Why It Happens and What It Means

In turbomachinery, one of the most interesting and dangerous patterns is when the journal bearing vibration and bearing metal temperature do not increase smoothly, but instead show a repeated saw-tooth trend:

temperature rises → vibration rises → sudden partial recovery → rises again → drops again → repeats

This is not random noise. In many cases, this pattern can be a strong field indication of unstable lubrication film behavior, and one important hidden cause can be oil varnishing inside the bearing, oil supply system, control valves, coolers, filters, and drain paths.

In simple words:

The machine is not continuously bad.
It is repeatedly entering and escaping a poor lubrication condition.

That is why the trend looks like a saw tooth.

---

1. What Does the Saw-Tooth Pattern Look Like?

A typical trend may look like this:

Bearing temperature

Bearing metal temperature gradually increases from, for example:

82°C → 88°C → 94°C → 99°C

Then suddenly drops back to:

90°C or 92°C

Then starts increasing again.

Vibration

At the same time, vibration may increase gradually:

30 µm → 45 µm → 60 µm → 75 µm

Then suddenly reduce:

50 µm or 55 µm

Then increase again.

The repeated pattern may happen over minutes, hours, or even days depending on:

• bearing load
• oil temperature
• oil viscosity
• oil flow rate
• varnish severity
• machine speed
• reservoir condition
• cooler performance
• valve stickiness
• thermal expansion
• oil degradation level

This is why one cannot diagnose it only from one vibration snapshot. It needs trend correlation.

---

2. Why Varnish Can Create a Saw-Tooth Pattern

Varnish is not only a cosmetic deposit. It is a thin, polar, sticky, oxidation-derived organic deposit that can form on metal surfaces and inside tight-clearance components.

In journal bearings, varnish can affect the system in four main ways:

1. Restricting oil flow
2. Changing bearing surface characteristics
3. Creating unstable oil film thickness
4. Causing intermittent stick-slip or thermal instability

The saw-tooth pattern often happens because varnish-related restrictions are not always steady. They can be temperature-dependent, pressure-dependent, and flow-dependent.

---

3. The Hydrodynamic Journal Bearing Is Very Sensitive to Oil Film Disturbance

A journal bearing depends on a stable hydrodynamic oil wedge.

The shaft does not normally run in the center of the bearing. It runs eccentrically, creating a converging wedge of oil. This oil wedge generates pressure and supports the rotor load.

The bearing needs:

• correct oil viscosity
• correct oil flow
• correct clearance
• correct shaft speed
• correct load direction
• clean oil
• smooth bearing surface
• stable oil temperature

When varnish affects any of these, the bearing film becomes unstable.

A small change in oil film thickness can cause a large change in:

• bearing metal temperature
• rotor eccentricity
• vibration amplitude
• phase angle
• shaft orbit shape
• oil whirl tendency
• pad temperature in tilting pad bearings
• babbitt distress risk

---

4. Mechanism 1: Varnish Restricts Oil Flow, Temperature Rises, Then Flow Partially Recovers

This is one of the most practical explanations.

Varnish can deposit in:

• oil supply orifices
• spray nozzles
• bearing inlet grooves
• oil distribution grooves
• control valves
• pressure regulating valves
• servo valves
• cooler passages
• filter elements
• fine clearances in hydraulic control systems

When the oil flow to the bearing becomes restricted, less heat is removed from the bearing.

So bearing metal temperature starts to rise.

As temperature rises, the oil viscosity drops. Lower viscosity can temporarily improve flow through partially restricted passages. Also, thermal expansion or pressure fluctuation may slightly change the restriction condition.

Then oil flow partially recovers.

Temperature drops.

But the root problem is still there, so varnish restriction again becomes dominant.

This creates:

flow restriction → temperature rise → viscosity drop / pressure change → partial flow recovery → temperature drop → repeat

This is a classic saw-tooth mechanism.

---

5. Mechanism 2: Varnish Changes the Effective Bearing Clearance

In journal bearings, clearance is critical.

Varnish thickness may look very small, but journal bearing oil films are also very small.

A deposit layer of only a few microns can be significant when the minimum oil film thickness is also in the micron range.

