Vibration Isolation: Design Method, Mount Selection, and the Mistakes That Undo Everything
Your job is not to put rubber under a machine. Your job is to break the mechanical path between the vibration source and everything around it. Here's the engineering behind that — and the field data to prove it works.
The Physics: Mass, Spring, and What Actually Isolates
Every vibration isolation system is the same thing underneath: a mass sitting on a spring. The machine is the mass. The mount is the spring. And between them, there's some damping — the material's ability to convert vibration energy into heat.
Engineers model this as a mass–spring–damper system with three parameters: mass \(m\) (kg), stiffness \(k\) (N/m), and damping coefficient \(c\) (N·s/m). From these three numbers, everything else follows.
Natural frequency: the number that determines everything
The most important parameter is the system's natural frequency — the frequency it would oscillate at if you pushed the machine down and released it. Lower stiffness or higher mass gives a lower natural frequency:
This number is everything. It determines whether your mounts isolate, do nothing, or make things catastrophically worse. The entire design process is about getting this number right relative to the machine's running frequency.
Transmissibility: how much gets through
The ratio of force transmitted to the foundation versus force generated by the machine is called transmissibility (\(T\)). In a simplified undamped form:
Where \(f_{exc}\) is the excitation frequency (machine running speed in Hz) and \(f_n\) is the isolator natural frequency. When \(T = 0.1\), only 10% of the vibration force reaches the foundation — that's 90% isolation. When \(T = 1\), you're transmitting everything. When \(T > 1\), the mounts are amplifying vibration.
The Three Zones — and Why One of Them Makes Things Worse
The transmissibility equation creates three distinct operating zones. Understanding them is the difference between isolation that works and mounts that make the problem worse.
Amplification zone
Resonance. The mounts amplify vibration instead of reducing it. This is the danger zone — if your mounts put the natural frequency near running speed, vibration gets worse than without mounts. Much worse.
No-benefit zone
Running speed is too close to natural frequency. Mounts don't help — vibration transfers with little or no reduction. You've spent money on rubber for nothing.
Isolation zone
Real isolation only begins when excitation exceeds 1.41× the natural frequency. For practical industrial use, target at least 3:1 or 4:1 ratio. A 4:1 ratio gives approximately 93% force reduction.
The single most common isolation failure I see is mounts that are too stiff. Someone puts thin rubber pads under a 1,500 RPM pump — the pads deflect 0.5 mm, giving a natural frequency around 22 Hz. Running speed is 25 Hz. Ratio: 1.14:1. You're sitting right in the amplification zone. The "isolated" pump vibrates worse than it would bolted directly to the floor. The fix: softer mounts with more deflection, or spring isolators.
| Frequency ratio (f_exc / f_n) | Transmissibility | Isolation effect |
|---|---|---|
| 1.0 | ∞ (resonance) | Amplification — dangerous |
| 1.41 (√2) | 1.0 | Crossover — no benefit |
| 2.0 | 0.33 | 67% reduction |
| 3.0 | 0.13 | 87% reduction |
| 4.0 | 0.07 | 93% reduction |
| 5.0 | 0.04 | 96% reduction |
Design Workflow: Sizing Mounts by Static Deflection
The practical way to size vibration mounts in the field uses static deflection — how much the mount compresses under machine weight. This sidesteps the need for stiffness tables and spring rate specifications. One number — millimeters of deflection under load — tells you the natural frequency.
Or reversed: \(\delta_{st} = \left(\frac{5}{f_n}\right)^2\) cm. This is the formula you'll use most.
Determine excitation frequency
Find the lowest operating RPM. Convert: \(f_{exc} = \text{RPM} / 60\). A fan at 1,500 RPM gives \(f_{exc} = 25\) Hz. A diesel generator at 750 RPM gives 12.5 Hz. Always use the lowest speed the machine runs at — that's where isolation is weakest.
Choose target natural frequency
Divide excitation frequency by 3–4. A 4:1 ratio provides 93% isolation — that's the standard industrial target. For the 25 Hz fan: \(f_n = 25/4 = 6.25\) Hz. For the 12.5 Hz generator: \(f_n = 12.5/4 \approx 3.1\) Hz.
Calculate required static deflection
For the fan at \(f_n = 6.25\) Hz: \(\delta_{st} = (5/6.25)^2 = 0.64\) cm = 6.4 mm. Select mounts that deflect 6–7 mm under the machine weight. For the generator at \(f_n = 3.1\) Hz: \(\delta_{st} = (5/3.1)^2 = 2.6\) cm = 26 mm. That's spring isolator territory — no rubber mount deflects 26 mm.
