Naim Headline

[kvik]

Hi Frans,

This is a basic upshot from the similiar thread in danish.

With a second Naim Headline incoming, I am considering to modify one of them. I am contemplating replacing the existing signal lead with a Naim HiLine signal lead. This would then be mounted/soldered internally as the original lead, with RCA plugs applied at the source end.

While I am at it (modifying things), I thought there might be something to gain by upgrading other parts of the amp. I have absolutely no knowledge in this area, so am really asking, if anything can be done, at what price, and with what expected outcome sound-wise?

I should mention that the idea is solely to refine/expand upon the existing sound signature, it is not to alter it fundamentally. The Naim HiLine I am familiar with already in this regard. The component side of things, however, is unchartered territory, and I am somewhat more cautious here.

I would appreciate any input from you in the above context. Be it specific proposals as to what to upgrade, or just general views on whether the project is feasible, with my reluctance against altering the fundamental sound signature in mind.

The danish thread has pictures of the actual model (the new one, which is what I have, is the one with the curved front), there is also a link to a pfm-thread containing what I understand to be the schematics for the amp, though I am unsure, if they cover the old or the new model.

Looking forward to reading your take on this, Frans. Bear in mind that I am a novice, and will likely hire someone more knowledgeable to perform the upgrade, if I decide to go ahead.


Regards Ketil

[Solderdude]

Well I do not 'believe' in the sound of parts and cables and that is simply because in my experience I have never been able to prove it to myself (and others) in (blind) AB tests.
Subjective influences I disregard by more or less educated experiences.

That said there are some things I would have designed differently.
For starters it is a single power supply (not a +/- rail voltage)
This doesn't have to be a problem though as many designs are.

The design is 'textbook standard' typical class-A-B design and doesn't appear to have any design aspects that are different or stand out as opposed to other designs.
It seems to 'match' best with low-impedance headphones and can drive 300 Ohm headphones to just sufficient levels (50mW).

If you intend to use it to drive higher Ohmic headphones and want it to be different (in that aspect) from the other one you've got, I see no reason to have 2 of the same, I would increase the power supply voltage to say 24V (will need to change parts) and get a little more juice and a possibility to play with different output resistances as well.
The feedback coupling cap (47uF) I would increase to 220uF (electro).

I see lots of tantalum capacitors are used. Although fine for decoupling in power supply circuits where limited currents are present it is common 'hifi' practice not to use them in the audio path as it is claimed to sound bad.

I have never been able to confirm this myself.

Tantalum have as good properties...
positive:
no ageing (electro's do age/dry out)

negative:
higher leakage currents (not so much a problem because of the output caps).
cannot withstand reverse polarity (single supply voltage here so no problem).
higher ESR (only important in power supplies where a low impedance for higher freq is needed, also not the case here).

In case you want to 'up' the power supply voltage I can give instructions.

[kvik]

I will have to just 'digest' your comments for a day or so.

About the power supply voltage, it is my understanding, that the amplifier already has the ability to use 18V (Naim Napsc power supply) as well as 24V (Naim Hicap). I plan to use a TeddyCap (24V) with the Headline unit in question, and have done so successfully in the past. Are you talking about something different?

Remember I am not well-versed, so will probably use wrong terminology occasionally.

[Solderdude]

All that 'extra' voltage is dissipated as heat in the regulator and not put to use in the amplifier circuit.
The internal LM317 is set to have 12.something Volt at the output and this determines the voltage that is used inside the amp, and subsequently determines the maximum output voltage as well.
Regardless of what is being fed into the regulator (within it's limits).
It is this voltage (the internal 12V) that needs to be changed but involves changing quite a few parts and some experimenting.

[kvik]

Ok, thanks. I will have a think about, whether it serves any purpose to proceed. I am not sure, if I will be using higher ohmic headphones in the future. Currently just using Beyerdynamic T70 and Spiral Ear 3-way Reference. The reason for getting the second Headline was primarily to have one at home as well as one at work. Thank you for your time and input.

[Solderdude]

Currently just using Beyerdynamic T70 and Spiral Ear 3-way Reference. The reason for getting the second Headline was primarily to have one at home as well as one at work. Thank you for your time and input.

The T70 = 250 Ohms, The T70P = 32 Ohm.
Which do you have and do you find them 'toppy' in the highs ?

For the T70 (250 Ohm = high impedance) it might be worthwhile exploring a higher voltage amplifier and add an output resistance.
This will bring out more warmth and slightly lower those OTT highs.

having 2 similar amps at 2 locations makes sense

[kvik]

The thing is I am perfectly happy with the amplifier as it is, which is also why I wanted another for my office setup. I use the 250 ohm T70 at home, and the CIEMs at work. I feel the Headline adds some warmth to T70, leaving them more balanced and without problems with 'toppy' highs (as you call it). It was mainly the luxury problem of having to identical units, that made me consider, if one could perhaps be improved upon. As it looks right now, the costs and logistics involved seem to outweigh the sound improvements, or rather the latter is too uncertain/unknown for me to proceed.