Varnish can reduce local clearance and disturb oil wedge formation. This can create:

• higher shear heating
• localized hot spots
• mixed lubrication risk
• unstable rotor position
• increased vibration
• orbit distortion

When the bearing gets hotter, some varnish or soft deposit may become more mobile or less rigid. It may partially dissolve, smear, shear, or relocate.

Then clearance improves slightly.

Temperature drops.

Then deposits rebuild or re-adhere on polar surfaces.

Again, the bearing enters an unstable condition.

This can create a repeated thermal-mechanical cycle.

---

6. Mechanism 3: Hot Oil Dissolves Some Varnish, Cooler Oil Precipitates It Again

Varnish precursors are often more soluble in hot oil and less soluble in cooler oil.

This is very important.

When oil temperature increases, some oxidation byproducts may stay dissolved in the oil. When oil cools down in the reservoir, cooler lines, coolers, or standby zones, these polar degradation products can come out of solution and form soft deposits.

So the machine can behave like this:

1. Oil gets hot near the bearing.
2. Some deposit becomes more soluble or mobile.
3. The restriction temporarily reduces.
4. Bearing temperature drops.
5. Oil cools in other parts of the system.
6. Varnish precursors precipitate again.
7. Deposits return to critical areas.
8. Bearing temperature and vibration rise again.

This is one reason why varnish problems are often cyclic, not linear.

---

7. Mechanism 4: Varnish Causes Stick-Slip Behavior in Oil Control Components

In many turbomachinery systems, bearing oil supply may be affected by:

• pressure control valves
• temperature control valves
• thermostatic valves
• bypass valves
• servo valves
• hydraulic actuators
• trip and control oil circuits
• lube oil cooler bypass valves

Varnish can make valve spools sticky.

A sticky valve does not move smoothly. It may stay in one position while temperature or pressure changes, then suddenly jump to another position.

This creates a saw-tooth trend.

For example:

• cooler bypass valve sticks partially closed
• oil temperature rises
• bearing temperature rises
• vibration increases due to lower viscosity and reduced film stiffness
• valve suddenly moves
• oil cooling improves
• bearing temperature drops
• vibration reduces
• valve sticks again

This is not only a bearing problem. It is a system contamination and oil chemistry problem.

---

8. Mechanism 5: Varnish Reduces Heat Transfer

Varnish is a poor heat transfer layer compared with clean metallic contact surfaces.

When varnish forms on bearing surfaces, cooler tubes, oil passages, or temperature probe pockets, it can reduce heat transfer efficiency.

This causes localized temperature increase.

In journal bearings, the bearing metal temperature may rise because:

• heat generation increases due to higher shear
• heat removal decreases due to lower flow
• thermal conductivity at the surface is impaired
• local hot spots develop

As the bearing temperature rises, oil viscosity drops. Lower viscosity reduces film thickness and damping, which can increase vibration.

Then if flow temporarily improves or the load condition changes, the system cools and vibration drops.

Again, saw-tooth behavior appears.

---

9. Mechanism 6: Oil Whirl / Oil Instability Triggered by Film Stiffness Changes

Journal bearing vibration is strongly affected by bearing dynamic coefficients:

• stiffness
• damping
• cross-coupled stiffness
• film thickness
• eccentricity ratio
• attitude angle

Varnish changes the bearing environment. It may not directly “create vibration” like unbalance, but it changes the support condition of the rotor.

When oil temperature rises:

• viscosity decreases
• film thickness decreases
• damping decreases
• rotor becomes less stable
• subsynchronous vibration can increase
• orbit may become larger or more elliptical
• oil whirl tendency may increase

Then if the temperature drops, viscosity increases again, film stiffness and damping partially recover, and vibration reduces.

So the saw-tooth vibration pattern can be the rotor responding to a cyclic change in oil film condition.

This is why it is important to check not only overall vibration, but also:

• 1X vibration
• subsynchronous components
• phase angle
• shaft centerline position
• orbit shape
• Bode plot
• polar plot
• bearing metal temperatures
• oil inlet and outlet temperature
• oil pressure and flow trend

---

10. Mechanism 7: Wiping / Smearing / Micro-Contact Events

In severe cases, varnish-induced oil starvation or clearance reduction can create intermittent mixed lubrication.

