Review and Measurements of Benchmark AHB2 Amp

[Panelhead]

2. Bi-amping will restore the damping factor in the bass to what it was during stereo use, so if (and it's a big if) damping factor was the cause of your dissatisfaction, then that's obviously the solution. However, you'll be wasting most of the extra power you've bought, because most demand is in the bass, where you'll be back to one channel + one channel, like you had before. The other two channels will be basically idling, in k up dacs, the face of HF demand, which is modest.

I hooked a pair of amps in a vertical bi-amp this week. Previous setup was stereo single amp. Like it. Will leave it alone.
Did double the input impedance of each amplifier so the load is the same as a single amp. Should not make difference, but wanted apples to apples.
Had previously ran bridged mode too. Sounded great also. The higher output impedance should have been inaudible. Think the rise time should increase. Definitely different sound than stereo connection. Sharper transients. But there was an issue.
After playing for a while, there was noise. Not heard when amps were used as single ended stereo amplifiers. Swapped to backup dac, changed cables, another pair of the same amps. Power cords, tried all three on separate outlets, all three in same outlet.
The noise was only audible if ear was right the tweeter horn. A hiss. Inaudible a foot sway. Both channels. Drove me crazy.
Tweeter is a B&C DE-120. Padded down a few dB. Maybe 105 dB.
Decided to check and noise floor was higher connected in bridged than stereo. Assume the residual noise, say 60 uV, doubled in bridged mode.
I had expected even order harmonics to decrease. Lower noise. The hiss is there in stereo just lower level. And it is temperature or on time related. Increases slightly after 10 - 15 minutes.
The amps are not AHB2, which are designed to operate in bridged mode. Assume the noise floor drops when bridged.

 

[DonH56]

2. Bi-amping will restore the damping factor in the bass <rest elided>

How does bi-amping "restore the damping factor in the bass"?

 

[Panelhead]

That statement is quoted, not mine.
Half the output impedance of the same amp bridged. Should not matter with milli-ohm levels.

 

[Inner Space]

How does bi-amping "restore the damping factor in the bass"?

He's quoting from my post # 1982. Bridging the AHB2s doubles the output impedance, thereby halving the so-called damping factor. Using two amps, both in stereo for bi-amping, "restores" the so-called damping factor to its original number. Unlikely to be a solution to his bass problems, but I thought worth mentioning for completeness.

 

[DonH56]

He's quoting from my post # 1982. Bridging the AHB2s doubles the output impedance, thereby halving the so-called damping factor. Using two amps, both in stereo for bi-amping, "restores" the so-called damping factor to its original number. Unlikely to be a solution to his bass problems, but I thought worth mentioning for completeness.

OK, thank you.

 

[Universal Cereal Bus]

At least you know how you can surreptitiously monitor sounds in the room when the amp's off...

...or jump an air gap https://www.pcmag.com/news/black-hat-researcher-shows-why-air-gaps-wont-protect-your-data

 

[Coach_Kaarlo]

I have been researching my question on ASR but as yet have been unable to find the answers.

Recently I re-tested my old Yamaha NS-2000 speakers in stereo versus with 2 amps in mono. While many might hold the view that an amplifier not clipping is working fine - my experience has been that two AHB2's in mono have far better attack, bass resolution, and clarity than in when used in stereo (with my old NS2k speakers). What has been difficult is finding an accurate enough measurement method to show the difference which is clearly audible. Free field measuring outside helped. But the challenge to measure the differences I hear is always there - the AHB2's are amazing but have audible quirks when used with certain speakers.



Currently I have the following setup.

Various digital sources
USB 2.0 audio cable

miniDSP SHD Studio w DIRAC 3.0 (CH 1&2 used/ CH 3&4 free)
XLR to BNC digital coax (balanced) cable

DAC2-HGC
XLR (balanced) interconnects (Ch 1 & 2)

AHB2 (mono-left)
NL2 to bare wire

AHB2 (mono-right)
NL2 to bare wire

Victor SX-1000 Laboratory | 180W nom. | 4 ohm nom. | 90W/dB/m | 25Hz to 80kHz | x-overs 440Hz / 5000Hz |


What I am seeking is sound reproduction that has:

• neutral sound quality and fidelity
• clarity, tonality, and space
• clear imaging, accurate reproduction of instrument sounds
• the best utilisation of currently owned hardware is the goal and new purchases are chosen to complement the existing system

My question is how to measure three different AHB2 configurations;

1. MONO: AHB2 #1 = ch 1 / AHB2 #2 = ch 2 with the full speaker analog crossover used
2. STEREO: amp = ch 1 & 2 with full speaker analog crossover used
3. BI-AMP: amp 1 stereo CH 1&2 LF only / amp 2 stereo CH 1&2 HF only / with the separate speaker crossovers used for each LF / HF band

If I change my speaker connectors and cables to NL4 with Canare 4S11 cables (4 x AWG 14) then I can measure / test the speakers HF and LF analog crossovers (apparently the crossover boards are completely separate). I am interested to measure these 3 basic use cases and also test what is audible.

