Chiller Sounds: What They Mean and How to Fix Them
- Alex Khachaturian

- Nov 5
- 8 min read

Promise: One guide to translate weird chiller noises into fast, confident fixes, plus a plain-English decoder for suction temp, superheat, subcooling, lift, and more.
TL;DR:
Most “mystery” noises map to 10 repeatable failure patterns. Name the sound → check one metric → fix.
Learn four numbers, suction temp, superheat, subcooling, lift, and 80% of confusion disappears.
Record a 60-second sound clip + 5 trending points; you’ll out-diagnose guesswork every time.
Key Takeaways
Sound is data. Rattle, hiss, whine, thump, whoosh each correlate with a known operating condition.
Numbers beat ears. Pair the sound with suction temp, superheat, subcooling, and lift to confirm.
Machine type matters. Centrifugal ≠ screw ≠ scroll/recip; their failure sounds differ.
Hydronics talk too. Cavitation, air, and water hammer sing their own songs.
Log now, argue never. A one-minute audio + trends capture kills debate and speeds approval.
Quick Links
Next Read
The First Chiller Sound I Heard Was Loud!
The call came in as an “ongoing issue”: the chiller was making really loud noises, then shutting down and restarting on its own. I was in a separate equipment room, multiple layers of concrete between me and the plant, when I caught a low, muffled vibration. I paused. It built into a wall-shaking roar.
I sprinted into the chiller room and got the full blast: a 500-ton screw chiller howling, then dropping out on protection, flow alarm, coast down, try again, repeat. Nothing about my current work should’ve been affecting it, so I started at first principles: sound + trends + valves + flow.
That’s when it clicked. The chilled-water isolation valve actuator wasn’t traveling, it was stuck mid-stroke. Flow was unstable, the machine was seeing erratic differential, and every time load shifted the chiller slipped into that ugly, loud instability and tripped to save itself. Final diagnosis: failed/stuck actuator creating intermittent flow and head conditions.
I told the customer exactly what was happening and why they were hearing it, then gave them the fix: replace the actuator and verify full stroke with a live trend. Post-swap, the room went from “earthquake” to a quiet hum, and the restarts stopped for good.
Quick Win (10-Minute Sound Log)
Goal: Turn noise into a decision in under 10 minutes.
Safety first: PPE, lockouts as required, guards on.
Capture 60 seconds of audio at the chiller and again by the pumps. Phone works fine.
Trend these five points (1-min resolution is plenty):
Suction pressure (or evap saturation temp)
Discharge pressure (or condensing temp)
Suction line temp
Liquid line temp (for subcooling)
% Load or IGV/slide-valve position
Name the sound (use the glossary below): hiss, whine, rattle, thump, whoosh, chatter, chirp, gurgle, roar, chuff.
Match the sound to the table in the next section, confirm with one number, act.
Sound → Fault Playbook
Use the “Hear → Check → Likely Fault → First Fix” map. Confirm with the one bold metric.
Before you think it, yes I understand that I'm making up words, but these are the best readable version of what it sounds like.
HISS (steady, not directional)
Check: Subcooling (°F/°C) and sight glass
Likely: Normal expansion/turbulence, or a small leak if localized.
First Fix: Verify subcooling is in spec; sweep with ultrasonic before calling it normal.
WHINE (rising pitch with load)
Check: Lift (condensing temp − evaporating temp)
Likely: High lift—dirty condenser, low condenser water flow, high ambient.
First Fix: Clean coils/tube bundle; verify condenser water setpoints & flows.
RATTLE/CHATTER (metallic tapping)
Check: % Load / IGV or slide-valve position
Likely: Unstable control loop, hunting TXV/EXV, loose baffles or panels.
First Fix: Stabilize loop gains; tighten hardware; damp TXV with a small superheat increase.
THUMP/POP (intermittent water hammer)
Check: Pump VFD ramp and valve timing
Likely: Rapid valve shutoff, no snubbers, or bad air pockets.
First Fix: Slow the ramp; add arrestors; bleed air at high points.
GURGLE (liquid line)
Check: Subcooling (low/negative)
Likely: Low charge or flashing liquid.
First Fix: Find/fix leaks; restore charge; confirm condenser approach.
ROAR/CHUFF (cyclic on centrifugal)
Check: Lift swings and IGV position
Likely: Surge—operating off the map at low flow/high head.
First Fix: Raise flow, open bypass, reduce lift (lower condensing temp, raise evap temp), adjust IGV.
BUZZ (electrical)
Check: Current THD or contactor/relay condition
Likely: Coil hum, loose laminations, failing contactor.
First Fix: Replace contactor; tighten transformer/mounts.
SQUEAL (belt/fan on air-cooled)
Check: Fan speed and belt tension
Likely: Slipping belts, failing bearings.
First Fix: Tension/replace belts; check pulley alignment and bearings.
