E_Notes :  (dB, SPL & SWL)

                                 A collection of Information relevant to Audio Engineering.
                                                         Compiled by Ken Leyva,  

Sometimes a simple question can turn into a test of will to find a meaningful answer.  Especially when the answers
you find only draw you deeper into the unknown.  

Each field of electronics place a different value on the properties of electric charges and what they can do  with them.  

The efficiency of the exchange of information within a group is increased by assuming some, in common,  broad baseline
knowledge base, (AC/DC, Ohm’s Law),  plus modified or enhanced baseline knowledge relating to their field, (Audio, Digital, Optical).

Within a field,  Icons can be used as pointers to their  common knowledge base.  

The exact technical boundary of any named field of electronics is not always clear to the electrons.    

Basically we’re logical people, ( it happens ),  electrons are all electronic,  and we basically all organize our knowledge bases along
similar lines.  

Same alphabet,  same language, different dialects.  Hopefully the following information will help those wanting to know what makes
the Audio Dialect different from theirs.

This page is an index into sections of 3 documents, (SPL.doc, SWL.doc, Decibel.doc).  Each doc consists of text
boxes containing text, tables and graphics.  I cut & pasted most of the information from Wikipedia.com,  per their
copyright.   The URL of the source page is at the top of each document.

These documents contain Hyper Text links to the Wikipedia.com site, “SPL.doc”,62 links and “SWL.doc”, 23 links,  
“Decibel.doc”, not counted.  Additionally there are External Links to URLs outside of Wikipedia.com that relate to the
document subject.  

The “Decibel.doc” has incorrect URL paths that point local.    You will find this note at the top of the documents ::
         If you find a bad Hyper Text Link in this doc, try adding,  (“http://” &)  “en.wikipedia.org” ,   to the link.


                                       SPL.doc
Sections :                        

    Sound pressure
    Measuring sound pressure levels
    Sound pressure p in N/m2 or Pa is:
    Sound pressure p:
    Sound Pressure Level
    Distance Law

    Table :
    Examples of sound pressure and sound pressure levels

    SPL in audio equipment :


    LINKS :
        External Links
        Converters
        Calculators


                                                   SWL.doc

Sections :

    Sound power level or acoustic power level

    Table:
    Sound power level and sound power of some sound sources
    LINKS :
       External Links
        Calculators
        Converters



                  dB? : dB? : Decibel.doc dB? : dB?
Sections :
    Decibel
    History of bels and decibels
    Definition
    Examples
    Electrical circuits :
    Standards
    Merits
    Difficulties



    USES :
    Acoustics
                Note that the SPL emitted by an object changes
                                       
                          Relation to Loud Speakers
           A practical example

    Frequency weighting
           In water

    Electronics
    Optics
    Telecommunications
    Seismology

    Typical abbreviations

    Note regarding absolute measurements

    Relative measurements

    Reckoning
                       Round numbers
          The 4 → 6 energy rule
          The "789" rule
          −3 dB ≈ ½ power
          6 dB per bit

    dB charts :
                       Commonly used dB values
          Other dB values

    LINKS :
                       External Links
          Converters
xit: Exit Sub
'--
xit: Exit Sub
'--
Hi,
This is my first attempt at a web site.  So things,  
will no doubt be, dynamic for a while. Change is
the only constant.
admin@testvector.com
xit: Exit Sub
'--
xit: Exit Sub
'--
xit: Exit Sub
'--
E_Notes_01
Graphical Notes as a "*.wmf"

This link should lead to a WMF of some audio related information
concerning  Ohms Law , SPL, RMS that I collected in AutoCAD.

About 1.5 Meg file size.
Background :

Math Stuff.
dB as voltage type ratio.
I was looking on the Web for some info on Bias Traps or Record Bias Traps.  

I was not concerned with tape recording only on information relating on the effect of HF signals on audio
transistor amplifiers.  I found information that was at best confusing at some points.  There were a lot of
questions about adjusting bias traps, where, why.

