The Nebula tip of the month (#38 - July 2012)

Nebula tips and tricks
User avatar
Posts: 2860
Joined: Sun Mar 28, 2010 9:00 pm
Location: Lodi | Madrid | Buenos Aires

The Nebula tip of the month (#38 - July 2012)

Post by enriquesilveti » Sat Jun 16, 2012 7:21 pm

Using old audio meters

The meters that can be considered obsoletes for a full digital environment are the volume unit meter (VU meter) and the peak program meter (PP meter), but due a lot of studios and broadcast companies work in hybrid environment we should understand it and know how to use it.

. The VU meter
A volume unit meter (VU) is a device displaying a representation of the signal level in audio equipment. The ASA standardised it in 1942 for radio broadcast stations. A VU meter which in effect integrates the signal with a rise time of 300 ms. This meter was designed not to measure the signal, but to let users aim the signal level to a target level of 0 dB the scale ranges from -20dB to + 3dB, with -3dB right in the middle. The reference for power is +4 dBu sine wave equal to 0 VU.
The VU meter is a unintentionally slow measurement, averaging out peaks and troughs of short duration and reflects more to a the perceived loudness of the material. A VU meter responds in a fashion very similar to how human ears perceive loudness. A true VU meter uses a decibel scale, and is carefully ballistics controlled to conform to specific ANSI specifications, this specifications are defined in ANSI C16.5-1942, British Standard BS 6840, and IEC 60268-17.
A VU meter should not be relied on to display accurate peak and transient levels so most modern gear with VU meters includes LED indicators as a supplementary peak display.

. The PP meter
A peak programme meter (PPM) is used for indicate the instant level of an audio signal. The principal standard for PPMs is IEC 60268-10.
A PPM in effect integrates the signal with a rise time of 10 ms so is a Quasi peak programme meter (quasi-PPM). PPMs do not provide effective loudness monitoring and in addition quasi peak programme meters indicate neither loudness nor true peaks but something between the two, so it is important to allow sufficient headroom when using them in the control of digital audio levels. The EBU recommendation r68 provides for this by defining alignment level as -18 dBFS thus a peak to the maximum permitted level as indicated on a quasi-PPM corresponds to -9 to -10 dBFS (the 9-10 dB margin allows for operator error). The true peak typically being several dB higher than the PPM indication, and that subsequent signal processing may increase the amplitude, for example using sample rate conversion, coding, etc.
SMPTE RP 155 recommends a different alignment level, corresponding to 0 VU equal of -20 dBFS. The two conventions result in line-up tone levels that differ by 2 dB, but in practise the level of programme modulation tends to be very similar. The SMPTE and the EBU agree that regardless of whether -18 or -20 dBFS is used as the alignment level, that level should be declared and that in both cases programme should peak to a maximum permitted level of -9 dBFS when measured on an IEC 60268-10 quasi-PPM with an integration time of 10 milliseconds.

There are many other different kinds of PPM, they fall into broad categories:
. True peak programme meter or TPPM. This shows the peak level of the waveform no matter how brief it is the duration.
. Sample peak programme meter or SPPM. This is a PPM for digital audio which shows only peak sample values, not the true waveform peaks which may fall between samples and be up to 3 dB higher in amplitude.
. Over sampling peak programme meter or OSPPM. This is a sample PPM in which the signal has first been over sampled, typically by a factor of four, to alleviate the problem with a basic sample PPM.

. Maximum peak program level
The maximum peak program level corresponds to -9 to -10 dBFS (the 9-10 dB margin allows for operator error) in Continental Europe and -6 dBFS in UK and Ireland and that means that you will need to use more a dynamic range compression in Continental Europe audio program than in UK and Ireland but this value has non sense in music production where the final product is a CD, the correct value should be -1 dBFS measured with over-sampling peak programme meter. As you can see broadcasting audio program are more over compressed due safe level value margin necessary for subsequent signal processing.

. RMS versus AES-17 RMS
In a pure sine wave file the RMS value will be 3 dB less than the peak value due the crest factor explained in the previous tip. AES-17 RMS value will add 3 dB to the RMS value in order to have the same value when you measure peak and RMS values in a pure sine wave file.
ExampleMetersWithA1kHzPureSineWave.jpg (642.97 KiB) Viewed 4101 times