Is the Benchmark AHB2, in mono mode, really better than my Mark Levinson No. 536 monoblocks?

[RichB]

This is exactly what is done.
And here is a technical explanation (in german): Link

And once again to remind:
After the stereo team has confirmed that the DF was measured properly at a 4ohms load and not bridged for the AHB2 ((63Hz/1kHz/14kHz): 48/45/19), I still wonder how the reduced control of the chassis and especially in bridge mode should not have an effect?
Where would the values be with 3ohms, ...?

ww.ta-hifi.de/en/audiosystems/series-200/a-200-power-amplifier/

Measurments:

https://www.ta-hifi.de/wp-content/uploads/AUDIO_Polen_7_2022_MP200_DAC200_A200.pdf

The T+A 200 specs for distortion specification is >0.002, power level and frequency are not stated.
The measurements show that the distortion is >0.1 at 1 watt and barely every below .02.
Pretty neat, being only 1 or 2 order of magnitudes off.

By all means, purchase this for the 800 DF specification, but I would not take it to the bank

- Rich

 

[PGAMiami]

Others may disagree, but adding a series resistor will effectively make this a worse amplifier. You may perceive more bass, but that’s because it will be sloppier bass. You could get the same effect using really bad spreaker cables with a high series resistance. Note this will affect all frequencies that are included in this circuit. If you’re not biwiring then expect unpredictable results for the mid and tweeter.

A better solution that would be much more predictable and adjustable would be to use a high quality digital EQ. I’m using an Audiolense convolution filter running on Roon. On high rez material the filter has 66,000 taps. A few years ago this would have been science fiction. This night and day from digital correction that’s available in a receiver. Mitch Barnett set it up for me. You can dial in the target curve that works best for you. I also run my TV through a convolution filter that’s on a PC. That one has zero latency although it’s not quite as precise as the one on Roon. The improvement of this is not just in the bass. It’s everything. You would think I have a center channel. Imaging improves quite a lot when the L and R speakers are matched.

 

[fpitas]

Others may disagree, but adding a series resistor will effectively make this a worse amplifier. You may perceive more bass, but that’s because it will be sloppier bass. You could get the same effect using really bad spreaker cables. Note this will affect all frequencies that are included in this circuit. If you’re not biwiring then expect unpredictable results for the mid and tweeter.

A better solution that would be much more predictable and adjustable would be to use a high quality digital EQ. I’m using an Audiolense convolution filter running on Roon. Mitch Barnett set it up for me. You can dial in the target curve that works best for you. I also run my TV through a convolution filter that’s on a PC. That one has zero latency although it’s not quite as precise as the one on Roon. The improvement of this is not just in the bass. It’s everything. You would think I have a center channel. Imaging improves quite a lot when the L and R speakers are matched.

Agreed. The only time a series resistor is appropriate is for some compression drivers that have lower distortion when current-driven. For your normal cone driver it's a bad idea.

 

[PGAMiami]

This is the FR of my speakers, before and after EQ. You can dial in what target curve you like. My speakers are near the corners but with a lot of absorption near them. Note how much this boosts the bass pre EQ. This of course has to be attenuated. Then it essentially becomes headroom.

 

[40th Anniversary]

Seems like you could just insert a series resistor if low DF is what you want?

That is what we used to do to see how it would affect low frequency response and impact a square wave when we used that method for some of the tests.

 

[40th Anniversary]

Going to test 1 (one) of them in stereo and see what happens!

More tests done and comparisons completed (not all the tests I wanted to do, but some). I added an additional more sensitive speaker to the tests, so now that makes 3 pairs of speakers tested in stereo and mono (bridge) mode with the AHB2s.

1. The least sensitive speaker (Dynaudio SP40) benefitted the most from bridge operation, especially in the mids and highs: a bigger sound stage so to speak.

2. The most sensitive speaker did not show any appreciable improvement. I'd be stretching it to say that 'maybe' there was more air and presence, but I was looking for it. Not a big impact and it would be a waste of money to use two AHB2s with sensitive speakers. Maybe the quality of the speaker is not good enough.

3. With the medium sensitivity speaker (the Triangle Comète 40th), it was all about the mids and highs being more 'there' as initially reported, and a "bit" more depth to the bass.

 

[PGAMiami]

More tests done and comparisons completed (not all the tests I wanted to do, but some). I added an additional more sensitive speaker to the tests, so now that makes 3 pairs of speakers tested in stereo and mono (bridge) mode with the AHB2s.