KNOCK (compressor)
Check: Superheat (too low)
Likely: Liquid slugging or floodback.
First Fix: Increase superheat; check TXV bulb and insulation; verify evaporator load.
WHOOSH (constant through tubes)
Check: ΔP across evaporator/condenser
Likely: Debris/partial blockage or air entrainment.
First Fix: Brush/rod tubes; purge air; verify strainers.
Terms Demystified (Suction Temp, Superheat, Subcooling, Lift…)
Plain English, field-useful, no jargon.
Suction Temperature
The actual temperature of refrigerant gas entering the compressor. It should be warmer than the evaporating (saturation) temperature. If it’s too close, liquid may be nearby; if it’s too high, you’re starving the coil.
Evaporating (Saturation) Temperature
What the suction pressure would be if the refrigerant were boiling at that pressure (use a PT chart). It represents coil “coldness.” We rarely measure it directly; we infer it from suction pressure.
Superheat
Suction line temp − evaporating temp.
It’s “how much warmer than boiling” the vapor is. Low superheat = risk of liquid; high superheat = starved evaporator.
Condensing (Saturation) Temperature
The temperature corresponding to discharge/condensing pressure. It’s your condenser’s “hotness.”
Subcooling
Condensing temp − liquid line temp.
It’s “how much cooler than condensing” the liquid is—protection against flashing. Low subcooling = flashing/low charge; high subcooling can mean overcharge or extra condenser capacity.
Lift
Condensing temp − evaporating temp.
The “height” the compressor must push against. High lift = hard work = heat, noise, amps. Reduce lift to quiet and save energy (clean heat exchangers, optimize water temps, trim ambient).
Approach (Evap/Cond)
Water leaving temp − refrig sat temp on that side. A health score for heat transfer. Large approach = fouled tubes or low flow.
Surge (centrifugal)
Flow instability where the compressor repeatedly stalls and re-primes. Sounds like a rhythmic “chuff” or “whoomp.” Fix by reducing lift and keeping operation inside the map.
Slide Valve (screw)
Internal capacity control. Rattling or buzzing as it hunts means the control loop or feedback is sketchy.
IGV (Inlet Guide Vanes, centrifugal)
Pre-swirls inlet air/refrigerant to control capacity. Whistle/whine at certain angles = flow regime + lift combo.
Centrifugal Chillers: Sounds & Fast Fixes
Common sounds: Whine with load, rhythmic chuff (surge), panel rattle.
Whine that grows with load → Lift creeping up.
Check: Condenser water is warm, towers dirty, or ambient high.
Fix: Drop condensing temp (clean, increase tower fan, lower setpoint within limits), or raise evaporator temp a touch.
Chuff/roar every 10–20 seconds → Surge. (This is the one from my story)
Check: IGVs nearly closed, lift high at low flow.
Fix: Increase flow (open bypass), reduce lift; review control limits so load shed doesn’t drive you into the surge line.
IGV whistle at mid angles
Check: Vane misalignment or excessive tip clearances.
Fix: Inspect vane linkage; recalibrate; check for obstructions.
Panel buzz/rattle
Fix: Tighten panels, add captive screws, install rubber isolators; verify that you’re not masking an actual mechanical vibration, log a quick FFT.
Screw Chillers: Sounds & Fast Fixes
Continuous growl, slide-valve buzz, high-pitch whine.
Growl that deepens with load
Check: Oil level/pressure and bearing temps.
Fix: Verify oil circuit; change filters; align couplings; baseline vibration.
Slide-valve buzz (rapid capacity hunting)
Check: Noisy EXV/TXV feedback, unstable PID gains.
Fix: Slow the loop; add derivative/damping; raise superheat 2–3°F to stop floodback that confuses the controller.
High-pitch whine + high amps
Check: Lift; fouled condenser or low condenser flow.
Fix: Same as centrifugal, reduce lift; clean and verify flows.
Knock on pull-down
Check: Superheat too low; liquid carryover.
Fix: Increase SH; insulate TXV bulb; confirm load balance across circuits.
Scroll/Recip Chillers: Sounds & Fast Fixes
Click/chatter, thump on start, gurgle in sight glass.
Check: Contactors/relays; low control voltage; worn scroll check valves.
Fix: Replace contactors; verify transformer taps; evaluate soft-start/VFD.
Thump at start/stop
Check: Liquid migration or rapid equalization.
Fix: Crankcase heat; delay valves; ensure proper off-cycle strategies.
Gurgle at liquid line
Check: Subcooling low; flashing.
Fix: Charge to spec after fixing the reason subcooling is low (condenser issues, airflow/waterflow).
Hydronics & Piping Noises
Cavitation (gravel-in-the-pump sound)
Cause: NPSH available < NPSH required; starved suction, clogged strainers, valves throttled on suction.
Fix: Open suction path; clean strainers; slow the pump; raise suction head.
