Xin Q. Diesel Engine System Design

Подождите немного. Документ загружается.

166 Diesel engine system design

2.11.3 Piston ring and piston pin wear

A piston ring wear model was briey introduced by Taylor (1998), which

was analogous to the model developed for cams and followers by Colgan

and Bell (1989). Pint and Schock (2000) simulated piston ring wear by using

a generalized total wear coefcient to account for all the wear mechanisms

involved, including adhesive, abrasive, and corrosive wear. The wear

mechanisms of piston rings and cylinder liner with the use of high-sulfur

diesel fuel were also studied by Takakura et al. (2005). They believed that

the wear mechanisms with the use of high-sulfur fuel were dominated by a

combination of corrosive wear and abrasive wear caused by the formation

of aqueous sulfuric acid via acid condensation at the EGR cooler outlet. The

wear at the piston pin and the connecting rod small-end bush was studied

by Merritt et al. (2008).

2.11.4 Cam wear

Cam wear mechanisms

The primary wear mechanism in valvetrain cam is adhesive wear. The central

issue in cam wear analysis is to predict the wear location on the cam surface

in the design phase. Previous experimental work has conrmed that a cam

having higher Hertzian stress and thicker oil lm thickness may exhibit less

wear than a cam having lower Hertzian stress but thinner oil lm thickness.

This indicates that cam fatigue (mainly related to Hertzian stress) and wear

(mainly related to oil lm thickness) have different mechanisms. This also

indicates the paramount importance of good lubrication design in cam wear

control.

Wear usually occurs around the cam nose and at certain locations on the cam

anks (about 40° from the cam nose). Cam lobes are prone to wear at certain

locations of the opening and closing anks where the elastohydrodynamic

lubricating oil lm thickness and lubricant oil entraining velocity reach

a minimum theoretically (near zero). Note that the entraining velocity

characterizes the amount of oil ‘pumped’ along the surfaces towards the

contact. Experimental evidence showed that wear on the cam lobe usually

occurs at the locations where the oil lm thickness reaches the lowest.

Narasimhan and Larson (1985) presented the fundamentals of wear and

a comprehensive review of the valvetrain wear and material performance.

Purmer and van den Berg (1985) measured cam wear from cam lobe proles

and suggested that the maximum cam wear occurred at the locations where

the elastohydrodynamic lubrication condition became critical. They pointed

out that relatively small changes in valvetain geometry and kinematics may

signicantly change the oil entraining velocity and hence the wear pattern

and the wear location on the cam.

Diesel-Xin-02.indd 166 5/5/11 11:44:55 AM



167Durability and reliability in diesel engine system design

Cam wear mechanism was experimentally researched by Ito et al. (1998)

by using friction and wear measurement. They found that the cam wear was

conned across the cam nose within the region between the two locations

having the theoretical minimum (zero) oil lm thickness which corresponded

to theoretical zero entraining velocity. They found that cam wear started

from the locations corresponding to the minimum oil lm thickness position

(about ± 40° cam angle) and propagated toward the cam nose direction. At

those two locations, the measured friction coefcient reached the peak value

within one cam cycle, indicating severe boundary lubrication. They found

that mixed and boundary lubrication regimes predominated around the cam

nose, while elastohydrodynamic lubrication occurred on the cam anks. Their

test results revealed that oil deterioration and high temperature of oil supply

were the two factors that strongly correlated to cam wear for the at-faced

follower.

It should be noted that in highly loaded engines the exhaust cam force

and stress at the timing of exhaust valve opening are very high due to the

high gas load acting on the exhaust cam. The gas load affects the Hertzian

stress and may also affect the oil lm thickness. As a result, the exhaust

cam lobe is prone to wear also at the location corresponding to the exhaust

valve opening timing.

Ceramic materials have excellent properties of wear and seizure resistance

compared to the steel parts under high contact pressure and poor lubrication

conditions. Ceramic materials are able to prevent micro-welding in adhesive

wear and scufng. The amount of wear at the cam–follower interface can be

signicantly reduced by using a ceramic roller to replace the conventional

steel roller (Kitamura et al., 1997). The wear characteristics of a roller

follower in a variable valve timing system was researched by Leonard et

al. (1995).