Distribute load across mount points
Determine total weight and center of gravity (CG). If CG is centered, load splits evenly across mounts. If the motor or gearbox shifts CG to one side, mount loads differ. The design target is equal deflection at every mount — that keeps the machine level and preserves shaft alignment. This can mean different stiffness at different corners.
Select mount type
Now match the deflection requirement to the mount technology. See the next section for a detailed comparison. The short version: rubber for small deflections (high-speed equipment), springs for large deflections (low-speed), air springs for ultra-low frequency (precision equipment).
Isolate all rigid connections
Install flexible connectors on pipes, ducts, and cable trays. This step is where most isolation projects fail — see the section on vibration bridges below.
Verify with vibration measurement
Measure vibration at the foundation before and after installation. The Balanset-1A in vibration meter mode reads mm/s directly — place the sensor on the support structure and compare the 1× running frequency component with and without the machine running. Target: 80–95% reduction.
Mount Types: Rubber, Springs, Air Springs, and Inertia Bases
Elastomeric (rubber-metal) mounts
Best for high-speed equipment: pumps, electric motors, fans above 1,500 RPM. The rubber provides built-in damping that limits motion during start/stop resonance pass-through. Small deflection means the machine stays stable. Downsides: limited isolation at low frequencies because deflection is too small; rubber ages and hardens over time, reducing effectiveness.
Spring isolators
Best for low-speed equipment: fans below 1,000 RPM, diesel generators, compressors, HVAC chillers, rooftop units. Large deflection gives low natural frequency. Many designs include rubber pads at the base to block high-frequency noise transmission through the coils — bare steel springs transmit structure-borne noise efficiently.
Air springs
Best for precision equipment: coordinate measuring machines, electron microscopes, laser systems, sensitive test benches. Extremely low natural frequency. Requires compressed air supply and automatic leveling control. Not practical for most industrial machinery — too soft, too complex, too expensive. But unmatched when you need sub-1 Hz isolation.
Inertia bases (inertia blocks)
Not an isolator by itself — a platform that adds mass. Bolt the machine to a concrete or steel inertia base, then mount the base on springs. This increases \(m\), lowers \(f_n\), reduces vibration amplitude, lowers the center of gravity, and improves lateral stability. Required when the machine is too light for stable spring mounting, or when large unbalanced forces cause excessive rocking.
Above 1,500 RPM: elastomeric mounts usually sufficient. 600–1,500 RPM: depends on required deflection — calculate and check. Below 600 RPM: spring isolators almost always. Below 300 RPM: large spring deflection + inertia base. The deflection calculation (step 3 above) always gives the definitive answer.
Foundation Effects and Vibration Bridges
Rigid vs flexible foundations
Isolation calculations assume the foundation is infinitely rigid — it doesn't move. Ground-level concrete slabs are close enough. But upper building floors, steel mezzanines, and rooftop frames are not. These are flexible foundations — they have their own natural frequency.
If you mount isolators on a flexible floor, the floor deflection adds to the isolator deflection. That shifts system frequencies in unpredictable ways. The combined "machine–isolator–floor" system can develop resonances that don't appear in the calculation. For flexible floors, you either need to account for the floor's dynamic properties (which requires structural analysis) or over-design the isolation with extra margin — aim for a 5:1 or 6:1 frequency ratio instead of 4:1.
Vibration bridges: the silent killer of isolation
This is the single most common reason that "properly designed" isolation fails in the field. You install beautiful spring mounts, calculate everything, measure the foundation — and vibration is still there. Why? Because a rigid pipe, duct, or cable tray connects the machine frame directly to the building structure, completely bypassing the mounts.
Every rigid connection is a vibration bridge. Pipes, ductwork, conduit, drain lines, compressed air lines — any of them can short-circuit the isolation. The fix is simple in principle and often painful in practice: install flexible connectors (bellows, braided hose, expansion loops) on every pipe and duct that attaches to the isolated machine. Provide slack in cables. Check that no rigid brackets or hard stops touch the machine frame after installation.
I've measured foundation vibration on machines with correctly sized spring mounts where 60–70% of the transmitted vibration came through the piping, not through the mounts. The springs were doing their job. The two cooling water pipes bolted directly to both the pump and the floor above were undoing it.
Field Report: Chiller Compressor on the Third Floor
A commercial building in Southern Europe had a 90 kW screw chiller installed on the third-floor mechanical room. The compressor runs at 2,940 RPM (49 Hz). Residents on the second floor complained of low-frequency hum and vibration transmitted through the concrete slab.
