The Two Most Interesting Power Amplifiers for the Audio Purist (Vol.2, No.1: Winter/Spring 1979)

The Two Most Interesting Power Amplifiers for the Audio Purist: Hafler DH-200; Rappaport AMP-1 (follow-up).

One of them gives you the cleanest, most accurate, most natural sound available so far, regardless of all other considerations. The other sounds almost as good, costs less than one fourth as much (22 cents on the dollar), and is thoroughly practical to boot.

If there's one subject that doesn't need further coverage here after our State of the Art seminar, it's power amplifier circuit design. In the aptly chosen words of Stew Hegeman, we beat it to death. If you want to know what theoretical design criteria are considered important by some of the best engineering minds in the business, please turn forthwith to the small-print transcript in this issue. We certainly won't add another word to theirs, except to re mind you that our laboratory test methods and listening setup were explained in some detail in the last three issues, with rationales and updates as we went along. Copies are still available to new subscribers. Let's proceed then directly to the reviews.

Hafler DH-200

The David Hafler Company, 5817 Roosevelt Avenue, Pennsauken, NJ 08109. Model DH-200 stereo power amplifier, $399.95 wired. (In kit form, $299.95.) One year warranty, manufacturer pays return freight. Tested sample on loan from manufacturer.

Our killer reputation notwithstanding, we can think of no greater editorial delight than to be able to recommend a truly superior piece of audio equipment at a reasonable price. The Hafler DH-200, either wired or as a kit, fills that bill about as neatly as anything we've tested so far. It's not only the best power amplifier value known to us; it's also one of the three or four best-sounding power amplifiers regardless of price. Now anyone with $400 can afford to be a finicky ultrapurist when it comes to power amps; if he's willing to solder two already assembled and wired channels to the power supply and do a few little screwdriver chores, he can swing it for $300. (According to Dave Hafler, that's not cheap; it's all the others that are overpriced.) In our listening tests, the $1800 Rappaport AMP-1 was the only power amplifier we found unequivocally superior to the DH-200, and not by a wide margin. Both were considered by our various auditioners to be amazingly trans parent, uncolored, sweet, solid, and focused in sound, but the DH-200 seemed to have just the slightest trace of “shimmer” on top by comparison-and only by comparison. Also, the Rappaport has a huge dual power supply, so it hangs in there on current-draining low-impedance loads a bit more consistently, with occasionally audible advantage. Switching to the Hafler after long exposure to the Rappaport is by no means a serious letdown, though; the minor retrogression in sonic purity is something we feel we could live with if we had to.

Against all other comers, the Hafler stood up brilliantly. The latest Futterman H-3aa, when happily feeding relatively high-impedance loads, was found marginally preferable on some program material, but the Hafler was preferred just as often if not more so; the two can certainly be considered sonically comparable. The excellent Audionics CC-2, our previous best-buy choice, sounded distinctly more “electronic” and constricted than the Hafler, although again only by direct comparison. The Mark Levinson ML-2 wasn't available, but our previous ranking of early versions of it as a hair below the Futterman should suffice as a guideline. All other power amps were hastily retired after a few minutes of comparison with the Hafler.

This kind of comfortable superiority is seldom due to good clean engineering alone; one begins to suspect some kind of very fundamental technological edge on the competition.

That may well be the case; the DH-200 is the first American amplifier to utilize Hitachi's new power MOS-FET's. What's more, Erno Borbely, the Hafler engineer responsible for the design of the DH-200, created a very different circuit to go with the power MOS-FET's than what you'll find in the few Japanese amplifiers that took early advantage of this new output device. It seems he was able to use quite a bit of feedback without ill effect, since the MOS-FET's introduce very little front-to-back delay (see also the seminar transcript on this subject). With 2 microfarads shunting the 8-ohm load resistor, we measured 30% over shoot on square waves, damped within less than 20 microseconds. With 1 microfarad across 8 ohms, the overshoot dropped to 15%. Into a purely resistive load, there was no overshoot.

Not that any of this proves a whole hell of a lot; it does indicate, though, that there's plenty of feedback in the circuit and that it's used in an intelligent, controlled manner. Whether that tiny residual shimmer on top is feedback related is a moot point. (Again, we refer you to the seminar transcript for the more sophisticated aspects of the matter.) Our routine program of measurements revealed absolutely no vices in the DH-200; THD was especially low at all frequencies and all power levels. We found the official power rating of 100/100 watts into 8 ohms and 150/150 watts into 4 ohms to be extremely conservative; you can definitely count on a bit more. The power supply (single transformer, two hefty electrolytics) becomes the ultimate limitation on current capability; with 4-ohm loads the purist may want to consider using two DH-200's with only one channel connected on each. That's still only $800 for a super stereo amplifier. We're also told that a $25 bridging device for converting the DH-200 into a 300-watt mono amplifier (at 8 ohms) will be available shortly. That should put the fear of God into makers of multi-kilobuck exotic amplifiers.

As far as reliability is concerned, the amplifier took in stride everything we could throw at it in the way of heavy use and/or abuse.

Its construction appears to be distinctly better than that of the Hafler DH-101 preamp. We foresee no problems.

