flat earth only a speaker theme?!

quote:

---

Originally posted by Todd Hutchinson:
It is not my theory, as you put it.

Orinally posted by James on page two:

"Naim uses filters deliberately to cut ultrasonic response down to around 40kHz for their olive range, and to around 70kHz for the new black." "Imposing a steep filter at, say 70kHz, will cause a phase shift within the audible range" "This phase shift in Naim electronics can enhance PRAT"

That's the best explanation so far.

Todd

---

Hy Todd ! i doubt if it is causal
that small bandwith will give more prat,
but I think big bandwith (if not done technical perfekt) is a possible source to loose prat and naims way is to limit this bandwith with filters well knowing that
less bandwith means less fault sources.
Ultrabandwith needs ultraquality (and therefore high degree selected and expensive
technical parts in an amplifier)
fm acoustics for example does 2mhtz in best everheard quality.Maybe halcro is also in this
league(i have never heard one)
mark levinson or jeff rowland are imo soundwise
definitly not in this class (i think rowland synergy preamp does 150000htz bandwith.)
krell bandwith i do not know.
imo it is not possible to go for more prat,it is only
possible not to loose the prat compared to the
the original(or to the given quantity on the software)playing music on your stereosystem.
O.J.

But there's nothing to filter! Certainly nothing that's part of the music.

Naim limit the bandwidth in their preamps to keep their power amps safe. I think there's nothing more to it than that.

Way back when the NAC72 came out they changed the type of filter to 'time aligned' (which implies less phase shift?), so maybe this is where the story is coming from?

Paul

quote:

---

Originally posted by Todd Hutchinson:
Hi OJ, that's a good way to put it. The small bandwidth employed by Naim filters isnt creating _more_ PRaT as much as it is filtering out other facets of the music that would otherwise mask it.

Naim focuses on musicality within a certain price point. To have both utltra bandwidth, round earth qualities and controlled pace, rhythym and timing, would be cost prohibitive for most enthusiast. Did I get that righ?

Todd

---

Hy todd!
1.I think this nearby correct but the thing is
that it is other facets of frequencies out of
the hearable range,I would not name these facets of music.(here started our paradoxon
in this discussion because thinking logical
music has its definition through the fact that it can be heard or felt(deep bass) by human
being,and therefore talking about music you think first about hearable frequencies.

2,Not every technican would go the way of Ultra
bandwith,because there are a lot of technicans
who do not believe that it is necessary.
Also if price is no limiting factor it is not
easy to handle the technical problems and there
are companies that have alot more experience
in going the bandwith way than others.i have
read a lot about this theme and having heard a lot of expensive ultrabandwith equipment it seems to me sure that high bandwith is one of the important steps in the direction of smooth artefactfree musicreproduction.
o.J.

quote:

---

Originally posted by Paul Ranson:
But there's nothing to filter! Certainly nothing that's part of the music.

Naim limit the bandwidth in their preamps to keep their power amps safe. I think there's nothing more to it than that.

Way back when the NAC72 came out they changed the type of filter to 'time aligned' (which implies less phase shift?), so maybe this is where the story is coming from?

Paul

---

hy paul! you wrote naim limit the
bandwith in their preamps to keep their power amps save.If i have a perfect signal to both edges of a highbandwith preamp it will get powered perfect in the power amp,if the signal
is distorted on the edges this distortion will multiplicate through the power amplifier and
maybe destroy the power amp or the speakers.
no distortion /no reason to keep the poweramp save with bandwithlimiting of preamp.
o.j.

You don't know what people are going to connect to your preamps, or what will happen if a stylus is dropped onto a record label. If your power amps don't have a substantial stability margin (as is the case with Naim, certainly classic Naim amps) then by restricting the input bandwidth you can reduce the risk of disaster.

The important point is that there is no audible information at ultrasonic frequencies. Anything that is there from a turntable, CD or tuner is noise. And it is sensible to remove it.

I don't really know what 'PRaT' is, but if it can be enhanced then that enhancement is false and should play no part in hifi.

Paul

quote:

---

Originally posted by Paul Ranson:
You don't know what people are going to connect to your preamps, or what will happen if a stylus is dropped onto a record label. If your power amps don't have a substantial stability margin (as is the case with Naim, certainly classic Naim amps) then by restricting the input bandwidth you can reduce the risk of disaster.

The important point is that there is no audible information at ultrasonic frequencies. Anything that is there from a turntable, CD or tuner is noise. And it is sensible to remove it.

I don't really know what 'PRaT' is, but if it can be enhanced then that enhancement is false and should play no part in hifi.

Paul

---

hy paul! i agree what you say about
prat.As i wrote to Todd i think you can not enhance it you can only try not to loose it.
(you cannot make exact timing more exact)
But i do not agree that anything out of the hearable range is noise.i also would not call this music (it is not hearable)but it is definitly a piece of electrical information running through the whole hifi system and damping these frequencies can change other informations. and i personally would not use any filter
that changes sound to minimise risk of desaster.
O.J.

quote:

---

Originally posted by Paul Ranson:
I don't really know what 'PRaT' is, but if it can be enhanced then that enhancement is false and should play no part in hifi.l

---

Even when you are having a good time with it?

Yes.

IMO of course...

Paul

quote:

---

Originally posted by Paul Ranson:
Yes.

IMO of course...

---

Gees. Paul.

Listening to tunes is supposed to be fun.

It is.

I just don't find any need for a 'turbo PRaT' switch. I want to get at the truth of the recording.

Paul

Colleagues,

I know that electronics engineers are generally hated and ridiculed on this site (in my case, I'm an ex-engineer but not immune I imagine) but it's a slow day in the office and I've got a few minutes for a reply.

