If you've been looking at my other recent threads (Oppo, Benchmark, Auralic, Harbeth) you'll know that I'm now using a pair of Benchmark AHB2 amps in bridged mono configuration to drive my speakers, the Harbeth Monitor 40.2. To my mind, these are unquestionably the best sounding and performing amps I've owned. As indicated in my post about the audio equipment I've owned, I've owned a few amps in my time before the Benchmarks, including:
Integrated Amps
Dynaco SCA-35 (my first amp and the only pure-tube component I've ever owned)
AR Amplifier (the ideal amp for Dynaquad surround because of its built-in difference circuit--owned twice, once during college and then again to drive vintage AR speakers)
Pioneer SA 9100
Lyngdorf TDAI-2170 (magnificent except for the RoomPerfect)
Receivers
Sony something from the 1980s. It lasted for 20+ years in my bedroom system and never sounded anything other than very good in that context.
Denon AVR5800 (basement home theater system)
Harmon Kardon something (nice sounding in den system until it overheated and failed)
Denon AVR something (den system--okay, but not as good sounding as the HK had been)
Arcam AVR600 (probably the best sounding receiver made to that point in time)
Power Amps
Audionics CC-2 (2)
Amber ST-70
Threshold SA-300
Mark Levinson No.23
Audio by Van Alstine FET-Valve 500 (2)
Cello Duet
Bryston 7B-ST (4)
Bryston 7B-SST (4)
Sanders Magtech Monos (4)
Lyngdorf SA-2400
The list of speakers I've owned and driven with those amps is longer yet. I've also auditioned at stores and shows many other "contenders" before purchasing what I did.
But I feel extremely confident now that I've used the Benchmarks for months now that the AHB2, at least in the bridged mono mode that I use them, is absolutely the finest sounding and otherwise performing power amplifier I know of. Prior reviews by others of the AHB2 are very helpful in describing the product and talking about its sound or lack thereof. I refer readers particularly to the reviews by:
The Absolute Sound
Stereophile
Enjoy the Music
Those reviews basically correctly describe the subjective sound of this amplifier, I think, especially Greg Weaver's early review for Enjoy the Music.
You should also take a look at Benchmark's own measurements and discussion of the THX Achromatic Audio Amplifier (AAA) circuit used in this amp. Benchmark is uniquely forthcoming and detailed about the measurable performance of all its products and this amp is no exception. The measured results published by others (e.g., Stereophile) bear out the honesty and in fact conservative nature of Benchmark's own evaluation and specifications. In terms of Benchmark's literature, see:
AHB Technology (then click on the Technology tab)
Links to Reviews
AHB2 Instruction Manual—performance plots start on page 26
Application Notes
Balanced vs. Unbalanced An alog Interfaces
The Importance of Star-Quad Microphone Cable
Benchmark Speaker Cable – NL2 to Banana – 2 Pole
Amp Set Up
I use three-foot-long Benchmark Studio&Stage Starquad XLR Cable for Analog Audio to feed sound to the AHB2s directly from my Benchmark DAC-3 HGC with its digital volume control. At the moment, there is no intervening preamp. That will change with the addition of the coming-in-June Benchmark HPA4, an all-analog unit which will bypass the DAC's digital volume control in favor of the relay gain control in the new unit, which will drive both the AHB2s and my Audeze LCD-4 headphones.
The speaker cables are the six-foot-long Benchmark Speaker Cable – NL2 to Banana- 2 Pole which have heavy-duty 40-amp SpeakOn connectors at the amp end and locking banana plugs at the speaker end. These are the only cables which Benchmark certifies as being able to deliver the amp's power to speakers with the rated noise and distortion figures. In Benchmark's words:
I do not use cables with SpeakON connections at both ends for the simple reason that my Harbeth M40.2 speaker binding posts are not compatible with the SpeakON connections but are eminently suited to locking bananas. Most audiophile-oriented speakers are not equipped with the SpeakOn jacks, which are very common on pro-audio-oriented speakers. The only speakers I've ever owned which had SpeakOn connections are the Gradient 1.5 Helsinki.
