Member
Joined 2009
Paid Member
I assume that I have bias and then the task becomes not to eliminate the bias (you only fool yourself) but to recognize what bias you have. One bias I tend to have is regarding treble 'smoothness' (call it what you like) as my ears were 'damaged' by listening to loud music and as I've aged that damage has lead to a sensitivity to exaggerated treble. Of course choice of speaker is a big factor but I've found that amplifiers that simulate with flat frequency response (and with global feedback there's really no excuse for anything else) don't all sound smooth. The usual problem with my amps is that the treble causes 'listening fatigue' or worse - plain irritation. Amplifiers that are no problem to me are my original AKSA 55 clone, my TGM7 and TGM8. However, a higher voltage version of that first amplifier made with dual output devices - call it AKSA 100 clone, is not as smooth (haven't finished optimizing it yet). My Naim clone is also not as smooth and this matches many reports from owners of original Naim amplifiers that they can be too forward. I don't believe this can be written off as amplifier oscillation due to inappropriate choice of speaker cable as there are too many reports of this nature from folk clearly using a recommended set-up. So I too find my Naim clone to be Airy in a way no other amp is, but it's also more fatiguing as a result and I may tweak mine further. Actually, I haven't used it for ages - I'm going to have to dust off my amp collection and do some listening tests. A friend of mine upgraded his Alpair 10.3 Pencil's to 10p (it's got a smoother published FR) which will give me a very 'pure' view into the sound signatures of my amps again.
Last edited:
IMO airy and fatiguing are not positively correlated in live sound. So this is an electronic side-effect. The Naims don't sound perfect but they do sound lively and have pace, but to Naim's shaim they do sound shrill/strident too which is fatiguing and is not correct. A retailer I know calls this the "10-minute effect"...Naims sound fantastic for about 10 minutes and then your brain starts to ache. I've not read any complaints like this about tube amps, for example, although I'm sure it is possible to make a tube amp sound shrill if you really try hard.
Last edited:
Member
Joined 2009
Paid Member
I studied the Naim circuit for a long time and eventually I realized what it is aiming to do. I also studied many other designs. I also built my own circuits and my own designs, both BJT and FET, and used my ears to judge the effect of seemingly minor circuit changes. I then "adjusted" (quite a big effort!) standard electronic circuit theory to explain the observations. After all this I designed the amp I am listening to as I write this; but it bears little resemblance to the Naim topology.
There are known ways to reduce shrill, like going with low feedback or using a single-ended class A topology, but this is no longer Naim. The challenge is to have startling clarity without shrillness but with realistic pace. Dull is easy.
I have not tried to clone a Naim and attempt to make it better than Naim can, by tweeking. I expect the stridency can't be moderated without costing PRaT within this topology. I'll give it some thought.
I agree with everyone's opinion.
Sometimes we are about to take the wrong approach to evaluating the musicality that an electronic or an audio setup can express.
As an alleged technician as I immodestly define myself, I certainly recognize the importance of numbers and objective measures for the evaluation of a project or a circuit design, but the musical seduction that an amplifier can express, certainly belongs more to our sensitivity and our emotions than to our rational and logical consciousness.
The NAIM design, with the necessary shrewdness related to the :
- matching and the choice of HFE for the transistors in the differential input stage,
- type of active components of the VAS stage,
- problems related to the VBE multiplier and thermal stabilization ,
- choice of pilot stage transistors,
is certainly a project that lends itself to a careful experimentation made in stages and experiments (which in my case lasts for more than five years) alternating with careful and relaxed listening.
Personally, I have never believed in plug and play or in "drums", which certainly allow you to assess macrospecies such as dynamics and power, but are absolutely misleading in assessing the real potential in terms of musical seduction.
I would also like to add that over the years, on NAIM design as well as on other projects, although I occasionally enjoy comparing them with their originals, I have never trivially tried to find the exact overlapping of sound and musicality, but rather what in my audio setup, with my speakers and in my ear, sounded right and better than the others.
Apart from my personal conviction that in a power amplifier, even more than in any other electronics intended for audio use, the power supply stage is just as important if not more important than the actual amplifier.
And with the NAIM clone circuit the adoption of a correctly sized, "fast" and low impedance power supply stage, both for the linear type for the driver and power stages and the stabilized one for the differential input stage and the VAS stage, are the things that have offered me the most concrete and exceptional results in terms of refined timbre, delicate detail, incredible realism and extraordinary sound-stage.
Everything I've always looked for in commercial electronics in 40 years of passion and I've never found expressed at the same time in the same amplifier.
The NAIM clone, even if with the effort of more than 5 years of attempts, modifications and improvements, has instead succeeded.
Sometimes we are about to take the wrong approach to evaluating the musicality that an electronic or an audio setup can express.
As an alleged technician as I immodestly define myself, I certainly recognize the importance of numbers and objective measures for the evaluation of a project or a circuit design, but the musical seduction that an amplifier can express, certainly belongs more to our sensitivity and our emotions than to our rational and logical consciousness.
The NAIM design, with the necessary shrewdness related to the :
- matching and the choice of HFE for the transistors in the differential input stage,
- type of active components of the VAS stage,
- problems related to the VBE multiplier and thermal stabilization ,
- choice of pilot stage transistors,
is certainly a project that lends itself to a careful experimentation made in stages and experiments (which in my case lasts for more than five years) alternating with careful and relaxed listening.