Cam wear modeling

The minimum oil lm thickness on the cam can be calculated based on

the elastohydrodynamic lubrication theory (detailed in Chapter 10). The

oil entraining velocity as a function of cam angle can be calculated based

on valvetrain kinematics. The entraining velocity largely determines the

effectiveness of the elastohydrodynamic lubrication.

Yang et al. (1996) developed a lubrication and surface temperature

model based on the ash temperature concept for a cam with a at-faced

follower. They indicated that only calculating the oil lm thickness and the

maximum Hertizain stress is not sufcient to explain or predict the cam

wear pattern observed from their experiments. Therefore, cam–follower

surface temperature must be calculated. They believed there was a strong

relationship between the surface temperature and cam wear because their

Diesel-Xin-02.indd 167 5/5/11 11:44:55 AM



168 Diesel engine system design

engine durability test results showed that the wear started approximately

from the locations corresponding to the maximum surface temperatures.

The calculated cam surface temperature reached multiple maxima (peaks)

at the two positions of zero oil lm thickness on the cam anks and at the

cam nose. They concluded that cam wear initiated from the locations having

the maximum surface temperature and then propagated toward the cam nose

direction where the surface temperature remained relatively high. This theory

was in line with the observations in their cam durability test results.

Colgan and Bell (1989) proposed an important valvetrain wear model to

calculate the wear on the cam and the follower. A simpler valvetrain adhesive

wear model was proposed by Coy (1997) based on the Archard wear law in

which the wear coefcient was assumed as a linear function of the oil lm

thickness throughout the mixed lubrication regime.

2.11.5 Cam ﬂash temperature

The method of calculating the ash temperature at cam–follower contact

was introduced by Dyson and Naylor (1960). The maximum cam surface

temperature strongly correlates to the location of cam wear (Yang et al.

(1996). Cam ash (or surface) temperature calculation needs to be included

in cam design and durability analysis for at-faced followers. The cam

ash temperature is one of the three most important design criteria in cam

design: the Hertzian stress, the ash (or surface) temperature, and the oil lm

thickness. The cam ash temperature usually reaches a maximum at three

locations on the cam lobe: the cam nose, and the two maximum eccentricity

locations that have zero cam acceleration. Cam ash temperature calculations

were also discussed by Ito et al. (2001) and Yang et al. (1996).

2.11.6 Valve stem wear

The interface between the valve stem and the valve guide and the interface

between the valve seat and the seat insert have become more challenging in

modern engine design. Lack of lubrication at the valve seat due to tighter

lube oil consumption control and using valve stem seals may increase valve

seat wear and stem–guide wear. The widely used solution for the valve stem

scufng problem is to use chromium plating on the stem. In fact, many of

today’s intake and exhaust valve stems are chromium plated to resist stem

scufng. The use of non-guided valve bridge for cost savings in the four-

valve head design may increase the side load and bending load acting on

the valve stem and hence induce or increase valve stem scufng (Schaefer

et al., 1997).

Diesel-Xin-02.indd 168 5/5/11 11:44:55 AM



169Durability and reliability in diesel engine system design

2.11.7 Valve seat wear

Valve seat recession and wear mechanism

Valve seat wear is a combined effect of impact wear and sliding (adhesive)

wear that leads to the issues known as seat recession or seat pounding. Valve

seat recession is a major problem in modern diesel engines. As pointed out by

Schaefer et al. (1997), higher contact stress and increased seat sliding motion

caused by increased peak cylinder pressures in combination with reduced

lubrication at the valve seating surface due to tighter lube oil consumption

control will increase the propensity of seat recession. Larger distortion of

the cylinder head and valve seat and the corresponding larger misalignment

also increase the sliding wear on the valve seat. Both impact and sliding can

be important for valve seat wear. The most inuential factors on seat wear

include valve seating velocity, valve mass, valve seat face angle, valve head

stiffness, hardness of seat material, peak cylinder pressure, and the impact

wear constant. The wear constant is related to the resistance to impact wear

of the seat material. The wear of the exhaust and the intake valve seat and

insert was investigated by Dissel et al. (1989), Wang et al. (1995, 1996),

and Lewis and Dwyer-Joyce (2002).

