1■ To enhance jet engine

operational reliability, Volvo

Aero Corporation applies
tungsten carbide coatings to
mid-span dampers of fan
blades. This will increase the
service life of these dampers.

Tungsten Carbide Coatings on Jet Engine Components:

Safe Aviation Tungsten carbide thermal spray coatings are used in the
JAN WIGREN aerospace industry to mitigate mid-span damper wear on
LARS PEJRYD jet engine blades. In some cases, however, the fatigue life of
VOLVO AERO the blade may actually be reduced if a coating is applied,
or the coating may fail. The Volvo Aero Corporation in
Trollhättan (Sweden) has comprehensively investigated
various coating systems in collaboration with the
University of Tulsa, Oklahoma (USA); the Sulzer Metco
process was found to be the best one. On the following
pages, we publish a short version of Jan Wigren’s report on
the investigation.

■ The vibration of fan blades

in jet engines (Fig. 1■) can be
controlled by the use of mid-span
continuous crack propagation
through the damper, culminating
in complete engine failure. Based
sistance and low-cycle fatigue test
results, since the characteristics of
which they are representative
dampers (Fig. 2■) as a point of con- on existing studies, residual accord with in-service observa-
tact between the blades. Tungsten stresses had to be suspected of tions and engine performance eval-
carbide (WC) thermal spray coat- being the most decisive cause of uations. The coating materials
ings are often applied to the con- these failures. Therefore, different selected for study were tungsten
tact surfaces of the dampers to factors contributing to this phe- carbide thermal spray coatings
reduce wear and, while the tech- nomenon have been studied in with cobalt (1), nickel (2), or
nique has proved successful in this depth. cobalt/chromium (3) binders. Sev-
respect, coating failures have been en HVOF (High Velocity Oxy Fuel)
observed due to cracking and SEVEN PROCESSES, THREE processes (A to G) and one plasma
spalling caused by cyclic fatigue COATING MATERIALS spray process (H) have been used
and impact. This can result in This article concentrates on the to apply the coatings, including a
impaired coating performance or residual stress, bending crack re- system by Sulzer Metco. As an

36 SULZER TECHNICAL REVIEW 2/2000 3945

example, the coating system desig-
nation A-1 indicates a WC-Co coat-
ing applied by the HVOF process
‘A’. The substrate was Ti-6Al-4V in
all cases.
To determine the through-thick-
ness residual stress distribution,
the layer has been removed from
the specimen, followed by measur-
ing the change in strain of the sub-
strate. Figure 3■ shows the typical
through-thickness residual stress 2■ Mid-span dampers (arrow) are used as a point of
distribution in the longitudinal contact between the fan blades to reduce vibration.
direction for the coating systems
E-3 and F-1. The E-3 result indi-
cates residual tensile stresses in
the coating and residual compres-
sive stresses in the substrate, close
to the interface. In the F-1 case, it STRESS/PERFORMANCE
is noteworthy that the residual CORRELATION
stress profiles of these two systems Certain observations can be made
are opposite in sign, although the regarding the correlation between
coating compositions and applica- residual stress and the perfor-
tion processes are similar. mance characteristics of a coating
system. Those coatings in which
RESISTANCE TO CRACKING the residual stresses were com-
IN BENDING pressive exhibited the highest
Figure 4■ illustrates the three- rankings, while those with tensile
point bending test used to deter- residual stresses displayed re-
mine the coating crack resistance duced crack resistance. These re-
under a bending load. The test was sults suggest that compressive
designed to simulate the bending- residual stresses will greatly Typical residual stress distribution in the coatings E-3 and F-1
strain configuration on a mid-span enhance the resistance of the coat- 400

damper subject to centrifugal ing to cracking under a bending 200

F-1 (HVOF, WC-Co)
forces. Every coating system has load. 0
Residual stress (MPa)

been evaluated by the extension of Regarding the substrate fatigue E-3 (HVOF, WC-Co/Cr)
–200
the cracked area and the crack life, which has been extensively
–400
length. investigated as well, a substrate
–600

–800

–1000
3■ Remarkable results have been found on residual stress measurement.
–1200
The through-thickness residual stress distribution is completely 0 0.5 1 1.5 2 2.5 3 3.5 4
Distance from the coating surface (mm)
different for the coatings E-3 and F-1, although similar coating
Coating Substrate
materials had been applied using similar HVOF coating systems.