Air hiss/gurgle in coils
Cause: Entrained air, poor purge.
Fix: Bleed high points; verify auto-vents; keep makeup water degassed where possible.
Water hammer (thump/pop)
Cause: Fast closing valves, long runs, no arrestors.
Fix: Adjust timing; install arrestors; add snubbers on actuators; tweak VFD ramps.
Tube “singing”
Cause: High velocity through partially fouled tubes.
Fix: Brush tubes; verify design gpm/ton; re-balance circuits.
Troubleshooting (Symptom → Cause → Fix)
Symptom: Chiller whooshes and amps spike in the afternoon.
Cause: Lift climbs with ambient/tower heat; condenser approach increases.
Fix: Clean coil/tubes; lower CW setpoint within design; increase tower fan; check water flow.
Symptom: Rhythmic chuffing on low-load nights.
Cause: Centrifugal operating near surge line (low flow, high lift).
Fix: Open bypass; raise evap temp 1–2°F; adjust IGV low-limit; coordinate with BMS to avoid starving flow.
Symptom: Metallic rattle near TXV.
Cause: EXV/TXV hunting, loose bulb, poor insulation.
Fix: Re-strap bulb at 3–4 o’clock, insulate fully; tune superheat; check for non-condensables.
Symptom: Compressor knock on morning startup.
Cause: Floodback/liquid slugging from off-cycle migration.
Fix: Verify crankcase heat; increase superheat; add liquid line solenoid with proper off-delay.
Symptom: High-pitch whine and hot discharge line.
Cause: High lift, dirty condenser, or low condenser water.
Fix: Restore flow; clean; check tower basin level and strainers.
Symptom: Gurgling sight glass with temperature swings.
Cause: Low subcooling; flashing liquid from pressure drop/undercharge.
Fix: Clear restriction; charge to spec; verify drier not plugged.
Symptom: Pump sounds like gravel.
Cause: Cavitation.
Fix: Open suction; reduce speed; verify NPSHa; raise suction head.
Symptom: Loud electrical buzz at starts.
Cause: Pitted contactor; low control voltage.
Fix: Replace contactor; confirm 24 VAC under load; tighten transformer lugs.
Symptom: Panel buzz that worsens with load.
Cause: Mechanical vibration telegraphing through panels.
Fix: Isolate source; add rubber isolators; correct misalignment; log vibration.
Symptom: Whistle from condenser fans (air-cooled).
Cause: Debris or blade angle mismatch.
Fix: Clean guards; replace damaged blades; match angles.
FAQ
What’s a “good” superheat on a chiller?
Depends on the design, but 8–14°F (4–8°C) at the compressor inlet is a common healthy window. Below ~5°F risks floodback; above ~18°F suggests starvation.
How much subcooling do I need?
Typically 8–12°F (4–7°C) at the receiver/outlet is stable. Low subcooling is the #1 reason for gurgle/hiss at the liquid line.
What’s a normal lift?
Many plants live happily around 60–80°F (33–44°C) lift; lower is better for energy/noise. If lift creeps, your noise and amps will too.
Is a steady hiss always bad?
No. Expansion makes noise. Use an ultrasonic to decide if it’s a leak or just flow.
Can BMS tuning really change the sound?
Yes. Hunting loops create chatter and rattle. Small derivative/damping changes make machines sound “calm.”
How do I separate electrical from mechanical noise fast?
Quick FFT (phone + puck or PicoScope). 60 Hz + harmonics = electrical; rotational/blade freq = mechanical.
Do I charge first when I hear gurgling?
Never. Fix the reason subcooling is low (heat rejection, restrictions) before adding refrigerant.
Field Checklist
Record 60 s audio at the chiller and pumps.
Trend suction temp, evap sat temp, condensing temp, liquid line temp, % load/IGV/slide.
Calculate superheat, subcooling, lift, approaches.
Name the sound; match to the playbook.
Inspect and tighten panels, guards, TXV bulbs, and mounts.
Brush/clean coils or tubes if approach is high.
Verify flows (condenser/evap) match design.
Tune PID to stop hunting; raise SH if floodback suspected.
Baseline vibration at identical load point; save to asset folder.
Log before/after values and attach audio to the work order.
Results & ROI
Fewer call-backs: One 60-second audio + 5-point trend cuts debate and repeat visits.
Energy savings: Lower lift = lower kW/ton; quiet machines are usually efficient machines.
Longer life: Early bearing/oil issues are audible; correcting them prevents major rebuilds.
Cleaner approvals: Objective decibels and trends move capital work forward.
Wrap-Up
One action this week: Add lift to your chiller dashboard and listen. If it gets loud when lift rises, you’ve found your lever.
Personal note: The day I started naming sounds out loud, “That’s surge chuff”, was the day customers started trusting my plan. Put a name on the noise; back it with numbers.








Comments