Lewis and Dwyer-Joyce (2002) conducted a comprehensive investigation

on the relationship between the engine design parameters and the valve seat

wear. They found that diesel engine intake valve recession increased with

combustion loading, valve closing velocity, and misalignment of the valve.

The seat recession originated from the impact of the valve during seating.

The recession also originated from the sliding of the valve against the seat

as the valve head deected and wedged into the seat under the combustion

pressure. The valve seating impact created a series of ridges and valleys

circumferentially around the axis of the valve seating face. The valve seating

impact also created cracking on the seat insert. The ridges and valleys were

caused by the deformation or gouging process during the valve impact

seating against the seat inserts. They found that the valve recession amount

increased as the valve seating velocity and the number of loading cycles

increased. Misalignment of the valve relative to the seat reduced the initial

contact area between the valve and the seat when valve seating occurred.

Therefore, the misalignment increased the deformation of the seat at seating

impact. They found valve recession (or wear amount) was approximately

proportional to the square of seating velocity for the cast insert material.

Valve seat angle

Valve seat angle is related to both engine breathing performance and valve

seat wear durability. Larger valve seat angle can improve valve ow but may

cause more lateral sliding movement and seat adhesive wear than smaller

Diesel-Xin-02.indd 169 5/5/11 11:44:55 AM



170 Diesel engine system design

(atter) seat angle under cylinder pressure loading. The intake valve seat wear

is more severe than the exhaust valve seat wear partly due to less lubrication

available at the interface between the intake valve and seat insert than the

exhaust valve. Therefore, the intake seat angle (30° as typical) is usually

smaller than the exhaust seat angle (30–45° as typical) in automotive diesel

engines. Although using better seat facings or seat inserts may reduce valve

seat wear, attening the seat angle is probably the most cost-effective way

to minimize valve seat wear, as pointed out by Schaefer et al. (1997).

Valve seat wear modeling

Very few studies exist on valve seat wear modeling in the literature. A valve

seat durability model allows evaluation of the effect of cylinder pressure and

valve seating velocity on valve seat wear and recession in the early design

stage. Such modeling can be useful to reveal the relative or qualitative

comparison between design concepts.

Lewis and Dwyer-Joyce (2002) proposed a semi-empirical model to predict

intake valve recession and wear. In their model, total valve seat wear is the

sum of the impact wear and the sliding wear. Impact wear was related to the

valve seating impulse. Sliding wear was related to peak cylinder pressure.

Therefore, the total valve seat wear was a function of valve seating velocity,

cylinder pressure, and the number of loading cycles. They found that the

contribution of impact wear to the total wear was greater than that of sliding

wear. However, the relative contributions varied with valve seat materials.

For instance, the contribution of sliding wear to the total varied from 30%

for a cast seat insert to 2% for a sintered seat insert.

2.11.8 EGR engine component wear

Impact of EGR on engine oil

During the expansion stroke carbon soot is exposed to the lubricant oil in

the cylinder. The effects of engine speed, load, air–fuel ratio, EGR rate and

blow-by all affect oil soot loading. Oil aging refers to an increase in the soot

level, an increase in oil viscosity and a decrease in the total base number

(TBN) of the oil. The viscosity increase may be due to the insolubles and

the soot (Sasaki et al., 1997; George et al., 2007). The increase in viscosity

due to the soot may negate the fuel economy gains from the light-weight

oil. The percentage of soot in the oil increases almost linearly with engine

operating hours (or vehicle mileage) and is affected by the soot ow rate and

the driving cycle. At very high mileage the engine oil may have a ‘gel-like’

appearance when the soot percentage in the oil is high. Previous theory in

the literature indicated that at low load conditions or mixed cycles, EGR has

Diesel-Xin-02.indd 170 5/5/11 11:44:55 AM



171Durability and reliability in diesel engine system design

no effect on oil aging and engine wear; but EGR may increase component

wear at high-load conditions (Dennis et al., 1999). Engine soot ow rate

usually increases with engine speed and load. At high speeds and high loads,

higher EGR rate generally tends to increase the soot ow rate.