SULZER TECHNICAL REVIEW 2/2000 37

 Bending test specimen
4■ Generation of coating cracks – during operation caused by
Fan blade centrifugal forces – has been simulated by means of a bending test.
Coating

Centrifugal
force Substrate

Load Typical coating cracks

Coating cracks Coating cracks Crack length

Spacing

Supports

with a compressive residual stress able and substrate fatigue life was
may be significantly superior to the most important design criteri-
one in which the stress is tensile – on. The relationships between re-
an interesting result since, in a sidual stress, coating crack resis-
substrate of that type, the coating tance and substrate fatigue life are
stress is likely to be tensile, reduc- plotted in Figure 5■.
ing the crack resistance.
The findings relating to coating TESTING UNDER
crack resistance in bending and SERVICE CONDITIONS
substrate fatigue life suggest that Three coatings have been selected
a coating system may be specified as examples because of their differ-
on the basis of the residual stress ent properties in this context. Sys-
in the coating and the component tems A-1 and F-1 exhibit superior
design criteria. For example, the coating crack resistance, while
fatigue life of the substrate in a H-1 possesses superior substrate
system with a relatively high coat- fatigue life. All three systems were
ing crack resistance (due to com- evaluated on fan blades with mid-
5■ For selecting a suitable coating system, the relation- pressive residual stress) may be span dampers in a jet engine,
ships between residual stress, coating crack resistance lowered by a tensile residual stress under test conditions designed to
and substrate fatigue life are most decisive. in the substrate itself. A system of simulate actual service conditions.
400
this type would be specified if The engine performance evalua-
1 3 cracking of the coating was unac- tions were used to validate the
200
ceptable and the reduction in sub- final coating selection.
0
strate fatigue life could be compen- The engine test evaluation figures
Residual stress (MPa)

–200 sated for by the design of the com- suggest that coating systems in
–400 1 Reduced coating crack resistance
ponent. On the other hand, a sys- which the residual stress in the
2
–600
tem with a relatively low coating coating is compressive are superi-
2 Improved substrate fatigue life
crack resistance (the result of ten- or to such in which the stress is
–800 3 Reduced substrate fatigue life sile residual stress) might have a tensile. Figure 6■, however, shows
–1000 4 4 Improved coating crack resistance higher substrate fatigue life as a a mid-span damper coated with the
–1200 result of compressive residual system F-1 – the one with the high-
0 0.25 0.5 0.75 1 1.25 1.5
Distance from the coating surface (mm)
stress in the latter. This type of est value of compressive coating
system would be the choice if residual stress – in which spalling
Coating Substrate
cracking of the coating was accept- failure of the coating occurred dur-

38 SULZER TECHNICAL REVIEW 2/2000

ing the engine test. The reason for 6■ Although a coating in which the
this failure may be that the high residual stress is compressive is
compressive residual stress in the normally superior in performance
coating is close to the ultimate to one in which the residual stress
compressive strength of the tung- is tensile, the coating system with
sten carbide material. This sup- the highest compressive stress (F-1)
ports the contention that compres- failed during a test under service
sive residual stress in the coating conditions due to spalling of the
is essential to satisfactory mid- coating.
span damper performance, provid-
ed that the sum of the in-service
and residual stresses is below the
level at which spalling occurs.
During the tests, the Sulzer Metco
type Diamond Jet Hybrid system for the purpose: The compressive INFO DIRECT
(Fig. 7■) was found to be the opti- residual stress in the coating was Sulzer Metco AG
mum coating system. It was select- shown to be sufficiently low to pre- Nils-Erik Svensson
ed for mid-span damper produc- vent spalling, while the coating Bottvid Center
tion on the basis of the bending and exhibited the best performance Fittja Backe 1
fatigue life test results, residual under engine operating conditions. SE-14584 Stockholm
stress evaluations and perfor- Till today, mid-span dampers coat- Sweden
mance test findings. The chosen ed with the selected system have Phone +46 (0)8-680 82 00
system gives exactly the residual sustained many operating hours Fax +46 (0)8-531 82 251
stress distribution which is ideal without failure. Ω E-mail [email protected]

7■ After extensive investigation, the HVOF installation type Diamond Jet Hybrid by Sulzer Metco was
found to be the optimum coating system. This result has been confirmed meanwhile by thousands of jet
engine service hours without failure.

SULZER TECHNICAL REVIEW 2/2000 39