Review and Measurements of Benchmark AHB2 Amp

[Coach_Kaarlo]

It should fine because there are number of users (and I have a friend) using the AHB2 bridged to drive the Salon2 and it dips below 4 Ohms.

View attachment 114876

- Rich

 

[RobS]

So how come Benchmark doesn't rate mono at 4ohm then?

 

[RichB]

So how come Benchmark doesn't rate mono at 4ohm then?

The documentation posted by @Coach_Kaarlo discusses the nominal speaker load that are often posted by manufacturers.
Nominal loads for a speaker at 6 Ohms can dip below 4 Ohms. In bridged mode, the AHB2 will trigger protection if your speaker dips below 2.6 Ohms for some short period at high output.

- Rich

 

[pogo]

It should fine because there are number of users (and I have a friend) using the AHB2 bridged to drive the Salon2 and it dips below 4 Ohms.

View attachment 114876

- Rich

A high damping factor is particularly noticeable in loudspeakers with a large mechanical size and low impedance. An impedance curve does not represent the whole truth, since the mechanical size that has to be guided is not visible.

 

[AudioExplorer]

A high damping factor is particularly noticeable in loudspeakers with a large mechanical size and low impedance. An impedance curve does not represent the whole truth, since the mechanical size that has to be guided is not visible.

For many amps including the AHB-2, the damping factor goes down as frequency increases. But, maybe this is not a problem because of what you said above, i.e., the mechanical size of the drivers is smaller at higher frequencies?

 

[RobS]

Thanks very much Rich for your help.

---

As an aside...

I would like to say a few things about my recent ownership of the AHB2.

First of all, I think it must be made clear that the sensitivity switch on the back of the amplifier is crucial in ones enjoyment of this amplifier..

In it's default position at the low-gain+22dbu setting it sounds worse than a cheap FiiO portable amplifier. I am serious. It's like a decaying corpse getting baked in the sun. Music is totally devoid of life (IOW no microdynamics whatsoever) and it's completely flat and dull. It has no balls. I would rather not listen to music and stick my head in a blender. Yet strangely, no matter how dull it sounded, it didn't sound sterile or clinical.

What could be wrong? Do we not have the right measurements that can demonstrate how lifeless this amplifier made music?

Then I got to thinking maybe too much feedback was being applied for the low gain setting, which is typical for headphone amplifiers in my experience. More feedback just kills the sound.

So I was doing some reading around and I realized that the low-gain setting was really designed for Benchmark's extremely hot outputting DACs. My current DAC has about 6.3VRMS-ish max output, so I don't necessarily have a gain problem, it was just a terrible match for this low-gain setting.

Ok so Benchmark recommended for unbalanced connections to use the high-gain setting. Well I don't have an unbalanced source. I'm going from Focusrite RedNet D16 -> Dangerous Music Convert-2 DAC -> JBL Nano Patch+ -> AHB2 Amplifier -> Dynaudio Heritage Special. So the entire chain here is balanced.

Well I ignored the recommendation by Benchmark to use either of the two bottom settings for the balanced and went straight for the high-gain, thinking that less negative feedback would be applied.

And voila, a much much better sound emerged from my speakers!

This has to be the cleanest amplifier I've ever heard. It sounds so even tonally with no hint of grit, grain, steeliness, hardness, etc. Treble is exceptionally clear and smooth with no harshness, the bottom end has tight with whip-cracking speed and goes really deep with lots of heft when the recording calls for it. Midrange has no artificial bloom or anything that would unnaturally draw attention to itself.

The real surprise for me is the transients and dynamics. I find the AHB2 to deliver all the raw ferocity and snap found on the debut Rage Against the Machine, or Prodigy's The Fat of The Land, and so on. It can get down and dirty. Lots of low level detail retrieval and textures, even more resolving than the Pass Labs XA25 amplifier I had previously. The Pass has bloomed over transients and doesn't slam hard with an articulate clean and attack like the AHB2 does.

Soundstage and imaging is superb. Convincing (re)creation/(re)production of a three dimensional stage, with laser precision where you can make out all the instrument placements and layering within the stage. Not the last word in depth, but there's enough to where I don't think I'm listening to aural wallpaper.

Timbre is very good, lifelike in some aspects.

But more than all this, I think it is the sheer effortlessness by which the AHB2 renders sound. It never once sounds like it is straining or compressed or distorted. I'm super sensitive to compression and distortion with two channel reproduction. There is an openness that lets the sound breath because it's not trying to strain itself. I like this quality of the AHB2 a lot.

So anyway, my beginning assessment with the AHB2 was how neutered and flat it sounded that was offensive when you consider the cost for it. But after a simple gain tweak, the AHB2 came into focus and revealed all the good things it does so good at.

