The Right Way to Sidechain in Logic Pro
Your bass breathes. Every four beats it drops out, swells back, drops out again. You built it that way because every tutorial you watched put a Compressor on the bass, routed the kick to its sidechain input, and called it done. But listen to the arrangement on a phone speaker and the low end disappears entirely between kicks — you didn’t make room for the kick, you punched a hole in the track.
Sidechaining is a frequency and timing problem with four distinct solutions. The compressor is one of them, and it’s the one that gives you the least control.
Sidechain Pumping Is the Sound of a Mistake
The pumping effect is a compressor artifact. When you set a threshold low enough that the kick pushes the bass 8–12 dB down, and a release long enough that gain recovery is still climbing when the next kick lands, you get a continuous rising-falling envelope. That’s the breathing. It has nothing to do with the kick fitting into the mix and everything to do with the compressor’s gain reduction curve being audible as a musical event in its own right.
There is a genre where this is the point. Late-90s French house and big-room EDM use pumping as a rhythmic layer — the sidechain is an instrument. That’s a deliberate aesthetic choice made at extreme settings. It is not what happens by accident when you copy a preset.
The problem with the compressor approach is that it’s program-dependent. Gain reduction depends on the incoming bass level, so a loud bass note ducks harder than a quiet one. Two identical kicks produce two different amounts of ducking depending on what the bass happens to be playing. You didn’t ask for that variation and you can’t hear it until it stacks up over 32 bars.
There’s a second failure mode nobody talks about: attack time. Logic’s Compressor in Platinum mode has a minimum attack around 0.1 ms, but if you leave it at the default 10–20 ms, the compressor doesn’t start ducking until the kick’s transient has already passed. You get the ducking after the collision, which means you’ve lost bass energy without gaining kick clarity.
Diagnose First: Masking, Transient Collision, or Phase?
Three different problems sound similar and need different fixes.
Energy masking is frequency overlap. Your kick’s fundamental sits at 50 Hz and your bass has a resonant peak at 55 Hz. Both are present continuously, so the low end sounds thick and undefined even in bars where the kick isn’t playing. Sidechain won’t fix this — static EQ will. Cut a narrow notch in the bass at the kick’s fundamental and leave it cut.
Transient collision is timing. The kick’s attack and the bass’s attack land on the same sample, and the summed peak clips your bus or eats your headroom. The low end sounds fine on its own but the mix loses punch on the downbeat. This is the one sidechaining solves.
Phase cancellation is the quiet killer. If the kick and bass sub content are moving out of phase in the 40–80 Hz region, the sum is quieter than either source alone. Adding sidechain to a phase problem makes the bass disappear and does nothing for the kick.
Check for it before you touch anything else. Solo the kick and the bass together, open Multimeter on the bus, and switch to the Goniometer. If the low-frequency correlation meter sits at or below zero when both are playing, you have cancellation. Then insert Logic’s Gain plug-in on the bass, enable Phase Invert, and listen again. If the low end suddenly gets louder and more solid, flipping phase was the whole fix. If nothing changes, phase wasn’t the issue — move on.
Do this test in mono. Phase cancellation is much easier to hear when you collapse the stereo field.
Method 1: Shaping with Volume Automation
This is the most controlled approach and the one nobody wants to do because it takes ten minutes.
Draw the volume envelope on the bass channel by hand. On each kick hit, the volume drops to a level you choose, holds for a duration you choose, and recovers along a curve you choose. No ratio, no knee, no program dependence. The 40th kick behaves exactly like the first.
In Logic, work with region-based automation rather than track automation. Select the bass region, press A to show the automation lane, then use the Track menu to switch that lane to Region automation. Now the envelope belongs to the region — copy the region and the ducking copies with it, move it and the ducking moves in time.
Draw one bar. Place a node just before each kick at 0 dB, a node at the kick position at your target reduction, and a recovery node roughly a 16th note later. Then loop that bar across the section.
The curve type matters more than the depth. Right-click a node and choose the curve shape: a linear recovery sounds mechanical, an exponential one — fast at first, slowing as it approaches unity — sounds like natural gain recovery. Hold Ctrl-Shift and drag the line between two nodes to bend it interactively.
The advantage over a compressor: the envelope is locked to the grid, not to the audio. Change the bass part and the ducking stays exactly the same.
Method 2: LFO-Based Ducking (Tremolo and LFO-Tool Logic)
Volume automation drawn by hand, automated. Logic’s Tremolo plug-in is a tempo-synced amplitude modulator, and because it modulates rather than compresses, it’s completely independent of the input signal level. Quiet bass note, loud bass note — same curve, same depth, every time.
Insert Tremolo on the bass channel and set Rate to 1/4 for a four-on-the-floor kick. The parameters that matter:
Symmetry shifts where in the cycle the dip occurs. At 50% the waveform is even; push it above or below and the down-phase gets shorter or longer. This is how you decide whether the bass is ducked for a 16th note or an 8th note.
Smoothing rounds the corners of the modulation waveform. At 0% with a square-ish shape you get an abrupt gate-like drop that will click on sustained bass. Raise it until the transition is inaudible — usually 20–40% is enough.
Depth is your reduction amount. Start low. 100% depth means full silence on every cycle.
Rate choice changes the character completely. 1/4 gives you one duck per kick on a standard house or trance pattern. 1/8 gives you a duck on the off-beat too, which adds rhythmic motion but starts competing with the hi-hats for attention. 1/16 is a texture effect, not a sidechain.
The catch: Tremolo doesn’t know where your kicks are. It follows the project tempo and the transport position. If your kick pattern has a missing beat or a syncopated hit, the LFO keeps ducking anyway. Use this on regular patterns; use automation on irregular ones.