Personally, I have never believed in plug and play or in "drums", which certainly allow you to assess macrospecies such as dynamics and power, but are absolutely misleading in assessing the real potential in terms of musical seduction.
I would also like to add that over the years, on NAIM design as well as on other projects, although I occasionally enjoy comparing them with their originals, I have never trivially tried to find the exact overlapping of sound and musicality, but rather what in my audio setup, with my speakers and in my ear, sounded right and better than the others.
Apart from my personal conviction that in a power amplifier, even more than in any other electronics intended for audio use, the power supply stage is just as important if not more important than the actual amplifier.
And with the NAIM clone circuit the adoption of a correctly sized, "fast" and low impedance power supply stage, both for the linear type for the driver and power stages and the stabilized one for the differential input stage and the VAS stage, are the things that have offered me the most concrete and exceptional results in terms of refined timbre, delicate detail, incredible realism and extraordinary sound-stage.
Everything I've always looked for in commercial electronics in 40 years of passion and I've never found expressed at the same time in the same amplifier.
The NAIM clone, even if with the effort of more than 5 years of attempts, modifications and improvements, has instead succeeded.
Last edited:
That's also true.
You're right. I used headphone in office PC and listened thru Youtube (HD) and I couldn't listen to it even with low volume.
Show me to the .asc of your current Naim clone, and if you intend to modify and listen to it I will try to make little modification (if possible) that I wish will remove fatigue a lot.
It is okay without transistors designations. Please?
We have heard many amps that supposed to be good quality. They do all sound similar (I'm not using the word the same). When an amplifier sounds different (usually impressive), its a suspect.
You mean low feedback or low gain?
The is no free lunch but there is lunch.
I have not heard an F5 but the schematic reveals a lot. I don't see any current feedback here; perhaps you meant the F7? I believe "positive current feedback" was added to the F7 after some disappointing reviews, to perk it up a bit.
Attachments
Low feedback gain.
I probably should not generalize because it is so dependent on the actual circuit and parts chosen. It is the case that too much NFB can harm the sound in certain circuits, and it can do so in different ways including making the sound fatiguing, strident, garbled, flat/shrunken/confused image and others, and a different combination of these depending on the circuit. The loudspeaker you use also has an influence. These sort of effects can also be caused by things other than NFB.
Member
Joined 2009
Paid Member
that would be interesting for sure - but we best move that to the proper thread because my clone was not from ePay
TGM10 - based on NAIM by Julian Vereker
The feedback on the F5 goes through R5, 6, 7 and 8 to modulate the Vds drop across the input JFETs. It's negative feedback (unlike on the F7 which does indeed have a pinch of positive feedback).
I put "current feedback" in quotes because it's technically low-impedance feedback, but all the high-speed opamp guys insist on calling it current feedback.
I put "current feedback" in quotes because it's technically low-impedance feedback, but all the high-speed opamp guys insist on calling it current feedback.
Member
Joined 2009
Paid Member
I have not found, in general, that I've NOT liked using a CFP type topology for gain. I know Hugh has tried it and liked it so that's not to say it's always true. I've tried it as a Vas in a couple of different amps and I've tried it in the input stage. It produced fatigue in all cases. However, I've not had any issues using it in the output stage, either as a compound driver or as the output power device itself. In both cases the amps sounded superb but the CFP was being employed as a follower. I believe the F5 has a kind of CFP type topology with gain so it's not a good choice for those sensitive to fatigue in my experience. In the case of the NAP 140 I would not ascribe any of the fatigue to the topology of the output stage (which contains a CFP); it's coming from the front-end.
Last edited:
Member
Joined 2009
Paid Member
The Romans once said:.................cum grano salis!
Although I am a faithful disciple of the great Matti OTALA, I actually think that Feedback, in the case of solid state electronics, used with parsimony and intelligence (......cum grano salis), is not necessarily the bearer of bad things.
But of course always the bare minimum and only to get spectacular THD bandwidth and distortion diagrams, as the Japanese did in the 70s and actually many continue to do today.
Musicality, otherwise, will always be like ................a magnificent and perfect greenhouse tomato, ...........tasteless and without any attraction!
Although I am a faithful disciple of the great Matti OTALA, I actually think that Feedback, in the case of solid state electronics, used with parsimony and intelligence (......cum grano salis), is not necessarily the bearer of bad things.
But of course always the bare minimum and only to get spectacular THD bandwidth and distortion diagrams, as the Japanese did in the 70s and actually many continue to do today.
Musicality, otherwise, will always be like ................a magnificent and perfect greenhouse tomato, ...........tasteless and without any attraction!
Last edited:
Member
Joined 2009
Paid Member
Low feedback gain.
Feedback gain is CLOSE loop gain, right? Some people think that NFB is bad and some people think that it is good. Only a few think that it is not the NFB itself but how it is used.
When we change the NFB, or any random resistors, there are more than one parameter changed also in the circuit. If only people know which parameter that is responsible with the change in sound when they change NFB (increase or decrease) then they will be more capable of creating good sounding audio systems.
In the old NAP circuit, I will predict that to reduce the 'shrill' the gain should be reduced (instead of increased). The THD may be increased as the result (instability too) but the perceived distortion may be lower. Have any of you heard about this 'perceived distortion'? May be not, but i believe that you know that some amps have high THD but do not sound like they have high distortion but the contrary.