[Solderdude]

The thing is I am perfectly happy with the amplifier as it is.

In that case there is not much reason to do mods.
If it were mine I would play with adding output resistors which is rather easy to do and see if you can get it to sound even better tinkering with that.
Also I would replace the 47uF tantalum for a 220uF electrolytic.
Perhaps replace the 10uF input tantalum for a 2.2uF or 3.3uF non-polar cap.
The tantalums in there now will not deteriorate over time as electrolytics do though.

Most sonic differences between amplifiers are differences between output resistance or output power limitations that 'compress' music.

[kvik]

Thank you, Frans, I will consider your proposals.

Taking the opportunity to learn a bit more.. I have had the one Headline for a while. So far it has been supplied with external power from Naim Napsc (18V), Naim FlatCap (24V), Naim HiCap (24V) and TeddyCap mkIII (24V). To my ears each psu adds a different flavour, maybe not night and day, but still noticeable. So, my question is why is this exactly? If the extra voltage is just dissipated as heat, it has to be something else doing the 'magic'. I am sure this is basic textbook amp/psu stuff, but I have often wondered about it anyway. If you have the time, can you explain the workings? It doesn't have to be an encyclopedia, just maybe cover the basics.


Edit: Flavour to be translated in a loose sense. It is rather perceived as increased resolution and control for the Naim PSUs, with the TeddyCap standing out a bit from the bunch, in that it seems to change perceived balance as well (leaner, smoother).

[Solderdude]

This is deadly ground for me to walk on.... subjective found differences that is.
It has been the start of many ugly threads in every hifi forum.

My opinion:
Whatever happens before the regulator, assuming input voltages are above the minimal required, thus the ripple bottom should be at least 3V above the output voltage, cannot have any TECHNICAL influence in the circuit behind it.
Viewed from the amplifiers side/perspective the power supply voltage it 'sees/receives'' is always exactly the same regardless of the input voltage.
Its what regulators do for a living so to speak.
For currents it is exactly the same as the maximum current drawn is determined by the regulator as well.
That is... if the connected power supply can at least deliver the current.
The heavily oversized trafo's in the picture should have no problems reaching the current limit of the LM317.
In this case the current limiter is the amplifier itself, which cannot reach high enough output currents the LM317 can provide.
There cannot be more or 'better quality' or 'faster' currents as it is determined by the regulator unless as said the power supply feeding that circuit is limited.
The other difference in power supplies that MIGHT have any technical influence is the leakage current.
This COULD influence audio when there are more mains fed (SMPS !) devices connected.
I expect Naim power supplies to be all 'normal' transformer type (linear).
For leakage currents to exist you do not even have to have electrical connections conducting those currents !

Leakage currents can cause garbage (well above the audible range) to enter the audio path.
It's only when it becomes high enough for semiconductors (transistors) to become detectors till it becomes audible.
The best way to show this is with the GSM signal.
900MHz or 1800MHz yet when rectified (when being called) the infamous sounds become audible by detection of HF.
When multiple SMPS garbage with different leakage currents enter the audio path by conduction between those devices via interlinks that this MIGHT influence audio/perception.

In the end:
When you are happy with the sound you have and feel spending funds on the gear and enjoying the perceived sonic improvements is worth the investment I am all for it though.

For me (designing analog electronics as a daytime job, and passionate audio lover in spare time) studying the realm of auditory perception was very entertaining and highly educative.
Studying perception is something that must be done in person though and with some rigour/open mindedness to understand the mechanism behind what's perceived and why.
Not suited to discuss in most forums though... highly divisive.
It is not possible to convey all I learned without upsetting people I am afraid.

[kvik]

Thank you for the technical insights. That was the point of creating this thread in the first place.

[Claus-DK]

This is deadly ground for me to walk on.... subjective found differences that is.
It has been the start of many ugly threads in every hifi forum.

Been there done that...

I do not know, but I actually think that people on hoved-fi would be capable of agreeing to disagree and then leave it that..

But it is from a forum perspective very dangegrous ground, Cables and DBX discussions, were not allowed at head-fi.org, some years ago, not sure if they are now ?

Or sistersite has just taken steps to shut those threads down..

Those ugly threads does not do anyone any good, only make trenches between friends and can create a not very pleasent tone on a forum, people litteraly sits ready at their keyboards so they can "get even" ASAP
The worst part is that it is so easy to get sucked in and join the idiots (whichever side) and yell with them..
So lets stay clear of those threads

However I have experienced improvement with better Power supplies on various of gear, but it does ofcourse depends on the PSU that was there before..

Also when making a mod the "I did factor" is not to be underestimated, I clearly remeber when Frans helped me with my Panda and it finally started playing, it was like hearing the gods singing..

[DavidB]

This is deadly ground for me to walk on.... subjective found differences that is.
It has been the start of many ugly threads in every hifi forum.