Based on the NO

–soot trade-off in diesel engines, one may think that the

retarded fuel injection timing and increased EGR rate used for NO

control

result in an increase in soot emission, and the soot is transported into the

lubricant oil to accelerate oil aging. However, such a trade-off may not

hold true when more advanced high-pressure fuel systems and air handling

systems are applied. In other words, modern high-EGR diesel engines may

achieve low engine-out soot level with high EGR rate via higher-pressure

fuel systems and advanced air systems without much degradation in oil soot

loading and wear. Engine-out soot control is an important system design

topic for engine wear durability. If it is difcult to reduce the engine-out

soot level at increased EGR rate, larger oil sump volume, reduced oil change

service interval and better oil lm thickness in tribological design can be

considered in order to accommodate the increased soot level in the oil. Some

chemically treated oil lters can keep soot from agglomerating in the oil for

soot dispersion so that a long oil drain interval can be maintained. Enhanced

oil lters may also help TBN retention.

If proper design measures are not taken to reduce, control, or accommodate

soot emission, increasing EGR usually increases the rate of oil aging and

hence increase engine wear on the tribological components to a certain extent.

In this case, a combination of both high engine load and high EGR rate may

increase engine wear. However, at low load or low EGR rate at full load,

the usage of EGR does not usually raise concerns about engine wear.

Impact of carbon soot on abrasive wear

Three commonly identied mechanisms of abrasive wear are plowing, cutting,

and fragmentation. It was found that the lubricant contaminated with carbon

soot resulted in increased wear (Mainwaring, 1997). Experimental work with

fundamental specimens was conducted by Green et al. (2006) on the effect

of soot-contaminated oil on valvetrain component wear. They conrmed that

the wear volume increased with the level of carbon black content in the oil

due to metal sliding wear, soot particle abrasion, and starvation of lubricant

from the contact. Ishiki et al. (2000) conducted engine tests to nd that the

piston frictional force increased with higher EGR rate in the latter half of the

compression stroke due to the soot deposited in the top ring groove, while the

piston frictional force decreases in the middle of the stroke with higher EGR

rate. They found that the piston ring wear essentially was a function of the

soot loading in the oil and the soot loading in the combustion chamber rather

than simply the EGR rate level. Dennis et al. (1999) reviewed the hypothetical

Diesel-Xin-02.indd 171 5/5/11 11:44:55 AM



172 Diesel engine system design

wear mechanisms of EGR, including antiwear additive absorption, abrasive

wear with carbon soot particles, corrosion/acid induced wear (i.e., corrosion

and abrasive wear mechanisms). It was believed that the corrosion-induced

wear with EGR was not signicant.

Impact of sulfur on corrosive wear

The total base number (TBN) of the oil decreases with oil aging due to

acids associated with EGR gas. A high TBN can neutralize the acid effects.

Reducing the sulfur content in the fuel can defer the decrease of the TBN.

High-sulfur fuel may increase engine wear due to the acid effect. Furuhama

et al. (1991) found that the low-temperature corrosive wear was not primarily

caused by the decrease in the oil TBN. Instead, it was caused by strong

acidity of condensed water at the top land space of the piston. Dennis et al.

(1999) experimentally found that fuel sulfur level had a large effect on power

cylinder wear at the piston ring and the liner, and the effect was aggravated

when the oil became aged. The researchers at the Musashi Institute of

Technology conducted a series of experiments to try to understand the causes

of the abnormal wear in diesel engines with EGR. They found that the acid

components in the EGR gas may result in ring wear for the given fuel tested

(Furuhama et al., 1991; Urabe et al., 1998; Ishiki et al., 2000).

2.12 Exhaust gas recirculation (EGR) cooler

durability

EGR is used for NO

control and sometimes for exhaust temperature control.

EGR cooler durability issues usually include fouling, plugging, condensation,

boiling, corrosion, and coolant leak.

2.12.1 EGR cooler fouling and plugging

EGR cooler fouling refers to the performance degradation over time in heat

transfer capability and gas-side ow restriction due to deposits. EGR cooler

plugging refers to a large increase in the gas-side pressure drop caused by

severe fouling. EGR cooler deposits are a combination of thermophoretic

carbon soot deposition, condensed hydrocarbons (HC), and acids occurring

on the cooled surface inside the cooler tubes.

