WIMA polystyrenes. All the circuits are said to operate in class-A; however, it is rare to find any preamplifier that doesn't operate in class-A, unless it uses IC op-amps in the signal path.

A relay mutes the output for approximately three seconds after turn-on to allow the circuitry to stabilize; the CMOS circuitry then chooses whatever source was selected when the unit was last turned off.

The power supply uses transistors to regulate the ±24V audio circuitry voltages, with IC regulators used to provide ±15V rails for the Toshiba switching ICs. All the in- and output sockets are gold-plated, and three utility mains sockets are provided, one unswitched and two switched.

The sound
After I had carried out some measurements and done some preliminary auditioning via the P/FET-900's line-level inputs, Parasound requested that they supply a second sample for review, the reason being that the first was faulty: it "popped" when the tone defeat button was switched in or out. I had noticed this but had also felt that the sound was not as good as I would have liked at this price level. The second '900, however, suffered from a dead left channel, due, I am informed, to a unique sample error which meant that the output muting relay wouldn't fully close on one channel when the line voltage was lower than average. I accordingly asked Parasound to send a third sample, with which all subsequent listening was done. This third sample, I am assured by Parasound's Richard Schram, is representative of current production.

The Parasound was allowed to warm up for 6 hours or so, set to unity gain, and inserted into the tape loop of the PS Audio 4.6 preamplifier I was using as a reference. Though its line stage didn't invert signal polarity with the tone defeat button in, switching the tone controls into circuit still caused a slight popping noise as the circuit became polarity inverting.

I listened first to the line stage on its own, comparing it with the straight-wire bypass. I reached for the Bernard Roberts Beethoven piano sonata CD (NI 5060), which features the "Pathetique." I am not sure what piano Mr. Roberts used for the recording, but via the PS Audio it sounds very much like a Steinway, the complicated and characteristic meshing and intermeshing of harmonics as notes decay rendered very obvious. The P/FET-900 was not quite as tonally neutral as the PS Audio, the sound taking on a slightly forward tonality in the upper midrange; this was certainly not to any extreme degree, but was identifiable as a faint nasality to the sound. The noise of the piano's action also became slightly more percussive, and the hammers appeared to have less felt on their heads. In addition, the midrange seemed just a little more uneven. In the first movement of the "Pathetique," after the grave opening statement, a long descending chromatic scale winds down from a high E-flat to the B-flat more than two octaves lower to prepare the way for the allegro. One of my favorite passages for examining a component's midrange, the reference presented this as perfectly even; the Parasound, however, seemed to present some notes as being more prominent than others.

This is a very cruel test, throwing departures from neutrality into sharp relief even with expensive preamplifiers. In actual fact, the Parasound was doing well. However, this slightly forward aspect of the low treble was also noticeable on the Sigiswald Kuijken recording of the Bach Partitas and Sonatas for solo violin (Deutsche Harmonia Mundi CDS 7 49290 2), where the violin image was pushed slightly forward in the image, sounding slightly louder, and its already astringent, period-instrument sound becoming slightly more so on its A and E strings when compared with the bypass. The bypass also had slightly more energy in the very top octave, giving a feeling of more "air" around the instrument.

How did massed orchestral forces reproduce? A recent recording recommended by Lewis Lipnick is the Telarc coupling of the Faur… and Duruflé Requiems (CD 80135), which features a wide, deep stage and natural voice and string tonality. I have to admit that any audible differences were much harder to hear than with the solo violin or piano: a slightly more "shut-in" quality in the highs; perhaps a very slight coarsening in the midrange.

It was time to move on to some nonclassical: Flim & the BB's (what else?). To my surprise, compared with the bypass, the P/FET-900 sounded, of all things, "faster." A sloppy adjective, but it is hard to think of a word more appropriate: transients seemed more pronounced, the music seemed more dynamic. The sound was more enjoyable in this respect, the "thunderclap" tom-toms in the "Tricycle" track subjectively knocking me out of my listening chair at the identical objective level to the PS Audio. The obvious assumption to make is that the Parasound was exaggerating things, but I was not so sure. Yes, the straightwire bypass should be more accurate, but remember that it actually passes the buck for driving all the cable between the CD player and the power amplifier back to the CD player's output stage. Maybe what I was hearing was the fact that the Parasound gets a better handle on the cable than do either of the CD players used for the tests.

This is, of course, conjecture, so I will change to LP as the signal source, the MC pickup amplified by the Tim de Paravicini–designed "Black Head" transformer. On went "The Look of Love" from my original 1966 English RCA release of Casino Royale, Rusty Springboard in full sotta voce. Tonally, the rundown went as follows: top octave usefully slightly lower in level than the PS Audio; midrange good, if again rather forward at the top of the region; lows a little fuller in the upper bass, and not as much ultimate extension. The more significant subjective difference, however, concerned the retrieval of fine detail. This track has a guiro accentuating the second beat of each bar in an implied claves rhythm. The PS Audio allows you just that little bit more easily to hear the return signal from an echo plate or chamber, localized in the same position as the guiro in the stereo stage. Similarly, the PS Audio presents images with more of a "palpable presence," to fall back on a cliche coined by the Audio Cheapskate.

This slight diminishing of soundstage depth was also apparent with the Radka Toneff Fairytales album (Odin LP03). Listening to my favorite track, "Nature Boy," Miss Toneff's voice was lighter in texture than with the PS Audio, but presented just a little more one-dimensionally, the image being foreshortened.

I finished my auditioning of the Parasound with the Chesky Sibelius Symphony 2. The string tone was again just a little more astringent than the reference, the upper bass was a little looser, and you couldn't "hear the walls" as well. However, more serious, for me at least, was a feeling that, in a manner similar to the Rotel's phono stage, the pulse of the music was somehow diminished; not by much, admittedly, but enough to disturb. In this respect, I preferred the sound of the phono section of the similarly priced NAD 1300.

Conclusion
Very slightly more forward in the upper midrange than the Rotel RC-850 that I also review in this issue, the Parasound P/FET-900 is less veiled overall, while not approaching the transparency of the more expensive PS Audio 4.6. It does reproduce more musical information than the cheap Rotel. I felt that the Parasound's line stage was better, overall, than the phono stage, though at the price, this preamplifier offers a good balance of virtues for those who prefer to use a good MM cartridge or a high-output moving-coil. A safe Class D recommendation in Stereophile's "Recommended Components."