I didn't understand exactly what the 'speakers were. The earlier post
says ILP TL4, but I can't find any references for these. I know of ILP
as amplifier module manufacturers, but not for 'speakers. I also haven't
seen anything from Arnie casting doubt on the metal faculties of the
builder. Possibly my News server misses the odd post.

S.

My typo -------- the speakers are IPL Acoustics STL4's
Re Arny K, his suggestion that I'd screwed the design up and that National
had idiot proofed the new LM4562's gave me the impression he thought I was
the idiot, maybe I read too much into his comments. I still say though,
these sound much better than the 5532's.

Quote
" 2. a few yrs back I got samples from Analog Devices of a
low noise audio switch and a dc controlled vol chip (
0.002% thd ) I knocked them together with a couple of
5532's and have been enjoying them since ( from tv/dvd or
Technics CD into 4 x 75wrms mosfet amps and onto a pair
of ILP TL4's, ground/door and window shaking stuff and
very clear, or so I thought..... last week I got hold of
samples of National's LM4562 and replaced the 5532's, the
attack on drums is now much better and everything sounds
tighter.

Anybody who thinks that a LM4562 sounds better than a NE5532, both in
well-designed circuits, is suffering from specificaion-induced
constructor's ear. They are both free of audible distortion unless someone
screws up the design.

As it was chucked together I used 7815 & 7915
regs, I had a trouble with rf on the -ve rail which could
be heard as a faint hiss, so I thought whilst it's in
bits I'll go over the power supply , no need! it's now
silent, better rejection ?

More than likely you screwed up the design by not properly bypassing the
5532 power pins in the origional design. Conventional wisdom is that a
suitable cap (classic 0.05 uF ceramic for example) between the power supply
pins will settle them down.

More than likely National idiot-proofed the 4562s."

Pete Cross wrote:
I didn't understand exactly what the 'speakers were. The earlier post
says ILP TL4, but I can't find any references for these. I know of ILP
as amplifier module manufacturers, but not for 'speakers. I also haven't
seen anything from Arnie casting doubt on the metal faculties of the
builder. Possibly my News server misses the odd post.

S.

My typo -------- the speakers are IPL Acoustics STL4's
Re Arny K, his suggestion that I'd screwed the design up and that National
had idiot proofed the new LM4562's gave me the impression he thought I was
the idiot, maybe I read too much into his comments. I still say though,
these sound much better than the 5532's.

Quote
" 2. a few yrs back I got samples from Analog Devices of a
low noise audio switch and a dc controlled vol chip (
0.002% thd ) I knocked them together with a couple of
5532's and have been enjoying them since ( from tv/dvd or
Technics CD into 4 x 75wrms mosfet amps and onto a pair
of ILP TL4's, ground/door and window shaking stuff and
very clear, or so I thought..... last week I got hold of
samples of National's LM4562 and replaced the 5532's, the
attack on drums is now much better and everything sounds
tighter.

Anybody who thinks that a LM4562 sounds better than a NE5532, both in
well-designed circuits, is suffering from specificaion-induced
constructor's ear. They are both free of audible distortion unless someone
screws up the design.


I've looked up the STL4s, I'm rather a fan of transmission line
'speakers. Generally, they need plenty of power, but I should think you
would have good reason to be happy with them.

As to the sound of different chips, I'm rather with Arnie on this one.
If the design is reasonably sound, and you get low distortion and noise,
then I don't think there would be any audible difference.

There are well accepted thresholds for the minimum noise, distortion and
frequency response variations which are audible. Once two amplifiers are
each below the threshold, whatever differences there are between them
will not be audible. For example, one amplifier having 0.05% distortion
and another having 0.01% won't sound any different, as it is generally
accepted that the threshold for distortion on programme material is at
least 0.1%. Analogue tape machines' output were rated as standard at the
3% THD level. Typical values were 1% at 0dB and 2-3% at 8dB above 0dB
which was considered peak level. At the time, few people complained that
the tape machine's distortion was ruining the music, and even today,
some artists like the analogue tape sound.

