U.S. patent application number 10/327054 was filed with the patent office on 2004-01-22 for nickel base superalloys and turbine components fabricated therefrom.
This patent application is currently assigned to General Electric Company. Invention is credited to Bouse, Gregory Keith, Henry, Michael Francis, Schaeffer, Jon Conrad.
Application Number | 20040011443 10/327054 |
Document ID | / |
Family ID | 22682071 |
Filed Date | 2004-01-22 |
United States Patent
Application |
20040011443 |
Kind Code |
A1 |
Bouse, Gregory Keith ; et
al. |
January 22, 2004 |
Nickel base superalloys and turbine components fabricated
therefrom
Abstract
A nickel base superalloy suitable for the production of a large,
crack-free nickel-base superalloy gas turbine bucket suitable for
use in a large land-based utility gas turbine engine, comprising,
by weight percents: Chromium 7.0 to 12.0 Carbon 0.06 to 0.10 Cobalt
5.0 to 15.0 Titanium 3.0 to 5.0 Aluminum 3.0 to 5.0 Tungsten 3.0 to
12.0 Molybdenum 1.0 to 5.0 Boron 0.0080 to 0.01 Rhenium 0 to 10.0
Tantalum 2.0 to 6.0 Columbium 0 to 2.0 Vanadium 0 to 3.0 Hafnium 0
to 2.0 and remainder nickel and incidental impurities.
Inventors: |
Bouse, Gregory Keith;
(Taylors, SC) ; Henry, Michael Francis;
(Niskayuna, NY) ; Schaeffer, Jon Conrad;
(Greenville, SC) |
Correspondence
Address: |
NIXON & VANDERHYE P.C./G.E.
1100 N. GLEBE RD.
SUITE 800
ARLINGTON
VA
22201
US
|
Assignee: |
General Electric Company
|
Family ID: |
22682071 |
Appl. No.: |
10/327054 |
Filed: |
December 24, 2002 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10327054 |
Dec 24, 2002 |
|
|
|
09794220 |
Feb 28, 2001 |
|
|
|
60185696 |
Feb 29, 2000 |
|
|
|
Current U.S.
Class: |
148/675 ;
420/448 |
Current CPC
Class: |
C22F 1/10 20130101; C22C
19/057 20130101; C22C 19/056 20130101 |
Class at
Publication: |
148/675 ;
420/448 |
International
Class: |
C22C 019/05 |
Claims
What is claimed is:
1. A nickel base superalloy suitable for the production of a large,
crack-free nickel-base superalloy gas turbine bucket suitable for
use in a large land-based utility gas turbine engine, comprising,
by weight percents: Chromium 7.0 to 12.0 Carbon 0.06 to 0.10 Cobalt
5.0 to 15.0 Titanium 3.0 to 5.0 Aluminum 3.0 to 5.0 Tungsten 3.0 to
12.0 Molybdenum 1.0 to 5.0 Boron 0.0080 to 0.0130 Rhenium 0 to 10.0
Tantalum 2.0 to 6.0 Columbium 0 to 2.0 Vanadium 0 to 3.0 Hafnium 0
to 2.0 and remainder nickel and incidental impurities.
2. The nickel base superalloy according to claim 1, wherein boron
is present in an amount of about 0.008-0.010 weight percent.
3. The nickel base superalloy according to claim 1, wherein boron
is present in an amount of about 0.009 weight percent.
4. The nickel base superalloy according to claim 1, wherein hafnium
is present in an amount of about 0.015-0.45 weight percent.
5. A nickel base superalloy suitable for the production of a large,
crack-free nickel-base superalloy gas turbine bucket suitable for
use in a large land-based utility gas turbine engine, comprising,
by weight percents Chromium 9.50-10.00 Cobalt 7.00-8.00 Aluminum
4.10-4.30 Titanium 3.35-3.65 Tungsten 5.75-6.25 Molybdenum
1.30-1.70 Tantalum 4.60-5.00 Carbon 0.06-0.10 Zirconium 0.01 max
Boron 0.008-0.010 Iron 0.20 max Silicon 0.20 max Manganese 0.01 max
Copper 0.10 max Phosphorus 0.005 max Sulfur 0.003 max Columbium
0.40-0.60 Oxygen 0.002 max Nitrogen 0.0015 max Vanadium 0.10 max
Hafnium 0.10-0.20 Platinum 0.15 max Rhenium 0.10 max
Rhenium+Tungsten 6.25 max Magnesium 0.0035 max Palladium 0.10 max
Nickel Remainder
6. A method of making a cast and heat treated article of a
nickel-base superalloy, comprising the steps of: (a) providing a
superalloy of the composition of claim 1; (b) heating the
superalloy to develop at least 60 percent solutioning of gamma
prime precipitate; and (c) cooling to room temperature.