1. The least sensitive speaker (Dynaudio SP40) benefitted the most from bridge operation, especially in the mids and highs: a bigger sound stage so to speak.

2. The most sensitive speaker did not show any appreciable improvement. I'd be stretching it to say that 'maybe' there was more air and presence, but I was looking for it. Not a big impact and it would be a waste of money to use two AHB2s with sensitive speakers. Maybe the quality of the speaker is not good enough.

3. With the medium sensitivity speaker (the Triangle Comète 40th), it was all about the mids and highs being more 'there' as initially reported, and a "bit" more depth to the bass.

If I recall from EE classes 40 years ago … bridging will double your slew rate … then again it’s so high for the AHB2, I don’t expect this improves the mids and highs. If I had to put money on where some of the improvements are coming from, I’d still go with the positive effects from doubling the voltage swing. Even though the mid and highs themselves are not taxing the power limits of the amplifier, these higher frequencies are still riding on top of the lower bass, which likely is using a lot of power with the less efficient speakers.

 

[fpitas]

That is what we used to do to see how it would affect low frequency response and impact a square wave when we used that method for some of the tests.

Fair enough. Seems odd to buy an amplifier that manipulates DF. Resistors are cheap.

 

[pogo]

Others may disagree, but adding a series resistor will effectively make this a worse amplifier. You may perceive more bass, but that’s because it will be sloppier bass.

But that's exactly what's happening here with AHB2s, isn't it?
AHB2 at a 4ohms load and not bridged (63Hz/1kHz/14kHz): DF 48/45/19 <-- measured
AHB2 at a 4ohms load and bridged (63Hz/1kHz/14kHz): DF 24/22.5/9.5 <-- derived

 

[PGAMiami]

But that's exactly what's happening here with AHB2s, isn't it?
AHB2 at a 4ohms load and not bridged (63Hz/1kHz/14kHz): DF 48/45/19 <-- measured
AHB2 at a 4ohms load and bridged (63Hz/1kHz/14kHz): DF 24/22.5/9.5 <-- derived

I haven’t read the article in German, but I read Benchmark’s article that states that at the speaker terminal the AHB2 has a DF of 370. These other measurements seem way off compared to the 370 spec at 8 ohm.

The difference in how bridged vs non-bridged sounds is almost surely not related to the DF in my humble opinion. The output impedance of the AHB2 is stated to be 0.0216 ohm. So when bridged that would be 0.0432 ohm, still vanishingly low compared to most speakers. According to Benchmark a 10' 11 gauge speaker cable will have an impedance of 0.0252 ohm, more than the AHB2. So if you want to model the change in DF from bridging, seems the best way would be to add about 9' to your speaker cable. Or alternatively, if by running mono blocks you can reduce the speaker cable by 9', then your system DF remains about the same as a stereo amp that's 9' further away.

 

[Geert]

These other measurements seem way off compared to the 370 spec at 8 ohm.

As already confirmed by others more than a month ago when Pogo first dropped these numbers:

1. https://audiosciencereview.com/foru...ments-of-benchmark-ahb2-amp.7628/post-1320948
2. https://audiosciencereview.com/foru...ments-of-benchmark-ahb2-amp.7628/post-1306179

Summary:
"Their measurements indicate that their load is not fit for the task - neither is their measurement equipment. And without even saying what they could achieve in bridged mode this looks more like an attempt to keep their usual advertizers happy... I for one had no problem (except for getting better loads first) to reproduce @amirm 's stellar results with my own AHB2".
"Stereo.de AHB2 damping factor measurements is an outlier. I have no idea what Stereo.de is doing, and perhaps they don't either".

John Siau confirmed on this forum they guarantee the performance of this amp up to loads of min. 1.4 ohm.

But Pogo can' t resist trolling.

 

[40th Anniversary]

If I recall from EE classes 40 years ago … bridging will double your slew rate … then again it’s so high for the AHB2, I don’t expect this improves the mids and highs. If I had to put money on where some of the improvements are coming from, I’d still go with the positive effects from doubling the voltage swing. Even though the mid and highs themselves are not taxing the power limits of the amplifier, these higher frequencies are still riding on top of the lower bass, which likely is using a lot of power with the less efficient speakers.

Is the slew rate high for the AHB2? I heard it was something like 16 v per.....

 

[PGAMiami]

But that's exactly what's happening here with AHB2s, isn't it?
AHB2 at a 4ohms load and not bridged (63Hz/1kHz/14kHz): DF 48/45/19 <-- measured
AHB2 at a 4ohms load and bridged (63Hz/1kHz/14kHz): DF 24/22.5/9.5 <-- derived

Is the slew rate high for the AHB2? I heard it was something like 16 v per.....