My two turntables both had 741s in the RIAA equaliser, which I replaced
with TL071s, only because I couldn't stand the idea of 741s. The TL071
is a direct plug-in replacement and the equalisers have their ICs on
sockets, so the replacement was easy. I didn't even have to solder
anything. I suppose the distortion of the TL071-equipped equaliser could
be lower than before (I didn't measure a before and after) but any
reduction would be swamped by the inherent distortion of vinyl and the
cartridge. I certainly haven't noticed any improvement, I just feel
better not having 741s in the signal path.

S.

As to the sound of different chips, I'm rather with Arnie on this one.
If the design is reasonably sound, and you get low distortion and noise,
then I don't think there would be any audible difference.

The LM4562's boast 20v/uS slew rate, the 5532's a measly 6v/uS ,
google for lm4652 and there are a few people saying same as me, there's more
detail and bass drums stand out amongst bass lines, I'm convinced they
didn't before, I guess I should put the 5532's back in but that's too much
messing and I have other fish to fry.........

I too would avoid 741's and LF351's too that seem to keep appearing...

Pete

Serge Auckland wrote:

My two turntables both had 741s in the RIAA equaliser, which I replaced
with TL071s, only because I couldn't stand the idea of 741s. The TL071
is a direct plug-in replacement and the equalisers have their ICs on
sockets, so the replacement was easy. I didn't even have to solder
anything. I suppose the distortion of the TL071-equipped equaliser could
be lower than before (I didn't measure a before and after) but any
reduction would be swamped by the inherent distortion of vinyl and the
cartridge. I certainly haven't noticed any improvement, I just feel
better not having 741s in the signal path.

Were the 741's used for high-level equalisation or as the input-devices?
My phono stage has an LM301, but only as a buffer after the long-tailed
pair and current mirror, and still sounds good. I don't see any reason
to replace it.

--
Eiron.

In article , Serge Auckland
wrote:


My two turntables both had 741s in the RIAA equaliser, which I replaced
with TL071s, only because I couldn't stand the idea of 741s. The TL071
is a direct plug-in replacement and the equalisers have their ICs on
sockets, so the replacement was easy. I didn't even have to solder
anything. I suppose the distortion of the TL071-equipped equaliser could
be lower than before (I didn't measure a before and after) but any
reduction would be swamped by the inherent distortion of vinyl and the
cartridge. I certainly haven't noticed any improvement, I just feel
better not having 741s in the signal path.

When I used to do such things I ended up designing a pre-amp using
HA12017's for all the gain stages. This despite the IC being intended by
Hitachi as an RIAA preamp for MM cartridge. I couldn't hear any real
differences between this and other ICs I tried - nor could anyone else I
tried the comparisons on at the time. However the HA12017 gave slightly
better results in terms of various specs, so coped with a wider range of
signals and output loads.

Took a little fiddling with the compensation to make it work, but turned
out OK. So far as I know, no-one else used the HA12017's like this, and I
guess they are long obsolete nowdays! :-)

FWIW When I tried using stabliser ICs like the 78/9XX familiy I also found
they tended to oscillate at a few MHz, so built PSU regulators with
discrete devices. So far as I recall, there were no audible effects I or
others noticed due to the oscillations. I had RF decoupled the power lines
on each individual IC, though. Just seemed a good idea to make a better
PSU.

Oh... and I don't really have a 'Desert Island Discs' list as I keep
changing my mind about what I am enjoying. Most recent has been the Rubbra
Symphonies (Hickock BBCNoW on Chandos) but before that it was a collection
of Benny Goodman Sextet on a Vocalion/Dutton CD. Before that I think it was
a recording of some Alain organ music from R3 performed by his sister in
1981 and 1986. Before that Pentangle... Special favourites from a while ago
include the Uchida sets of Schubert and Mozart sonatas, and some of the
re-issues of Duke Ellington with extra tracks.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

Eiron wrote:
Serge Auckland wrote:

My two turntables both had 741s in the RIAA equaliser, which I
replaced with TL071s, only because I couldn't stand the idea of 741s.
The TL071 is a direct plug-in replacement and the equalisers have
their ICs on sockets, so the replacement was easy. I didn't even have
to solder anything. I suppose the distortion of the TL071-equipped
equaliser could be lower than before (I didn't measure a before and
after) but any reduction would be swamped by the inherent distortion
of vinyl and the cartridge. I certainly haven't noticed any
improvement, I just feel better not having 741s in the signal path.