I spent about a dozen years testing prototype tape machine designs and seem to remember a few things this
way.

Relating to Bias Traps :


1) The most pronounced effect of a mistuned bias trap is that the machine will not record.
2) Less significant is over or under expected recorded levels, distortion.
3) Then mistracking of noise reduction systems.

The bias oscillator for good quality consumer machines range from about 100 to 200 kHz or about 10 *
maximum audio band pass.  Varying from maybe 30v to 100v.  A shock feels like a burning sensation, not a
buzz like from the wall.

Depending on a 2 or 3 head machine,  there will be a bias trap for each channel between the bias oscillator
output / record head connection and the record amplifier output, and if 3 heads also between the play head
and playback amplifier.  Two head machines do not have record bias present during normal playback.

The cause of 1&2 above is that the high frequency bias signal, if it gets to the record amplifier transistors, will
get rectified into a DC that will change the DC bias of the transistors.  I have seen this cause a record
amplifier to completely shut off.  The record amplifier will work fine with the bias oscillator turned off.  With the
bias oscillator on, it is difficult to see the audio without a sharp multi pole filter due to the magnitude of the
record bias signal.  If the change in transistor bias doesen’t shut the amplifiers off,  it can change the bias
operating point away from the intended design values and by definition that is distortion.
.
Item 3 fakes out complementary noise reduction encoders and decoders. The processors see the bias signal
as a low level high frequency audio component and adapts. During playback there is no record bias present
and the NR decoder mistracks,  sounds bad. A similar situation occurs when recording from FM Stereo where
the 19k stereo pilot tone causes noise reduction to mistrack.

Adjusting the Bias Traps :
The easiest way to adjust a bias trap,  is with a scope and a non ferric tuning wand.

Put the machine  in record.  Look for the bias signal at the output of the record amplifier, (or input to the
playback amplifier, 3-head), adjust the trap for minimum bias signal.

You want as little as possible of the high frequency bias signal getting into the record or playback
circuitry.

The same would apply with a dynamic bias servo like HX or HX Pro.  

The exact frequency of the bias oscillator is not much of a concern,  the trap should cover the proper
range.  

When to adjust.
It is a good idea to check the tuning of the traps if oscillator components are moved or changed.  It is
also a good idea to let the oscillator get up to
normal operating temperature before adjusting the traps
because that would affect the bias current for the tape itself.

The high frequency saturation point for common cassette tape Fe, FeCr, FeCo, Me at 4.75 cm/s are such
that there is no significant recorded bias signal to be filtered out during the play back only reproduction
process.

The
most critical thing for tape recording is, “tape head contact”.

Cassette Tapes :

Cassette tape @ 4.75 cm/s, 1-7/8 in/s, record EQ corner frequencies are, using 1/(2 * Pi * TC) :

        
3180us, about 50 Hz for the low frequency corner.
        
120us, about 1,326z and 70us, about 2,273.6 Hz for the high frequency corners.
        (only one at a time, please).

 
High recording bias is usually associated with 70us of EQ but not necessarily so.

Fe = Ferric or plain iron tape.  Usually  recorded at 120us.

FeCr = Ferrichrome has a chrome layer above a plain Fe layer that holds high frequency audio
 better.  There is a bit of cross over distortion at the frequency where the upper Chrome or lower
 Ferric layers become dominant in holding the audio.  Typically recorded at 70us.

FeCo = Ferric Cobalt these are the Maxell Type II's and TDK SA's and normally recorded with 70us of
 EQ.

Me = Metal tapes.  Almost imposable to erase low frequencies from, like bass guitar and drums.  I
 never found Me tape to be very interesting.  Typically recorded with 70us of EQ.

My personal favorite cassette tape  at 4.75 cm/s or 9.5 cm/s was
TDK AD.  AD-223 for develpoment.
 See :
http://www.btgallery.com/tdk/ad.html.  It had as much headroom as a high bias tape
 and was quieter on the CCIR/ARM 2k curve than any other tape I can remember at this point.

