Audio By Van Alstine Super-PAS Three preamplifier

If I had to pick one amplifier designer as having had the greatest continuing influence on the high-end market, as much as I admire John Curl, Audio Research's Bill Johnson, and Krell's Dan D'Agostino, the name of David Hafler inexorably springs to mind. Not because he challenged the very frontiers of hi-fi sound, but because he combined a fertile, creative mind (footnote 1) with a need to bring good sound to as wide an audience as possible, both by making his products relatively inexpensive and by making them available as kits. (The Major Armstrong Foundation apparently agrees with me—they presented David with their "Man of High Fidelity" award at the summer 1988 CES.) It remains to be seen if the Hafler company will continue in this tradition, now that David has sold it to Rockford-Fosgate. But there is no doubt that many audiophiles were first made aware of the possibilities of high-end sound by Hafler products in the late '70s, and by Dynaco in the '60s.

Which brings me to the subject of this review. In those far-off days when tubes ruled, the Dynaco PAS series of preamps, launched in 1958, was consistently regarded as being second only to the classic Marantz 7 when it came to sound quality. In only the third issue of Stereophile, J. Gordon Holt described the PAS-2 as having "very low noise, rich, transparent sound, and incredible tube life," and listed the much more expensive Marantz 7 and McIntosh C20 as being the PAS's only competition.

The PAS-2 and PAS-3 were identical apart from having different front panels and knobs; the -X designation was added in the mid-'60s to indicate that the tone-control circuit had been redesigned to ensure a flat response when the controls were centered. (JGH had no time for the tone controls of the PAS-3X, feeling that they had far too dramatic an action.) The '3X also had a three-position stage-width control, full stereo and mono being joined by a position giving a partial blend of the two channels, this intended to compensate for the exaggerated soundstage width typical of the "ping-pong" stereo recordings that were more predominant then than now. In addition, either channel could be piped to both outputs.

The construction was to a relatively good standard, despite the low cost, and the circuitry was contained on two printed circuit boards, one for the phono stage, the other for the line stage following the volume control, each carrying two 12AX7 tubes. As a kit design, easy access to the boards and controls was important; despite its small size, the chassis was spacious and component layout was rational.

Hi-fi components are no more immune to the ravages of time than any other device, and the PAS's switches, controls, and sockets seemed to enter senility earlier than most. Also, the PAS needed to see a load of 100k ohms or so, and by 1969, when it was joined by the transistorized PAT-4, solid-state was all the rage. (Gordon gave the PAT-4 a rave review, feeling that it was accurate rather than euphonic, plus ca change . . ..) It should be no surprise, then, that this excellent component had handed the tube-preamp baton to Audio Research by 1973 when The Absolute Sound started publishing, the SP-3 being HP's reference preamplifier at that time.

The PAS-3X kit, however, did remain in limited production until the mid-'70s, when Dynaco was sold to the ESS corporation, and its good-sounding circuit and well-laid-out chassis attracted the early attention of modifiers. The first article on improving the sound of the basic PAS to appear in print, at least as far as I am aware, was by David A. Vorhis in the 1/1974 issue of The Audio Amateur, who basically set the groundrules for such modifications: replace passive components—switches, potentiometers, resistors, and capacitors—with premium-grade equivalents, and upgrade the power supply, in this case adding shunt regulation to the B+ supply. Vorhis even recommending bypassing the line stage altogether, a now-common option for those wanting the best sound from CD. A follow-up article from Vorhis in the 2/1976 issue of TAA suggested RIAA EQ improvements—JGH had found the original PAS-3X to have an excess of low bass—as well as describing a series-pass series transistor regulator for the HT, while Ed Dell suggested increasing the HT voltage ahead of the regulator to lower overall distortion (footnote 2).

