Audio Levels

[poulkirk]

This thread is started with the objective of clarifying what Audio Levels and Audio Metering is about in simple terms

In the Broadcast and Professional Recording industry, three types of Audio Level Meters are in common usage

The IEC Type I PPM Peak Programme Meter, The IEC Type II PPM and the VU Meter

The IEC Type I PPM comes with two different scales, the IEC Type I PPM Scale, and the Nordic Scale, both calibrated in decibells

The IEC Type II PPM also comes with two different scales, the Type IIa BBC Scale, calibrated 1 to 7, and the Type IIb calibrated in decibells

The VU Meter is calibrated in decibells from -20 to +3 decibells

There are a number of other Audio Level Meters fitted to various pieces of equipment, some are OK, most indicate, Who Knows what ?

It is not proposed here to go into the merits of the three basic Professional Audio Level Meters, except they all have one thing in common

That is they all have an agreed on Alignment Level

This varies from country to country

In Australia, it was +8 dBu, or 2 Volts, actually 1.9457 Volts RMS

It is now, in common with USA, and many other countries where VU Meters are in commom use, +4 dBu or 1.2277 Volts

In France, it is +6 dBu, or 1.5455 Volts

In many other countries, particularly Europe, PPM's are in common use with the Alignment Level 0 dBu, or 0.7746 Volts

For many years, the common Hi-Fi Alignment level was -8 dBu, or 0.3084 Volts, or say 300 millivolts

This is still the common Alignment Level for Hi-Fi Equipment using Unbalanced RCA type interconnects

For Audio Equipment to operate and give satisfactory sound quality, a certain amount of Headroom above Alignment Level is necessary

Traditionally, Professional Equipment was expected to have a headroom of 18 to 20 dB to accomodate normal Audio peaks on Voice and Music, and is still the expected norm

For Domestic Hi-Fi Equipment, this headroom was somewhat lower, and in the order of around 10 dB for Analog Equipment with clipping occouring at around 1V RMS

Specifications for CD's are allowing for 15 dB of Headroom above Alignment Level

Headroom means Maximum Level or Full Scale Modulation, and is the absolute maximum level possible to put on a CD, and for most CD Players quoted as 2 Volts RMS

15 dB below 2V RMS is fairly close to the original 300 millivolts, and for all practical purposes the same level used for years on Analog Hi-Fi Equipment

CD's would sound better if these standard levels, set many years ago by Sony and Philips were observed and respected by Record Producers

300 millivolts is the standard Alignment Level, 2 Volts is the absolute maximum possible level on peaks from Standard Unbalanced Hi-Fi Equipment

[mrfeedback]

Most modern music is produced using Digital Audio Workstation software.
All of this software that I have seen incorporates a 'normalise' function where an audio passage, track or complete album project can be automatically scaled such that peaks reach a nominated peak level.
With modern compressed and peak limited music with very limited dynamic range, this causes the average level to become rather higher than the CD alignment level given above.
Perhaps producers should be educated to 'normalise' to a level lower than 0dBfs as is the usual case.
DAW software usually has 0dBfs as the default normalise setting, and perhaps at least partly because of this we have the current situation.
In the race to make the next record 'louder' than the last, peaking to full scale has become standard.
So long as all gear downstream from the CDP is capable of handling 2V peak signals there should be no problems and maximum SNR is attained through the system.
Surprisingly however not all modern equipment does handle this peak level and this could be a contributing factor to 'bad' sound, and not necesarily the fault of the CD medium or CD recordings.

Eric.

[dhaen]

Can of worms:

In the past, have been able to agree levels with all countries except Oz

Cheers,

[dhaen]

Hi,

That is they all have an agreed on Alignment Level

This varies from country to country

In Australia, it was +8 dBu, or 2 Volts, actually 1.9457 Volts RMS

That means nothing at all - sorry.

The only things that mean anything, are the flux density (for analogue recordings) and the number of dB below max (for digital recordings).

Volts could mean anything:
for enviroment, it depends on the local standard: -8, 0, +4, +8
For equipment, it depends on where measured.

After all, the subject IS levels. And the starting point is reference level.

Cheers,

[poulkirk]

John

Sorry you cannot come to grips with Audio Levels as used in Australia

The Alignment Level is the Voltage Level expressed in dBu with reference to 0.775 Volts, ( Used to be a Power Level expressed in dBmW where 0 dBmW equals 1 milliwatt across a 600 Ohm Resistor, or 0.775 Volts across a 600 Ohm Resistor)

Current Standard Broadcast Alignment Level in Australia and the USA is +4 dBu, and I believe in the UK and most European Countries 0 dBu

+4 dBu corresponds to 0 VU on a "Real" VU Meter on steady tone
when the meter is connected in series with 3.6 K Ohms

0 dBu corresponds to 4 on the IEC IIa BBC type PPM

You have other levels to consider

In Australia, Distortion and Noise is measured at Reference Level, this being 8 dB above Alignment Level

In the UK and Europe, you have PML, or Permitted Maximum Level, being 9 dB above Alignment Level, or a whisker above 6 on the BBC PPM Scale

I do not know what level is used in the UK for Distortion and Noise Measurements

Above that, you have what is commonly referred to as "Crash Level" preferable 18 dB + above Alignment Level

Levels you can put on Magnetic Tape, depends very much on the type of Tape you are using, with gain and other factors in the recorder set to give you a agreed maximum distortion level at say 9 dB above Alignment Level at a frequency of 400 Hz

On CD, the Maximum Level is determined by the Digital Recording Format, and for 16 Bit, Sony and Philips consider Alignment Level to be 15 dB Below Maximum Digital Level

Please remember, Magnetic Tape "Bends Gracefully" on overload

Digital hits a "Brick Wall"

[mrfeedback]

Rod Elliot gives some VU And PPM Audio Metering explanations here.