My understanding is that the Benchmark AHB2's have one Speakon output which has both channel 1 on 2 wires/ poles, and channel 2 on 2 wires/ poles - meaning this bi-amps the speakers. One amp can drive the HF crossovers, one amp driving the LF crossovers (not sure if this is vertical or horizontal).

Gaps in my understanding are painfully numerous, but what is the best way to use the DAC2-HGC to drive the bi-amped setup for measurement?

There are only 2 analog channels (balanced XLR's) coming out of the DAC but is seems we need 4 analog channels to correctly amplify the 4 signals required?

I understand that XLR splitters are a poor compromise. And buying another DAC seems excessive just to do some playing around.

What are the options? Bearing in mind I am interested in getting the cleanest signal to the speaker crossovers.


Thanks.

 

[restorer-john]

So the AHB-2 in mono is fine on the NS2Ks? Have you tried any other >200W@8 amplifiers on the same speakers, same content?

The NS2Ks are not a forgiving speaker and are demanding of amplifiers. It took the top Yamaha at the time (the MX-1000 260+260@8R) to really make them sing and even it could have done with a bit more.

I'm not at all surprised a single AHB-2 runs out of steam fast on the NS2Ks. The NS-1000Xs also needed the MX-1000. Why not beg/borrow another big (>200WPC) power amplifier and compare to the bridged AHB?

 

[witwald]

...the AHB2's are amazing but have audible quirks when used with certain speakers.

This is a trifle confusing. First, there's the use of the "amazing" superlative. Second, there's the presence of its antithesis, "audible quirks". Which is it to be? Then I'm left wondering how free-field measurements are applied to an amplifier?

 

[witwald]

The NS2Ks are not a forgiving speaker and are demanding of amplifiers. It took the top Yamaha at the time (the MX-1000 260+260@8R) to really make them sing and even it could have done with a bit more.

What makes the NS-2000s such an unforgiving loudspeaker? Their minimum impedance drops to just a tad below 6 ohms around 100 Hz. That frequency region requires a fair bit of power, but most speakers have low impedances in that frequency range. From the brochure, their sensitivity is 90 dB/W/m. 100 W will produce levels at 1m of 110 dB. That's going to be quite loud, even at 3–4 meter listening distances in a typical room. Is it their relatively low distortion levels that is causing them to be pushed to such high listening levels that any amplifier would be taxed to its limits? Before doing any testing of the amplifier, maybe it's worthwhile getting out an SPL meter and seeing how loud things are going on normal unclipped porgramme material?

 

[restorer-john]

Is it their relatively low distortion levels that is causing them to be pushed to such high listening levels that any amplifier would be taxed to its limits?

They can be played at levels that many other speakers cannot. The clean midrange and treble encourage you to turn the volume up in order to get the bottom end bass slam you feel is missing. And you run out of power.

They are an incredible speaker. If you ever get a chance to listen to a pair of 2000s or the 1000Xs on a really muscular amplifier, it'll re-scale clean sounding speakers for you. No two way comes remotely close.

 

[Coach_Kaarlo]

So the AHB-2 in mono is fine on the NS2Ks? Have you tried any other >200W@8 amplifiers on the same speakers, same content?

The NS2Ks are not a forgiving speaker and are demanding of amplifiers. It took the top Yamaha at the time (the MX-1000 260+260@8R) to really make them sing and even it could have done with a bit more.

I'm not at all surprised a single AHB-2 runs out of steam fast on the NS2Ks. The NS-1000Xs also needed the MX-1000. Why not beg/borrow another big (>200WPC) power amplifier and compare to the bridged AHB?

Fine relatively speaking, nothing on the XX of course. There are clear and audible differences between mono and stereo, I have measured both setups, but the resolution of the differences measured is less than what you would expect (my bias) from the listening experience. They are a great speaker, sad to see them go, but they are very difficult to drive well.

