The Hegel soundengine explained

[haraldo]

Everything about Hegel is scientific

Bent Holter of Hegel explains how to kill distortion the Norwegian way.

The number one target is distortion: not a steady state aim (as per test tones) but a dynamic one that rises and falls with the up and down nature of analogue music signals and particularly problematic at higher frequencies. Inside Hegel amps, distortion is tackled in real time.

Within each gain stage, Holter applies a summer to the original and inverted signals to constantly measure distortion. Should that distortion cross a predetermined threshold, the circuit will inject signal corrections to remove it.

Holter says this is especially important for the high-order harmonics generated by transistors that lend solid-state push-pull circuits their occasionally unfavourable reputation as sounding hard or cold. Holter uses a similar feed-forward loop to dynamically correct for crossover distortion in his amplifier’s transistor-based output stages. The result is (reportedly) superior micro- and macro-dynamics.

 

[PENG]

The gentleman did a good job in describing how their feed forward circuit and "engine" reduces distortions. I don't like the way he explained the bias vs crossover distortion part. To me, he misstated (I assume unintentionally) a couple things related to bias adjustment vs crossover distortions and the odd harmonics. Assuming he's from the Hegel design team, he should know better, so I assume it's just something lost in translation.

 

[Out-Of-Phase]

Interesting presentation.

Now let's test it.

Let's listen to this power amp versus a placebo in a DBT.

Oh, and the DBT cannot be performed internally by Hegel. It has to be performed on neutral turf by an outside source.

If it still sounds better after the objective listening test, fair enough. Maybe they're on to something in their R&D.

But if you can't tell which amp you're listening to, their correction process for crossover distortion just doesn't matter. We can't hear it. Buy something cheaper.

 

[Verdinut]

The problem with crossover distortion happening with Class B amplifiers was resolved a long time ago with appropriate output circuit biasing enabling them to perform in a Class A/B configuration.

Even if a serious DBT by a totally independent group was done and proved that Hegel amplifiers produce less distortion than all other existing good power amplifiers, it seems to me like a futile work because all audiophiles except the ones with "golden ears" won't be able to hear the difference anyway. Why would someone pay a lot more to get a minute reduction in distortion which cannot be noticeable?

Of course, you have to pay for the complex circuit that measures distortion in real time and supposedly removes it. Personally, I prefer to pay less for a more simply designed and reliable Class A/B amplifier with excellent specs to an amplifier with complex circuits that add to the cost and which could affect its reliability with time.

Buy good affordable Class A/B or Class D amplifiers. With the money saved, use it for getting better speakers with which you will hear a really obvious difference.

 

[TLS Guy]

Everything about Hegel is scientific

Bent Holter of Hegel explains how to kill distortion the Norwegian way.

The number one target is distortion: not a steady state aim (as per test tones) but a dynamic one that rises and falls with the up and down nature of analogue music signals and particularly problematic at higher frequencies. Inside Hegel amps, distortion is tackled in real time.

Within each gain stage, Holter applies a summer to the original and inverted signals to constantly measure distortion. Should that distortion cross a predetermined threshold, the circuit will inject signal corrections to remove it.

Holter says this is especially important for the high-order harmonics generated by transistors that lend solid-state push-pull circuits their occasionally unfavourable reputation as sounding hard or cold. Holter uses a similar feed-forward loop to dynamically correct for crossover distortion in his amplifier’s transistor-based output stages. The result is (reportedly) superior micro- and macro-dynamics.

kCfEuAY[/MEDIA][/QUOTE]

Well I don't know what he means by an analog computer. However his long winded explanations sounds just like Peter Walker's current dumping approach from 1976.

He is correct about distortion being dynamic, especially crossover distortion, to which the ear is very sensitive.

D.T.N. Williamson was he first to apply negative feedback to HI-Fi amplifiers. He and Peter Walker were lifelong friends. Peter let Williamson have all the credit. However both right from the beginning were acutely aware of the limitations of negative feedback. So they worked together to develop the feed forward correction of the output stage. These are the renowned current dumpers. With standard testing they don't measure differently with standard tests because it is very hard to test a signal dynamically and at very low level. The thing that has been the Achilles heel of transistor A/B biased output stages with -ve feedback has been variable dynamic distortion not revealed with standard static testing.