This means the shaft and bearing are no longer fully separated by a stable oil film at every moment.

The machine may experience:

• localized rubbing
• temporary hot spots
• babbitt distress
• wiping marks
• deposit smearing
• high friction events
• sudden temperature spikes

After a small rubbing or wiping event, the deposit may be removed locally, and temperature temporarily decreases.

But the surface is now more vulnerable.

Then the cycle repeats.

This is very dangerous because the machine may appear to “recover” several times before a major failure.

A saw-tooth pattern can therefore be an early warning before:

• bearing wiping
• babbitt damage
• rotor rub
• trip
• high vibration shutdown
• journal scoring
• seal damage
• thrust bearing distress

---

11. Why Temperature and Vibration Rise Together

In a varnish-related journal bearing problem, temperature and vibration may rise together because both are connected to the oil film.

Temperature increases because:

• oil flow is restricted
• heat removal is reduced
• friction/shear increases
• hot spots develop
• cooler efficiency may be reduced
• oil viscosity becomes too low

Vibration increases because:

• film stiffness changes
• damping reduces
• rotor eccentricity changes
• shaft centerline moves
• oil whirl may be triggered
• rubbing risk increases
• bearing support becomes unstable

So the common root is usually not “temperature problem” or “vibration problem” alone.

It is:

unstable hydrodynamic lubrication caused by degraded oil chemistry and varnish deposits.

---

12. Why the Pattern Is Saw-Tooth Instead of Continuous Increase

This is the key question.

If a bearing is mechanically damaged, the trend may continuously worsen.

But varnish-related problems can be reversible for a short time.

Varnish is affected by:

• temperature
• oil solubility
• flow velocity
• pressure drop
• chemical equilibrium
• load
• viscosity
• additive condition
• polar attraction to surfaces

Therefore, the system may repeatedly move between two conditions:

Condition A: unstable / restricted

• varnish restricts flow
• oil film becomes thin
• temperature rises
• vibration rises

Condition B: temporary recovery

• higher temperature reduces viscosity
• flow improves slightly
• deposit softens or moves
• valve suddenly changes position
• oil film partially recovers
• temperature drops
• vibration drops

Then Condition A returns.

That cyclic instability gives the saw-tooth shape.

---

13. What Varnishing Pattern May Be Seen During Inspection

During inspection, varnish-related journal bearing issues may show:

• amber, brown, orange, or dark sticky deposits
• lacquer-like film on babbitt
• deposits in oil grooves
• blocked or partially restricted oil holes
• staining on bearing shells
• darkened drain oil
• deposits on cooler tubes
• sticky control valves
• filter plugging with brown residue
• dark deposits in low-flow zones
• varnish on reservoir walls above oil level
• discoloration around hot zones
• abnormal MPC patch color

In more severe cases:

• wiping marks
• smeared babbitt
• localized overheating
• shaft rub marks
• polish marks
• coking-like dark deposits near high-temperature zones

The bearing may not always look catastrophically failed. Sometimes the evidence is subtle.

---

14. Oil Analysis Indicators That Support This Diagnosis

The best way to support a varnish-related saw-tooth diagnosis is to combine vibration, temperature, and oil analysis.

Important tests include:

MPC — ASTM D7843

Measures varnish potential. High MPC indicates high tendency of polar degradation products to form deposits.

The patch photo is very important, not just the number.

A dark brown, orange, grey, or black patch can give clues about oxidation, thermal degradation, soot-like contamination, or high-temperature stress.

RULER / LSV — ASTM D6971

Measures remaining antioxidant health.

Low phenolic and/or aminic antioxidant levels indicate reduced oil oxidation defense.

RPVOT — ASTM D2272

Measures oxidation stability reserve.

A reduced RPVOT value supports oil aging and oxidation risk.

TAN — ASTM D664

Increasing TAN indicates formation of acidic oxidation products. Acidic polar species can contribute to varnish chemistry and deposit formation.

Particle count

May not always detect soluble varnish precursors. A clean particle count does not mean low varnish risk.

FTIR

Can indicate oxidation, nitration, additive depletion, or contamination patterns.

