Sludge and Varnish in Turbine

Systems

Jim Fitch, Noria Corporation
Tags: turbine lubrication, varnish, contamination control
A wide range of fluid degradation issues can affect lube oil systems. One growing concern is the
presence of sludge and varnish. This condition can occur in even the most well-maintained
machines. Surprisingly, it can also arise when oils are not particularly old or contaminated, as
well as in thermally robust synthetic lubricants and hydraulic fluids.
Figure 1. Varnish Formation on Inlet Guide
Vane (IGV) Valve from a Gas Turbine
Sludge and Varnish Defined
The products of lubricant degradation are called sludge and varnish. These products are
generally unstable in the oil and, as such, are prone to form deposits or separate from the bulk
oil. n certain instances, deposits form on machine surfaces at the e!act location where the oil
has degraded. n other cases, the oil degrades in one location but the insoluble degradation
products are carried elsewhere by the moving fluid forming deposits on surfaces.
Figure 2. Varnish Formation on lain !earing and
Shaft ("ourtesy "erestar Deuts#hland)
Over time, some deposits can thermally cure to a tough enamel-like coating. Other types of
deposits, generally in cooler "ones, remain soft or gummy, and may in some cases appear clear
and grease-like. The following are e!amples of where sludge and varnish might occur in turbine
lube oil and electrohydraulic control systems:
black crusty deposits on mechanical seals
gold adherent films on valves
charcoal-like deposits on babbitt sleeve bearings
gooey-brown accumulations on oil filters
black scabby deposits on mechanical seal surfaces thrust-bearing pads
carbonaceous residue on mechanical surfaces
The deposits that form on sensitive machine surfaces interfere with the flow of fluid and the
machine#s mechanical movements, contribute to wear and corrosion or may simply $ust cling to
surfaces and impair heat transfer. %or e!ample, deposits on the spool and bore of a servo
control valve can tighten the interference fit.
Other types of sludge and varnish-induced failures include silt lock valve failures, plugged
orifices, damaged mechanical seals, plugged discharged ports, $ournal-bearing wear &disrupted
hydrodynamic film production', premature plugging of oil filters and impaired oil cooler
performance.
Turbine Systems
n turbine systems, few failure conditions can disrupt operation as (uickly and completely as a
varnished and sei"ed-up control valve operation. Sludge can gum up flow controls, strainers
and critical oil pathways. n recent years, there has been an increasing number of reported
cases associated with varnish and sludge formation in turbine-generator applications.
n turbine applications where reliability is essential and operating conditions are stressful, there
is a need for incipient advisories that report developing varnish potential. The risks are
especially pronounced in high-pressure, servo-controlled hydraulic systems operating in
continuous and high-duty service.
Degradation $ethods and %nalysis
)efore discussing the methods for analy"ing sludge and varnish in turbine systems, it is helpful
to understand the lubricant degradation methods in turbine lube oil and hydraulic systems.
!ul& 'il '(idation
O!idation causes degradation of the bulk oil over time. *nder mild machine operating conditions
and clean environment, o!idation will occur gradually once the o!idation inhibitor additives have
depleted. +owever, elevated temperatures accelerate the o!idation process , the general rule
of thumb is that for every -./0 &-1/%' increase in operating temperature, the rate of o!idation
doubles &Arrhenius 2ate 2ule'. 3ater, metals &such as iron or copper particles' and aeration
also act as catalysts to speed up this process. O!idation by-products develop and coalesce into
suspended insolubles in the oil as the rate of o!idation increases, resulting in progressive
formation of sludge and varnish in the system.
Thermal and "om)ressive !ase 'il Degradation
Thermal degradation occurs due to adiabatic compression from entrained bubbles or when oil
comes in contact with a hot surface. 3hen the machine surface is greater than 4../0 &5../%',
depending on oil type, thermal degradation can start. Such heat can come from gas
combustion, steam and highly loaded frictional surfaces. Often the aeration occurs due to tank
agitation, plunging oil returns or surface lapping. Suction line leaks, pump seal leaks and venturi
"ones &vena contracta regions' can also introduce air into circulating fluids.