7. The method according to claim 6 wherein the article is heated to
a temperature of about 2260.degree. F.-2300.degree. F. but at least
about 25.degree. F. below the incipient melting temperature of the
superalloy.
8. The method according to claim 6 wherein the article is cooled by
a furnace cool at a rate of about 35.degree. F./minute to about
2050.degree. F.
9. The method of claim 6 wherein the wherein the article is cooled
by a gas fan cool at a rate of about 100-150.degree. F./minute from
below about 2050.degree. F.
10. The method of claim 6, wherein said heating is carried out in
an argon atmosphere.
11. The method of claim 6 wherein said heat treating comprises the
steps of: (a) heating said article to a temperature of about
1400.degree. F. at a rate of 25.degree. F./minute and holding for
about 10 minutes; (b) heating said article in (a) to a temperature
of about 2225.degree. F. at a rate of 25.degree. F./minute and
holding for about 8 hours; (c) heating said article in (b) to a
temperature of about 2250.degree. F. at a rate of 25.degree.
F./minute and holding for about 4 hours; (d) heating said article
in (c) to a temperature of about 2280.degree. F. at a rate of
30.degree. F./minute and holding for about 2 hours; and (e) cooling
to room temperature.
12. The method according to claim 11 wherein the article is cooled
by a furnace cool at a rate of about 35.degree. F./minute to about
2050.degree. F.
13. The method of claim 11 wherein the wherein the article is
cooled by a gas fan cool at a rate of about 100-150.degree.
F./minute from below about 2050.degree. F.
14. The method of claim 6 wherein said article is a large turbine
bucket.
15. The method of claim 6 wherein said article is a large aero
engine turbine blade.
16. An article produced by the method of claim 6.
17. An article of claim 16 which is directionally solidified.
18. An article of claim 16 which is conventionally cast.
19. An article of claim 16 which is single crystal cast.
20. A gas turbine engine containing an article of claim 16.
Description
[0001] The present invention relates to directionally solidified
nickel-base superalloys alloys having improved heat treat
characteristics, good high temperature longitudinal and transverse
creep strength properties, good hot corrosion resistance and
resistance to oxidation. The invention also relates to the use of
the alloys in the fabrication of turbine components, particularly
large turbine buckets and turbine blades for aircraft engines.
BACKGROUND OF THE INVENTION
[0002] It is known to employ nickel base superalloys in the
fabrication of aircraft engine components. To be acceptable, such
alloys must exhibit good castability with no heat treat cracking,
good high temperature longitudinal and transverse creep strength
properties and good hot corrosion resistance.
[0003] One such nickel base superalloy employed as a turbine
blading material in aircraft engines is single crystal (SC) Rene N4
alloy. A form of SC Rene N4 is described in U.S. Pat. No. 5,154,884
as a nickel-base superalloy composition comprising, by weight,
7-12% Cr, 1-5% Mo, 3-5% Ti, 3-5% Al, 5-15% Co, 3-12% W, up to 10%
Re, 2-6% Ta, up to 2% Cb, up to 3% V, up to 2% Hf, the balance
being essentially nickel and incidental impurities. U.S. Pat. No.
5,399,313 describes a modified version of SC Rene N4 as comprising,
by weight, 9.5-10.0 Cr, 7.0-8.0 Co, 1.3-1.7 Mo, 5.75-6.25 W,
4.6-5.0 Ta, 3.4-3.6 Ti, 4.1-4.3 Al, 0.4-0.6 Cb, 0.1-0.2 Hf,
0.05-0.07 C and 0.003-0.005 B, the balance being nickel and
incidental impurities.
[0004] Typically, aircraft engine blades are small, on the order of
a few inches long, and weigh a few ounces, or a few pounds at most.
Power turbine buckets, by contrast, are typically up to about 36
inches long, and weigh up to about 40 pounds. It has been found
that use of single crystal alloys for such large parts is
impractical. A need exists for a superalloy for use in the
fabrication of large turbine blades which exhibits good castability
with no heat treat cracking, good high temperature longitudinal and
transverse creep strength properties and good hot corrosion
resistance. The present invention seeks to satisfy that need.
SUMMARY OF THE INVENTION
[0005] The present invention is directed to an alloy and high
temperature heat treatment for buckets fabricated from nickel base
superalloys that will allow the buckets to be used for extended
periods, typically up to about 72,000 hours in a power turbine. It
is has been found that such an extended turbine life can be
achieved if approximately 60-80% solutioning of the gamma-prime
precipitates in the alloy occurs. The gamma-prime precipitates
provide the strengthening phase for the alloy. Moreover, it has
been discovered according to the invention that adjusting the level
of boron in the alloy of the invention to within the range of about
70-130 ppm, generally about 80-130 ppm, more usually about 80-100
ppm (about 0.0080-0.01 weight %), for example about 90 ppm (about
0.009 weight %), results in a reduction in the incidence of heat
treat cracking in the cast buckets.