EGR cooler fouling may cause signicant deterioration in heat transfer

performance, sometimes in the order of 20–30% (Hoard et al., 2008). This

causes an increase in the intake manifold gas temperature and affects the

emissions. Gas-side pressure drop may increase slightly (e.g., a few

percent) or signicantly (plugging). When EGR cooler plugging occurs, if

the engine control system cannot compensate for the high ow restriction

Diesel-Xin-02.indd 172 5/5/11 11:44:55 AM



173Durability and reliability in diesel engine system design

in the EGR cooler, lower EGR ow than desirable results, and the NO

emissions target cannot be met.

EGR cooler fouling characteristics change over time. The deposits

initially build up quickly with a fast deposition rate and then stabilize after

50–200 hours (Hoard et al., 2008). As a result, the rate of change in cooler

performance degradation due to fouling is initially quite large. However, the

cooler effectiveness may asymptotically approach a stabilized constant value

after some time (Zhang et al., 2004). The stabilization is believed due to

certain mechanisms related to deposit removal and deposition rate change.

The mechanism of EGR cooler fouling is complex. It depends largely

on both design and engine operating conditions such as the amount and

composition of soot/HC/acid, condensate, exhaust gas temperature, gas ow

velocity, and ow pressure pulsation. Severe EGR cooler fouling and plugging

may occur when excessive carbon soot is generated during combustion, or an

excessively large amount of HC is produced by low-load operation, misre

or late post-injection used to assist DPF regeneration. A larger amount of

condensate makes the cooler deposits worse in fouling since heavy wet soot/

HC deposits are worse than dry uffy soot deposits. The problem of soot and

HC fouling may be aggravated by the presence of liquid lm in the cooler

such as water condensate, leaking coolant, unburnt and condensed HC.

EGR cooler conguration (n type or tube-in-shell type) affects its fouling

characteristics. High gas ow velocity in the cooler tube may reduce the

deposition but it generally causes a larger pressure drop. Lower gas ow

velocity increases the amount of deposits. The most dangerous condition

for soot fouling is at the low speeds/loads. Deposit accumulation is worse

at the lower speed or load conditions where the EGR ow rate and gas

temperature are lower. At these conditions, it is also more difcult to blow

off the deposits with the low gas ow velocity. The higher exhaust gas

velocity under some driving conditions may have an abrasive self-cleaning

effect to remove the soot deposits. This event can help avoid tube clogging

and maintain sufcient performance of heat transfer and pressure drop.

The EGR fouling and plugging problems are complex issues whose

solution requires coordination of the efforts from the teams of cooler design,

combustion development, emissions calibration, aftertreatment and electronic

controls. The selection of EGR cooler n density is critical for alleviating

the fouling problem. The cooler size in terms of cooling capacity and ow

restriction needs to be determined based on the stabilized fouled (seasoned)

condition rather than the clean condition with a sufcient design margin

reserved. Hoard et al. (2008) suggested that in order to consider the worst-

case service operation condition the EGR coolers need to be over-sized in

the order of 30% to obtain the required cooling capacity when fouled. In fact,

the specic percentage margin reserved for fouling should be determined

based on specic cooler design details and engine operating conditions. A

Diesel-Xin-02.indd 173 5/5/11 11:44:55 AM



174 Diesel engine system design

simulation model of cooler soot/HC fouling coupled with engine–vehicle

driving cycle analysis is desirable in order to provide guidance on cooler

sizing in system design. The simulation model may predict the deterioration

of cooler performance. Using DOC and DPF in the EGR system upstream

of the EGR cooler may remove a large amount of HC and soot to reduce

the cooler deposits. Moreover, the use of EGR can be minimized to reduce

the EGR cooler usage in order to alleviate the cooler fouling problem. For

example, at cold intake manifold temperatures or in the scenarios that are

not frequently encountered during driving (e.g., outside the FTP, SET, or

NTE zones) the usage of EGR may be reduced in order to achieve better

durability of the EGR cooler.