I use the stock power cords which came with the Benchmark amps. They are a special locking design which locks the IEC end into the IEC jack on the back of the amp to prevent accidental disconnection.
Each of the two AHB2 amps sits on its stock feet, centered near the front on a 4-inch-thick Mapleshade solid natural maple platform. Between the platform and the wood floor are three Mapleshade Isoblock 1 footers, one under each front corner and one in the center back of the platform. This arrangement leaves enough room behind each amp for cable dressing so as to avoid contact with the hard foam diffusors behind the amps. None of the speaker cables, interconnects or power cables contact the carpeting.
The amps are both fed from a single 20-amp dedicated circuit feeding this system through a single quad outlet. The line level equipment is fed from another dedicated 20-amp circuit quad outlet. All accessible non-soldered connections are treated with Caig Deoxit Gold G100L brush-on liquid contact enhancer.
The top plate of the amp rings a bit in response to the tap-it-with-my-fingernail test. I damp this with an issue of The Absolute Sound placed atop it.
Aesthetic Qualities
As other reviewers have noted, the first impression is how small this thing is, considering its rated power. It is handsomely finished without looking over-the-top elegant or trying too hard to impress the owner with an overly thick faceplate. The amp is very solid feeling, like a brick, with very uniform weight distribution. The back panel is fairly crowded, but given the recommended-by-Benchmark type of connections I'm using (Benchmark XLR, stock power cord, and Benchmark speaker cables with SpeakOn connectors), the back panel is more than commodious for making such connections.
The amps run only slightly warm to the touch on the heatsinks. This is even with the magazine atop the cover. There is no audible hum or other mechanical noise coming from the chassis.
Sonic Qualities
Inaudible noise: For the first time ever in my more than 50 years of experience with audio systems, there is absolutely no audible noise coming from the speakers with my ear pressed against any of the speaker drivers in a quiet room. No buzz, hum, hiss—nada! This is with an active digital source that is either paused or not playing music, at any setting of the DAC-3 HGC volume control from full mute to wide open.
All other set-ups I've ever owned produced at least some background hiss/hum/buzz with this test. One system I had (which shall remain nameless) produced so much quiescent hiss that I could hear it when I entered the listening room! This is a testament both to the inherently low noise of the combination of the DAC-3 HGC plus AHB2, the proper Benchmark XLR interconnects and speaker cables, plus the proper gain structuring of the active electronic components.
In this context, proper gain structuring means maximizing the signal-to-noise-and-distortion ratio at the loudspeaker input terminals. To do this, you basically run the front-end components as "hot" as possible, voltage-wise, without clipping any device's output or the following input, and then apply the minimum gain in your amplifier to allow the amp to produce just a bit more than its rated output into the speakers with the system volume control (the volume control of the DAC-3 HGC) set at maximum.
To do this, with the Benchmark DAC-3 HGC, you eliminate any output pads by moving, if necessary, some jumpers inside the chassis. Then, on the back panel of the AHB2, you set the amplifier gain/sensitivity control to its minimum level of 22 dBu (the lowest position of the three-position switch.)
The use of all-balanced connections and truly balanced internal electronics in the Benchmark equipment surely also contributes to this, as discussed in the Balanced vs. Unbalanced Analog Interfaces link. Also contributing, I'm sure, is the use of star-quad interconnecting wire in the Benchmark XLR interconnects and speaker cables, such cable being the only type which can successfully lower magnetic field interference, as discussed in The Importance of Star-Quad Microphone Cable link.
The sonic consequence of the absolute absence of audible background electronic noise is that all the low-level detail present in any recording is fully presented since the sound to noise and distortion level of the electronics is greater than that in the recording and is thus not the limiting factor. The electronics thus cannot mask or distort the program source. Reverberant tails of concert hall ambiance, for instance, last longer than with any other amps in my experience. With suitable program material (that lacking inherent recorded background hiss or other electronic or venue noise) the background is supremely black.
