Stereo field analysis: width, phase & mono safety

Stereo field analysis looks at how your mix uses the left-right image — how wide it is, whether the sides stay in phase, and whether the low end collapses cleanly to mono. Get it wrong and the mix hollows out on a phone or loses its sub in a club. Unveil measures stereo width, phase correlation, and mid/side balance, and flags low-end that's too wide to translate.

What stereo width is and why it needs measuring

A stereo mix places sounds across a left to right image. Width is how far that image extends and how different the left and right channels are from one another. Two channels that are identical produce a mono image fixed in the centre; two that are completely different produce a wide, diffuse image.

Width is easy to misjudge by ear because panned sounds and widened effects feel impressive on good headphones. But a mix must also work on a phone speaker, a mono club system, and a Bluetooth speaker that sums to mono. Measurement catches problems that only appear when the two channels combine.

Mid/side and what the meters show

Mid/side processing re-expresses a stereo signal in two parts. The mid is what the channels have in common, calculated as (L + R) divided by 2. The side is what differs, calculated as (L - R) divided by 2. A centre-panned vocal lives entirely in the mid. A hard-panned guitar appears in both mid and side. A wide stereo reverb shows up mostly in the side.

The ratio of side level to mid level is a direct measure of width. If the side channel is close to the same level as the mid, the mix is very wide. If the side is 12 dB or more below the mid, the mix is narrow and centre-focused. Many commercial masters have side levels between roughly 6 and 12 dB below the mid, though the range varies by genre.

Width is not constant across frequency. Low frequencies are usually near-mono and wide content lives above a few hundred hertz. A per-band width reading is more useful than a single full-mix figure.

Phase correlation and the goniometer

The phase correlation meter shows the relationship between left and right on a scale from -1 to +1. At +1 the channels are identical, which means mono. Around 0 the channels are unrelated, which is typical of a wide stereo image. Toward -1 the channels are in opposite polarity, and summing to mono will cancel them.

A healthy full mix usually averages between about +0.3 and +0.9, with brief dips lower. Short excursions near zero on a wide chorus are normal. A sustained negative reading is a warning that elements will vanish or thin out in mono.

The goniometer (also called a vectorscope) plots left against right as a cloud of points, rotated so that mono falls on the vertical axis. A narrow vertical line is mono. A fat, roughly circular cloud is a wide stereo mix. A cloud stretched along the horizontal axis indicates out-of-phase content. A tall oval standing upright with some spread to both sides is the usual picture of a healthy mix.

Mono compatibility and low-end in mono

Mono compatibility asks: what happens when left and right are added? Identical content adds by 6 dB. Unrelated content adds by about 3 dB. Out-of-phase content cancels. If a section is wide and out of phase, it can lose most of its level, which the listener hears as a hollow chorus or missing guitars.

Low frequencies deserve special attention. Bass notes have long wavelengths, and phase differences between channels can cause cancellation that changes the perceived level of the bass dramatically depending on where you stand. Club systems and many speakers sum the lowest frequencies. The convention is to keep everything below roughly 100 Hz at or near the centre. Some engineers use a high-pass on the side channel at 80 to 120 Hz for this purpose, and a value of about 100 Hz is a typical starting point.

Widening techniques and their risks

Panning: the safest approach. Pairs of double-tracked guitars hard left and right widen the image with no phase risk.
Stereo delay (Haas effect): delaying one side by roughly 5 to 30 ms. It widens, but produces comb filtering on a mono sum. A 1 ms delay, for example, creates a first notch at 500 Hz, and a 10 ms delay creates notches every 100 Hz.
Chorus and modulation: widens through slightly different pitch on each side. Check mono, as modulation creates moving cancellations.
Stereo reverbs and delays: widen the ambience, not the source, and are usually mono-safe because the dry signal stays in the centre.
Mid/side EQ: raising the high end of the side channel adds air and width without changing the centre. It increases the risk of harshness on the sides.
Stereo widener plugins: many use phase tricks or side boosts. Boosting the side too far can reduce mono compatibility. Always check the correlation meter and the mono sum afterwards.

Step-by-step workflow

Open a correlation meter and goniometer on the master bus.
Play the chorus and note the average correlation and the shape of the cloud.
Press the mono button and listen. Note anything that drops in level or changes character.
If something vanishes, mute channels in groups to find the source, usually a widened synth, a stereo guitar, or a wide effect.
Inspect the bass region. Use a width reading per band, or put a mono-maker on the bass and kick bus below about 100 Hz.
Correct the cause: reduce the widener amount, shorten a Haas delay, or invert polarity on the offending side if a polarity problem exists.
Re-check the correlation and the mono sum.

Worked example, mistakes and checklist

A chorus measures an average correlation of +0.05 with dips to -0.3, and in mono the lead synth drops by about 8 dB. The synth uses a stereo widener set to 150 percent plus a 12 ms Haas delay. Removing the delay and setting the widener to 100 percent raises average correlation to +0.45, and the mono level loss on the synth falls to about 2 dB. The image is slightly narrower, but the chorus no longer loses its main hook in mono.

Typical mistakes are widening everything, which leaves no contrast; leaving sub-bass in stereo; judging width only on headphones; and relying on a single widener on the master instead of widening individual elements with purpose.

Average correlation positive, with no sustained negative readings.
Mono sum keeps the vocal, bass and drums at roughly the same balance.
Below about 100 Hz is at or near mono.
Width is used for contrast, with some elements left narrow.
Checked on headphones, speakers and phone.

What you get

Stereo width and mid/side ratio — how much of your energy is in the sides vs the center.
Phase correlation: why sustained negative correlation destroys your mix in mono.
Low-end mono compatibility — keeping sub and bass mono below ~80–120 Hz for club and phone translation.
A free stereo readout that tells you if you're too narrow, too wide, or phase-risky.

Questions

What is phase correlation?

A −1 to +1 measure of how similar your left and right channels are. Near +1 is mono-safe; sustained negative values mean parts of the mix cancel out when summed to mono.

Should bass be in mono?

Generally yes below ~80–120 Hz. Wide stereo bass causes phase cancellation on club systems and mono playback, robbing the low end of punch.

Analyze your mix — free