Benchmarking the Benchmark... AHB2 Amplifier

Collapse
This topic is closed.
X
X
 
  • Time
  • Show
Clear All
new posts
  • JonMarsh
    Mad Max Moderator
    • Aug 2000
    • 15259

    Benchmarking the Benchmark... AHB2 Amplifier

    OnTheBench.key by Jon Hancock, on Flickr

    You know, sometimes you know right away good things are happening. In this case, there is no mystery... I've seen the other reviews... I've seen John Atkinson's measurements on Stereophile...


    The Benchmark AHB2 is a nicely proportioned little amp that seeing just a picture of the front may not tip you off to how diminutive it is... but seeing how those XLR and SpeakOn connectors fill the rear panel, you realize this is pretty itty bitty for an analog power amplifier (to be fair, it does use a very sophisticated resonant switching power supply, though).


    20150726132016_BenchmarkAHB2AmpFrontWeb by Jon Hancock, on Flickr


    And I've listened to these for hours and hours since just before Thanksgiving, last November. But nonetheless, when I finished connecting one of these to the AP, set the gain control switch to the mid point (adjustable between true Pro levels, a compromise level, and consumer input level- I went for the middle ground) and fired it up, and default it's showing a moderate level output signal (7VRMS with 1VRMS input), reading out the levels and THD+N, with a default FFT window, and right away, you know there's something special here... something one might have not believed could exist commercially for under $10K or $20K, much less for $3K.






    FirstCheck_SmallWindow by Jon Hancock, on Flickr




    No, you didn't misread that; at 6 watts output power, the THD+N (Total Harmonic Distortion + Noise) is just 0.00042%. Just think about how quiet the front end on this has to be to reach that level of performance... and about how your preamp does or doesn't match up to that!


    So, is this a "First Watt" design? optimized just for low Power? Let's take a look. (Keep in mind I DIDN'T use the Pro level input setting, which requires even more front end drive, and improves the SN ratio further- this middle ground setting will work fine with most good stereo preamps, say, units comfortable driving up to 4 or 5VRMS out). So don't expect to see as low numbers as JA published. This is a little more real world... :W




    First, we'll look at the swept sine behavior starting from a level a bit below 100mW and proceeding up to the clip point, at 1kHz.


    THD+N Ratio vs Measured Level Benchmark AHB2 by Jon Hancock, on Flickr.


    It's possible with with more careful approach to lead dress and the test setup, I could have gotten better numbers than this, but let's leave that for the future for now.


    One of the things one ought to do to appreciate this amp is compare what it's THD+N figures look like at 100mW compared to may other "high end" amps out there- here, we're coming in at 0.003%, and I've seen many amps for 10X the money coming in at 0.02 to 0.07% at 100mW. For a speaker that's 90 dB efficient, 100mW is a conversational playback level.


    By 1W we're down to 0.001% THD+N, and 20W we've reached 0.0003%! Yes, three zeros in front of that 3!


    OK, that's impressive, but what about that bridged mono mode? Benchmark supports in directly with a separate SpeakOn output connector, and using the balanced inputs and stereo-mono switch. Move the SpeakOn output connector (you did get some of the Benchmark Speaker cables, didn't you?) flip a switch, and voila, a miniature bridged mono block amp.




    THD+N Ratio vs Measured Level Benchmark AHB2 Balanced Mono by Jon Hancock, on Flickr


    Now we're seeing a bit past 350W at the clipping point- because the two channels are being added in series, the noise floor is doubled, but we're still looking at less than 0.001% at 10W output.


    Let's widen the frequency range of investigation- but keep in in bridged mono. It's a very competitive $2995 mono block, given the overall performance.


    Next up some frequency sweeps at different power levels.


    First, at 50W RMS.


    THD+N Ratio 50W Bridged Mono by Jon Hancock, on Flickr


    Just under 0.0005% up to 5 kHz, rising smoothly to about 0.0015% at 20 kHz.


    Let's push it harder- 200W now.


    THD+N Ratio 200W Bridged Mono by Jon Hancock, on Flickr


    Now we're seeing the effect of the noise floor dropping further, with lower THD +N below 3 kHz, but perhaps at this higher output level and higher frequency range, the feedforward error correction can't quite work as well, and the distortion is rising to a peak level of 0.003% at 20 kHz. Tsk, tsk. :W


    I personally like FFT's more than summed distortion measurements like this- the latter show the large trends, the former the nitty gritty detail. Let's try some low and high frequency examples.