A few commenters feel that comparing the noise/distortion level produced by the Benchmark AHB2 amp to that of other amps is not fair since it is not an apples-to-apples comparison. The Benchmark, when set up for proper gain structure, has much lower gain than most other amps. Setting it up for the higher gain of most consumer amps would reduce or even possibly eliminate its edge in quietness.
Such arguments are beside the point. Most consumer amps do not allow proper gain structuring at all since they have no way to adjust the sensitivity of the amplifier. If the amplifier doesn't need more than a low amount of gain to produce its rated output when driven by a properly "hot" source, why make or set up the amp with more gain than needed? All that accomplishes is lowering the designed sound-to-noise-and-distortion ratio and create the real possibility of input clipping at the amplifier. It is not Benchmark's fault that most consumer front-end electronics are not capable of producing the suitably "hot" pro-audio-level voltages without clipping their outputs and therefore cannot be used with the AHB2's low sensitivity input setting. Most consumer audio front-end electronics produce about 2 volts out from an unbalanced analog output, and about 4 volts out from balanced outputs. Benchmark's DAC-3 HGC produces the much-higher-yet pro-audio level of +24 dBu out.
Uncolored response: I hear no frequency response issues whatsoever. Program material varies a lot from one program to another with no detectable "cast" which might be caused by coloration of the electronics. See the comments in the Enjoy the Music review.
This detail retrieval I talk about above is true, that is, these amps truly reveal more recorded detail than others. The audible detail is not enhanced by any added brightness in the upper ranges. Instruments all sound as they should in the mids and highs and the bass is a remarkable combination of control, depth, and fullness—just right.
More than enough power: While not rated as high in power as, say, the Sanders Magtech Monoblocs I once owned, the bridged AHB2s sound at least as powerful into the Harbeth M40-series speakers as those Sanders amps. There is an easy, rock-solid quality to the sound unequalled by any other amps I've owned. The dynamic range seems altogether superior to that of the Sanders, probably because of the vastly lower noise level of the AHB2s. The stock combination of the Sanders amps and Harbeth M40.1 speakers produced hiss/hum from the speakers audible from a couple of feet away.
My Harbeth M40.2 speakers show audible signs of stress far before the amplifier runs out of steam. Even the levels at which the speakers begin to show signs of stress are far beyond what I can listen to comfortably without ear plugs for more than a few moments.
Three-dimensionality: Unequaled in my experience, pure and simple. Rock solid and rounded images; wide, deep, and stable staging; fully developed height illusion; finely graduated depth perspective.
Shortcomings?
The quibbles noted at the end of the Enjoy the Music review are actually strengths of the design. First, the lack of inclusion of single-ended inputs is intentional so as to strongly encourage the user to use balanced connections in order to take full advantage of the low-noise-and-distortion capabilities of the amplifier. See the discussion in the Balanced vs. Unbalanced Analog Interfaces Benchmark application note mentioned above.
Second, the difficulty of using the safety series WBT speaker cable binding posts on the back of the amp with spade lugs is of no consequence if you are using Benchmark's recommended connection methods. For best sound/lowest measurable distortion with this amp, Benchmark recommends using as the best choice the SpeakON connectors, and as a second choice, locking banana plug connectors. Benchmark's literature specifically mentions that spade lugs cannot be relied on to deliver a low distortion signal at high power levels; see the above quote from the Benchmark literature about this point.
Conclusion
There is actually very little to say, sonically, about the Benchmark AHB2. As far as I can detect, it does nothing wrong and more right than any other amp I have ever heard or owned. As other reviewers have said, nothing made by the hands of man can be perfect, but these amps, for all practical purposes, are as close to perfection as one could want and far closer than most others. An admirable achievement at a price which, while certainly not cheap, seems very fair for what you get.