I believe we read the same thing from their website which I pasted below and it's more than enough bandwidth.

Finally, the square wave response shows a very good transient behavior, without any particular problem. Slew rate was estimated at 16V/uS and rise time at 3.5mS, a value that corresponds to a conventionally calculated bandwidth of 100kHz."

 

[pogo]

Summary:
"Their measurements indicate that their load is not fit for the task - neither is their measurement equipment.

I refer only to measurable facts and listening experience and do not troll! As you can see from another example, the measurement equipment seems to perform very well: Link

On inquiry with STEREO the measured values were confirmed to me again, since I did not expect such low values.
A response from benchmark is still pending: Link

 

[Geert]

I refer only to measurable facts and listening experience

You only have one measurement with questionable results. For the rest you mainly respond to observations made in uncontrolled listening tests.

and do not troll

The last year you cross post the same nonsense about DF in a ton of threads. Every time exactly the same information about that amp with switchable DF, 'the time domain' and a few references to marketing brochures. People have spends days explaining DF to you from different angles. Nevertheless you continue repeating the same story like nothing happened. All of that without any result, your story about DF didn't get any stronger. On the contrary.

So yes, you are trolling and different people already pointed that out. Again, you continue like that didn't happen.

As you can see from another example, the measurement equipment seems to perform very well: Link

They don't document their testing procedure so this link doesn't make us any wiser. I suggest you wait for confirmation of the test results by a second party, or, since you are so passionate about DF, try to manage a test yourselves and document it so others can review the procedure. Anyone with an AHB2, a DVM and a resistor can do an attempt. A ball park figure would already be a good start since the stereo.de measurements are very far off spec.


Edit: found back a relevant comment from John Siau: https://forum.audiogon.com/posts/1648374:

• "The AHB2 has a high damping factor and this allows excellent performance in bridged mono. The damping factor is 350 into 8 Ohms stereo and 175 into 8 Ohms bridged mono."
• "the AHB2 is well suited for bridged mono operation into 4-Ohm nominal impedances and the performance is virtually identical to stereo mode except that the power is nearly 4 X higher. Dips in the speaker impedance curve are not a problem and the AHB2 drives these cleanly."

I would be surprised if the manufacturer of technically one of the best amps in the world, who provides a manual with a list of AP measurements with the product, published technical articles about damping factor and is know for performing controlled listening tests completely missed the fact that their amp delivers a sub-par DF with 4 ohm loads.

 

[pogo]

the fact that their amp delivers a sub-par DF with 4 ohm loads

Another independent measurement could really clarify this. As far as I know, only STEREO has measured at 4 ohms!?
However, the audible impressions already indicate that the STEREO measurements can't be that far off the truth.

 

[Geert]

However, the audible impressions already indicate that the STEREO measurements can't be that far off the truth.

That conclusion lacks objectivity, especially when the listening tests are uncontrolled. Your believe in the effect of DF makes you want to believe everything that suites your agenda.

Did you consider the fact that the AHB2's unique feed-forward error correction output stage and its low negative feedback might disturb certain types of DF measurements (like the basic open circuit voltage versus loaded output voltage test)?

 

[lateralous]

Another independent measurement could really clarify this. As far as I know, only STEREO has measured at 4 ohms!?
However, the audible impressions already indicate that the STEREO measurements can't be that far off the truth.

What is the basis of your obsession with this single measurement of a single amplifier? You completely ignore valid lines of questioning around your obsession, and immediately come back to the very same measurement.

 

[pogo]

Did you consider the fact that the AHB2's unique feed-forward error correction output stage and its low negative feedback might disturb certain types of DF measurements (like the basic open circuit voltage versus loaded output voltage test)?

Maybe, but it could also be that the AHB2 behaves the same way as this one here: Link

 

[Geert]

Maybe, but it could also be that the AHB2 behaves the same way as this one here: Link

What they did there is exactly the type of measurement I warned for in my post you quoted ... "I checked this with an alternate method, comparing the open-circuit voltage vs. the voltage when loaded with 4 ohms, and got the same result."

There's always a chance something's wrong, but how strong do you think that chance is when talking about technically the best amplifier in the world, made by one of the most respected companies in the industry? Do you have any idea what went into designing this amp? In this case I would be a bit more hesitant with drawing conclusions from nondisclosed and non peer reviewed measurements from a Hifi magazine.