Audio amateurs, however, were not the only ones to see the potential in the basic PAS design. Audio Research briefly offered an upgrade service in 1974, while Frank Van Alstine, the owner of Jensen's Stereo Shop in Minnesota, entered the fray in 1979 with his Super-PAS, based on the PAS chassis but with new circuit boards and an improved power supply. The Audio by Van Alstine version could either be based on the customer's own Dynaco PAS, or made from new parts.

Early reports indicated that the AVA Super-PAS was a serious contender, and Stereophile accordingly organized a review due to appear in 1984, sending Larry Archibald's own decrepit PAS-2 to Jensen's Stereo Shop for appropriate treatment. For some reason, a review of the resulting Super-PAS failed to appear in print that year, and the preamp was brought up to the then-current AVA specification in the summer of 1985. In September 1985, Mr. Van Alstine realized that he had been too pessimistic about some of the circuit parameters, and requested that this unit be returned to him for further modifications (increasing the value of the output cap to 2.2µF and decreasing the line-section compensation capacitors to 33pF). Upon return of Larry's now much-traveled PAS, Stereophile sent it to Steve Watkinson for a formal review, whereupon SWW promptly lost it!

Two years passed, with Mr. Van Alstine understandably getting increasingly hot under the collar about the nonappearance of the Super-PAS review (footnote 3) until, miraculously, SWW found it in a closet earlier this year. Covered in shame, we sent it back to Minnesota, to be brought up to the 1988 standard of performance, and I embarked upon this review.

Technical details
The AVA Super-PAS 3 covers a number of options. The least expensive is for an owner of a PAS to order the AVA upgrade kit. Costing $200, this includes new power supply, phono and line-level pcbs, new volume and balance controls, and new audio and psu parts. The customer uses his own vacuum tubes, chassis, and transformer, but has to do his own wiring and installation. Alternatively, for $350 he can send his PAS to AVA to have the same modifications done professionally. An additional $150 obtains a new brushed-aluminum AVA faceplate (black with silver lettering is a no-extra-cost option), ceramic selector switch, and a new set of in- and output jacks installed on a ground plane. (This latter is also available separately for $50.) Specifying Tiffany gold-plated jacks adds another $40, while AVA supplies a set of four Audioquest Sorbothane tube damping rings for $12. Replacing the original 12AX7 tubes with four selected, high-gain GE 5751 tubes costs another $40.

All these options are worth exploring if you have an old PAS or if you buy one at a garage sale for a few tens of dollars. AVA, however, offers new Super-PAS preamps, assembled from parts supplied by Stereo Cost Cutters of Dublin, Ohio (who bought up all existing stocks of Dynaco metalwork and other parts), for $595. The difference between having your existing PAS fully modified by AVA and buying a Super-PAS new is a fraction under $50, so it would not be worth buying a used preamp for more than that.

Larry's PAS-2 had had all the AVA upgrade options installed, including the Sorbothane damping rings and Tiffany-equipped socket board, which would have brought the price of the new equivalent to $635.

The result with the AVA front panel is a handsome-looking piece of kit. From left to right: the six-position selector switch is above the source monitor switch and next to the volume control, this a stereo 100k stepped component from Noble. The steps run from around 6dB at the bottom of the range to less than 1dB when fully up; with the VTL 100W monos used for auditioning, which are 6dB or so more sensitive than the norm, the volume control was used around its 12 o'clock position, where the steps are a little large, at approximately 1.5dB. On the right of the volume control is the Dynaco stereo width and mono control, retained from the original, with the balance control underneath, this again a Noble component. On the far right of the panel is a slide on/off switch and a red LED power indicator. The original holes in the chassis for the now-deleted tone controls are covered by the new panel.

Footnote 1: As with Peter Walker in the UK, you often find that David had been there first when you examine the possibilities of what you thought was a new idea.

Footnote 2: Back issues of The Audio Amateur can be obtained from PO Box 576, Peterborough, NH 03458. Tel: (603) 924-9464, Mon-Fri, 9-4 EST. Web:

Footnote 3: See "Manufacturers' Comments," Vol.10 No.9, p.169.