Limiting the bandwidth of the signal fed into a power amplifier is not an unusual approach. It is pretty much a standard entry in any "hints and tips" list.

Here's why you do it, or at least here's why I do it:

Most of the musical information from any source will peter out at quite a low frequency, maybe before 10kHz. The next octave or so contains harmonics and all the high frequency information, which, while saying little about the music says a lot about how the instruments sound. Over 20kHz there is no useful information, of either variety, from a CD, a record will provide information to a higher frequency. Over about 30kHz, I should imagine that no source provides any useful information.

There will, however, be signal present at these extremely high frequencies, from a CD there will be digital noise and from a record the higher frequency components of tracing distortion, maybe surface noise too.

It's no problem though, you can't hear it anyway. Or is it a problem?

For a well designed pre-amplifier, huge bandwidths are no problem, but for a power amplifier they could be and usually are.

Another "hint and tip" is do not provide too much negative feedback over a number of amplifying stages. Modern practice (leaving aside no-feedback designs) is to arrange each stage to be inherently linear with perhaps a little local feedback and to provide only minimum of overall feedback to tidy things up.

Lets look at how the overall feedback loop works. In comes a signal, goes through the stages reaches the output and gets fed-back to the input in such a way that a signal is passed through the stages again which corrects any errors. After a very short time, the output is an exact replica of the input but with the gain determined (substantially) by the feedback loop.

With slow moving signals the time taken for the feedback to settle is insignificant. For fast moving signals the time-delay becomes significant. Say hello to transient intermodulation distortion.

So what do we do. Well, this is what I do, and as I say it's a fairly standard approach. I design the power amplifier to have a massive power-bandwidth, up to 1MHz. I like to see a good 100 kHz square wave generated across a difficult load at power. For such a power amplifier, audio signals are hardly moving at all, it can eat them for breakfast, burp and be hungry again.

Now I need to make sure that this power amplifier never sees anything it can't handle, so I precede it with a low pass filter rolling off at 60kHz. Such a filter won't have too much effect upon the useful audio signals, up to, say, 30kHz, and ensures that the amplifier never runs into trouble trying to cope with fast moving signals.

Anyway, that's why I precede power amplifiers with low-pass filters while at the same time striving for huge bandwidths elsewhere.

As Naim use a low pass filter before their power amplifiers, I imagine that they have the same approach.

I hope that was of some interest.

Regards,

Paul.

quote:

---

Originally posted by paul99:
Colleagues,

I know that electronics engineers are generally hated and ridiculed on this site (in my case, I'm an ex-engineer but not immune I imagine) but it's a slow day in the office and I've got a few minutes for a reply.

Limiting the bandwidth of the signal fed into a power amplifier is not an unusual approach. It is pretty much a standard entry in any "hints and tips" list.

Here's why you do it, or at least here's why I do it:

Most of the musical information from any source will peter out at quite a low frequency, maybe before 10kHz. The next octave or so contains harmonics and all the high frequency information, which, while saying little about the music says a lot about how the instruments sound. Over 20kHz there is no useful information, of either variety, from a CD, a record will provide information to a higher frequency. Over about 30kHz, I should imagine that no source provides any useful information.

There will, however, be signal present at these extremely high frequencies, from a CD there will be digital noise and from a record the higher frequency components of tracing distortion, maybe surface noise too.

It's no problem though, you can't hear it anyway. Or is it a problem?

For a well designed pre-amplifier, huge bandwidths are no problem, but for a power amplifier they could be and usually are.

Another "hint and tip" is do not provide too much negative feedback over a number of amplifying stages. Modern practice (leaving aside no-feedback designs) is to arrange each stage to be inherently linear with perhaps a little local feedback and to provide only minimum of overall feedback to tidy things up.

Lets look at how the overall feedback loop works. In comes a signal, goes through the stages reaches the output and gets fed-back to the input in such a way that a signal is passed through the stages again which corrects any errors. After a very short time, the output is an exact replica of the input but with the gain determined (substantially) by the feedback loop.

With slow moving signals the time taken for the feedback to settle is insignificant. For fast moving signals the time-delay becomes significant. Say hello to transient intermodulation distortion.

So what do we do. Well, this is what I do, and as I say it's a fairly standard approach. I design the power amplifier to have a massive power-bandwidth, up to 1MHz. I like to see a good 100 kHz square wave generated across a difficult load at power. For such a power amplifier, audio signals are hardly moving at all, it can eat them for breakfast, burp and be hungry again.

Now I need to make sure that this power amplifier never sees anything it can't handle, so I precede it with a low pass filter rolling off at 60kHz. Such a filter won't have too much effect upon the useful audio signals, up to, say, 30kHz, and ensures that the amplifier never runs into trouble trying to cope with fast moving signals.

Anyway, that's why I precede power amplifiers with low-pass filters while at the same time striving for huge bandwidths elsewhere.

As Naim use a low pass filter before their power amplifiers, I imagine that they have the same approach.

I hope that was of some interest.

Regards,

Paul.

---

Hy Paul!whats about naim preamps?
are they bandwith limited as paul ranson wrote
before?if this is true naims preamps would be
far away from high bandwith,and as i understood
your technical approach is to limit the entrance of the
poweramp and not the output bandwith of preamp.
thinking about an preamp/poweramp combi this seemes to me not the same.
O.J.

O.J.

It' the same thing. A low pass filter between the pre-amp and the power-amp could be considered to be at the output of the pre-amp or the input of the power-amp.

I just wanted to explain the principle in general terms. Which box you put the filter in is not so important as far as the general principle is concerned. What Naim do in detail I don't know.

Regards,

Paul.