2egardless of the means of entrainment, the forcing function that leads to sludge and varnish is
in place. The failure will now proceed along one of two pathways. )oth involve adiabatic
compression in either the load "one of a lubrication system or the pressured "one of a hydraulic
control system. Adiabatic compression occurs when air bubbles travel from low pressure to high
pressure. The air bubble compresses rapidly &implosion', resulting in the entrapment and
concentration of heat and e!treme rise in the oil#s temperature. The temperatures reached
&typically greater than -,.../%' are often more than sufficient to thermally degrade the oil.
*le#trostati# Dis#harge
Studies have been conducted on the effects of static discharge in hydraulic systems since the
-67.s. Static discharge is a form of locali"ed thermal degradation, which has been discussed
previously. 2ecently, attention has been directed to fluid electrification and static discharge as a
prominent contributor to sludge and varnish formation in turbine systems.
8lectrostatic charge generation occurs in fluid systems as a result of internal molecular friction
and electric potential between the fluid and machine surfaces &particularly where no boundary
films develop, such as the interstices of an oil filter'. 9any factors contribute toward the
magnitude of the static charge within the oil: however, grounding of the machine itself has little
impact toward mitigating charge propagation. This is due to the oil being nonconductive, which
effectively self-insulates the charged fluid "ones from grounded surfaces. Once these charges
build up in the working fluid "ones, including reservoirs, the subse(uent static discharging,
similar to lightning strikes through the fluid, may cause locali"ed thermal-o!idative oil
degradation.
Figure +. %#id ,umber (%,) for Turbine 'il *()osed to Stati# "harges
after -ero. Si( and ,ine $onths in Storage
According to research conducted by ;r. Sasaki, spark discharges can reach temperatures as
high as 4.,.../0. Additionally, he has found that although the initial failure is locali"ed, the
chemical degradation process is autocatalytic. n one study, a turbine oil sample was e!posed
to spark discharges. Acid number &A<' was measured on the oil samples $ust after initial
e!posure and after si! and nine months of oil storage. 2esults of this test are shown in %igure =.
A< was measured immediately after the tests and were the same, regardless of the number of
sparks e!posed to the turbine oil, suggesting that the bulk oil degradation had not yet started.
+owever, there was a significant change in A< after being stored, indicating that the oil
o!idation process had advanced past the induction period, even when stored at room
temperature in the dark.
Stages of
Failure
"onditions of
the Fluid to be
Tested
"orres)onding
%nalyti#al $ethod
Stage 'ne
&precursor stage'
, 2oot causes
that propel the
fluid into the ne!t
stage
9easurable
o!idation root
causes:
-. >article
contamination
4. 3ear debris
=. 3ater
contamination
5. Antio!idant
-. ?arious
laboratory particle
counting methods
&optical, pore
blockage and
microscopic'
4. 8lemental
analysis,
ferrography, ferrous
additive depletion density analysis
=. ?arious
laboratory methods
including @arl
%ischer and %T2
5. 9ethods include
%T2, 2>?OT,
2*A82 and
+>;S0
Stage T/o
&transitory stage'
, The beginning
of base oil
degradation and
increasing risk of
varnishing
formation
The presence of
soluble o!idation
products to form:
-. 0arbo!ylic
acids, formic
acids, etc.
4. +igh-density
resins, polymers,
etc.
=. 0hromophoric
compounds &color
bodies'
-. %T2 &-,7=. cm-
-', A<
4. 0oagulated o!ide
insolubles
=.