    First, 50Hz at 100W output in bridged mode. This is interesting because it can reveal both signal related harmonic distortion and power supply related noise, with harmonics related to the line frequency. Keeping in mind the output signal is at +30 db rA, 2nd harmonic is essentially none-existent, and 3rd harmonic at 150Hz is -135dB from the signal output level. There are some power line related spuriae; at 180Hz, 300Hz, 420Hz, etc, but generally not above -130 dB relative to the 100W output.


    FFT Spectrum 50Hz 100W AHB2 by Jon Hancock, on Flickr




    Considering the rise in high frequency distortion, an obvious avenue of investigation is two tone CCIF Intermodulation distortion, with 19 + 20 kHz. They're mixed at equal levels, and setup to reach a peak output of 40V, equivalent to the peak level for 100W RMS output. Needless to say, this would be very quick death to almost any tweeter, but it's a good stress test for HF distortion, both harmonic and intermodulation.


    Here is what the test signal looks like:




    Scope_AHB2_CCIF by Jon Hancock, on Flickr




    And this is an FFT of the results, plotted from DC to 40kHz. Ideally there would only be the two spikes reading out at 19kHz and 20kHz, but we can see a variety of side band components at 1 kHz intervals, both around the test signal, and at an image at the 2nd harmonic of the test signal. There are also difference tones present at 1 and 2 kHz.


    FFT Spectrum 19+20kHz 100W Pk AHB2 by Jon Hancock, on Flickr


    Let's keep in mind that most of these are below the -100dB point, which means they are 122.5db below the actual signal! The worst case components, at 18 and 21 kHz, are about -95 dB relative to the actual signal peaks. This calculates to a percentage basis of 0.0017%. This is a very low number compared with most power amplifiers.




    Of course, what really matters is what it sounds like- or if it sounds like nothing at all, just the proverbial piece of wire with gain. I haven't heard or measured anything else that comes as close to achieving this ideal, so if this characteristics is what you want in your system, give these an audition before spending money elsewhere.




    Positives


    • Very compact for the amount of output power, and not being Class D
    • Run very cool, not a high bias design
    • Transparent as all get-out, with tight articulate bass- but with no extra fat
    • Stunningly good measurements- state of the art
    • Available with choice of either black anodized or silver anodized front panels
    • SpeakOn and binding post output connectors
    • XLR input connectors






    Concerns
    • No unbalanced input connectors - you'll have to use cable adapters
    • Best performance does require a high performance preamp-
    • Most preamps can't match the S/N or THD performance of this amplifier
    • It would be strikingly unfortunate if your preamp is bigger and hotter than your power amp - this is completely plausible with this unit.
    • Not available in Rose Gold...
    • Where am I going to come up with the money for a pair of these after spending so much on test equipment this year? (this pair is on loan from their owner).
    Last edited by JonMarsh; 17 April 2016, 18:42 Sunday.
    the AudioWorx
    Natalie P
    M8ta
    Modula Neo DCC
    Modula MT XE
    Modula Xtreme
    Isiris
    Wavecor Ardent

    SMJ
    Minerva Monitor
    Calliope
    Ardent D

    In Development...
    Isiris Mk II updates- in final test stage!
    Obi-Wan
    Saint-Saƫns Symphonique/AKA SMJ-40
    Modula PWB
    Calliope CC Supreme
    Natalie P Ultra
    Natalie P Supreme
    Janus BP1 Sub


    Resistance is not futile, it is Volts divided by Amperes...
    Just ask Mr. Ohm....
Working...
😀
😂
🥰
😘
🤢
😎
😞
😡
👍
👎
Searching...Please wait.
An unexpected error was returned: 'Your submission could not be processed because you have logged in since the previous page was loaded.

Please push the back button and reload the previous window.'
An unexpected error was returned: 'Your submission could not be processed because the token has expired.

Please push the back button and reload the previous window.'
An internal error has occurred and the module cannot be displayed.
There are no results that meet this criteria.
Search Result for "|||"