The feed forward approach from a very good low powered class A amp to dynamically correct the output stage is brilliant. I bet this Hagel is either virtually identical or a somewhat different means to the same end.

Current dumpers really do sound better than amps relying on -ve feedback to lower distortion. They sound noticeably smoother and relaxed than other designs. I personally will not use anything else. I would be interested to hear these new amps, as I feel a feed forward approach rather then negative feedback used in almost all other amps is superior.

I have long been amazed that negative feedback schemes with all of their inherent problems have stuck around so long. I have long been of the view that those amps are well past their sell by dates.

Here is a Current dumper circuit, one channel of a Quad 909.

Note the extreme elegance of the circuit and very low part count. The output is fed to the input pin2 of IC1 and the output of from IC1 pin 6 is fed feed forward to the voltage amplifier.

Hegel need to publish their circuit for peer review, which Peter Walker did right away before production of the first iteration of the 405.

 

[haraldo]

kCfEuAY[/MEDIA]
Well I don't know what he means by an analog computer. However his long winded explanations sounds just like Peter Walker's current dumping approach from 1976.

He is correct about distortion being dynamic, especially crossover distortion, to which the ear is very sensitive.

D.T.N. Williamson was he first to apply negative feedback to HI-Fi amplifiers. He and Peter Walker were lifelong friends. Peter let Williamson have all the credit. However both right from the beginning were acutely aware of the limitations of negative feedback. So they worked together to develop the feed forward correction of the output stage. These are the renowned current dumpers. With standard testing they don't measure differently with standard tests because it is very hard to test a signal dynamically and at very low level. The thing that has been the Achilles heel of transistor A/B biased output stages with -ve feedback has been variable dynamic distortion not revealed with standard static testing.

The feed forward approach from a very good low powered class A amp to dynamically correct the output stage is brilliant. I bet this Hagel is either virtually identical or a somewhat different means to the same end.

Current dumpers really do sound better than amps relying on -ve feedback to lower distortion. They sound noticeably smoother and relaxed than other designs. I personally will not use anything else. I would be interested to hear these new amps, as I feel a feed forward approach rather then negative feedback used in almost all other amps is superior.

I have long been amazed that negative feedback schemes with all of their inherent problems have stuck around so long. I have long been of the view that those amps are well past their sell by dates.

Here is a Current dumper circuit, one channel of a Quad 909.

Note the extreme elegance of the circuit and very low part count. The output is fed to the input pin2 of IC1 and the output of from IC1 pin 6 is fed feed forward to the voltage amplifier.

Hegel need to publish their circuit for peer review, which Peter Walker did right away before production of the first iteration of the 405.

There is of course mixed opinions on the Hegel approach if you ask guys who sell amplifiers, I never listened to any of the Hegel amps to understand if there is any credit to this.

I don’t have enough tech insight to understand and compare with other approaches, but I certainly found the video interesting.

By the way, soundengine is patented which means the technical details should be publicly avaliable.

Some claims that amplifier doesn’t matter is pretty far fetched for me .... to say the least....

 

[TLS Guy]

There is of course mixed opinions on the Hegel approach if you ask guys who sell amplifiers, I never listened to any of the Hegel amps to understand if there is any credit to this.

I don’t have enough tech insight to understand and compare with other approaches, but I certainly found the video interesting.

By the way, soundengine is patented which means the technical details should be publicly avaliable.

Some claims that amplifier doesn’t matter is pretty far fetched for me .... to say the least....

I agree, totally far fetched.

I will see if I can dig up those Sound Engine patents. The only other feed forward approach I am aware of is some designs of Nelson Pass, which many thought infringed, Peter's patents. Anyhow Peter's patents have now expired and there are now Chinese knock off boards.

 

[lovinthehd]

So has anyone here "heard" these Hegel amps and found them noticeably superior?

 

[Verdinut]

So has anyone here "heard" these Hegel amps and found them noticeably superior?

Earlier this year at one of my acquaintance's place, I heard some Hegel monoblock amps that replaced a pair of Bryston monoblocks. I didn't find a noticeable difference but the owner's brain must have convinced him that there was one.

Well, that guy also had equipped his speaker cables with those audiophool elevators. When I saw these, I almost laughed at his face. You know, he is really convinced that they improve the sound.

 

[lovinthehd]

Earlier this year at one of my acquaintance's place, I heard some Hegel monoblock amps that replaced a pair of Bryston monoblocks. I didn't find a noticeable difference but the owner's brain must have convinced him that there was one.