No need to worry - we will behave

[Claus-DK]

the problem is that we have 400 members and if just one or two wants to stir up a trenchwar, it is a known fact that others will join the pack, it is human nature and before we even get a chance to react people writes stuff they really do not mean and friends becomes enemies, it goes downhill so fast that it is hard to understand..

[kvik]

Hi Frans,

I want to pick up on this old thread. I have swopped the Beyerdynamic T70 for a T1 (mostly out of curiosity), and with your previous comments in mind, I am considering carrying out the modifications you proposed (or rather to have someone more skilled carry them out for me).

I ended up with only one Headline after all. I don't anticipate using it with anything other than T1 (and maybe T70 again some day). I am therefore basically free to modify it in whatever way serves these two headphones the best.

As mentioned previously I was fundamentally happy with Headline+T70. Having spend a few hours now with Headline+T1, I am simply curious what kind of improvements can be achieved by implementing your suggested modifications.

To summarise your suggestions this is what I get:

1. Change from 12V single power supply to 24V (dual?) power supply.
You mention quite a few parts need changing to do this, which ones?
You mention some experimentation will be needed here, what kind?
With 24V internally will this have any significance for the use of external power supply (Naim Napsc being 18V, and Naim HiCap being 24V)?
Will 24V internally make it possible to change the single-endet output for a balanced one? Would there be any point in this, the Headline getting the signal via DIN (using only three of five pins - plus/negative/ground)?

2. Replace the 47uF tantalum feedback coupling cap for a 220uF electrolytic.
Towards what effect?

3. Possibly replace the 10uF input tantalum for a 2.2uF or 3.3uF non-polar cap.
Towards what effect?

4. Play with output resistors.
More specifically what does this mean?

If you have the time, would you give instructions to psu change, and comment on the external psu role in this? Also if you could instruct more specifically on the output resistor part?

With this information I will be able to 'attack' the modification in a more real way. I will be able to gather information about costs involved, and begin to search for someone capable of doing the modification for me.

Thank you in advance.


Regards Ketil

[Solderdude]

To summarise your suggestions this is what I get:

1. Change from 12V single power supply to 24V (dual?) power supply.
You mention quite a few parts need changing to do this, which ones?

some capacitors might need a higher voltage rating (most likely) and the LM317 circuit values need to change.
perhaps even some values in the amp itself.

You mention some experimentation will be needed here, what kind?

Some measurements and perhaps a change of some values.

With 24V internally will this have any significance for the use of external power supply (Naim Napsc being 18V, and Naim HiCap being 24V)?

With 24V internally you will need at least 28V DC input voltage

Will 24V internally make it possible to change the single-endet output for a balanced one? Would there be any point in this, the Headline getting the signal via DIN (using only three of five pins - plus/negative/ground)?

Balanced operation requires double the amount of amps and a phase shifter circuit or 1 stereo amp with small mods to create a single channel.

2. Replace the 47uF tantalum feedback coupling cap for a 220uF electrolytic.
Towards what effect?

better bottom end, tantalum does not age, electrolytics do but admittedly takes many years.

3. Possibly replace the 10uF input tantalum for a 2.2uF or 3.3uF non-polar cap.
Towards what effect?

getting rid of a tantalum as coupling cap. They are semiconductors and have higher distortion than electrolytics or film caps.

4. Play with output resistors.
More specifically what does this mean?

Beyer themselves drives the T1 from 100 Ohm. You can increase the output R (now 4.7 Ohm) to 100 Ohm BUT as it is now, with 12V PS voltage 100 Ohm is not possible.

If you have the time, would you give instructions to psu change, and comment on the external psu role in this? Also if you could instruct more specifically on the output resistor part?

The output R is 4.7Ohm (in series with the 1000uF output cap in the schematic) but as said it can only be increased IF the internal voltage is at least 24V

With this information I will be able to 'attack' the modification in a more real way. I will be able to gather information about costs involved, and begin to search for someone capable of doing the modification for me.

Thank you in advance.


Regards Ketil

[kvik]

Thanks. That means a different external psu will be needed as well (≧28V). The HiCap is dual-rail 24V, will this change anything in this regard?

[Solderdude]

If you use a dual rail (+/-12V) you could also increase the internal voltage from 12V to 20V in which case you will have more output voltage.

In case you want to know how much output power that will give AND you like to hear what output resistance does to your T1 you could audition the Starlight amp.
The modified Headline will come very close to it.

Maybe the mods will only be changing one or 2 resistors and a few capacitors.

[kvik]

As I understand it, the HiCap psu (and the TeddyCap psu as well) has a 2x 24V power output. To utilise the dual-rail I assume the current 2-pin connection point (on Headline) for the external psu would have to be changed as well?

It is probably a good idea to have a listen to Starlight to see, if this is all worth it. Unfortunately the one on tour won't be an option, since I am in Norway. I will see what I can work out. Thanks, again.

[kvik]

A small addition.. Wouldn't the Horizon amp be better suited for T1? (I might just buy to try )