Densen amp

www.hifisonix.com
Joined 2003
Paid Member
" . . . with tubes, but also "solid state" amplifiers like "ZEN/ALEPH" series of master Nelson Pass . . . "

I don't think you can campare these low feedback NP designs with the subject of this thread. One is about minimalist design and voicing, the other is about idiosyncratic design practice (good word that 'idiosyncratic' - I like it)
 
" . . . with tubes, but also "solid state" amplifiers like "ZEN/ALEPH" series of master Nelson Pass . . . "

I don't think you can campare these low feedback NP designs with the subject of this thread. One is about minimalist design and voicing, the other is about idiosyncratic design practice (good word that 'idiosyncratic' - I like it)

I do not know a lot about NP´s designs, what I think I do know is that they are very simple and normally also low powered. In some sense I think that they could also be called idiosyncratic :D, as NP is looking another way than the obvious logical one.
I.e if you need good grip on your speakers you need several pairs of output devices and so on. evt. even modest amounts of NFB could be a good solution. But me saying so, is also in some sense idiosyncratic :D.

But the most pronounced idiosyncracy in this thread, I think, is that some do favour distortion, and calls it good sound, airguitar factor, PRAT, musicallity or what do I know. And basically this often leads to endless discussions, because the words simply do not have the same meaning or deffinition.

To me "reproduction" is the buzzword. In my world, an amp is neither ought to ad anything nor subtract anything, just reproduce the signal @ a higher voltage.

Sonic behaviour will then tend to become very different from the idiosyncratic gear, differences between recordings tends to be a lot more pronounced, even between tracks on the same record, just as 3D and dynamics can become surprisingly natural.
Where the idiosyncratic way often, more or less, tends to make everything sound the same.

As a good example which many do know of is the Linn Sondek LP12 record player. It simply sounds in a funnyhumming way always. From Elvis to Stravinsky you alway have this hmmmmmmmmmm sound :D

Then try out a direct drive system or a reference system as i.e. Technics SP10 or SME 30, and listen to the difference. The idiosyncratics would be disturbed by the silence and roaring transients, and even have their attention taken to other things than the music, i.e. coughing, non musical instrumental sounds as i.e. clicking noises, ambience and so on. And that might disturb your PRAT feeling, because the resonances are gone.

Thats somewhat the same with amps, resonance in the electronic domain cause distortion, and can only be faught by keeping control of every stage what so ever. Low impedances, almost endless power delivery and wide bandwith are necessary to avoid these resonances, just as smooth distortion behavior. But this is neither easy nor cheap, and definately not high waffing.
 
Last edited:
When the outside loop power darlington leaves class A, at the same moment it starts to produce huge amount of higher harmonics. I know what I am speaking about, as I have been designing class A output stages for 7 years. Compared to this, properly designed global NFB amp would have much lower amplitude of higher harmonics.

There is a way how to design low distortion, class A out-of-the-loop output stage. CFP, high idle current and multiple output devices. Then, one needs a proper heatsink, which is not the Densen case. This Densen amplifier is really poorly designed.

It is actually their present top of the line model @ 7.500€ which is the subject for the measurements shown earlier.
The schematics shown earlier and now removed, I think was for the B110 2*60/120 Watt @ 8/4 Ohms integrated amp, or maybe an older B100.

But I can see that in germany an even bigger model is marketed @ 14.300 €, which should be capable of 375 Watts pr unit. They may feature at least 2 pairs of output devices pr. ch. I think.
I hope this new design will be of more mature nature, with less idiosyncratics incorporated.
 
Last edited:
Hi Mooly,
I always remember, something like "It's so easy isn't it"
I have concluded you probably won't change peoples opinion on here, the same topologies appear, in one form or another, with sound quality often in inverse proportion to the complexity of the design. I accept too that perceived "sound" is so very personal, yet I'm sure many haven't really seen (or heard) lol the light when it comes to audio reproduction.
Of course, subjective opinions are not an appropriate subject to discussion, but it is possible to be specific about this matter.
I meant the well-established British tradition of single-ended input topology, producing a tonal quality (with higher measured distortion) that balanced topologies cannot do. Simplicity is another criterion for a transparent sound, primarily meaning small number of stages.
 