What else is there to say? We're more impressed that we ever figured we could be by an inexpensive product and predict that the amplifier business will never be the same again.

Rappaport AMP-1 (follow-up)

A. S. Rappaport Co., Inc., Box 52, 530 Main Street, Armonk, NY 10504. Model AM P-1 stereo power amplifier, $1800. Three-year warranty. Tested early production samples, owned by The Audio Critic.

In our long introductory review of the AMP-1 in the last issue, we covered the basic points of the design in some detail but could evaluate only the earliest prototypes as to laboratory performance and listening quality.

We're now able to report that the production version represents a significant step forward sonically, well beyond what we expected. The production Rappaport sounds cleaner, more transparent, more detailed, more neutral, more ‘un-electronic’ than all other power amplifiers we're presently aware of, regardless of price or origin. No exceptions, no qualifications.

Switching to any other power amp, even as good as our previous top choices, reveals tiny but discernible flaws that might have gone un noticed without the direct comparison. As a consequence, the Rappaport AMP-1 is now our exclusive ‘Reference A' power amplifier.

That said, we must add that our reservations are legion. Although the production version doesn't reach, let alone exceed, 100° C (212° F) in temperature on any part of the chassis as did the prototype, the problem hasn't been solved to our full satisfaction.

The amplifier is still much too hot, especially with high line voltage (ours occasionally rises to 122 or even 123 volts). There can be no question of anything but open-shelf housing; children and pets must still be kept away from it.

In at least one sample, hum was distinctly audible through an efficient speaker system, even with the inputs shorted, and we had considerably more trouble with the AMP-1 than with any other amplifier trying to avoid ground loops in our complex multichassis system. But that's the least of it.

Our main reservation is about operating reliability. We had one massive failure after another. Output transistors blew on three or four different occasions; some of the lower level transistors in the circuit boards also went sour. Even when the amplifier didn't shut down, there was grossly excessive DC offset in one or both channels that needed fixing, again at least three or four times. Grounds opened up inside the amplifier, causing monstrous amounts of hum. And so on and so forth. Andy Rappaport was, needless to say, extremely prompt and cooperative in giving us the required service and repairs; in fact, our original two production models were completely re placed, and three of the four channels in our two replacement amplifiers are no longer the original ones, either.

The question is, was all this typical or a statistical freak? Andy Rappaport claims that out of 60 other production AMP-1's in the field, only 3 have given the slightest trouble, and one of these was the victim of a horrendously fouled-up AC power line in a store, leaving only 2 internally caused defects out of 60 units, a defective rate of 3.3%. That's generally considered acceptable in a totally new and different product. (We asked him to put these figures officially into writing for our files, but he hasn't done so as of press time.) Is it possible that our high line voltage, combined with our constant plugging and unplugging of both inputs and outputs, plus our propensity for clipping amplifiers with piano master tapes, etc., created a totally abnormal operational environment for this amplifier? Or is the no-feed back design of the AMP-1 a destabilizing influence on bias conditions and other device parameters, so that the amplifier tends to live dangerously? We haven't got the answer.

Since Andy Rappaport himself defended the no-feedback philosophy at considerable length in our State of the Art seminar and was critiqued in depth by all the other participants, we'll add only one little tidbit here on the subject. At 20 volts output into 8 ohms (i.e., 50 watts, which is just a hair over half power for an AMP-1 channel), we measured THD of the order of 0.6% at any frequency below 1 kHz, rising to approximately 0.8% in the neighborhood of 10 kHz and 1.15% at 20 kHz. How about that? Remember, this is the world's best-sounding amplifier. What would it sound like if 40 dB of feedback could be applied to reduce these figures to the vanishing point? Like a DB Systems amplifier maybe? There's food for thought.

It should also be added that an outrigger device for bridging the two channels of the AMP-1 will be available shortly. The resulting mono amplifier will be able to deliver over 54 volts rms into any load down to 3 ohms or so. Into 8 ohms, that's a 370-watt amplifier; into 4 ohms, a 740-watter. Landlords, beware.

As for our overall assessment of the Rappaport AMP-1, we're of two minds about it.

We'd like to tell every red-blooded audiophile to rush out and buy one. We put our money where our mouth is and bought two. But what if you live in the middle of the prairie or up in the mountains and the damn thing conks out on you? Or suppose you have a curious three-year old who can't keep his hands off large black objects. Just don't say we didn't present both sides of the question. We have our priorities and you have yours.

Recommendations

The following reflects our sifting of a large number of highly touted power amps since the beginning of this series of tests, but doesn't include two important Mark Levinson candidates. One is the current version of the ML-2, which has been reported to us by reliable parties as distinctly superior to the early ML-2's we had tested and reviewed; the other is the new ML-3, a 200/200-watt (into 8 ohms) class AB monster amplifier with an unusually large power supply. We shall see.

Best-sounding power amplifier tested so far, regardless of price, but with numerous caveats (see review above): Rappaport AMP-1.

Close to the best at a spectacularly lower price (and one of the three or four best regard less of price): Hafler DH-200.

---------

[adapted from TAC]

---------

Also see:

Speaker Systems, Large and Small: Updates and New Developments

Various audio and high-fidelity magazines