The diagram shows the actual difference: the compressor path runs through level detection, so the bass signal itself influences how much it gets ducked. The LFO path bypasses that entirely.
Method 3: Frequency-Specific Sidechain — Duck Only the Problem Band
The kick and bass collide somewhere between 40 and 120 Hz. Above that, they mostly coexist. So why are you turning down the entire bass — including the 800 Hz harmonics that carry its note identity and its presence on small speakers?
Ducking full-range is why sidechained bass sounds like it’s disappearing. Perceived loudness lives in the midrange. Drop the mids and the bass vanishes from the mix; drop only the sub and the kick gets its space while the bass stays audible.
Logic’s Multipressor does this. Insert it on the bass, then open its Side Chain menu in the plug-in header and select your kick channel or a bus carrying the kick.
Set the crossover points around the kick’s fundamental. If the kick’s fundamental is 55 Hz, put the first crossover at roughly 120 Hz — high enough to cover the kick’s body, low enough to leave the bass’s harmonic content untouched.
Now enable compression only on band 1. Set its threshold so the kick triggers 3–5 dB of reduction, attack fast (1–5 ms), release around 80–150 ms depending on tempo. Set bands 2, 3, and 4 to a ratio of 1:1 so they pass through unchanged.
The result: the bass’s fundamental steps aside for the kick, and everything above 120 Hz keeps playing at full level. Perceived bass loudness barely moves.
Method 4: Dynamic EQ and Spectral Ducking
The most surgical option. Instead of ducking a band, duck a frequency — a narrow notch that opens only while the kick is sounding.
First, find the kick’s fundamental. Solo the kick, insert Channel EQ, and enable the Analyzer. Don’t read the transient — the attack is broadband and will show energy everywhere. Read the sustain portion, the 30–100 ms after the initial hit, where the pitched body of the kick rings. That’s where you’ll see a clear peak, typically between 40 and 70 Hz for an electronic kick. Set the Analyzer resolution to a longer window so the display settles enough to read. If the kick has a pitch envelope, the fundamental will drift downward over the decay — take the frequency where it settles, not where it starts.
Then on the bass, insert Logic’s Channel EQ isn’t enough — you need a dynamic band. Use a dynamic EQ plug-in with a sidechain input, set a bell at the kick’s fundamental, Q around 4–8, and a threshold that produces 3–4 dB of reduction when the kick hits.
This is better than a compressor when your bass has significant content above the collision zone and you need it fully intact — plucked bass, reese bass with strong upper harmonics, anything where the note’s identity lives above 200 Hz.
It’s overkill when your bass is a pure sine sub. A sub has almost nothing above the fundamental, so a narrow notch at the fundamental is functionally identical to ducking the whole channel. Use Method 1 or 2 instead and save the CPU.
Choosing the Right Method for the Job
| Situation | Method | Why |
|---|---|---|
| Trance, sustained sub, regular 4/4 kick | LFO ducking (Tremolo) | Consistent curve every bar; sub has nothing to preserve above the fundamental |
| Techno, long 909 tail kick | Volume automation | Recovery curve must outlast the kick tail; needs hand-shaping |
| House, plucked bass | Dynamic EQ, or nothing | Short notes may not overlap the kick at all — check before treating |
| Pop or hip-hop, irregular kick pattern | Multipressor with sidechain | Follows actual kick timing; preserves bass midrange for small speakers |
| Pumping wanted as a musical effect | Compressor, deliberately extreme | Program dependence is part of the character here |
Two questions decide most cases. Does the kick pattern repeat on a fixed grid? If yes, LFO ducking works and costs you nothing. Does the bass carry information above 150 Hz? If yes, duck the band, not the channel.
And check the obvious case first: a plucked bass playing 16ths with gaps on every downbeat may already leave room for the kick. Sidechaining it makes the mix worse.
How to Verify Your Sidechain Actually Works
Your ears adapt to the pumping within about thirty seconds. Measure instead.
Measure the reduction depth. Put a Level Meter or Loudness Meter on the bass channel. Play eight bars with the kick muted and note the RMS. Unmute the kick and play the same eight bars. The difference tells you how much average level the sidechain is costing you. Between 3 and 6 dB of peak gain reduction is generally enough to clear space for a kick. If you’re reading 10–12 dB, you’re not making room — you’re pumping, and the perceived bass level will have dropped noticeably.
Watch the gain reduction meter, not the output. On the Compressor or Multipressor, the GR meter should return to zero between kicks. If it never reaches zero, your release is too long and the ducking is cumulative.
Listen in mono. Press the mono button on the output channel. Sidechain problems in the sub region show up immediately in mono because the low end is usually mono content anyway.
Listen quietly. At low monitoring levels, the Fletcher-Munson effect reduces your perception of bass. If the bass still holds up at whisper volume, the ducking is reasonable. If it disappears, you ducked too hard.
Check on a phone speaker. A phone has no sub reproduction, so what you hear is the bass’s harmonic content. If the bass vanishes on a phone whenever the kick hits, your sidechain is eating the midrange — which means you used a full-range method where you needed a band-specific one.
What to Do Next
Open a project you’ve already finished. Find the bass channel, bypass the sidechain compressor, and pick an eight-bar section with a steady kick pattern.
Build the same ducking four times: hand-drawn region automation, Tremolo at 1/4, Multipressor with the kick sidechained into band 1 only, and a dynamic EQ notch at the kick’s fundamental. Target the same 4 dB of reduction in each. Bounce all four plus the bypassed original — five files, same section, same level.
Import them into a new project on separate tracks and A/B them at low volume, in mono. You’ll hear differences that don’t show up while you’re tweaking parameters.
Whichever one wins, save it as a channel strip setting and put it in your template. Name it by use case — “SC sub trance,” “SC band house” — not by plug-in. You’re building a decision, not a preset.