Adcom GFP-565 preamplifier Page 2
The tape outputs are buffered using IC op-amps in a unity-gain voltage-follower configuration. As with the GTP-400, I applaud Victor Campos's insistence on the use of tape output buffers. They are absolutely essential if the integrity of the internal signal path is to be maintained. The loading of the circuitry caused by tape recordersparticularly if they are switched offand cable capacitance can seriously impair the performance of an otherwise excellent design. Every preamp claiming high-end performance should have actively buffered tape outputs, but many, including the older GFP-555, the $900 Forté 2, and the $5000 Audio Research SP-11 II, do not.
The GFP-565's power supplies provide ±18V to the circuitry, with extremely low output impedances. Discrete solid-state regulators are used, along with 6800µF Panasonic TS series electrolytic capacitors at the inputs of the regulators. The outputs of the regulators are bypassed with 470µF Panasonic HF series low-ESR (equivalent series resistance) electrolytics (footnote 3). Three additional pairs of the 470µF HF series capacitors are used for bypassing near the phono preamp, line-level, and tape-out buffer circuitry. These capacitors are placed as close to the circuitry as is physically possible. It is worth noting that all of the 470µF bypass capacitors have a 35V rating even though 25V capacitors would have been more than adequate for the ±18V supply rails. This was done in the interest of lower power-supply impedance, since the 35V caps have a lower ESR than their 25V counterparts.
One severe limitation of power-supply output impedance is normally the resistance of the circuit-board traces. No matter how good the supply, its ultimate output impedance, as far as the circuitry is concerned, is limited by the resistance of the conductors connecting the supply to the circuitry. Adcom has taken the unique approach of paralleling the power-supply traces with three heavy brass busses, insuring virtually unmeasurable resistance between the regulators and the circuitry. The power transformer is encased in a mu-metal shield which, given the vanishingly low noise level of the 565, is highly effective.
The 565 preamp employs separate power-supply regulators for the headphone amp. Your first reaction might be: "how thoughtful of Adcom to devote so much attention to the headphone amp." Wrong! This was done so the headphone amplifier would not put an additional, unnecessary load on the main power supply. In fact, the headphone amplifier is nothing special, being based on a dated 4556 dual op-amp.
Adcom spared no expense on the quality of the passive components used in the GFP-565. All resistors are 1% metal film manufactured by Roederstein, which are among the finest resistors available for audio use. The RIAA feedback network uses 1%-tolerance Roederstein film capacitors, normally very difficult to obtain in the United States, and very expensive. Panasonic polypropylene capacitors are used in the tone-control and filter circuits (which are not connected when the bypass output is used). Both the volume and balance controls are premium Alps pots, the same grade used by Audio Research in the SP-11 II. (Incidentally, judging by the physical appearance of the SP-11 II's resistors, I believe they are also Roedersteins.)
All of the GFP-565's circuitry is contained on one large circuit board, with the exception of one tiny circuit board next to the power transformer which contains the jumpers necessary for configuring the 565 for various line voltages. The main circuit board is double-sided, with the component side covered with a copper ground plane. This is a beautifully manufactured board, worthy of the attentions to detail taken in designing the circuitry which it contains. The recording and listening selector switches are also made by Alps. The working parts of these switches are soldered directly to the PC board, and are operated by mechanical remote connections on the front panel. This eliminates the need for unnecessarily long leads between the input and output jacks, the switches and the circuitry. Gold-plated connectors are used for all input and output jacks. Overall, the quality of construction is excellent.
One final touch which separates the GFP-565 from the competition in this price range is their choice of material for the bottom plate. Many audio designers are aware of the sonic problems encountered when a chassis made of permeable metal is placed close to the circuitry. Adcom uses an aluminum bottom plate to eliminate these problems. If the full sonic benefits of the aluminum bottom plate are to be realized, the 565 should not be placed on top of another component with a steel top. The 565's cover is made of steel, but the distance between the cover and the main circuit board is said to be large enough to prevent any sonic problems.
The same free interconnect Adcom supplies with their other preamplifiers is included with the GFP-565. As I stated in my GTP-400 review, the cable is far better than those normally given away, and many users will find it perfectly satisfactory. I conducted my evaluations with my own cables, however. The 565 manual, at least the version I received, is a bare-bones affair consisting of loose pages stapled together in one corner, a presentation hardly worthy of this fine preamp. They didn't even bother to include a list of specifications. Victor Campos tells me that the manual is being rewritten.
The first GFP-565 sample I received was a pre-production prototype which Adcom supplied in June 1989. When the first production units rolled off the assembly line in September, I was promptly sent a sample. The two were very similar in their sound, with the production unit being subtly better in the areas of inner detail and imaging. The pre-production prototype used some substitute parts, not in value but in origin of manufacture, due to the lack of availability at that time. Roederstein has a six-month backlog on certain resistor values. The production samples contain all of the specified components.