It is a very neutral amp. I'm not hearing any tilt towards warmth or brightness (no timbral warmth either). This might be boring to some folks who want a more colored or "fun" sound" I try to get razor sharp accuracy with my two channel setup because I want music to sound dynamic as hell which is what live music is, and I don't want a sleepy, mushy, gooey sound signature. Headphones I have different priorities but I'm trying to get as close to life-like accuracy as I can.

Anyway, highly recommend this amplifier for its clarity, timbre, imaging, transients, even tonality, lack of grain. If you like the colored sound ala Pass, this is definitely not the amplifier for you.

Thanks Amir for your measurements and everyone else at ASR that has contributed to this thread. It has been very helpful in finding my two-channel amplifier.

 

[RichB]

Here is a paper from Benchmark concerning damping factor:

Audio Myth - "Damping Factor Isn't Much of a Factor" - Benchmark Media Systems

When you bridge, there could be more than .1 dB with a speaker that dips below 4 Ohms but that can be address by using 11 or 12 gauge wire.

Bridged into @RobS speakers (minimum impedance 4 Ohms) computes the following maximum attenuation of 0.13 dB. This will happen in bass frequencies doubtful to be heard.

- Rich

 

[pogo]

For many amps including the AHB-2, the damping factor goes down as frequency increases. But, maybe this is not a problem because of what you said above, i.e., the mechanical size of the drivers is smaller at higher frequencies?

In my setup this is a problem. See also post #2161 & 2167.

 

[pogo]

Bridged into @RobS speakers (minimum impedance 4 Ohms) computes the following maximum attenuation of 0.13 dB. This will happen in bass frequencies doubtful to be heard.

The 0.13db is the minor problem. Rather, the swing out behavior may not be properly suppressed, which leads to a washed-out low-frequency range.

 

[RichB]

In my setup this is a problem. See also post #2161 & 2167.

What are your speakers?

- Rich

 

[GoldenOne]

Thanks very much Rich for your help.

---

As an aside...

I would like to say a few things about my recent ownership of the AHB2.

First of all, I think it must be made clear that the sensitivity switch on the back of the amplifier is crucial in ones enjoyment of this amplifier..

In it's default position at the low-gain+22dbu setting it sounds worse than a cheap FiiO portable amplifier. I am serious. It's like a decaying corpse getting baked in the sun. Music is totally devoid of life (IOW no microdynamics whatsoever) and it's completely flat and dull. It has no balls. I would rather not listen to music and stick my head in a blender. Yet strangely, no matter how dull it sounded, it didn't sound sterile or clinical.

What could be wrong? Do we not have the right measurements that can demonstrate how lifeless this amplifier made music?

Then I got to thinking maybe too much feedback was being applied for the low gain setting, which is typical for headphone amplifiers in my experience. More feedback just kills the sound.

So I was doing some reading around and I realized that the low-gain setting was really designed for Benchmark's extremely hot outputting DACs. My current DAC has about 6.3VRMS-ish max output, so I don't necessarily have a gain problem, it was just a terrible match for this low-gain setting.

Ok so Benchmark recommended for unbalanced connections to use the high-gain setting. Well I don't have an unbalanced source. I'm going from Focusrite RedNet D16 -> Dangerous Music Convert-2 DAC -> JBL Nano Patch+ -> AHB2 Amplifier -> Dynaudio Heritage Special. So the entire chain here is balanced.

Well I ignored the recommendation by Benchmark to use either of the two bottom settings for the balanced and went straight for the high-gain, thinking that less negative feedback would be applied.

And voila, a much much better sound emerged from my speakers!

This has to be the cleanest amplifier I've ever heard. It sounds so even tonally with no hint of grit, grain, steeliness, hardness, etc. Treble is exceptionally clear and smooth with no harshness, the bottom end has tight with whip-cracking speed and goes really deep with lots of heft when the recording calls for it. Midrange has no artificial bloom or anything that would unnaturally draw attention to itself.

The real surprise for me is the transients and dynamics. I find the AHB2 to deliver all the raw ferocity and snap found on the debut Rage Against the Machine, or Prodigy's The Fat of The Land, and so on. It can get down and dirty. Lots of low level detail retrieval and textures, even more resolving than the Pass Labs XA25 amplifier I had previously. The Pass has bloomed over transients and doesn't slam hard with an articulate clean and attack like the AHB2 does.

Soundstage and imaging is superb. Convincing (re)creation/(re)production of a three dimensional stage, with laser precision where you can make out all the instrument placements and layering within the stage. Not the last word in depth, but there's enough to where I don't think I'm listening to aural wallpaper.