[0006] In a first aspect, there is provided a nickel base
superalloy suitable for the production of a large, sound,
crack-free nickel-base superalloy gas turbine bucket suitable for
use in a large land-based utility gas turbine engine, comprising or
consisting essentially of, by weight percents:
[0007] Chromium 7.0 to 12.0
[0008] Cobalt 5.0 to 15.0
[0009] Carbon 0.06 to 0.10
[0010] Titanium 3.0 to 5.0
[0011] Aluminum 3.0 to 5.0
[0012] Tungsten 3.0 to 12.0
[0013] Molybdenum 1.0 to 5.0
[0014] Boron 0.0080 to 0.013
[0015] Rhenium 0 to 10.0
[0016] Tantalum 2.0 to 6.0
[0017] Columbium 0 to 2.0
[0018] Vanadium 0 to 3.0
[0019] Hafnium 0 to 2.0 and
[0020] Remainder nickel and incidental impurities.
[0021] A typical nickel base alloy of the invention comprises or
consists essentially of, in weight percent:
[0022] Chromium 9.50-10.00
[0023] Cobalt 7.00-8.00
[0024] Aluminum 4.10-4.30
[0025] Titanium 3.35-3.65
[0026] Tungsten 5.75-6.25
[0027] Molybdenum 1.30-1.70
[0028] Tantalum 4.60-5.00
[0029] Carbon 0.06-0.10
[0030] Zirconium 0.01 max (no min)
[0031] Boron 0.008-0.010 (also expressed as 80-100 parts per
million (ppm))
[0032] Iron 0.20 max (no min)
[0033] Silicon 0.20 max (no min)
[0034] Manganese 0.01 max (no min)
[0035] Copper 0.10 max (no min)
[0036] Phosphorus 0.005 max (no min)
[0037] Sulfur 0.003 max (no min)
[0038] Columbium 0.40-0.60
[0039] Oxygen 0.002 max (no min)
[0040] Nitrogen 0.0015 max (no min)
[0041] Vanadium 0.10 max (no min)
[0042] Hafnium 0.10-0.20
[0043] Platinum 0.15 max (no min)
[0044] Rhenium 0.10 max (no min)
[0045] Rhenium+Tungsten 6.25 max (no min)
[0046] Magnesium 0.0035 max (no min)
[0047] Palladium 0.10 max (no min)
[0048] Nickel Remainder
[0049] In a further aspect, there is provided a method of making a
cast and heat treated article such as a large power turbine bucket
of a nickel-base superalloy of the invention, wherein the article
is heated in an argon atmosphere or in vacuum to develop 60-80
percent solutioning of gamma prime precipitate, followed by cooling
to room temperature. Typically, the article is heated to a
temperature of about 2260.degree. F.-2300.degree. F., but at least
about 25.degree. F. below the incipient melting temperature of the
superalloy. The article may be cooled by a furnace cool at a
cooling rate of about 35.degree. F./minute to 2050.degree. F.,
followed by gas fan cooling at nominally 100.degree. F./minute to
1200.degree. F., and then any cooling rate to room temperature.
[0050] In yet a further aspect, the invention provides an article,
such as a large turbine bucket, produced according to the method of
the invention.
[0051] In a further aspect, there is provided a gas turbine engine
containing an article of the present invention.
[0052] The alloy of the invention exhibits several advantages.
First, at 90-130 ppm boron the alloy of the invention has better
castability (for large turbine buckets) than SC Rene N4 at 30-50
ppm boron. Secondly, at 90-130 ppm boron in DS form the alloy of
the invention has an improved yield over SC Rene N4 at 30-50 ppm
boron. In regard to "yield", SC Rene N4 implies one grain per part.
SC Rene N4 is typically used to make small turbine blades. As small
parts go, it is possible to have a true "single crystal." However,
for large components, it is difficult to actually produce a part
with only one grain. Thus, "yield" for a SC part would be near zero
(i.e. it is not possible to fabricate any). By changing the
composition of SC Rene N4 primarily by adding more boron, it is
possible to make a multi-grained DS component. This multi-grained
DS component is designed to accommodate many grains across the
cross-section of the part. Made in this manner, the "yield"
increases to 80-100%.