Spectrophotometry,
patch colorimetry
Stage Three
&attack stage' ,
9ore advanced
o!idation phase
where significant
risk
of varnish,
corrosion and
impaired
lubrication is
present
The presence of
both soluble and
insoluble
o!idation
products to form:
-. 9etal
carbo!ylate
soaps and sludge
4. Suspended
o!ide insolubles
=. 2ising oil
viscosity
-. Analysis of
sludge &bottom
sediment, organic
filter sediment, etc.'
by elemental
analysis, %T2
&-,7=. cm--',
B0C9S
4. *ltracentrifuge,
patch test,
patch colorimetry
=. ?iscometry
Table 1. $ethods for %naly0ing
'(idation1indu#ed Sludge and Varnish
%nalysis Te#hni2ues for $onitoring Sludge and Varnish
Analytical techni(ues for measuring sludge and varnish vary depending on the failure
mechanism. These methods fall into three categories, according to the stage of contaminant or
degradation product targeted. The first stage is a precursor condition. Specifically, it relates to
the contaminants that give lubricants a propensity to form o!ides and other oil breakdown
products. Stage two is a transitory stage. Typically, it consists of soluble o!ides and other
dissolved impurities that might later synthesi"e to form insolubles. The condition that aids the
transition from stage two &o!ide soluble' to stage three &o!ide insoluble' may be cold oil
temperature or simply a supersaturated state &a concentration of o!ides in e!cess of the varnish
saturation point'.
Stages of
Failure
"onditions of
the Fluid to be
Tested
"orres)onding
%nalyti#al $ethod
Stage 'ne
&precursor stage'
9easurable root
causes:
-. Air release tests,
&AST9 ;=547'
, 2oot causes
that propel the
fluid into the ne!t
stage
-. mpaired air-
release
properties
4. 0ross
contamination
&mi!ed lubricants'
=. 0ontamination
with polar
impurities &dirt,
water, o!ides,
etc.'
4. %T2, elemental
analysis, etc.
=. ?arious standard
lab tests for such
impurities
Stage T/o
&transitory stage'-
The beginning of
base oil
degradation and
increasing risk of
varnishing
formation
The presence of
soluble thermal
degradation
products:
-. <itric o!ides
4. 0hromophoric
compounds &color
bodies'
=. Aocali"ed
thermal failure
conditions
-. Thick-cell &D..
Em' %T2 for
nitration &-,F=6 cm-
-'
4.
Spectrophotometry,
patch colorimetry
=. >attern testing
&combining above
tests with 2>?OT,
A<, viscosity, etc.'
to differentiate from
bulk oil degradation
Stage Three
&attack stage' ,
9ore advanced
degradation
phase where
significant risk of
varnish is present
The presence of
both soluble and
insoluble thermal
degradation
products to form:
-. Suspended
high molecular-
weight nitric
o!ides and
similar resinous
insoluble
compounds
4. Sludge and
deposits
-. *ltracentrifuge,
patch test
colorimetry,
coagulated
insolubles
4. Analysis of
sludge &bottom
sediment, organic
filter sediment, etc.'
by elemental
analysis, %T2,
B0C9S
Table 2. $ethods for %naly0ing Thermal
Degradation1indu#ed Sludge and Varnish
Stage three is an attack stage. 3hile conditions in stages one and two may not damage the
machine, stage three contaminants are surface active, potentially causing deposits, corrosion or
mechanical wear. Stages one and two are more proactive, meaning that analytical methods
detect aberrant conditions before damage or harm to the machine occurs. Stage three is a
predictive maintenance stage where an incipient or impending machine failure is already in
progress.
Stages of
Failure
"onditions of
the Fluid to be
Tested
"orres)onding
%nalyti#al $ethod
Stage 'ne
&precursor stage'
2oot causes:
-. 8!ceedingly
-. @arl %ischer, dew
point meters
, 2oot causes
that propel the
fluid into the ne!t
stage
dry oil
4. 8!ceedingly
clean oil
=. *no!idi"ed oil
5. ;rop in oil
conductivity
4. >article counters
&various', total
insolubles
=. A<, 2>?OT,
%T2 &-,7=. cm--'
5. 0onductivity
meters, dielectric
breakdown testers,
dielectric constant
meters, etc.