Integrated Amps
Dynaco SCA-35 (my first amp and the only pure-tube component I've ever owned)
AR Amplifier (the ideal amp for Dynaquad surround because of its built-in difference circuit--owned twice, once during college and then again to drive vintage AR speakers)
Pioneer SA 9100
Lyngdorf TDAI-2170 (magnificent except for the RoomPerfect)
Receivers
Sony something from the 1980s. It lasted for 20+ years in my bedroom system and never sounded anything other than very good in that context.
Denon AVR5800 (basement home theater system)
Harmon Kardon something (nice sounding in den system until it overheated and failed)
Denon AVR something (den system--okay, but not as good sounding as the HK had been)
Arcam AVR600 (probably the best sounding receiver made to that point in time)
Power Amps
Audionics CC-2 (2)
Amber ST-70
Threshold SA-300
Mark Levinson No.23
Audio by Van Alstine FET-Valve 500 (2)
Cello Duet
Bryston 7B-ST (4)
Bryston 7B-SST (4)
Sanders Magtech Monos (4)
Lyngdorf SA-2400
The list of speakers I've owned and driven with those amps is longer yet. I've also auditioned at stores and shows many other "contenders" before purchasing what I did.
But I feel extremely confident now that I've used the Benchmarks for months now that the AHB2, at least in the bridged mono mode that I use them, is absolutely the finest sounding and otherwise performing power amplifier I know of. Prior reviews by others of the AHB2 are very helpful in describing the product and talking about its sound or lack thereof. I refer readers particularly to the reviews by:
The Absolute Sound
Stereophile
Enjoy the Music
Those reviews basically correctly describe the subjective sound of this amplifier, I think, especially Greg Weaver's early review for Enjoy the Music.
You should also take a look at Benchmark's own measurements and discussion of the THX Achromatic Audio Amplifier (AAA) circuit used in this amp. Benchmark is uniquely forthcoming and detailed about the measurable performance of all its products and this amp is no exception. The measured results published by others (e.g., Stereophile) bear out the honesty and in fact conservative nature of Benchmark's own evaluation and specifications. In terms of Benchmark's literature, see:
AHB Technology (then click on the Technology tab)
Links to Reviews
AHB2 Instruction Manual—performance plots start on page 26
Application Notes
Balanced vs. Unbalanced An alog Interfaces
The Importance of Star-Quad Microphone Cable
Benchmark Speaker Cable – NL2 to Banana – 2 Pole
Amp Set Up
I use three-foot-long Benchmark Studio&Stage Starquad XLR Cable for Analog Audio to feed sound to the AHB2s directly from my Benchmark DAC-3 HGC with its digital volume control. At the moment, there is no intervening preamp. That will change with the addition of the coming-in-June Benchmark HPA4, an all-analog unit which will bypass the DAC's digital volume control in favor of the relay gain control in the new unit, which will drive both the AHB2s and my Audeze LCD-4 headphones.
The speaker cables are the six-foot-long Benchmark Speaker Cable – NL2 to Banana- 2 Pole which have heavy-duty 40-amp SpeakOn connectors at the amp end and locking banana plugs at the speaker end. These are the only cables which Benchmark certifies as being able to deliver the amp's power to speakers with the rated noise and distortion figures. In Benchmark's words:
The SpeakON™ connectors provide a reliable low-impedance connection that can withstand high-currents. Of the more traditional speaker connectors, we have found that locking banana plugs provide the next best connections. In contrast, spade lugs and pins often provide poor connections. Spring-type (non-locking) banana plugs almost always provide poor connections and should be avoided.
The Benchmark AHB2 power amplifier is so clean that poor speaker connections can easily create more distortion than the entire power amplifier. We have seen this many times in the lab. The effects of poor speaker connections can easily be measured when testing the AHB2 power amplifier. We have learned that we cannot accurately measure the AHB2 unless we use SpeakON™ connectors or high-quality locking banana connectors.
SpeakON™ connectors also provide convenience and safety. No speaker connector is easier to use, they cannot short out, and they cannot give you a shock.