Hi Mooly,

Of course, subjective opinions are not an appropriate subject to discussion, but it is possible to be specific about this matter.
I meant the well-established British tradition of single-ended input topology, producing a tonal quality (with higher measured distortion) that balanced topologies cannot do. Simplicity is another criterion for a transparent sound, primarily meaning small number of stages.

I think these, by some people highly praised, single ended topologies are the very essence of idiosyncratic design. It will never become reproducing, but the way of colouration and harmonics are tweaked in a direction desired by the designer.

Some examples are Sugden, MF, Linn, Naim, Dennis Morecroft and surely others. They are not necessarily single ended, but often idiosyncratic in other ways.

Btw. non single ended does not necessarily mean ballanced, but more often push pull.
 
Kurt von Kubik,
Btw. non single ended does not necessarily mean ballanced, but more often push pull.
Non-single-ended has to mean balanced, as there is not a third possibility. Push-pull is a balanced mode.
I understand, but you actually first have to define good and bad meaurements.
To me the lowest THD+N is not the best, I like it more when THD+N are reasonable but completely linear, just depending on voltage as the ideal amplifier.
If it is 0,1%THD or 0,0000001% I don´t care, but I do care if it is 0,1% @ 1KHz and 1,5% @10KHz. And in addition I do seldom like amps linearised by NFB.

But large amounts of THD+N, narrow bandwith and failure to deliver large currents, are always bad, and sounds bad, as they also are indicators of design flaws.
When it comes to harmonic distortion, to me it is fundamentally about spectrum not level. Harmonic distortion is a very benign distortion, including the more discordant and more easily detectable high orders. I just don`t care of it. Trying to "remove" harmonic distortion will inevitably cost you precious values, since once appeared, distortions cannot be removed only transformed.
If you look at music, it contains even order harmonics, and none what so ever uneven.
Uneven harmonic distortions, has simply nothing to do with music, therefor the uneven are the most mallicious ones. In addition it is mostly the lower order uneven harmonics, which are these closest to the basic tone, that are the worst ones. But I do think higher order unevens are uneasy to the ears as well.
This indicates big confusion.
 
Harmonic distortion is a very benign distortion, including the more discordant and more easily detectable high orders. I just don`t care of it. Trying to "remove" harmonic distortion will inevitably cost you precious values, since once appeared, distortions cannot be removed only transformed.

It's now several years since you are preaching the same **. Over time, you were offered arguments, measurements, math, physics, reasoning, and apparently everything failed. I have to conclude it would probably take brain surgery to pull this out of your head.
 
Last edited:
Kurt von Kubik,

Non-single-ended has to mean balanced, as there is not a third possibility. Push-pull is a balanced mode.

When it comes to harmonic distortion, to me it is fundamentally about spectrum not level. Harmonic distortion is a very benign distortion, including the more discordant and more easily detectable high orders. I just don`t care of it. Trying to "remove" harmonic distortion will inevitably cost you precious values, since once appeared, distortions cannot be removed only transformed.

This indicates big confusion.

Non single ended is not ballanced.
Ballanced audio is always utilising 2 amplifiers in stead of one. One non inverting and one inverting amplifier.
The signal is defined as the difference between the non inverting and the inverting amp. Ground is floating, and has nothing to do with the signal itself.

Push pull amplifiers are usually designed as one amplifier with complementary semiconductors, amplifying each half of the signal waves. This is not a ballanced topology, it is most often referred to as single ended, as it is referring to ground.

Real Single ended amplifiers are designed with only one semiconductor (or with no complementary semiconductors at least) amplifying both halves of the signal wave, and will normally output the supply voltage in conjunction with the signal, referring to ground. It runs on a single pole supply, i.e. + xx V.

