Audio
Speaker Level Calibration Speaker channel level calibration (trim levels)
Also known as: speaker level matching, channel level calibration, SPL meter calibration, AVR test tone levels, trim levels
Speaker level calibration sets each channel's trim so that the same test signal measures the same SPL at the listening position. SMPTE ST 202 uses 85 dB SPL C-weighted, slow for cinema, while home AVR guides use 75 dB per speaker with an SPL meter. Denon receivers offer a -12 dB to +12 dB trim range per channel, and auto-EQ systems such as Audyssey set the trims from microphone measurements.
What speaker level calibration is
Speaker level calibration matches the loudness of every channel at the seat. Speakers differ in sensitivity, sit at different distances, and radiate into different room boundaries, so the same electrical signal produces different SPL from each. The receiver compensates with a per-channel trim, and the job is to find the trim values by playing a test tone through one speaker at a time and measuring it. The cinema reference this descends from is covered in reference level.
SMPTE ST 202:2010 section 5.4 gives the cinema numbers: a suitable single loudspeaker auditorium level with wide-band pink noise is 85 dB SPL, C-weighted, slow reading. Home guides drop the target. Bob Pariseau's trim-balancing guide says movie theaters use 85 dB for reference level and calls that too loud for home use, aiming at 75 dB SPL for each main speaker and for the combined subwoofers.
How the procedure works
The meter settings are consistent across the guides I fetched: C-weighting, slow response, a range with 75 dB near the middle of the scale, and the microphone at seated ear height pointing at the ceiling, not at the speaker. Pariseau holds the meter at arm's length and warns against moving it between measurements. Start every trim at 0 dB, set the master volume so the front left reads 75 dB, leave that trim alone, then adjust the other channels to match.
Denon documents the control side. The Levels menu sets the test tone volume to be the same at the listening position from each speaker, with a range of -12.0 dB to +12.0 dB and a 0.0 dB default. Levels cannot be changed with headphones connected, and a separate Channel Level Adjust menu sets levels per input source.
If a trim has to go positive and might clip, Pariseau's rule is to lower every trim by the same amount and raise the master volume by that amount, which keeps the balance and restores headroom. His guide also prefers test tones from a calibration disc, such as the LPCM channel ID tracks on the AIX Blu-ray, over built-in tones, because disc tones pass through the crossover processing the way a movie does.
Subwoofer level and the LFE +10 dB
The sub is the awkward channel. ITU-R BS.775-4 notes that a band-limited LFE pink noise signal reproduces 10 dB above the other channels, and that a wideband SPL meter will not read +10 dB because the signal is band-limited. In practice, home guides sidestep this by trimming the sub to the same 75 dB reading as the mains using the receiver's sub test tone. Pariseau tests each sub alone and sets its volume knob to match; with two subs he targets 72 dB each so they combine near 75 dB, and he notes readings swing even on slow, so you average them mentally. An AVForums thread reports a 4 to 5 dB needle swing on the sub tone and suggests aiming for a 72 to 77 dB swing centered on 75.
The same thread makes a point worth keeping: a cheap meter's offset from true SPL does not matter as long as every speaker reads the same. I found no source that measures phone-app accuracy, so take app readings as relative matching tools. Mehlau's calibration notes go further for narrow cases: they describe the SPL meter plus band-limited pink noise, limited to 500 Hz to 2 kHz, as only suitable for comparing satellite speaker levels.
Manual vs auto-EQ
Audyssey MultEQ XT32 automates the trims. Denon's manual says it uses the main listening position measurement to calculate speaker distance, level, polarity and the optimum crossover for the subwoofer. It recommends six or more measurement positions, allows up to eight, requires each position after the first to be within 2 ft (60 cm) of the first, and wants the microphone at least 20 inches (50 cm) from walls. Marantz's SR7009 manual confirms measurements in up to eight positions. Auto-EQ applies the result, and a meter check afterward is still the way to confirm it. See Audyssey MultEQ XT32 and gain staging for how trims interact with clipping.
Worked example
You set the front left to read 75 dB at the main volume. The center reads 72.5 dB, so its trim goes to +2.5 dB. A surround reads 78 dB, so it goes to -3 dB. The sub reads 70 dB, so raise its own volume knob by 5 dB (the trim is not the place to hide a 5 dB shortfall). The largest positive trim is +2.5 dB. If that might clip, Pariseau's rule applies: lower every trim by 2.5 dB and raise the master volume by 2.5 dB, which leaves the balance unchanged.
Keep the +/-12 dB range in mind. A speaker that needs most of that range deserves a second look at placement and sensitivity before a trim papers over it. The slip Pariseau warns about is moving the meter between channels; another is measuring at one seat and sitting at another, which is the case multi-position Audyssey measurement addresses. For a final check, the blog guide how to verify surround sound walks through channel ID tests.
Sources
- [1]
- [2]
- [3]
- [4]
- [5]
- [6]
- [7]
- [8]