CCIR/ARM filter :
Found @ http://www.sweetwater.com/expert-center/glossary/t--CCIRARM-weighting.  

In part :  CCIR/ARM-weighting or CCIR 2 kHz-weighting — A type of weighting filter for sound level
measurements. This curve derives from the CCIR 468-curve we’ve covered in WFTD previously. Dolby
Laboratories proposed using an average-response meter with the CCIR 468-curve instead of the costly true
quasi-peak meters used by the Europeans in specifying their equipment. They further proposed shifting the
0 dB reference point from 1 kHz to 2 kHz (in essence, sliding the curve down 6 dB). This became known as
the CCIR ARM (average response meter), as well as the CCIR 2 kHz-weighting curve. (See: R. Dolby, D.
Robinson, and K. Gundry, "A Practical Noise Measurement Method," J. Audio Eng. Soc., Vol 27, No. 3, 1979)

As I remember :
The CCIR 468 weighting curve was originally developed in France under the CCIT.  It was developed for
measuring the annoyance of rotary dial phone clicks, they being a form of impulse noise,  very different from
recording tape noise.  Ray thought the CCIR curve worked better with his A and B type NRs.  But in order to
make the subjective change in noise level resulting from his NR systems more favorably compare with the
test equipment,  he moved the unity gain point from 1kHz to 2kHz.  Usually referred to as a CCIR-ARM 2k.  It
was the basis for the Dolby Labs Cat.98 noise filter,  and adopted as the company standard noise filter.













On occasion I would receive requests from manufacturers and the general public for a schematic of a CCIR-
ARM 2k filter.  I drew up the schematic above to send out to the public and OEMs.




















Ray Dolby :
See Wikipedia @ http://en.wikipedia.org/wiki/Ray_Dolby
A blurburization.
Differential Inputs :
In my words and a few borrowed from Americanise  :

Mode –
Condition of being.

Common –
The lack of demarcation,  lack of distinguishable characteristics.
No definitions.
Where A = B = C & C = A.

Difference –
The demarcation where something becomes somethings.
Where somethings are defined to be so.  
At where, A is not like B.

Differential –
Influenced by a difference.

Input –
Point of a circuit that responds desirably to the difference between that node and a reference node.

Differential Input –
Input node, (+),  of a circuit that responds desirably to the difference between that node and an additional
input node referred to as (-).

This input has high sensitivity to the difference in unit values of two input nodes.
It is less sensitive to common unit values at the same two input nodes.

With multi wire conductors that are at the same impedance,  the susceptibility of each conductor to external
noise influences is basically the same for a given physical location.  This means that each conductor can
carry similar noise characteristics.  This would be common mode noise.  This type of signal is typically
rejected by a differential input.  

If two conductors are fed a bi-phased signal of equal unit magnitudes.  The differential input will respond with
an analog of one phase of the input signal.

This assumes that the difference measured between the inputs is the intended difference.  And that the
common mode signal if any is the result of a common signal path.  

The easiest way to assure a common path is to twist the wires.  On a PCB this means that the two
conductors must follow the exact same path so that they pick up the same amount of external noise so it can
be removed.  

If the two input nodes receive the same signal but from a different path some of the common mode signal
energy becomes differential and passes right  through the input section as it should.    This characteristic
would be more evident at relatively higher frequencies than what the circuit is expected to deal with.   A
difference in reactive loading of either conductor can produce the same effect as traveling a different path,
because it would be.
xit: Exit Sub
'--
Listed below are some notes I have collected including :

SPL                  Sound Pressure Level
SWL                 Sound Power Level
dB                    Decibel 1/10 * Bell.
Diff. Inputs      Differential Inputs
Bias Traps      Tape recorder bias traps
E_Notes_02
Graphical Notes as a "*.wmf"  
This link should lead to a WMF of some audio related information
concerning  Ohms Law , SPL, RMS that I collected in AutoCAD.

About 2.8 Meg file size.
E_Notes