They can be played at levels that many other speakers cannot. The clean midrange and treble encourage you to turn the volume up in order to get the bottom end bass slam you feel is missing. And you run out of power.

They are an incredible speaker. If you ever get a chance to listen to a pair of 2000s or the 1000Xs on a really muscular amplifier, it'll re-scale clean sounding speakers for you. No two way comes remotely close.

Indeed they can be played loud without difficulty, but the bottom end really seems to need big current and voltage swing capacity in the amplifier to get the most from the speaker. They are honest to source however, clean and neutral. I'd say the woofer is slightly mismatched with the mid, but they are pretty good once you adjust the tone controls way way down (see attached).

Edit: [Once tone controls are set as per pic - they measure fairly flat, and tonally sound much more balanced]

Painful beryllium brightness otherwise.

 

[Coach_Kaarlo]

This is a trifle confusing. First, there's the use of the "amazing" superlative. Second, there's the presence of its antithesis, "audible quirks". Which is it to be? Then I'm left wondering how free-field measurements are applied to an amplifier?

I guess your POV is all amps sound the same? So not further discussion needed really. My POV is something can be other than black or white, like infinite shades of grey perhaps. The amp is great, but far from perfect for all speakers.

What makes the NS-2000s such an unforgiving loudspeaker? Their minimum impedance drops to just a tad below 6 ohms around 100 Hz. That frequency region requires a fair bit of power, but most speakers have low impedances in that frequency range. From the brochure, their sensitivity is 90 dB/W/m. 100 W will produce levels at 1m of 110 dB. That's going to be quite loud, even at 3–4 meter listening distances in a typical room. Is it their relatively low distortion levels that is causing them to be pushed to such high listening levels that any amplifier would be taxed to its limits? Before doing any testing of the amplifier, maybe it's worthwhile getting out an SPL meter and seeing how loud things are going on normal unclipped porgramme material?

All this forum [Edit: less inflammatory for me to use the word "'debate"" perhaps] chatter about amplifiers measuring / sounding / performing the same is interesting - but what real world dynamic measurement methods have been applied to these amplifiers? Where are the measurements of actual musical dynamic performance in changing conditions? Or do some just listen to steady state test tones at home (joking a little being serious a little).

In much the same way Benchmark highlighted the need to dynamically measure headphone amplifiers with real headphone mimicking loads, one might imagine that loudspeakers with wildly differing impedances and phases might produce wildly different measurements, listening experiences, and performance - when measured dynamically with the SAME amplifier. To some this seems self evident, to others who lack of an acceptable metric or measurement it seems ludicrous. My point is simply just because you are unable to measure something or explain something does not preclude it from existing, or from being measured or explained at some point in the future when science and engineering catch up.

One of the great scientific methods is to reproduce an experiment, and get the same result. Let's see the dynamic measurements and tests which say all amplifiers which measure the same sound the same with real musical dynamic test content.......




My question is how to measure three different AHB2 configurations;

1. MONO: AHB2 #1 = ch 1 / AHB2 #2 = ch 2 with the full speaker analog crossover used
2. STEREO: amp = ch 1 & 2 with full speaker analog crossover used
3. BI-AMP: amp 1 stereo CH 1&2 LF only / amp 2 stereo CH 1&2 HF only / with the separate speaker crossovers used for each LF / HF band

If I change my speaker connectors and cables to NL4 with Canare 4S11 cables (4 x AWG 14) then I can measure / test the speakers HF and LF analog crossovers (apparently the crossover boards are completely separate). I am interested to measure these 3 basic use cases and also test what is audible.

My understanding is that the Benchmark AHB2's have one Speakon output which has both channel 1 on 2 wires/ poles, and channel 2 on 2 wires/ poles - meaning this bi-amps the speakers. One amp can drive the HF crossovers, one amp driving the LF crossovers (not sure if this is vertical or horizontal).

Gaps in my understanding are painfully numerous, but what is the best way to use the DAC2-HGC to drive the bi-amped setup for measurement?

There are only 2 analog channels (balanced XLR's) coming out of the DAC but is seems we need 4 analog channels to correctly amplify the 4 signals required?

I understand that XLR splitters are a poor compromise. And buying another DAC seems excessive just to do some playing around.

What are the options? Bearing in mind I am interested in getting the cleanest signal to the speaker crossovers.


Thanks.

 

[restorer-john]

All this forum chatter about amplifiers measuring / sounding / performing the same is interesting - but what real world dynamic measurement methods have been applied to these amplifiers?