I do not use cables with SpeakON connections at both ends for the simple reason that my Harbeth M40.2 speaker binding posts are not compatible with the SpeakON connections but are eminently suited to locking bananas. Most audiophile-oriented speakers are not equipped with the SpeakOn jacks, which are very common on pro-audio-oriented speakers. The only speakers I've ever owned which had SpeakOn connections are the Gradient 1.5 Helsinki.
I use the stock power cords which came with the Benchmark amps. They are a special locking design which locks the IEC end into the IEC jack on the back of the amp to prevent accidental disconnection.
Each of the two AHB2 amps sits on its stock feet, centered near the front on a 4-inch-thick Mapleshade solid natural maple platform. Between the platform and the wood floor are three Mapleshade Isoblock 1 footers, one under each front corner and one in the center back of the platform. This arrangement leaves enough room behind each amp for cable dressing so as to avoid contact with the hard foam diffusors behind the amps. None of the speaker cables, interconnects or power cables contact the carpeting.
The amps are both fed from a single 20-amp dedicated circuit feeding this system through a single quad outlet. The line level equipment is fed from another dedicated 20-amp circuit quad outlet. All accessible non-soldered connections are treated with Caig Deoxit Gold G100L brush-on liquid contact enhancer.
The top plate of the amp rings a bit in response to the tap-it-with-my-fingernail test. I damp this with an issue of The Absolute Sound placed atop it.
Aesthetic Qualities
As other reviewers have noted, the first impression is how small this thing is, considering its rated power. It is handsomely finished without looking over-the-top elegant or trying too hard to impress the owner with an overly thick faceplate. The amp is very solid feeling, like a brick, with very uniform weight distribution. The back panel is fairly crowded, but given the recommended-by-Benchmark type of connections I'm using (Benchmark XLR, stock power cord, and Benchmark speaker cables with SpeakOn connectors), the back panel is more than commodious for making such connections.
The amps run only slightly warm to the touch on the heatsinks. This is even with the magazine atop the cover. There is no audible hum or other mechanical noise coming from the chassis.
Sonic Qualities
Inaudible noise: For the first time ever in my more than 50 years of experience with audio systems, there is absolutely no audible noise coming from the speakers with my ear pressed against any of the speaker drivers in a quiet room. No buzz, hum, hiss—nada! This is with an active digital source that is either paused or not playing music, at any setting of the DAC-3 HGC volume control from full mute to wide open.
All other set-ups I've ever owned produced at least some background hiss/hum/buzz with this test. One system I had (which shall remain nameless) produced so much quiescent hiss that I could hear it when I entered the listening room! This is a testament both to the inherently low noise of the combination of the DAC-3 HGC plus AHB2, the proper Benchmark XLR interconnects and speaker cables, plus the proper gain structuring of the active electronic components.
In this context, proper gain structuring means maximizing the signal-to-noise-and-distortion ratio at the loudspeaker input terminals. To do this, you basically run the front-end components as "hot" as possible, voltage-wise, without clipping any device's output or the following input, and then apply the minimum gain in your amplifier to allow the amp to produce just a bit more than its rated output into the speakers with the system volume control (the volume control of the DAC-3 HGC) set at maximum.
To do this, with the Benchmark DAC-3 HGC, you eliminate any output pads by moving, if necessary, some jumpers inside the chassis. Then, on the back panel of the AHB2, you set the amplifier gain/sensitivity control to its minimum level of 22 dBu (the lowest position of the three-position switch.)
The use of all-balanced connections and truly balanced internal electronics in the Benchmark equipment surely also contributes to this, as discussed in the Balanced vs. Unbalanced Analog Interfaces link. Also contributing, I'm sure, is the use of star-quad interconnecting wire in the Benchmark XLR interconnects and speaker cables, such cable being the only type which can successfully lower magnetic field interference, as discussed in The Importance of Star-Quad Microphone Cable link.
The sonic consequence of the absolute absence of audible background electronic noise is that all the low-level detail present in any recording is fully presented since the sound to noise and distortion level of the electronics is greater than that in the recording and is thus not the limiting factor. The electronics thus cannot mask or distort the program source. Reverberant tails of concert hall ambiance, for instance, last longer than with any other amps in my experience. With suitable program material (that lacking inherent recorded background hiss or other electronic or venue noise) the background is supremely black.