Karl Fischer water

Water accelerates oxidation, additive depletion, corrosion risk, and can worsen deposit formation.

Membrane patch weight / gravimetric deposit tendency

Useful when varnish potential and insoluble loading are high.

---

15. Why Particle Count Alone May Mislead You

This is a very important point.

A varnishing system can have acceptable ISO cleanliness code but still have severe varnish potential.

Why?

Because many varnish precursors are dissolved or submicron soft contaminants, not hard particles counted by standard particle counters.

You may have:

• good particle count
• low visible contamination
• acceptable filter differential pressure

but still have:

• high MPC
• sticky valves
• varnished bearing surfaces
• temperature instability
• vibration saw-tooth pattern

This is why varnish is a chemistry and solubility problem, not only a filtration problem.

---

16. How to Confirm the Root Cause in the Field

A strong investigation should include the following.

A. Trend correlation

Overlay:

• bearing metal temperature
• shaft vibration
• oil inlet temperature
• oil outlet temperature
• oil pressure
• oil flow
• filter differential pressure
• cooler inlet/outlet temperature
• load
• speed
• ambient temperature
• valve position
• reservoir temperature

Look for time correlation.

For example:

• Does vibration rise after bearing temperature rises?
• Does temperature rise after oil pressure drops?
• Does the drop happen when cooler valve changes position?
• Does vibration reduce after oil temperature changes?
• Is there a time lag?

The time lag is very important.

B. Vibration analysis

Check:

• 1X amplitude and phase
• subsynchronous components
• shaft orbit
• shaft centerline
• Bode and polar plots
• bearing-specific vibration
• horizontal versus vertical response
• proximity probe gap voltage trend

If it is varnish-induced hydrodynamic instability, you may see changes in orbit and shaft centerline, not only overall vibration.

C. Thermal analysis

Check:

• individual bearing thermocouples
• loaded zone temperature
• pad temperature differences
• oil inlet and outlet delta T
• sudden changes in bearing metal temperature
• thermal lag between oil temperature and metal temperature

A bearing metal temperature rise without matching inlet oil temperature rise may suggest localized bearing heat generation or poor flow.

D. Oil analysis

Perform:

• MPC with patch photo
• RULER
• RPVOT
• TAN by ASTM D664
• FTIR oxidation
• Karl Fischer water
• particle count
• elemental analysis
• ferrography or analytical microscopy if rubbing is suspected

E. Mechanical inspection

During shutdown, inspect:

• bearing oil grooves
• inlet holes
• drain paths
• babbitt surface
• journal surface
• oil rings if applicable
• cooler surfaces
• control valves
• filter elements
• reservoir deposits

---

17. Differential Diagnosis: Do Not Blame Varnish Too Quickly

A saw-tooth pattern can also be caused by non-varnish issues.

Important alternatives include:

• defective temperature control valve
• cooler fouling
• unstable cooling water flow
• oil pressure control valve hunting
• low oil level
• air entrainment
• foaming
• water contamination
• wrong oil viscosity
• misalignment and thermal growth
• rotor rub
• bearing clearance issue
• damaged babbitt
• plugged oil orifice by debris
• unstable process load
• compressor surge
• steam turbine load swings
• generator electrical load variation
• instrumentation problem
• loose thermocouple
• proximity probe issue
• grounding or signal noise

The correct approach is not to say:

Saw tooth = varnish.

The correct approach is:

Saw-tooth vibration and temperature may indicate cyclic instability of the lubrication film. Varnish is one possible and important root cause, especially when supported by MPC, RULER, TAN, visual deposits, sticky valves, or oil flow instability.

---

18. Practical Field Example

Imagine a steam turbine journal bearing with normal bearing metal temperature of 82°C.

After months of service, the trend starts showing:

• bearing temperature cycling between 86°C and 98°C
• vibration cycling between 35 µm and 70 µm
• oil inlet temperature mostly stable
• oil pressure slightly fluctuating
• filter differential pressure normal
• particle count acceptable
• MPC high
• RULER antioxidants depleted
• TAN increasing
• brown deposits found on filter element and oil control valve

This is a classic situation where the machine may not have a simple mechanical fault.

The oil chemistry is damaging the lubrication system performance.