Timbre is very good, lifelike in some aspects.

But more than all this, I think it is the sheer effortlessness by which the AHB2 renders sound. It never once sounds like it is straining or compressed or distorted. I'm super sensitive to compression and distortion with two channel reproduction. There is an openness that lets the sound breath because it's not trying to strain itself. I like this quality of the AHB2 a lot.

So anyway, my beginning assessment with the AHB2 was how neutered and flat it sounded that was offensive when you consider the cost for it. But after a simple gain tweak, the AHB2 came into focus and revealed all the good things it does so good at.

It is a very neutral amp. I'm not hearing any tilt towards warmth or brightness (no timbral warmth either). This might be boring to some folks who want a more colored or "fun" sound" I try to get razor sharp accuracy with my two channel setup because I want music to sound dynamic as hell which is what live music is, and I don't want a sleepy, mushy, gooey sound signature. Headphones I have different priorities but I'm trying to get as close to life-like accuracy as I can.

Anyway, highly recommend this amplifier for its clarity, timbre, imaging, transients, even tonality, lack of grain. If you like the colored sound ala Pass, this is definitely not the amplifier for you.

Thanks Amir for your measurements and everyone else at ASR that has contributed to this thread. It has been very helpful in finding my two-channel amplifier.

Great write-up.
I also found low gain to sound pretty flat compared to medium/high.
I really don't know why this is, especially given as its actually quite a noticeable difference, but I did find that when in low gain the amp has only one relay click on startup/shutdown. Whereas in medium/high gain it has two. I'm not sure what this implies but it does seem something is being done differently mechanically inside the amp when in low gain.

BUT, given as the performance is so crazy, there's not really much reason not to just stick it in medium gain anyway.
Even with ~30dB of attenuation on the input signal you'll still be getting better performance than most other speaker amplifiers.

 

[RichB]

The AHB2 uses a regulated power supply so it may provide better bass where a linear power supply sags and produces more distortion.

Here is search of all @John_Siau on this thread:

(4) Search results | Page 3 | Audio Science Review (ASR) Forum

Some of my favorites are below.

Review and Measurements of Benchmark AHB2 Amp
The AHB2 uses feed-forward correction and some feedback. The feed-forward correction is not subject to loop delays and can provide correction at very high frequencies. In contrast, feedback systems have a limited correction bandwidth. If the bandwidth of a feedback system is not adequately...


Review and Measurements of Benchmark AHB2 Amp
The amplifier will stay clean when driving very low impedances. The published rating is the long-term continuous output power and this is a function of heat sink area. Driving 3-Ohm, 2-Ohm or 1-Ohm impedance dips are not a problem. What is unique about the AHB2 is that is stays clean when...


Review and Measurements of Benchmark AHB2 Amp
The AHB2 uses feed-forward correction and some feedback. The feed-forward correction is not subject to loop delays and can provide correction at very high frequencies. In contrast, feedback systems have a limited correction bandwidth. If the bandwidth of a feedback system is not adequately...

Review and Measurements of Benchmark AHB2 Amp
Your assumptions are well thought out but incorrect. The AHB2 power supply is tightly regulated. The transients are not regulated. If the clip lights on the AHB2 are not turning on, the transients are not being clipped, attenuated, or distorted. The clip lights have a timer that keeps them on...

- Rich

 

[pogo]

What are your speakers?

- Rich

In my possession for 30 years and without the need to change

Extract from special print (1988)

 

[DonH56]

Ok so to give you an example here:

Would this result in a mismatch with having monoblock AHB2?

I am not sure how you define "mismatch". The AHB2's specs for damping factor are:

DAMPING FACTOR

• 350 at 20 Hz, 8-Ohms
• 254 at 1 kHz, 8-Ohms
• 34 at 20 kHz, 8-Ohms
• 7 at 200 kHz, 8-Ohms

Looking at the plot, there is a high peak at 20 Hz that is off the chart (use 30 ohms), an 18-ohm peak at 60 Hz, a min of ~4 ohms around150 Hz, and it's about 9 ohms at 20 kHz. I don't feel like trying to interpolate the DF for this so just using some guesses and looking at the voltage division:

Zout(20 Hz) = 8/350 = 0.023 ohms => dB change = 20log[30/(30+0.023)] = -0.0066 dB
Zout(70 Hz) ~ 8/300 = 0.027 ohms => dB change = 20log[18/(18+0.027)] = -0.013 dB
Zout(150 Hz) ~ 8/300 = 0.027 ohms => dB change = 20log[4/(4+0.027)] = -0.058 dB
Zout(20 kHz) = 8/34 = 0.235 ohms => dB change = 20log[9/(9+0.235)] = -0.22 dB

So you have about 0.22 dB spread from 20 Hz to 20 kHz, much of that at the highest frequencies where I really doubt it matters. This is for a single channel so you must double Zout for bridged operation. That is about -0.013 dB at 20 Hz to -0.44 dB at 20 kHz, and about -0.12 dB at 150 Hz.

