We Can't Go On Metering Like This!
VU and PPM meters explained.
There is still a certain amount of confusion as to the difference between VU meters and peak reading level meters. Hopefully, this short article will clarify the subject.
When considering audio signal levels, we often talk in terms of dBs, but strictly speaking we should refer to volume units (VUs for short). The decibel may be used to measure the ratio between two powers or voltages and if we're looking at audio tones of a constant level, its use is valid for comparing these as well. The problem occurs when the audio signal is constantly changing in level, as is the case with recorded speech or music. Peaks in an audio signal may be far above the average signal level but, because of the way in which the human ear works, may not contribute greatly to the perception of loudness as the ear tends to perform some sort of averaging process when evaluating a subjective sound level. What all this adds up to is that the signal with the highest peaks does not necessarily sound the loudest.
When looking at the moving coil meter in the context of sound level measurement, it can be appreciated that the relatively slow meter needle is never going to keep up with the short term signal peaks anyway and that any dB measurements will only be accurate on a steady state signal such as the tone from a line-up oscillator. The VU meter however, is designed not to follow the peaks in the signal but to represent as closely as possible the subjective loudness as perceived by the ear which falls somewhere between the peak and the average signal level of a complex audio waveform. The ballistics, or mechanical response time of the meter is consequently tailored to match the response of the ear, at least in theory, but if you check out VU meters built by different manufacturers, you are likely to notice a difference in response time performance, even though they might all read the same on a steady state signal.
Because the VU meter tends to act like the ear in evaluating loudness, it is a valuable aid to recording but there are pitfalls of which you need to be aware.
Remember that any peaks contained in the programme material will be quite a lot higher than the meter readings and it's not uncommon to have peaks that are 10dB above the VU reading in a typical recording situation. This is particularly true when you are recording sounds of a transient nature such as percussion as you could be overloading the signal electronics or even the tape itself whilst the VU meter is happily telling you that everything is alright. With analogue recordings, this may not be as serious as you think because things don't suddenly fall to pieces when you exceed 0VU, the system usually has a bit of headroom to spare and tape tends to saturate gracefully rather than clip. Some tape saturation is acceptable and on some types of programme material, it can even improve the sound so the best bet is to do a few experiments and see what VU indication you are at when the sound quality starts to go off. This way you can establish a safe VU level for recording different types of sound and still keep away from the much hated noise floor.
On some occasions though, it's useful to know what the peaks are really up to and this is particularly true if you intend to indulge in any digital recording as there you run out of headroom with a bang. This is because there is no tape compression to save you, you simply run out of bits.
"...high frequencies need to be treated with respect and soon show signs of distortion when recorded at high levels."
The design of the peak reading meter, as its name implies, allows it to more closely track the true programme peaks and to this end, the traditional moving coil meter is now tending to give way to the LED ladder display which does not suffer from the mechanical inertia of its predecessor. Because programme peaks may be very short, it's normal to introduce an artificially slow decay characteristic into the meter drive circuitry so that the peak level is displayed long enough for the operator to see it. In some designs, this may even take the form of a single illuminated section which independently registers the peaks whilst the rest of the display responds quickly to the programme dynamics.
As you might imagine, this type of meter is most useful when you are recording drums or other percussive instruments, but if you adhere too rigidly to its findings, you may well end up with excessive background noise. This is again due to the fact that a little tape compression is generally acceptable, so it's up to you to determine how far you can push your recorder. A further complication is that whilst you may be able to get away with a fair degree of overload on low frequency signals, high frequencies need to be treated with respect and soon show signs of distortion when recorded at high levels.
Both types of metering have their place but they are only there to advise you; it's up to you to make the best use of their information if you really want to get the optimum performance out of your equipment.
Feature by Paul White
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!