Removing distortion is impossible, transmitting it can result in problems indeed.
But a reproducing unit is to transmit every incomming signal @ any level precisely what so ever. Anything else is simply untruthfull.
Thus both spectrum and level is of utmost importance.
Smooth spectrum AND low values are both very important parameters.
 
Last edited:
It's now several years since you are preaching the same **. Over time, you were offered arguments, measurements, math, physics, reasoning, and apparently everything failed. I have to conclude it would probably take brain surgery to pull this out of your head.


Of course no distortion is a lot better than any distortion, what ever the order or spectrum might be, any diversion from that path, must be considered idiosyncratic:D

But opinions are as common as AH´s, so why don´t we at least agree upon that ? :grouphug:
 
Last edited:
syn08,
It's now several years since you are preaching the same
And I will (with God's help) continue to do so.
Over time, you were offered arguments, measurements, math, physics, reasoning, and apparently everything failed.
I can only remember some worthless Spice figures and graphs.
I have to conclude it would probably take brain surgery to pull this out of your head.
Well, maybe an extensive brain surgery...
 
syn08,

And I will (with God's help) continue to do so.

I can only remember some worthless Spice figures and graphs.

Well, maybe an extensive brain surgery...

Instead of emotional reactions, I´d find objective response a lot more appreciable.

So:
Push Pull still has nothing to do with ballanced topology.
Ballanced audio can be of push pull topology, and are mostly, but do not have to be so.

I think you are mixing signal types with amplifier topology.
Audio signals can be only ballanced or unballanced, where the latter sometimes flawlessly is reffered to as single ended.

Single ended audio is in fact an amplifier topology, mostly used in tube amplifiers.

Connections between equipment are thus either ballanced or unballanced, and never single ended.
 
Last edited:
Kurt von Kubik,
Non-single-ended has to mean balanced, as there is not a third possibility. Push-pull is a balanced mode.
This indicates big confusion.

"Push-pull is a balanced mode.".....????? and what is bridged mode?

The English language is often less precise than other European languages. This is an advantage for, and is also the reason, why pop stars prefer English for composing their songs. By the English language, you can compose the same in a variety of word choice.

But it has the disadvantage that many things can be ambiguous understood or interpreted. The terms "single ended" and "balanced" e. g. are ambiguous in the context of amplifier technology in any case.

English manufacturers calls their amps often "single ended". Here this term was chosen because the input stage instead of a LTP (differential amp) only use one input transistor. In the output stage a normal PP darlington or PP CFP power buffer is in use

Another manufacturer calls their amp "single ended", because its VAS stage operates in this mode (it works by most of all models always in this mode)

Only by "Passlabs" and some of the very few other manufacturers is "single ended" really meant the output power stage (single ended as the opposite of mostly used "PP" or "push-pull") By the way, an overview of all me known amplifiers with single ended output stages you find here:
http://www.diyaudio.com/forums/soli...ended-integrated-power-amplifier-devices.html

Therefore, it should be mentioned always individual in such intense discussions, what we understand by themselves.

So it is with the term "balanced" and "unbalanced" concerning to amplifier technologies.
Not only for myself but also for those who have English as their main language, it is difficult to find clear descriptions, so those that are clearly understood by others.
 
Kurt von Kubik and tiefbassuebertr,
Non single ended is not ballanced.
Ballanced audio is always utilising 2 amplifiers in stead of one. One non inverting and one inverting amplifier.
The signal is defined as the difference between the non inverting and the inverting amp. Ground is floating, and has nothing to do with the signal itself.

Push pull amplifiers are usually designed as one amplifier with complementary semiconductors, amplifying each half of the signal waves. This is not a ballanced topology, it is most often referred to as single ended, as it is referring to ground.

Real Single ended amplifiers are designed with only one semiconductor (or with no complementary semiconductors at least) amplifying both halves of the signal wave, and will output the supply voltage in conjunction with the signal, referring to ground

Push Pull still has nothing to do with ballanced topology.
Ballanced audio can be of push pull topology, and are mostly, but do not have to be so.

I think you are mixing signal types with amplifier topology.
Audio signals can be only ballanced or unballanced, where the latter sometimes flawlessly is reffered to as single ended.