You also need to include the resistance of the cables to the speakers as well (single wire times two since it is the round-trip resistance that matters for this). That is true for any amplifier.

You'll have to decide if that is too much "mismatch". It would not bother me.

I don't see Benchmark rating 4ohm monoblock either. Is that because it can hit protection or clip early?

You'd have to ask them. But John Siau, their Director of Engineering, is on ASR and has responded before that driving 4 ohms bridged is OK (you'd have to search this thread).

Really, if you are uncomfortable with them, why not just choose another amplifier?

EDIT: I did not see the rest of the posts after this. Given the exchanges, this post is superfluous, and no point in my responding further.

EDIT 2: Plots of damping factor (DF) and output impedance (Zout) versus frequency. Note in reality these are complex values.

 

[NTK]

Floyd Toole wrote an article in 1975 trying to dispel the myth of the damping factor. Seemingly he wasn't totally successful. Damping factor has nothing to do with damping of anything. Frequency response modulation: Yes. Damping: No.
http://diyaudioprojects.com/Technic...ping-Factor-and-Damn-Nonsense-Floyd-Toole.pdf

An example:
Here is the cross-over of the Purifi's SPK5 reference design. Notice the padding resistor R1 (a 10Ω resistor) right at the input to the tweeter side of the cross-over? So no "damping" for the tweeter.

For the woofer side, right after the input, there is a parallel branch. The top branch has a 12Ω resistor (R3). The bottom branch has a 1.5 mH inductor (L3). Notice that even with big fat 14 AWG wire, the DC resistance of the inductor is 0.34Ω? Therefore, the inlet DC resistance to the woofer is minimum 0.34Ω.

And none of the above even consider the DC resistance of the drivers. So, with regard to damping, what does it matter how tiny the output impedance of the amp is?

 

[Coach_Kaarlo]

Great write-up.
I also found low gain to sound pretty flat compared to medium/high.
I really don't know why this is, especially given as its actually quite a noticeable difference, but I did find that when in low gain the amp has only one relay click on startup/shutdown. Whereas in medium/high gain it has two. I'm not sure what this implies but it does seem something is being done differently mechanically inside the amp when in low gain.

BUT, given as the performance is so crazy, there's not really much reason not to just stick it in medium gain anyway.
Even with ~30dB of attenuation on the input signal you'll still be getting better performance than most other speaker amplifiers.

Main reason I believe Benchmark Media advise the low gain setting (if possible with one's system gain setup) is to reduce noise.

 

[RobS]

Main reason I believe Benchmark Media advise the low gain setting (if possible with one's system gain setup) is to reduce noise.

Probably from noise of an active preamp. With my passive pre, I can hear only the noise floor of my DAC if I have the pot turned to 90%+ with the high gain setting. So even at it’s high gain setting, the AHB2 has no audible noise floor. Might make for a great headphone amp because of this (heck maybe even super sensitive IEMs?)

 

[pogo]

Floyd Toole wrote an article in 1975 trying to dispel the myth of the damping factor. Seemingly he wasn't totally successful. Damping factor has nothing to do with damping of anything. Frequency response modulation: Yes. Damping: No.
http://diyaudioprojects.com/Technic...ping-Factor-and-Damn-Nonsense-Floyd-Toole.pdf

Interesting article that shows in fig. 3 on the last page what I mean. In the last half of the diagram (division 7) you can see that a lower damping factor slows down the mechanical movement of the chassis more poorly, resulting in a more washed-out bass range. And an even higher damping factor would probably come even closer to the ideal.

 

[pogo]

Sure. I have my AHB2s and the NAD C298s as well as a pair of PSAudio M1200s (ICE Edge). Report in due time.

I actually expect in this listening comparison an audible difference, provided it is carried out on mechanically large chassis which must be slowed down. It is also a matter of taste whether you prefer an intense bass range, generated by a lower damping factor, or a cleanly contoured reproduction.

 

[misterdog]

I couldn't hold out any longer, I have one arriving in a couple of hours. I will be feeding it (digital) from SMSL M400 into Topping Pre 90 then Quad 989 Electrostatics. The Quads are run with the dust covers and protective metal grilles removed. I have fitted the panels into bespoke steel frames for maximum rigidity.

I currently have Neurochrome Mod 86 Parallel Monoblocks, which are close in the SINAD charts so it will be interesting to compare.