It's the experiential exposure to multiple amplifiers and multiple speakers where you hear sometime stark differences between amplifiers. Same speakers, same source, same volume, different amplifiers. Switched quickly on a dedicated comparator.

I've never heard a bridged amplifier that sounds as good as it did in normal operation. Sure, there is tons more voltage swing, but the detail, noise and speaker control goes out the window with real loudspeaker loads in a listening situation. Not on my bench, it all looks good.

You have a measurably world class power amplifier pair and yet something isn't right- to your ears. I don't think measurement is going to show anything you don't already know. Try another amplifier of equal power to the BTL AHB-2 is my suggestion, before attempting to go down the high powered measurement path for dynamic loads.

 

[witwald]

Indeed they can be played loud without difficulty, but the bottom end really seems to need big current and voltage swing capacity in the amplifier to get the most from the speaker.

Well, they are a sealed enclosure, with a measured –3 dB point of about 52 Hz. That's not a lot of bass extension for a 12" driver, but I guess that's the trade-off that was made for that size of enclosure and sensitivity. The impedance around 50 Hz seems to be around 9 ohms. That's not a low impedance at all, and is rising to its maximum at 30 Hz of around 60 ohms.

They are honest to source however, clean and neutral. I'd say the woofer is slightly mismatched with the mid, but they are pretty good once you adjust the tone controls way way down (see attached). Painful beryllium brightness otherwise.

They do seem to have a step down in the bass response compared to the upper bass and midrange. That will no doubt make them sound bass light, unless placed relatively close to a wall boundary. There's also a very noticeable boost of a few dB around 2 kHz. Taken together, it would seem that these factors are producing the subjective "beryllium brightness" that you describe, or probably more accurately just "brightness", as it will not have anything to do with the use of "beryllium". I can see why application of the tone controls will help, as it can make them much flatter, and it appears that you've toned things down more or less in keeping with the excesses in the measured frequency response.

 

[witwald]

All this forum chatter about amplifiers measuring / sounding / performing the same is interesting

I wouldn't deem to call it chatter.

- but what real world dynamic measurement methods have been applied to these amplifiers? Where are the measurements of actual musical dynamic performance in changing conditions? Or do some just listen to steady state test tones at home (joking a little being serious a little).

Well designed (and built) amplifiers are linear devices when operated within their design limits. As a result, their dynamic response is governed by their frequency response, as the nonlinearities are quite small in good examples. Their performance with different input signals and loads can be quite readily predicted.

In much the same way Benchmark highlighted the need to dynamically measure headphone amplifiers with real headphone mimicking loads,

Testing of amplifiers with simulated loudspeaker loads has been done for decades. The fact that the same approach could be applied to headphone amplifiers seems to be a quite logical extension, as headphones probably have a varying impedance load from one headphone to another. I can see that some models of headphones may cause some headphone amplifiers to run out of voltage swing and/or current delivery capability, thus running into nonlinear operating ranges.

...one might imagine that loudspeakers with wildly differing impedances and phases might produce wildly different measurements, listening experiences, and performance - when measured dynamically with the SAME amplifier.

Not if the amplifier is operating in its linear operating range. If the amplifier is "overloaded", then it will create distortion, but then it's no longer operating linearly.

To some this seems self evident, to others who lack of an acceptable metric or measurement it seems ludicrous. My point is simply just because you are unable to measure something or explain something does not preclude it from existing, or from being measured or explained at some point in the future when science and engineering catch up.

Due to psychoacoustics, it would seem that perceiving that one has heard a difference is not the same as there actually being a difference. Keep in mind that a small change in level of 0.2 dB can be audible, and cause a listener reaction in terms of changes to the perceived sound quality. Whether those changes are labelled as "good" or "bad", well that depends on the reference I suppose.

One of the great scientific methods is to reproduce an experiment, and get the same result. Let's see the dynamic measurements and tests which say all amplifiers which measure the same sound the same with real musical dynamic test content.......

Have you heard of the Null Test, which was applied to amplifiers by David Hafler? Might not it be applicable to such a set of circumstances? Distortion mechanisms in amplifiers seem to be relatively well understood, and many amplifiers can reduce them to levels that would seem to make them inaudible. To take such an amplifier and then perceive that it has problems seems to be difficult to understand from the point of view of the engineering that has gone into the design and building of that amplifier.

 

[Panelhead]

I like your setup. Suspect the DRC or X-over is the weak link in the chain.
Try without the DRC to see if it resolves the issues. Too much correction can cause issues. Many are limited in boost or cut to address high Q nulls.
I am a Benchmark fanboy.