The bearing is receiving oil, but not always in the right flow, cleanliness, chemistry, and film condition.

The saw-tooth pattern is the machine saying:

“Sometimes I can build a good oil film, sometimes I cannot.”

---

19. Why Varnish Removal Can Stabilize the Trend

A correct varnish mitigation strategy should remove both:

• insoluble deposits
• soluble varnish precursors

If only insoluble particles are removed, the root chemistry may remain in the oil.

For turbine oils, a good varnish management strategy should aim to:

• reduce MPC
• stabilize antioxidant consumption
• reduce acidic degradation products
• improve oil solubility balance
• restore valve movement reliability
• improve oil flow stability
• reduce bearing temperature cycling
• reduce vibration instability

When varnish potential reduces, the system usually becomes more stable because oil passages, valves, and bearing surfaces are less affected by sticky polar deposits.

The saw-tooth trend may become flatter.

---

20. Recommended Diagnostic Action Plan

Step 1: Confirm the trend

Do not rely on one alarm event. Extract historian data for at least:

• 7 days
• 30 days
• 90 days if possible

Overlay vibration, temperature, oil pressure, oil temperature, and load.

Step 2: Check if vibration follows temperature

If vibration increases after temperature increases, the primary driver may be thermal/oil-film related.

If temperature increases after vibration increases, the primary driver may be rubbing or mechanical instability.

Step 3: Check oil inlet versus bearing metal temperature

If oil inlet temperature is stable but bearing metal temperature rises, suspect:

• restricted bearing oil flow
• local bearing distress
• clearance issue
• deposit insulation
• oil film instability

Step 4: Perform varnish oil analysis

Minimum package:

• MPC ASTM D7843 with patch photo
• RULER ASTM D6971
• TAN ASTM D664
• RPVOT ASTM D2272
• FTIR oxidation
• Karl Fischer water
• particle count

Step 5: Inspect system restrictions

Check:

• bearing oil inlet orifice
• filters
• strainers
• valves
• coolers
• bearing oil grooves
• drain line
• reservoir deposits

Step 6: Analyze vibration deeply

Do not only look at overall vibration.

Check:

• 1X amplitude and phase
• subsynchronous vibration
• shaft orbit
• shaft centerline
• bearing-specific trends
• phase stability

Step 7: Decide corrective action

Depending on severity:

• operate varnish removal system online
• improve oil temperature control
• inspect and clean sticky valves
• check oil flow to affected bearing
• flush local bearing supply lines
• replace severely varnished filters
• inspect bearing during outage
• review oil type and oxidation condition
• prevent water and air ingress
• improve reservoir breathing control
• improve sampling and monitoring frequency

---

21. Important Warning

A saw-tooth pattern is dangerous because it can create false confidence.

The machine may recover several times.

Operators may think:

“It came down again, so it is okay.”

But repeated recovery does not mean the root cause disappeared.

It means the machine is cycling between acceptable and unacceptable lubrication regimes.

Every cycle may cause additional thermal stress, additive depletion, deposit movement, and bearing surface damage.

In journal bearings, this can progress from:

temperature fluctuation → vibration fluctuation → oil film instability → intermittent rubbing → wiping → trip/failure

---

22. Khash Summary

When journal bearing temperature and vibration show a saw-tooth pattern in a varnished turbine or compressor oil system, the most probable technical explanation is cyclic instability of the hydrodynamic oil film.

Varnish can create this instability by:

• restricting oil flow
• sticking control valves
• reducing effective bearing clearance
• insulating heat transfer surfaces
• changing surface energy and oil wettability
• disturbing oil film stiffness and damping
• promoting intermittent mixed lubrication
• dissolving and re-precipitating with temperature changes

The correct diagnosis requires combining:

vibration analysis + temperature trend + oil flow/pressure data + varnish oil analysis + inspection evidence

Never diagnose varnish from MPC alone.

Never diagnose bearing failure from vibration alone.

The machine, the bearing, and the oil must be analyzed as one reliability system.

Final thought:

A saw-tooth trend is not a normal fluctuation.
It is the fingerprint of an unstable system repeatedly trying to recover.
In varnished turbine oil systems, it may be the early warning before the bearing stops forgiving us.