The chiller was sitting on OEM rubber mounts — thin pads that deflected about 1 mm under load. That gives a natural frequency of approximately \(f_n = 5/\sqrt{0.1} \approx 16\) Hz. Frequency ratio: 49/16 = 3.1:1. Barely adequate on paper, but the flexible floor slab pushed the effective system frequency higher. And three refrigerant pipes ran rigidly from the compressor to the header — classic vibration bridges.
We replaced the rubber pads with spring isolators (25 mm deflection, \(f_n \approx 3.2\) Hz, ratio 15:1) and installed braided flexible connectors on all three refrigerant lines. Before/after vibration at the second-floor ceiling, measured with a Balanset-1A on the slab underside:
90 kW screw chiller, 2,940 RPM, third-floor installation
OEM rubber pads replaced with spring isolators (25 mm deflection). Rigid refrigerant pipes replaced with braided flexible connectors. Measurement point: second-floor ceiling slab, directly below compressor.
The complaints stopped. The measured 0.3 mm/s at the floor is below the ISO 10816 perception threshold for most people. The springs alone would not have achieved this — about 40% of the original transmitted vibration was coming through the rigid piping. Both fixes were necessary.
Need to measure vibration before and after isolation?
The Balanset-1A works as both a vibration meter and a balancer. Measure mm/s at the foundation, verify your isolation design, and balance the machine if needed. One device, two functions.
Common Mistakes That Undo Isolation
1. Mounts too stiff (not enough deflection). This is the most frequent error. Thin rubber pads with 0.5–1 mm deflection under heavy equipment give a high natural frequency. If it's near running speed, you get amplification, not isolation. Always calculate deflection first — don't just "put rubber under it."
2. Rigid piping connections. See above. Every rigid pipe, duct, and conduit that touches both the machine and the building structure is a vibration bridge. Flexible connectors on all lines. No exceptions.
3. Soft foot. If the machine frame is twisted or the mounting surface is uneven, one or two mounts carry most of the load while others are nearly unloaded. This creates unequal deflection, tilts the machine, stresses the shaft alignment, and shortens mount life. Check the frame with a feeler gauge before installing mounts. Shim if needed.
4. Lateral instability. Vertical-only springs can rock sideways, especially if the machine has high CG or large horizontal forces. Use housed spring mounts with built-in lateral restraint, or add snubbers. For machines with very high starting torque (large motors, compressors), lateral stability is critical.
5. Start/stop resonance pass-through. Every machine passes through the isolator's natural frequency during acceleration and deceleration. If the machine ramps slowly (VFD-driven, or diesel generators warming up), it spends significant time in the resonance zone. Solution: mounts with higher damping (elastomeric elements or friction dampers on springs) to limit resonance amplitude during pass-through.
6. Ignoring the floor. Putting spring mounts on a flexible mezzanine without accounting for the floor's dynamic response creates a coupled system with unpredictable resonances. Either stiffen the floor, increase the frequency ratio margin, or do a proper structural dynamic analysis.
Verification: How to Prove It Works
Design calculations tell you what should happen. Vibration measurement tells you what did happen. Always verify.
The test is simple: place a vibration sensor on the foundation or support structure. Measure with the machine off (background). Measure with the machine running at full speed. Compare the vibration velocity at the 1× running frequency. Effective isolation shows 80–95% reduction compared to the pre-isolation condition (or compared to a rigid-mount reference).
A Balanset-1A in vibration meter mode does this directly. Set it to display mm/s, place the accelerometer on the support structure, and read the value. If you also need FFT spectrum analysis — to distinguish the 1× component from other sources — the Balanset-1A includes that mode.
| Foundation vibration (mm/s) | Interpretation | Action |
|---|---|---|
| < 0.3 | Below perception threshold | No complaints expected |
| 0.3 – 0.7 | Perceptible to sensitive occupants | Acceptable for industrial, marginal for commercial |
| 0.7 – 1.5 | Clearly perceptible | Investigation needed — check mounts and connections |
| > 1.5 | Complaints likely, possible structural concern | Redesign isolation — softer mounts, flexible pipes, or inertia base |
Frequently Asked Questions
Measure it. Prove it. Fix it.
Balanset-1A: vibration meter + spectrum analyzer + rotor balancer in one kit. Verify your isolation design, diagnose the source, balance if needed. Ships worldwide via DHL. 2-year warranty.
EN 
AR 
AZ 
BG 
ZH 
HR 
CS 
DA 
NL 
ET 
FI 
FR 
DE 
KA 
EL 
HE 
HI 
HU 
ID 
IT 
JA 
KO 
LV 
LT 
MS 
NB 
FA 
PL 
PT_BR 
PT 
RO 
SR 
SK 
SL 
ES 
SV 
TH 
TR 
UR 
UK 
VI 
0 Comments