Drum Recording: Mic Techniques, Phase Alignment & Bleed
Tracking an acoustic drum kit is the ultimate test of audio engineering. Because 8 to 16 microphones capture one acoustic instrument simultaneously, phase cancellation can destroy low-end punch and smear stereo imaging. This guide covers kick and snare miking, overhead configurations, Glyn Johns geometry, and phase alignment.
1. The Physics of Acoustic Phase Cancellation on Drums
Sound travels through air at approximately 1,130 feet per second (343 meters per second). When a drummer strikes the snare drum, sound reaches the close snare top microphone in under 0.2 milliseconds. That same sound wave takes 3 to 4 milliseconds to reach the overhead microphones positioned 4 feet above.
When the close snare track is summed together with the overhead tracks in your DAW mixer, the slight arrival time delay causes acoustic phase cancellation. Frequencies whose half-wavelength matches the arrival time difference cancel out completely, turning punchy snare drums thin and hollow.
2. Kick Drum: Kick In vs. Kick Out Pairing
Professional tracking setups use a complementary dual-microphone system on the bass drum:
- Kick In (Shure Beta 91A boundary or AKG D112 / Audix D6): Placed inside the drum shell pointing 2 to 3 inches from the beater impact point. Captures the high-frequency transient click attack (3 kHz to 5 kHz) that cuts through dense rock and pop mixes.
- Kick Out (Yamaha Subkick speaker mic, Neumann U47 FET, or Solomon LoFrequenc): Positioned 2 to 4 inches outside the resonant front head. Captures the deep sub-bass fundamental weight (40 Hz to 80 Hz).
3. Snare Top & Snare Bottom Polarity Inversion
When the drumstick hits the top snare head, the top head moves downward (pushing air toward the Snare Top mic and creating positive air pressure). Simultaneously, the bottom resonant head pushes downward (pulling air away from the Snare Bottom mic, creating negative air pressure).
Because the physical diaphragms move in opposite directions, the two microphones are 180 degrees out of phase electrically. Always engage the Polarity Invert (Ø) switch on the Snare Bottom preamp channel to bring both signals into coherent phase alignment, instantly restoring full low-end body.
4. Overhead Microphone Geometries: A/B vs. X/Y vs. ORTF
Overhead microphones capture the entire drum kit as a cohesive stereo acoustic instrument:
Spaced Pair (A/B)
Two matched small diaphragm condensers spaced 3 to 5 feet apart. Delivers a wide, lush stereo image. Must maintain equal distance to the center of the snare drum using a mic cable tape measure.
Coincident (X/Y) & ORTF
X/Y crosses capsules at 90 degrees with zero time arrival difference, guaranteeing 100% mono compatibility. ORTF spaces capsules 17cm apart at 110 degrees for natural human ear stereo realism.
5. The Glyn Johns 3-Microphone Technique
Pioneered by legendary Led Zeppelin and The Who producer Glyn Johns, this minimalist setup creates an enormous, punchy drum sound using only three microphones:
- Overhead 1: Positioned 3 to 4 feet directly above the center of the snare drum, pointing straight down at the kit.
- Overhead 2 (Side Mic): Positioned just above the floor tom, pointing horizontally across the snare and hi-hat.
- The Crucial Measurement: Measure the exact distance from the center of the snare drum head to both microphone capsules using a piece of drum string. Both distances must match down to the millimeter (e.g. exactly 40 inches). Pan Overhead 1 slightly left and Overhead 2 hard right for a massive three-dimensional image.
- Kick Mic: Add a single dynamic kick mic for low-end reinforcement.
Measured: how deep the drum phase cancellation goes, and what a millimetre is worth
We ran the comb filter test of the dialogue guide on 2026-10-01 with the numbers this page gives. The signal was six seconds of white noise, summed with a delayed copy through FFmpeg 8.1.2 (adelay in whole samples at 48 kHz, amix with normalising off). Overheads 4 ft up are 3.55 ms from the snare and a close mic about 5 cm away is 0.15 ms, so the offset is 3.4 ms or 164 samples. No drum, microphone or room was involved.
| Level of the late copy | Expected dip | Deepest dip measured |
|---|---|---|
| Equal level | complete | -29.4 dB |
| 0.7 of the close mic's level | -10.5 dB | -10.5 dB |
| 0.5 of the close mic's level | -6.0 dB | -6.1 dB |
| 0.3 of the close mic's level | -3.1 dB | -3.1 dB |
| 0.1 of the close mic's level | -0.9 dB | -0.9 dB |
The offset is as the page says, 3 to 4 ms, and the first notch fell at 146 Hz with another every 293 Hz. "Cancel out completely" needs more than the timing, though. The dip is only as deep as the level match: at half the level it is 6 dB, at a tenth it is under 1 dB, and only an equal level copy gave a dip beyond 25 dB (the analysis resolution is the limit). If the overheads are much quieter than the close mic, the notches in the sum are shallow and the thinning comes from many of them across the spectrum, and not from deep nulls. The test uses noise and says nothing about a real kit's bleed.
Glyn Johns distances "down to the millimeter" are finer than the sum needs. Two equal overheads off by 21 mm (3 samples) put the first notch in the mono sum at 8.0 kHz, 50 mm at 3.4 kHz and 100 mm at 1.7 kHz. A millimetre is 2.9 microseconds. Matching to about 1.5 cm keeps the first notch above 10 kHz, and the page's string method is still a good way to get there. Hard panned, the mismatch is an image shift and not a comb, which this test did not examine.
For the snare bottom, we took the page's premise as given: the bottom mic hears the top mic's signal inverted, at 0.6 of its level and 14 samples (about 0.3 ms) later. Summed without touching the polarity, the model gave -7.4 dB at 100 Hz against the top mic alone, and 4.0 dB with the polarity switch engaged, so the page's advice to invert follows if the premise holds. The page's description of the pressure looks reversed to us, though. A head moving down moves away from a mic above it and toward a mic below it, so the top mic sees a pull and the bottom mic a push, and the two are still opposite in polarity. We did not record a drum, so the real size of the effect depends on the mics and the head.
Validated 2026-10-01 on white noise with FFmpeg 8.1.2, for the offsets and levels stated. The snare bottom case is a model of the page's premise. No drum, microphone or room was used.
Working Vocabulary
Key drum tracking and acoustic engineering terms introduced in this guide:
- Phase Cancellation. The acoustic phenomenon where two sound waves with identical frequencies but opposing phase angles cancel each other out when summed.
- Polarity Inversion (Ø). An electrical switch that flips the positive and negative terminals of a balanced audio signal by 180 degrees.
- Glyn Johns Method. A classic drum miking technique utilizing two overhead microphones equidistant from the snare drum plus a kick mic.
- ORTF. A near-coincident stereo miking standard developed by French radio using two cardioid capsules spaced 17 cm apart at a 110-degree angle.
- Boundary Microphone. A flat pressure-zone microphone designed to sit inside a kick drum or on a floor to eliminate reflective phase distortion.
- Transient. The initial high-amplitude, short-duration peak of a percussive sound (such as the stick strike on a drum head).