During my evaluations of the pre-production prototype, a problem developed which I must report. At one point in my listening I noticed my woofers moving forward as I advanced the volume control. This occurred only when the recording and listening switches were set to the same input. Suspecting a DC offset problem in the tape-output buffers, I disconnected the preamp, hauled it over to my office, and put it on the bench. Sure enough, there was 18V DC on both channels of the tape 1 outputs. Further investigation showed that the suspect dual op-amp had the same 18V on its inputs as well, explaining the appearance of DC at the main preamp outputs when the recording selector was moved to the same position as the listening selector. On my system, this problem was potentially damaging since my tri-amplified system has a DC signal path for the woofers from system input to output (the midrange and tweeters do not, since bandpass and high-pass filters are, by the nature of their design, capacitor-coupled). Most power amplifiers do have input coupling capacitors, so such a failure is unlikely to damage any associated equipment (footnote 4).
I phoned Victor Campos, who immediately rushed me several replacement op-amps via Federal Express. While Victor was recovering from heart failure, I desoldered the bad IC and installed a replacement. (Adcom was ready and willing to fix the preamp for me, but I felt it would be a waste of time and shipping expense to send the unit back when I was perfectly capable of repairing it myself.) The pre-production sample has performed flawlessly ever since, and it has seen more than four months of continuous operation since the IC was replaced. (I loaned it to a friend when I received the production sample; she intends to buy one.)
I have not experienced a single problem with the production unit. The pre-production sample was hand assembled; perhaps excessive heat during the soldering of the bad IC caused the failure. I am not the least bit concerned about this problem, and report it only because I believe a reviewer has an obligation to document any and all difficulties for the edification of the reader.
All of my comments on the sound refer to the actual production sample connected to my system via the bypass outputs, which I believe is essential if the full potential of this preamp is to be realized. I also recommend making sure the tone-control and high filter switches are left out, even when the bypass outputs are used. Even though the tone-control and filter circuits are not in the signal path, their passive components are still connected to the bypass outputs when they are switched in, which can result in a very subtle degradation of the sound.
In common with other solid-state preamps, the sound of the GFP-565 improves after warm-up. As is my normal practice with preamps, I left both samples of the 565 powered continuously. I should point out that the sonic improvement after warm-up is quite subtle; the 565 sounds extremely good even from a cold start. Phono preamp evaluations were made with my custom-built belt-driven turntable, Grado Signature tonearm with both Grado Signature 10MR and Adcom XC-MR2 cartridges. The line-level section was fed by my extensively modified Magnavox CDB-650 CD player.
The GFP-565 is the best under-$1000 preamp I've had the pleasure of hearing. The phono preamp is one of the very best I've ever heard, period. The total absence of background noise in the phono preamp is immediately striking. If you hear any noise, you can rest assured that it is from your LP playback system and not from the GFP-565. The inner detail and clarity of the 565 phono section is stunning. Try any heavily scored, well-recorded LP, such as Respighi's Pines of Rome with Dutoit and the Montreal Symphony (London), and you'll be able to penetrate the densest portions of the score with ease.
The 565's phono section does not deliver such impressive inner detail at the expense of warmth and hall ambience. It simply delivers an accurate account of what is on the recordclarity, warmth, ambience, and whatever else exists in the program material. The string sound on Chesky's reissue of Fritz Reiner's Scheherazade emerges with characteristic sweetness and delicacy. The 565's phono preamp reproduces a wide and deep soundstage. On Reiner's Scheherazade, the boundaries of the stage in Orchestra Hall are clearly evident, particularly the back wall of the stage. Similarly, the smaller room boundaries of the MGM soundstage studio used for Sheffield's Wagner disc, conducted by Erich Leinsdorf, are realistically reproduced.
One "depth test" I constantly use as a reference occurs just prior to the Forging Scene proper on side 3 of Solti's Siegfried, on my pre-DMM German Teldec pressings. The scraping of the sword against the anvil is deep in the soundstage, behind the orchestra, and that is precisely where it appears when played on the GFP-565. The dynamics of the 565 phono are excellent, to say the least. Siegfried's Funeral Music, on side 9 of Solti's Götterdämmerung, contains a crescendo in the brass of shattering impact, and it is delivered most impressively by the Adcom phono section.
Footnote 3: I have used many Panasonic HF series capacitors in my own equipment and have found them to be the best electrolytic capacitors available.
Footnote 4: This is not strictly true as, depending on the preamp volume setting and the power amplifier's intrinsic LF extension, the 18V DC offset will be reproduced as a full-power pulse of relatively long duration when the input and record selector switches are set to the same input. This happened to me some eight years ago with a preamp possessing a similar problem: by the time I realized what had happened, the power amplifier, which was capacitor-coupled, had still managed to fuse both woofer voice-coils.JA