Stage T/o
&transitory stage'-
The beginning of
base oil
degradation and
increasing risk of
varnishing
formation
The presence of
soluble thermal
degradation
products:
-. 0hromophoric
compounds &color
bodies'
4. Aocali"ed
thermal failure
conditions
-.
Spectrophotometry,
patch colorimetry
4. >attern testing
&combining above
tests with 2>?OT,
A<, viscosity, etc.'
to differentiate from
bulk oil
degradationG
Stage Three
&attack stage' ,
9ore advanced
degradation
phase where
significant risk of
varnish is present
The presence of
soluble and
insoluble thermal
degradation
products:
-. Suspended
high molecular-
weight similar
resinous
insoluble
compounds
4. Sludge and
deposits
=. Bas evolution
-. 0oagulated total
insolubles
4. Analysis of
sludge &bottom
sediment, filter
sediment, etc.' by
elemental analysis,
%T2 &-,7=. cm--',
B0C9S
=. %lash point, gas
chromatography for
acetylene and other
gasesG
*Note: Analytical methods for clearly isolating oil
degradation in conditions to electrification and static
discharge has not been fully developed.
The methods listed here are general in nature and may aid
in the differentiation.
Table +. $ethods for %naly0ing Stati# Dis#harge1
indu#ed Sludge and Varnish
O!idation is a well-known oil degradation mechanism and is relatively simple to monitor with
traditional oil analysis methods compared to other failure mechanisms. Thermal degradation,
although less understood, is also common and can be analy"ed with traditional oil analysis. The
effects of static discharge are still relatively new to the oil analysis field. t is more difficult to
analy"e &isolate as the root cause' with traditional oil analysis techni(ues.
The degradation stages and analytical scheme for o!idation, thermal and compressive base oil
degradation and static discharge conditions are presented in Tables - through =.
<umerous analytical tests have been identified in Tables - through =. t is of interest to discuss
a couple of these tests and how their results can be used to distinguish the various failure
mechanisms.
Fourier Transform Infrared S)e#tros#o)y
As o!idation increases, common reaction by-products are carbon-o!ygen double bonds, also
called the carbonyl group. 0arbonyl peaks on %T2 spectra in the -,75. cm-- region &%igure 5',
easily identifying o!idation. As o!idation increases, the absorbance peaks will increase in this
region. Additionally, phenol inhibitors used as antio!idants in the oil show peaks around =,FD..
0hanges in this peak are also noteworthy.
Figure 3. FTI4 S)e#trum Sho/ing
'(idation ea& around 1.536 #m11 4ange
)ecause thermal degradation can occur without significant amounts of o!ygen, different
degradation by-products are often observed. Therefore, the -,75. cm-- peak is less likely to be
significant. nstead, the by-products of thermal base oil degradation show up in the -,F.. to
-,F5. cm-- region , also known as the nitration peak due to the nitrogenous by-products, which
is more pronounced using a thick-cell &D.. Em path length' spectrometer.
%#id ,umber (%ST$ D753 or D883)
As the number of acids increase due to the o!idation process, A< will increase accordingly. n
large turbine systems, an A< change as low as ..= to ..5 above the new oil baseline is often
sufficient to condemn an oil.
4otating ressure Vessel '(idation Test (%ST$ D2252)
2>?OT measures an oil#s resistance to o!idation. This information indicates the oil#s remaining
o!idative useful life &2*A' and is calculated by dividing the in-service sample result by the new
oil result. 2>?OT values are influenced by the type and (uantity of antio!idants present in the
oil and the o!idative robustness of the base oil. 0autionary and critical limits for turbine oils are
usually at F. percent and 5. percent 2*A, respectively.
Vis#osity (%ST$ D339)
;uring o!idation, cleaved oil molecules recombine to form higher molecular weight species. An
increase in absolute viscosity can indicate when o!idation becomes advanced. n some cases,
oil can be thermally cracked during degradation, where the oil molecules are severed into
smaller molecules. As a result, a decrease in viscosity can be detected.