Were the 741's used for high-level equalisation or as the input-devices?
My phono stage has an LM301, but only as a buffer after the long-tailed
pair and current mirror, and still sounds good. I don't see any reason
to replace it.

Both. In the EMT 948, the Equaliser has three stages, all with 741s. The
first stage is the equaliser proper, stage 2 is the active rumble filter
and stage 3 is an active low-pass filter at 25kHz. The signal is then
passed to a balanced line driver with a 741 input buffer, an LM301
driver and gin adjust stage and a push-pull output driver with a
balanced output transformer.

In the AEG, the first gain and equalisation stage is a TDA2310, then
followed by two 741s as an active rumble filter with a further 741 as a
gain adjustment stage. The signal then passes to a line driver stage
with a 741 as input buffer and LM378 as output driver again driving a
balancing transformer.

In both cases I have by-passed the rumble filter.

S.

"Pete Cross" wrote in message

As to the sound of different chips, I'm rather with
Arnie on this one. If the design is reasonably sound,
and you get low distortion and noise, then I don't think
there would be any audible difference.

The LM4562's boast 20v/uS slew rate, the 5532's a measly
6v/uS ,

So what? As long as the op amp's slew rate is far less than the slew rate of
the audio signal, no problems.

I'll leave it as an exercise for you to calculate the slew rate of a 20 KHz
wave at whatever voltage you think there is in your project's circuit.

google for lm4652 and there are a few people saying same
as me, there's more detail and bass drums stand out
amongst bass lines,

There are people who will say anything. Apparently there are people who
thought that Anna Nicole Smith was beautiful!

I'm convinced they didn't before, I
guess I should put the 5532's back in but that's too much
messing and I have other fish to fry....

That's how everybody who has avoided the obvious pitfalls of sighted
evaluations thinks - before they rip out perfectly good 5532s!

I too would avoid 741's and LF351's too that seem to keep
appearing...

It takes a lot of ignorance to compare 741s to LF 351s. There's about a
13:1 difference in slew rate, and a 741 is so slow that it could conceivably
get into trouble.

You really need to do some good double blind tests.

"Pete Cross" wrote in message


My typo -------- the speakers are IPL Acoustics STL4's
Re Arny K, his suggestion that I'd screwed the design up
and that National had idiot proofed the new LM4562's gave
me the impression he thought I was the idiot, maybe I
read too much into his comments.

You did read too much into my comments.

"idiot proofing" is one of those things that benefits all of us because we
all mistakes.

I still say though,
these sound much better than the 5532's.

If it gets you through the day...

Trust me, better ears than yours have done far more carefully controlled
listening tests, and come up empty.

Beware of constructor's ear.

In article , Pete Cross wrote:
As to the sound of different chips, I'm rather with Arnie on this one.
If the design is reasonably sound, and you get low distortion and
noise, then I don't think there would be any audible difference.

The LM4562's boast 20v/uS slew rate, the 5532's a measly 6v/uS ,

Erm.. I'm not sure what kind of music you listen to, or what speakers you
are using, but if I've estimated it correctly, 6v/microsec translates into
a power bandwidth of the order of 95kHz for +/-10V waveforms. :-)

google for lm4652 and there are a few people saying same as me, there's
more detail and bass drums stand out amongst bass lines,

How many recordings of bass drums give transients approaching
6v/microsecond with peak levels of less than 10V?

I'm afraid that if you read magazines, forums, etc you will see all kinds
of 'opinions' given - often contradictory.

FWIW rather than simply comparing raw 'slew rate' values I think it would
be better to take note of the load used (as this will have an effect on
the slew rate delivered) and the level of distortion for an audible HF
test signal of a kind which could be expected to arise in real use.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

ok, I'm biased.......
there must be something though.......
http://www.diyaudio.com/forums/showt...threadid=86276


or maybe my ears are shot as I spent 8yrs working under Hugh Banton ( VDGG )

Pete