EGR cooler fouling was reviewed by Hoard et al. (2008). The mechanisms

of soot deposition and EGR cooler fouling control were experimentally

investigated by Zhang et al. (2004), Bravo et al. (2005, 2007), Zhan et

al. (2008), Mulenga et al. (2009), and Chang et al. (2010). EGR cooler

fouling models were developed by Abarham et al. (2009a, 2009b), Teng

and Regner (2009), and Teng (2010). The effects of liquid lm, gas and

coolant temperatures and Reynolds number on the deposits in coolers were

discussed by Lepperhoff and Houben (1993).

2.12.2 EGR cooler boiling

EGR cooler boiling is a failure mode where lm boiling occurs in the cooler

so that the cooler tube metal temperature becomes unacceptably high. The

boiling is the result of a mismatch between a high heat rejection and a low

coolant ow rate.

When the cooler tube is hot enough or the coolant boiling temperature

is low enough, the liquid coolant that is in contact with the tube instantly

vaporizes. This creates a lm of vapor. The vapor lm barrier between

the tube metal surface and the coolant results in poor heat transfer. As a

result, the tube metal temperature quickly approaches the gas temperature

and causes damage (e.g., tube annealing). At the same time, EGR cooler

outlet gas temperature increases rapidly as well. Film boiling occurs in the

following situations: (1) the EGR cooler coolant ow rate is too low; (2)

the coolant pressure is too low; (3) the coolant temperature is too high; and

(4) the heat rejection is too high.

Film boiling is different from nucleate boiling. The latter improves heat

transfer and does not result in excessively high metal temperature. The

maximum allowable cooler tube metal temperature is restricted by nucleate

boiling, beyond which failure occurs. Between nucleate and lm boiling

there is an unsteady transition boiling condition, where the coolant vapor

lm forms and collapses in an unsteady manner. The transition boiling can

also cause high metal temperature failures.

Diesel-Xin-02.indd 174 5/5/11 11:44:55 AM



175Durability and reliability in diesel engine system design

The theoretical boiling temperature of the coolant can be calculated

based on coolant pressure. An increased coolant pressure may increase the

boiling temperature threshold. A less restrictive coolant circuit in the EGR

cooler may help preserve desirable high boiling threshold (e.g., using less

restrictive bafe design in the EGR cooler). Coolant composition also affects

the boiling temperature. For example, water as the coolant is more prone

to boiling than the 50/50 mixture of water and ethylene glycol. A sufcient

margin is required in the EGR cooler design between the theoretical boiling

temperature threshold of the coolant and the actual coolant temperature in

order to cover product variability and extreme ambient operations in the

entire engine speed–load domain.

Film boiling is usually prone to occur at engine full load and low or

medium speed conditions, for example, near peak torque, where the heat

rejection is relatively high but the coolant ow rate is relatively low. Boiling

may occur at both steady-state and transient conditions. In the transient

thermal cycles when the engine load is reduced from high load to low load

or idle, the thermal inertia in the cooler tubes may cause boiling. Proper

protective measures are needed in order to prevent coolant boiling during

such transients.

The maximum allowable heat load for a given EGR cooler is the heat

rejection at which coolant boiling occurs. The worst case of cooler boiling

usually occurs at the conditions of low coolant pressure and excessively

high exhaust temperature, for example at high altitude or with high intake or

exhaust restriction. The EGR cooler needs to be designed to accommodate

the required heat rejection without lm boiling. This should be done by

supplying a sufciently high coolant pressure and ow rate under all

operating conditions. On the other hand, the coolant ow rate should not

be over-designed too high with an excessively large margin with respect to

boiling.

2.12.3 EGR cooler corrosion and condensation

Acidic corrosion and condensate

Modern diesel engines use high EGR rate and low EGR gas temperature via

EGR cooling to control NO

. The EGR ow contains corrosive gases having

sulfur and nitrogen compounds produced from combustion. Corrosion occurs

with the onset of condensation in the EGR cooler.

Fuel sulfur level affects the corrosion in EGR cooler and in engine power

cylinder components. Most of the sulfur in the fuel is converted into gaseous

or absorbed onto the particulate matters. The gaseous SO

reacts with

oxygen in the exhaust to form SO

. With cooled EGR these corrosive gases

are returned to the intake manifold. If the EGR cooler outlet gas temperature

Diesel-Xin-02.indd 175 5/5/11 11:44:55 AM