A few commenters feel that comparing the noise/distortion level produced by the Benchmark AHB2 amp to that of other amps is not fair since it is not an apples-to-apples comparison. The Benchmark, when set up for proper gain structure, has much lower gain than most other amps. Setting it up for the higher gain of most consumer amps would reduce or even possibly eliminate its edge in quietness.
Such arguments are beside the point. Most consumer amps do not allow proper gain structuring at all since they have no way to adjust the sensitivity of the amplifier. If the amplifier doesn't need more than a low amount of gain to produce its rated output when driven by a properly "hot" source, why make or set up the amp with more gain than needed? All that accomplishes is lowering the designed sound-to-noise-and-distortion ratio and create the real possibility of input clipping at the amplifier. It is not Benchmark's fault that most consumer front-end electronics are not capable of producing the suitably "hot" pro-audio-level voltages without clipping their outputs and therefore cannot be used with the AHB2's low sensitivity input setting. Most consumer audio front-end electronics produce about 2 volts out from an unbalanced analog output, and about 4 volts out from balanced outputs. Benchmark's DAC-3 HGC produces the much-higher-yet pro-audio level of +24 dBu out.
Uncolored response: I hear no frequency response issues whatsoever. Program material varies a lot from one program to another with no detectable "cast" which might be caused by coloration of the electronics. See the comments in the Enjoy the Music review.
This detail retrieval I talk about above is true, that is, these amps truly reveal more recorded detail than others. The audible detail is not enhanced by any added brightness in the upper ranges. Instruments all sound as they should in the mids and highs and the bass is a remarkable combination of control, depth, and fullness—just right.
More than enough power: While not rated as high in power as, say, the Sanders Magtech Monoblocs I once owned, the bridged AHB2s sound at least as powerful into the Harbeth M40-series speakers as those Sanders amps. There is an easy, rock-solid quality to the sound unequalled by any other amps I've owned. The dynamic range seems altogether superior to that of the Sanders, probably because of the vastly lower noise level of the AHB2s. The stock combination of the Sanders amps and Harbeth M40.1 speakers produced hiss/hum from the speakers audible from a couple of feet away.
My Harbeth M40.2 speakers show audible signs of stress far before the amplifier runs out of steam. Even the levels at which the speakers begin to show signs of stress are far beyond what I can listen to comfortably without ear plugs for more than a few moments.
Three-dimensionality: Unequaled in my experience, pure and simple. Rock solid and rounded images; wide, deep, and stable staging; fully developed height illusion; finely graduated depth perspective.
Shortcomings?
The quibbles noted at the end of the Enjoy the Music review are actually strengths of the design. First, the lack of inclusion of single-ended inputs is intentional so as to strongly encourage the user to use balanced connections in order to take full advantage of the low-noise-and-distortion capabilities of the amplifier. See the discussion in the Balanced vs. Unbalanced Analog Interfaces Benchmark application note mentioned above.
Second, the difficulty of using the safety series WBT speaker cable binding posts on the back of the amp with spade lugs is of no consequence if you are using Benchmark's recommended connection methods. For best sound/lowest measurable distortion with this amp, Benchmark recommends using as the best choice the SpeakON connectors, and as a second choice, locking banana plug connectors. Benchmark's literature specifically mentions that spade lugs cannot be relied on to deliver a low distortion signal at high power levels; see the above quote from the Benchmark literature about this point.
Conclusion
There is actually very little to say, sonically, about the Benchmark AHB2. As far as I can detect, it does nothing wrong and more right than any other amp I have ever heard or owned. As other reviewers have said, nothing made by the hands of man can be perfect, but these amps, for all practical purposes, are as close to perfection as one could want and far closer than most others. An admirable achievement at a price which, while certainly not cheap, seems very fair for what you get.
Last edited: