|
|||||||||
|
Thread Tools | Search this Thread |
April 21st, 2008, 10:08 AM | #1 |
Regular Crew
Join Date: Jan 2008
Location: Winnipeg Manitoba Canada
Posts: 75
|
What's the importance of reading peak v.s. VU?
Hello,
I was wondering if someone could give me a quick primer on "peak" v.s. "VU" readings. An option on my Sound Devices 302 is to set the meter such that it shows both peak and VU readings. Till now, just using my camera, I've been setting sound levels with the camera's VU meter. That seems to have worked out pretty well. I've followed the rule of: set between -20 and -12, try to average around -12 and go over 0 as little as possible. I guess I'm wondering why I shouldn't just continue to do this on the mixer? (I can set it to just show VU.) I've also only recently learned how to read a mixer's meter correctly (mixer +6 = camera -12, mixer 0 = camera -20, mixer +20 = camera 0). The 302 only reads to +20. If that's the case all peak readings would be below the camera 0 cut-off. Thanks. John |
April 21st, 2008, 01:46 PM | #2 |
Major Player
Join Date: Jun 2004
Location: McLean, VA United States
Posts: 749
|
"VU"s (volume units) refer both to a particular reference level and the "ballistics" (the way in which the meter responds over time) of the indicator. In the early days VU meters were D'Arsenval movements with sensitivities adjusted to read 0 VU when 1 mW (0 dBm) was being delivered by a 1 kHz tone to a 600 ohm load. The response time of the movement was 300 milliseconds. The response/decay time meant the meter averaged the voltage (not the power) over that time so that if the meter read 0 VU on a cymbal clash or drum beat the peak voltage could be well over that. The overal reading of the meter was indicative of the impression of loudness. So the engineer rode the levels to try to keep the meter near 0 VU and used his experience to keep peaks under control. In one FM station where I worked there was a red light which hung from the ceiling over the console which flashed if the transmitter deviation meter indicated that peak deviation was reached. This is a sin in the eyes of the FCC.
In today's equipment mechanical meters (and their attendant problems) have gone away to be replaced by LEDs. LEDs are "inertialess" (it was the mechanical inertia of the pointer/coil/spring assembly in the D'Arsenval meters that gave the 300 mS response time so that it is possible to indicate the level instantaneously and continuously. Usually the electronics holds a peak reading until either a new higher peak is encountered or a predetermined time (usually half a second but often adjustable) passes at which time it moves to the next peak. This gives you time to read the peak. VUs are now indicated by using an electronic circuit to do the 300 mS averaging with the filtered signal being applied to the LEDs. Thus it is possible (in equipment which displays peak and VU) to have simultaneous indication of the relative loudness and to see if any peaks have busted the rail in an A/D system or overdeviated the transmitter in a broadcast system (or driven the limiter into compression or whatever depending on how the levels are set for the particular equipment). The manuals for the gear should indicate what the various levels on the meter mean and what the implications of exceding particular levels are. |
April 21st, 2008, 02:30 PM | #3 | |
Inner Circle
Join Date: Nov 2004
Location: Baltimore, MD USA
Posts: 2,337
|
Quote:
Tone (continuous sine wave) really doesn't have peaks like normal audio. So setting tone at -20 is just fine. But you DO want to watch peak levels. I leave my SD mixer in RMS/Peak mode. When shooting, I'm always watching the peaks to keep them safely below 0 on the camera or +20 on the mixer. Every once in a while I slip up and accidentally move the switch to RMS only mode. Then I can't figure out why things sound squashed. That's my mixer limiter kicking in. If I'm running RMS levels to +16, the peak levels are definitely past 0. Doesn't happen very often, and when it does, my brain usually says, "Hey, stupid, what's the metering set to?" Regards, TyFord |
|
April 21st, 2008, 11:37 PM | #4 |
Major Player
Join Date: Apr 2007
Location: Espoo Finland
Posts: 380
|
With analog systems suddenly overloading peaks were not a total disaster they are with digital recorders. Analog tape saturates quite gracefully when overdriven, thus metering with VU only is possible. With digital recorders overloading the analog to digital converter totally ruins the recording, for that reason it is imperative to know where the real peak levels are and that they never go over 0 dBFS (+20 dBVU). For that reason peak reading meters are a must with digital.
|
| ||||||
|
|