If ever there’s a chance , would love to let you have a listen to my pairing 
but I am on the other part of the globe … anyway , If you have chance to demo , demo the Norse 2 range , especially tyr 2 loom… I demoed them( tyr2) on naim gears w the naim cables , I inched more towards the nordost sound . Also I want to have a change of some sound DNA 
1 Like
Is it the impedance curve? I thought it was the frequency response…
Stability requirements are much more than closed loop frequency response, in fact depending on how it’s done it can decrease an amps stability. To check for stability you need the Nyquist plot or the Bode plot as the phase relationships are also important.
There can also be other problems if the gain is rolled off using the feedback loop, this causes ultrasonic issues, particularly when looking at transient performance and other oddities like driver latch up problems.
Given Naim’s transient performance they are not just using bandwidth limits implemented in the feedback loop.
To ensure a good transient performance the forward gain bandwidth of the gain block may well be restricted to below it’s ‘nominal’ capability.
I really don’t see what everyone has against putting capacitors in the feedback loop. The whole Naim-clone crowd seem to think it’s a mortal sin too. Sure, you can’t just shove anything in there but these days we have excellent, free, Spice modelling which will tell you very accurately what’s going on. There is a capacitor there anyway (and in a few other places), you just don’t know what its value is or what behaviour it is responsible for. Why not put in one that you know, and define the bandwidth of the amplifier? I’d say there is an argument for bandwidth limiting every amplifier stage; not least from the perspective of noise. Why have 300MHz of noise (yes there are op amps that fast in audio use) when you only need a defined response to 100kHz, or rather less.
Incidentally, I’m not sure what you’re getting at when you say “Naim’s transient performance”. I think we should stop using the word “transient” everywhere because it’s not what we mean. You can’t “improve” the transient response by limiting the bandwidth, and getting everything to happen at the same time (ie. zero group delay) isn’t the key to getting what we call “transient attack” - ie. the leading edge being in the right place but also with sufficient weight and body to the sound to make it realistic and meaningful. One could conceivably look at a Bessel transfer function and deduce that it might well be more precise at high frequencies, but there is nothing there that gives you a clue to how full-bodied it’s going to sound. (In fact, nor am I at all convinced that “timing” in the music comes from time-accurate circuitry.)
Could we please abide by forum rules and not import discussion etc from other forums such as ASR etc… thanks.
This is not an example.
For years, I used my Isobariks DMS with 2x 0.30cm of Naca 4 then 2x0.30 cm in Naca5. My nap 135s were placed just below the DMSs in the metal brackets. The 135s never heated up, and the system played well.
Transient is what I mean.
I assume you are aware of the differences of transient performance between forward gain bandwidth limitation and closed loop bandwidth limitation of an amplifier?
Are you also aware of the transient ‘driver latchup’ situation that can sometimes occur in amplifiers?
Then consider the effect of the DA of a capacitor in the feedback loop has on the closed loop characteristics of a high gain amplifier when the bandwidth is being limited in the feedback loop.
These give a good indication why forward gain bandwidth limitation is often preferred in audio power amplifiers.
My post seems to have been deleted, but I only mentioned that the Stereophile measurements shown above, which @Christian_Thomas said he struggled to believe, where in line with those on that on that particular forum.
I’m not sure I quite understand how that equals importing a discussion. Or is the rule that we’re simply not allowed to mention, or hint at the existence of, other fora at all? Or just not that particular one?
Could you please clarify? Thank you.
I think you are going to have to explain to me what you’re getting at, starting with your definition of “transient”. I’m aware of the gain bandwidth product and of various situations where transistors can latch to the rails, either momentarily or for longer. And I have no idea why the DA of a circa 100pF film capacitor (usually either PP or Polystyrene) should matter since I imagine it’s close to negligible. So no, overall, I have no idea what point you are trying to make or, more importantly, what important thing I am apparently overlooking.
I suspect you mean 0.3 of a metre rather than centimeter. Lol. 
1 Like
Two of mine have gone too - taking with them the most important philosophical point I have ever made on a public forum about the limitations of using measurements as a metric. I shan’t repeat it as I anyway thought it might be giving away too much to anyone who was going to give it a moment’s thought. Equally, I would have left it up had it not been excised.
Importing discussion from other social media and forums such as ASR, etc… is not allowed here. I removed the posts and any follow ups.
Referencing established and legitimate publications subject to editorial control and responsibility such as Stereophile is OK within forum AUP limits.
Having looked at the guidelines I suppose there is some justification for removing our comments. I think your reliance upon editorial control of ‘legitimate publications’, however, is a little hard to take seriously given the content is hifi.
2 Likes
Hi Richard,
thanks for the response, but unfortunately it hasn’t made things clear to me. I knew importing discussions is not allowed, but AFAICT I didn’t import any discussion. Yet you seem to think I did.
What I did was mention it’s existence, and I mentioned it had a plot similar to the Stereophile one. No mention of any discussion or even anything anyone said on there.
What is your (Naim’s?) definition of “importing discussion” ?
Mentioning another forum (or commercial site)? Acknowledging the existence of any piece of information on any other forum, social media or commercial site ?
I’m sure I’ve seen both being done on here regularly, I even did the second one recently in my speaker build thread. I’m happy to edit that if not allowed of course, just trying to understand what the boundary is.
Thanks.
Making reference to another forum or social media site with regard to a discussion or thread there is likely where the boundary lies. These things are never black and white though so it often comes down to the moderators discretion. As ever, usually best to avoid mention or reference - i can only moderate on what’s here, not what’s elsewhere!
Discussing commercial entities is fine so long as posters have no personal or professional interest or involvement, although linking to their websites, social media or YouTube is usually not permitted without Naim’s OK to do so.
1 Like
Transient: characteristics present in short term signals that are no present in steady state conditions such as sine waves, or input signals that exceed the closed loop bandwidth of the amplifier when present in combination to in-band input signals.
Concerns:
Small signal transistors in the gain block and also the driver transistors are many times faster than the closed loop bandwidth of the amplifier and the output structure of the amplifier. If the bandwidth is primarily limited by the negative feedback loop, when a fast impulse signal appears at the input of the amplifier, then until the output structure and feedback loop apply closed loop correction the gain block will amplify this with the full open loop gain. This is likely to cause the driver transistors to swing close to full excursion risking latch up. If there is energy storage in the feedback loop (DA of capacitors, stray inductance etc.) this will delay the negative feedback signal, which, with the fast response of the gain block, further compounds the issue.
Even without drive latch up, this still causes a temporary inaccurate tracking of the in-band signal throughout the amplifier. One notable effect of this is an increase in TID.
Sorry for the translation error.
The length of my Naca was 12 inches
1 Like
DA is a LOSS, not a storage mechanism! The storage components are the capacitance and the inductance.
I see you are talking about “energy” storage. It was a dark day for science when Linn products started littering their (usually duff) explanations with the word “energy”. This is the very, very worst way of looking at almost anything in hi fi. When people have learnt their basic Physics, or Electronics, from Linn Products - or from Hi Fi Answers who learnt it from Linn Products - they are usually way beyond rescuing.
Yes, I am aware of all the points you make - though that’s not really latching so much as a “stickiness” due to capacitances needing to discharge.