 

[Coach_Kaarlo]

Well, they are a sealed enclosure, with a measured –3 dB point of about 52 Hz. That's not a lot of bass extension for a 12" driver, but I guess that's the trade-off that was made for that size of enclosure and sensitivity. The impedance around 50 Hz seems to be around 9 ohms. That's not a low impedance at all, and is rising to its maximum at 30 Hz of around 60 ohms.

They do seem to have a step down in the bass response compared to the upper bass and midrange. That will no doubt make them sound bass light, unless placed relatively close to a wall boundary. There's also a very noticeable boost of a few dB around 2 kHz. Taken together, it would seem that these factors are producing the subjective "beryllium brightness" that you describe, or probably more accurately just "brightness", as it will not have anything to do with the use of "beryllium". I can see why application of the tone controls will help, as it can make them much flatter, and it appears that you've toned things down more or less in keeping with the excesses in the measured frequency response.

Measurements show mono frequency response.

 

[waynel]

I have been researching my question on ASR but as yet have been unable to find the answers.

Recently I re-tested my old Yamaha NS-2000 speakers in stereo versus with 2 amps in mono. While many might hold the view that an amplifier not clipping is working fine - my experience has been that two AHB2's in mono have far better attack, bass resolution, and clarity than in when used in stereo (with my old NS2k speakers). What has been difficult is finding an accurate enough measurement method to show the difference which is clearly audible. Free field measuring outside helped. But the challenge to measure the differences I hear is always there - the AHB2's are amazing but have audible quirks when used with certain speakers.



Currently I have the following setup.

Various digital sources
USB 2.0 audio cable

miniDSP SHD Studio w DIRAC 3.0 (CH 1&2 used/ CH 3&4 free)
XLR to BNC digital coax (balanced) cable

DAC2-HGC
XLR (balanced) interconnects (Ch 1 & 2)

AHB2 (mono-left)
NL2 to bare wire

AHB2 (mono-right)
NL2 to bare wire

Victor SX-1000 Laboratory | 180W nom. | 4 ohm nom. | 90W/dB/m | 25Hz to 80kHz | x-overs 440Hz / 5000Hz |


What I am seeking is sound reproduction that has:

• neutral sound quality and fidelity
• clarity, tonality, and space
• clear imaging, accurate reproduction of instrument sounds
• the best utilisation of currently owned hardware is the goal and new purchases are chosen to complement the existing system

My question is how to measure three different AHB2 configurations;

1. MONO: AHB2 #1 = ch 1 / AHB2 #2 = ch 2 with the full speaker analog crossover used
2. STEREO: amp = ch 1 & 2 with full speaker analog crossover used
3. BI-AMP: amp 1 stereo CH 1&2 LF only / amp 2 stereo CH 1&2 HF only / with the separate speaker crossovers used for each LF / HF band

If I change my speaker connectors and cables to NL4 with Canare 4S11 cables (4 x AWG 14) then I can measure / test the speakers HF and LF analog crossovers (apparently the crossover boards are completely separate). I am interested to measure these 3 basic use cases and also test what is audible.

My understanding is that the Benchmark AHB2's have one Speakon output which has both channel 1 on 2 wires/ poles, and channel 2 on 2 wires/ poles - meaning this bi-amps the speakers. One amp can drive the HF crossovers, one amp driving the LF crossovers (not sure if this is vertical or horizontal).

Gaps in my understanding are painfully numerous, but what is the best way to use the DAC2-HGC to drive the bi-amped setup for measurement?

There are only 2 analog channels (balanced XLR's) coming out of the DAC but is seems we need 4 analog channels to correctly amplify the 4 signals required?

I understand that XLR splitters are a poor compromise. And buying another DAC seems excessive just to do some playing around.

What are the options? Bearing in mind I am interested in getting the cleanest signal to the speaker crossovers.


Thanks.

You have not found an answer to your question as to why your system sounds different with bridged amps because there is no scientific explanation, they should sound the same when the extra power is not needed. Perhaps you are reacting to the 6dB higher gain in bridged mode?

You will not find any difference in acoustical measurements with the 3 configurations because there are none when the amp is not clipping.
You have top notch equipment already, I suggest focusing on room acoustics and better speakers/subs.

 

[DonH56]

Bridging changes output impedance, current and voltage characteristics, frequency response, noise, etc. etc. etc. All measurable. If you don't care about the science and measurement "chatter" this is the wrong place for you. @restorer-john gave the best answer -- try another amp that delivers the power you need and move on.