Single ended audio is in fact an amplifier topology, mostly used in tube amplifiers.

Connections between equipment are thus either ballanced or unballanced, and never single ended.

"Push-pull is a balanced mode.".....????? and what is bridged mode?

The English language is often less precise than other European languages. This is an advantage for, and is also the reason, why pop stars prefer English for composing their songs. By the English language, you can compose the same in a variety of word choice.

But it has the disadvantage that many things can be ambiguous understood or interpreted. The terms "single ended" and "balanced" e. g. are ambiguous in the context of amplifier technology in any case.

English manufacturers calls their amps often "single ended". Here this term was chosen because the input stage instead of a LTP (differential amp) only use one input transistor. In the output stage a normal PP darlington or PP CFP power buffer is in use

Another manufacturer calls their amp "single ended", because its VAS stage operates in this mode (it works by most of all models always in this mode)

Only by "Passlabs" and some of the very few other manufacturers is "single ended" really meant the output power stage (single ended as the opposite of mostly used "PP" or "push-pull") By the way, an overview of all me known amplifiers with single ended output stages you find here:
http://www.diyaudio.com/forums/solid...r-devices.html

Therefore, it should be mentioned always individual in such intense discussions, what we understand by themselves.

So it is with the term "balanced" and "unbalanced" concerning to amplifier technologies.
Not only for myself but also for those who have English as their main language, it is difficult to find clear descriptions, so those that are clearly understood by others.

You can`t offer such a muddy explanation.

The definition is simple and explicit and you should accept it:

In the single-ended unbalanced circuit the signal is referred to ground and is asymmetric with reference to ground, in a balanced circuit, two devices operating in phase opposition, the signal is balanced to ground and is symmetric with reference to ground.
 
Administrator
Joined 2007
Paid Member
Hi Lumba

syn08,

And I will (with God's help) continue to do so.

I can only remember some worthless Spice figures and graphs.

Well, maybe an extensive brain surgery...

:) You are funny Lumba


May I ask, I'm just curious what type of amp you have at home... your favourite.
What class ? What topology ? power etc. Bjt FET ?

I think you understand the points I make about subjective sound quality now and again... and how certain topologies have a sort of "family" sound that you can't get away from.
I believe the spectrum of harmonics has a lot to do with perceived sound quality, it's one reason I think the OPA604 is such a good opamp for audio.

So do tell :)
 
Kurt von Kubik and tiefbassuebertr,

You can`t offer such a muddy explanation.

The definition is simple and explicit and you should accept it:

In the single-ended unbalanced circuit the signal is referred to ground and is asymmetric with reference to ground, in a balanced circuit, two devices operating in phase opposition, the signal is balanced to ground and is symmetric with reference to ground.

Well!
You may think it is so, but it is not.:deerman:
There is no such thing as single ended signal. It is called unballanced.
Single ended is an amplifier topology, not a signal transfer format.
Look here: Single-ended triode - Wikipedia, the free encyclopedia

Ballanced is not the same as push pull: Balanced audio - Wikipedia, the free encyclopedia

Unballanced audio is not the same as single ended: Sound Info: Balanced Vs Unbalanced Audio

The single ended term did by error end up in signal transfer, because unballanced or assymetrical obviously seemed to be "no go terms".
RCA cables are not single ended, they are unballanced or assymetrical.

This is, and probably always will be, the deffinition of ballanced/unballanced audio.

Luckily single ended audio is something completely different, maybe some reading of the SE master NP will help. If anybody, then he knows everything about SE audio.

Ballanced audio never refers to ground, which is the main advantage of ballanced audio in i.e. noisy studios and PA invironments.
 
Last edited:
Kurt von Kubik and tiefbassuebertr,
You can`t offer such a muddy explanation.
The definition is simple and explicit and you should accept it:
In the single-ended unbalanced circuit the signal is referred to ground and is asymmetric with reference to ground, in a balanced circuit, two devices operating in phase opposition, the signal is balanced to ground and is symmetric with reference to ground.
:rolleyes: