Audio, hi-fi and car audio - Intelligence and RIAA

Andre Jute wrote:
Don Pearce wrote:
On Tue, 15 May 2007 17:31:24 -0500, John Byrns
wrote:

In article ,
(Don Pearce) wrote:

John, are you still insisting that RIAA playback requires high
frequency boost? It doesn't. An RIAA phono preamp has a feedback
mechanism that provides high frequency cut. I have designed several
myself, and studied the circuits and operation of many. Had I (and
every other designer on the planet) been getting it wrong all the
time, our systems would be muffled and entirely without top. They are
not; they play back just fine, and certainly for my own, when I play a
white noise track on a test disc (recorded with standard pre-emphasis
before you say anything), I recover noise which is flat within about
1dB from 30Hz to 20kHz.

*Please* go and do some reading so you can back away gracefully from
this ridiculous position you are placing yourself in.

Don, yes I am still insisting that RIAA playback requires high frequency
boost. Why are you suggesting that I might want to back away from this
position?

Let me attempt to explain, I'm going to assume that you have some
knowledge of math and know what differentiation is. Let's consider an
LP recording which has had a music signal cut into it. Now in our
playback system we need to read the amplitude of the signal cut into the
disc and convert it into an electrical signal of varying amplitude to
drive our speaker system, while along the way undoing any amplitude
equalization that was incorporated when the music signal was originally
cut into the disc using the RIAA record equalization. Now you are
insisting that RIAA playback equalization involves a large high
frequency cut approximating some 38 dB, while I claim that RIAA playback
equalization involves the boosting of the amplitude of the high
frequency signals cut into the disc by approximately 12 dB. What
accounts for the difference in our perspectives? The difference is
simply explained by the fact that you are lumping two separate
equalization curves together while I am talking about only the
equalization necessary to counter the RIAA amplitude equalization
applied when the music was cut into the grooves of the record.

You are assuming that the LP is being played with a "magnetic" pickup.
It is a characteristic of "magnetic" pickups that they differentiate the
amplitude of the music signal cut into the record groove to produce the
electrical output. The differentiation of the recorded amplitude causes
the signal output of the "magnetic" pickup to be tilted upwards towards
the high frequencies at a rate of 6 dB per octave, which results in a
very tinny sound unless this effect is compensated for. To restore the
output of the "magnetic" pickup back to a flat representation of the
recorded amplitude on the disc, we must pass its output through an
integrator circuit. An integrator produces a response which falls
towards the high frequencies at a rate of 6 dB per octave, falling
approximately 50 dB at 15 kHz vs. 50 Hz, this is the first part of your
equalizer. The second part of your equalizer is the same as my RIAA
amplitude equalizer and consists of shelving the high frequencies up by
approximately 12 dB using the time constants of 318.3 usec. and 75
usec.. When you combine the "magnetic" pickup equalizer and the RIAA
amplitude equalizer into a single composite circuit you have what you
call "RIAA equalization". This equalization is the sum of a 50 dB high
frequency cut for "magnetic" pickup compensation and a high frequency
boost of 12 dB for RIAA amplitude equalization, giving a net high
frequency cut of 38 dB for the combined network.

Using a pickup that is directly responsive to the recorded groove
amplitude, like say an FM pickup, or a strain gauge pickup, eliminates
the need for the pickup compensation integrator required with a
"magnetic" pickup, and leaves us with the need to provide only the 12 dB
high frequency boost required by the RIAA cutting curve.

Get it, it's simple once you understand it, the "RIAA phono preamp" you
are describing is really doing two equalization jobs, pickup
compensation and compensation for the RIAA amplitude response.

Regards,

John Byrns
John, I stopped reading "let me explain", I'm afraid. Don't take this
badly, please. I did that because I knew that whatever followed was
going to be a catalogue of misunderstanding and error. It isn't too
important really what those errors are. What is important is that they
are errors, which thirty seconds of research (google for phono preamp
sche,matic - that should do it) will show you. You will then be in the
enviable position of knowing something that you have been getting
completely wrong for years, and being able to learn something new.

Please make this small effort before you post again. I promise you
won't find it wasted. And do listen and understand when I tell you
that those of us who have designed audio gear have never, ever
designed an RIAA preamp that boosts rather than reduces high
frequencies, Do some web research and see if you can find one.

d

--
Pearce Consulting
http://www.pearce.uk.com

Don, baby:

You amused me with your barns and rods (should they be roods?) in this
thread where Sander fed his slug Amstel, made me wonder if I shouldn't
give you another chance, if I hadn't misjudged you as just another
humourless purveyor of excessive negative feedback who should be
kicked on sight. So, just in case you really aren't an enemy of
fidelity, I'm going to give you a tip and hope St Peter is watching
and inscribing my incredible generosity in the Big Book Before the
Pearly Gates.

Save yourself a lot of grinding frustration and anger and either:
a) do not argue with John Byrns on this, meaning drop out now, don't
even tell him to look it up
or
b) accept that what you think you know has some pinholes in it to
which John has already taken a reamer and, before this is over, will
take a bloody great big angle grinder, and therefore go look it up
yourself with your prejudices (what you might prefer to call your
education and knowledge) put firmly aside in a locked box

I've seen John grind down the graduate engineers before, politely,
persistently. He never hesitates to apologize when he is wrong, and he
will always give your argument full consideration and your goodwill
the benefit of the doubt, but I have never seen him fail to understand
the warp and weft of something thoroughly before he starts. You might
note that Chris Hornbeck, a guy who sees through bull**** and
encrustations of hallowed practice to the true fundmentals beneath,
has decided that John is right, giving you the key to why John is
right: "differences between amplitude and velocity, and *why* they're
historically treated differently in cutter-head amplifiers". (Thanks,
Chris. I was struggling with whether that is it or whether it is more
complicated.). Or, in pure self-protection, Don me old gabbas, you
might look up some old RAT threads in which John (ever so politely!)
wiped the floor with that toe-rag Pasternack, admittedly a dullard,
but a dullard who claims to have a Stanford MSEE and observably has a
glib way with the math that often borders on deceit about professional
matters, and sometimes deliberately steps over that limit, after which
Pasternack usually claims that John drove him to betraying his
profession or, even more laughably, "I did it in my zeal to flame
Andre". See above for either of two simple acts you may perform to
save yourself from landing up in the same position as Plodnick vis a
vis John.

There, my duty is done. My money is on Mr Byrns to find all the tees
that aren't crossed and all the eyes that aren't dotted, and to slot
them into a Teflon-covered, Kevlar-armoured argument.

Thanks again for the chuckle.

Andre Jute
The trouble with most people is not what they don't know, but what
they know for certain that isn't true. --- Mark Twain

There are more things in Heaven and Earth, Horatio, than in thy
fondest dreams. --- Will the Shake

I've read and re-read John Byrne's arguments and still think he's wrong.
Every RIAA amplifier I've ever designed and every one I've measured has
a voltage amplitude response that boosts the bass end and cuts the
treble end. The RIAA curve calls for a 19.36dB boost at 20Hz, and a
19.95dB cut at 21kHz. Both are relative to 1kHz. The IEC curve is
identical to the RIAA curve with the exception of the extreme low end
which is boosted less on replay to act as a built-in rumble filter.

No curve I've ever seen has a 12dB boost to the treble.

If John is so precise, I can't understand for the life of me what curve
he is referring to. You only have to put a generator to any RIAA input
stage to see that the curve is as above, with 19 odd dB boost at the
bess end and almost 20 dB cut at the top.

S.

--
http://audiopages.googlepages.com