Today’s frequent use of VU meters (Volume Unit meters) on professional equipment is, for various reasons, a bit of a mystery to me. Developed in the late 1930s/early 1940s, they were designed for measuring telephone lines and levels for broadcasters. They were cheap. They have a range of around 23dB (from a reading of -20dB to +3dB, relative to 0VU which is +4dBu), and were only accurate at a reading of 0VU (and this calibrated with a 1000Hz sine wave), the lower values becoming increasingly unreliable. They have a very slow rise and fall time, around 300ms, so give an average reading for music input, largely missing peaks. This is useful in that it indicates the perceived loudness of what it is measuring.
I see a number of reasons why VU meters are of very limited use:
• They are only accurate at 0VU/+4dBu
• They do not measure above +7dBu
• They do not measure peaks
• Most modern studio equipment has maximum input/output levels of +21dBu, +26dBu or more
I therefore don’t understand why they are so widely used on pro-audio equipment; apparently one reason is that they look great.
DACS is developing a mastering compressor and we will be including VU meters on the output. They will be supplemented by some peak metering. My feeling is that this is a sensible use for a VU meter. The function of a compressor is in some ways similar to the VU, taming peaks, and the needle activity of a VU will reflect this process nicely.
Our biggest area of indecision is how we make it useful without discouraging people from making the most of the device’s headroom. No point in the needles permanently banging against the stops if 0VU is +4dBu or not reaching 0VU at all if we set 0VU at a higher value. With a device like ours we have a wide useful range of signal measured by a meter that has a really limited range of measurement.