Flash oint (%ST$ D72)
%lash point may be used to identify thermal degradation if oil molecules have been thermally
cracked. As the percentage of lower molecular-weight oil fractions increases due to thermal
cracking, the flash point will drop accordingly.
Gas "hromatogra)hy (G")
One of the advantages of gas chromatography is its ability to separate similar molecules based
on their si"e. Thermally cracked oil will show an increased number of light-end molecules
compared to the new oil baseline.
at#h "olorimetri# Test
The patch test has been used for many years in oil analysis as a (ualitative test to assess the
condition of oil. 2ecently, several commercial laboratories have developed (uantifiable scales to
trend the varnish potential of oil. %or e!ample, on Analyst nc.#s scale from . to -.., a varnish
potential rating of . to 5. is considered acceptable. The range 5- to F. is a reportable condition,
meriting the need for fre(uent oil monitoring. 2eadings greater than F. are considered
actionable and should trigger work plans to (uickly remediate the condition.
"on#lusion
A number of degradation methods have been discussed in this article. n order to differentiate
between the failure mechanisms and assess varnish potential severity, turbine owners should
consider an ensemble of tests to monitor the o!idative or chemical degradation of turbine oils,
such as:
%T2
2>?OT
2*A82
>atch colorimetry
*ltracentrifuge
0oagulated insolubles
Acid number
References:
1. Livingstone, G., Clarus Technologies and Thompson, B., nal!sts, "nc. #Ne$
%arnish Test "mproves &redictive 'aintenance &rogram.( Practicing Oil
Analysis maga)ine, Jul! *++,.
*. Fitch, J., Noria Corporation. #-em!sti.!ing /ludge and %arnish.( Machinery
Lubrication maga)ine, Januar! *++*.
0. Fitch, J., Noria Corporation. #1sing 2il nal!sis to Control %arnish and
/ludge.( Practicing Oil Analysis maga)ine, 'a! 1333.
4. Noria Corporation. #The Lo$do$n on 2il Brea5do$n.( Practicing Oil Analysis
maga)ine, 'a! *++0.
,. /T' -40678+0. /tandard &ractice .or "n8/ervice 'onitoring o. 'ineral
Turbine 2ils .or /team and Gas Turbines, nnual Boo5 o. /T' /tandards
*++,, 9est Conshohoc5en, &enn.: American Society for Testing and
Materials, *++,.
:. ;amamoto and <irano. #=lectrical Behavior o. &hosphate =sters.( Wear,
volume ,+, 1367, p. 043>0,,.
6. #=lectro5inetic 9ear.( BFPR Journal, Fluid &o$er ?esearch Center, /till$ater,
25la. 1377, vol. *1, p. 106>14,.
7. /asa5i, . et al. #Generation o. /tatic =lectricit! -uring 2il Filtration.(
Lubrication ngineering maga)ine, /eptember 1333, p. 148*1.
3. -ucho$s5i, J. et al. #2vercoming =lectrostatic -ischarge in <!draulic and
Lubricating pplications B! "ncorporating Novel Filter 'edia.( NCF& "+,8
13.1, NF& Technical &aper /eries, &resented at "nternational Fluid &o$er
=@position, 'arch *++,.
1+. /asa5i, ., et al. #Free ?adicals and 2il uto82@idation -ue to /par5
-ischarges o. /tatic =lectricit!.( Lubrication ngineering, 'a! 1337, p. *48
*6.
11. 9hite, L. #/nap Crac5le &op,( <illiard Corporation, =lmira, Ne$ ;or5 An.d.B.
1*. Fitch, J. #<!drocarbon nal!sis,( Chapter *4, Fuels and Lubricants
<andboo5: Technolog!, &roperties, &er.ormance, and Testing, George =.
Totten, 9est Conshohoc5en, &enn.: /T' "nternational, *++0.
10. #Fourier Trans.orm "n.rared /pectroscop!.( Practicing Oil Analysis maga)ine,
'arch *++*.