PPM: A standard PPM has a 5ms integration time, so that only peaks wide enough to be audible are displayed. This translates into a response that is 1dB down from a steady state reading for a 10ms tone burst, 2dB down for a 5ms burst, and 4dB down for a 3 ms burst. These requirements are satisfied by an attack time constant of 1.7ms. The decay rate of 1.5 seconds to a -20dB level (IEC specified) is met using a 650 ms time constant.

VU: A VU meter is designed to have a relatively slow response. It is driven from a full-wave averaging circuit defined to reach 99% full-scale deflection in 300ms and overshoot not less than 1% and not more than 1.5%. Since a VU meter is optimised for perceived loudness it is not a good indicator of peak performance.

So a VU meter is quite blind to high frequency single spikes, and the BBC PPM is already down 1dB in sensitivity to spikes of 10ms (?) duration, and 4dB under-indicating for a spike of 3ms (330Hz?) and further drooping response due to attack and integration times.
This means that neither of these meters will accurately indicate the peak level of a very short transient.

DAW softwares provide PPM displays that are a sample by sample accurate display of the instantaneous voltage of the audio signal waveform.
These 'virtual' meters seem to have an immediate attack time with sometimes adjustable decay and peak hold times, and a resettable 'digital over/clip' indication.
With systems such as digital with a defined absolute peak capability, this presumably is the only correct way to display levels.

On analog systems peak voltage of the audio signal waveform is absolute also.

What is wrong with a led peak meter with an 'instantaneous' attack time and suitable peak hold time ?

With 0dB indicating a peak 'clean' or peak allowable voltage, and generous positive display range, the degree of peak crushing (if desired) would be clearly apparent, as would headroom if levels are set at below the 0dB reference peak voltage.
As I understand it, the VU meter and the BBC PPM 'smoothing' specifications are useful to see average or percieved loudness levels but are useless actually to accurately display very short non repetitious pulses.

Eric.

[poulkirk]

Eric

You may have missed something

This discussion was not meant to be about Audio Level Metering, but Audio Levels

A good "Real" VU Meter, and the recognised PPM's used in the Broadcast, Film and Recording Industry are perfectly workable and meaningful instruments in the right hands, provided they are interpreted correctly

Maybe the "Spurious" Audio Level Indicators found on Semi Pro and Domestic Recording Equipment, in the hands of Would Be Record Producers lacking the necessary skill and taste, coupled with inadequate monitoring facilities, operated at excessive levels are some of the causes of bad recordings

The discussion was actually meant to be about Audio Levels, particularly Alignment Levels, and the usual headroom provided on good professional equipment above Alignment Level

Suggest you strar a thread about Audio Level Meteringl

[mrfeedback]

Originally posted by poulkirk
This thread is started with the objective of clarifying what Audio Levels and Audio Metering is about in simple terms......

.......For Audio Equipment to operate and give satisfactory sound quality, a certain amount of Headroom above Alignment Level is necessary

.......Traditionally, Professional Equipment was expected to have a headroom of 18 to 20 dB to accomodate normal Audio peaks on Voice and Music, and is still the expected norm

Originally posted by poulkirk
This discussion was not meant to be about Audio Level Metering, but Audio Levels

Yes, the two two are closely related.

Ok, so the VU and BBC meters are usefull for keeping an eye on average lower frequency repetitive signal levels, and are assuming accepted crest factors for different kinds of audio signals.

For example a typical bassy pop music passage with limited dynamic range will indicate rather higher without causing peak overloads than say a voice or snare drum signal on a VU meter.

The skill part is practice in correlating differing signal types (with differing crest factors) to VU and BBC PPM indications, and incorporating a mental 'correction' factor.
This is not always intuitive and has caught me out, and plenty of recording engineers also I expect, and is a likely cause of bad sounding recordings.

The discussion was actually meant to be about Audio Levels, particularly Alignment Levels, and the usual headroom provided on good professional equipment above Alignment Level

I accept that in skilled (practiced) hands these meter standards are workable and accepted, and indeed the norm, but by definition advance knowledge of the 'smoothing' characteristics and a mental 'fudge' factor are required if momentary overload is to be reliably avoided.
So IOW the accepted meter standards are not 'idiot' proof if momentary overload (wrt the accepted headroom for the particular equipment) is to be avoided.

In my previous post is actually a question - "What is wrong with a led peak meter with an 'instantaneous' attack time and suitable peak hold time ?" to ensure that agreed system peak levels are not exceeded momentarily, and calibrated according to the headroom spec of the particular equipment.

Eric.