Most of the fuss I've seen over Hegel has to do with their integrated amps, too. Not sure there's supposed to be any distinction in that regard either....

 

[haraldo]

I am in the process of getting sorted in a new flat and I have some speakers for refitting of drivers. When all that is sorted I will try to take a Hegel 590 home to see how it fares towards a Krell KAV 400xi. It will be a few months probably.

Hegel is really popular, well regarded and sell well around here.

 

[PENG]

He is correct about distortion being dynamic, especially crossover distortion, to which the ear is very sensitive.

Agreed, but he's incorrect in the way he explained the cons of adjusting the bias higher. Obviously there are caveats in applying even slightly higher bias than necessary for class AB, but the way he drew the resulting signal waveform is so weird, take a look of the video again (from approx. 17:50 to 19:30) and his graphs of bias vs harmonics distortions. I have read many things on bias for class AB amp and electronic text books, never seen graphs like those (ffwd to 18:00 to see that graph). Again, assuming he's one of Hegel's design engineer/designer, something might have gotten lost in translation and I misunderstood what he's trying to show.

 

[PENG]

I never listened to any of the Hegel amps to understand if there is any credit to this.

I have listened to their integrated amps and high end power amps. Sounded fantastically transparent to me, but then so do my own amps. If I didn't have my Halo and Bryston amps, I might consider the H20, thought it still be difficult as they are just so many choices.

 

[PENG]

I agree, totally far fetched.

Of course amps matter, its like power amp's THD specs matter, but to say 0.0001% is audibly better than 0.005%, or amps that employ some sort of feed forward scheme is audibly better than one that does not, would be even more far fetch. There is a bottleneck in the chain from the recording to the listener so while everything matter, it's the real bottleneck int he chain that matter most. Wires matter too!

The only other feed forward approach I am aware of is some designs of Nelson Pass, which many thought infringed, Peter's patents.

I bet there are a few, Benchmark has one too.

 

[haraldo]

There is also a DAC with specs that I certainly have never seen before ....

Distortion: Typical 0.0005%
Frequency response: 0 Hz - 50 kHz
Noise floor: Typically -150 dB

There is NOT a typo here..

 

[Verdinut]

There is also a DAC with specs that I certainly have never seen before ....

Distortion: Typical 0.0005%
Frequency response: 0 Hz - 50 kHz
Noise floor: Typically -150 dB

There is NOT a typo here..

Yeah, you can get one at a ridiculously high price. Nobody needs that. It would go well with the Hegel amps.

 

[haraldo]

Yeah, you can get one at a ridiculously high price. Nobody needs that. It would go well with the Hegel amps.

Why nobody needs that?

 

[lovinthehd]

Yeah, you can get one at a ridiculously high price. Nobody needs that. It would go well with the Hegel amps.

Audiophiles don't need a lot of things, LOL.

 

[Verdinut]

Why nobody needs that?

A lot of people who don't know what to do with their money, buy very expensive equipment because they can afford it and they are fooled by the audiophile shop salesmen because of ignorance. There has been no scientific proof that specs that exceed a certain level provide a superior musical quality. This means a waste of money at the end.

 

[haraldo]

A lot of people who don't know what to do with their money, buy very expensive equipment because they can afford it and they are fooled by the audiophile shop salesmen because of ignorance. There has been no scientific proof that specs that exceed a certain level provide a superior musical quality. This means a waste of money at the end.

Do you really believe people like to throw money out the window?
I have a friend who spent $10k+ on a pickup alone, and he is one of the more clever people I know, I never listened to his rig yet, but will ....

I spent a significant amount of money on amplifier and also on power filter, both of these make some significant audible difference in my home.

According to your thinking then I am an absolute idiot.

On expensive equipment, how can you for instance possibly imply that anyone spending $100k on speakers are idiots?

Did you ever audition $350k speakers?
The KEF Muon shocked me to the core of my soul, nothing is even on the same planet .... it is 5 times better than any regular speaker.

I will never be able to afford them, but I dont call people buying these things with ugly names.

Is for instance a KEF Muon worth that insane amount of money, probably yes.... thing is, there is no point of diminishing returns here. (It’s actually being written that Kef is losing money by an artificially low retail price on Kef Muon)

Can be same with electronics