[0053] Thirdly, at 90-130 ppm boron, the alloy of the invention has
nominally equivalent mechanical properties (in the longitudinal
direction) to the SC Rene N4 at 30-50 ppm boron. Fourthly, at
90-130 ppm boron, the alloy of the invention has better transverse
creep properties than SC Rene N4 at 30-50 ppm. Fifthly, at 90 ppm
boron, the alloy of the invention has better resistance against
heat treat cracking than either the SC Rene N4 at 30-50 ppm boron
or the 130 ppm boron DS alloy of the invention. The alloy with 130
ppm boron has a lower melting point (approx. 2301.degree. F.) than
DS Rene N4 or DS Rene N4 with 90 ppm boron (m.p. approx.
2315.degree. F.), or SC Rene N4 which has a melting point near
2334.degree. F. (Melting points: DS Rene N4 with 130 ppm
boron--2301.degree. F.; DS Rene N4 with 90 ppm boron--2315.degree.
F.; SC Rene N4 with 30-50 ppm boron--2334.degree. F.).
BRIEF DESCRIPTION OF THE DRAWINGS
[0054] The invention will now be described in more detail with
reference to the accompanying drawings, in which:
[0055] FIG. 1 is a series of plots showing the effect of different
processing conditions on crack length in a MS7001 H turbine bucket;
and
[0056] FIG. 2 is a regression plot showing creep strength as a
function of temperature;
[0057] FIG. 3 is a regression plot showing transverse creep
strength (%) as a function of boron content (ppm);
[0058] FIG. 4 is plot showing creep elongation as a function of
test temperature;
[0059] FIG. 5 is a plot showing the effect of varying amounts of
boron on incipient melting of SC or DS Rene D4;
[0060] FIG. 6 shows a third and fourth stage bucket fabricated from
an alloy of the invention; and
[0061] FIG. 7 is a gas turbine engine showing the location where
buckets of the invention are used.
DETAILED DESCRIPTION OF THE INVENTION
[0062] It has been found, according to the invention, that
increasing the boron from about 30-50 ppm in the SC Rene N4
specification to no greater than 130 ppm boron, along with several
changes in part configuration, including bucket shape, essentially
eliminates casting cracks in large turbine buckets. The additional
boron may create a "M.sub.5B.sub.3" phase where M is Ni or
Ni.sub.5B.sub.3 eutectic phase in the grain boundaries and
elsewhere within the alloy matrix (as determined by Auger
Spectrometry and Microdiffraction analyses), and the melting
properties of the alloy have been attributed to the presence of a
"M.sub.5B.sub.3" boron phase. The presence of this eutectic phase
lowers the incipient melting point (the point at which the metal
starts to melt) from 2334.degree. F. to 2301.degree. F. (as
determined by Differential Thermal Analysis (DTA)). Thus, after
application of a 2320.degree. F. heat treatment (normal for SC Rene
N4), the DS alloys begin to melt at locations within the eutectic
pools where the boron as Ni.sub.5B.sub.3 is concentrated. Many of
these eutectic pools are in the grain boundaries, and can be more
segregated than those eutectic pools elsewhere within the grains.
When the eutectic melting starts and the bucket cools back down to
room temperature, a linear imperfection defined as a crack may be
created. These cracks, called heat treat cracks, may be several
inches long but may not be visible to the unaided eye. The heat
treat cracks may be found by use of fluorescent penetrant
inspection (FPI), a nondestructive inspection technique.
[0063] The inventors have carried out work to determine parameters
with respect to the boron content of the alloy. It has been found
that boron at 30-50 ppm in the alloy of the invention is not
particularly suitable for castability of large buckets. At this
level of boron, a 2320.degree. F. heat treatment fully solutions
the gamma-prime phase and provides optimum longitudinal mechanical
properties for long bucket life. However, at this low level of
boron, the transverse creep properties are less than optimum for
large buckets.
[0064] In contrast, boron at 130 ppm in the alloy has been found to
be suitable for castability, but is not particularly suitable for a
full solution heat treatment. The melting point of such an alloy is
reduced to about 2301.degree. F., and the highest heat treatment
that may be reliably applied is 2280.degree. F. if melting is to be
avoided. Heat treatment at a temperature of 2280.degree. F.
provides only about 60-80% solutioning of the gamma-prime phase,
but this is generally acceptable for a full-life bucket. Thus, the
gamma-prime phase in the 130 ppm boron material cannot be fully
solutioned because the alloy starts to melt before full solutioning
can be achieved.
[0065] The transverse creep properties are acceptable with this
higher level of boron of 130 ppm. However, at this level of boron,
a 5% failure rate for heat treat cracking has been observed.
[0066] It has been found that a level of boron of about 80-100 ppm,
i.e. about 90.+-.10 ppm, is optimum for castability. In order to
improve the longitudinal creep properties for an improved margin
for bucket life, an increase in the percent gamma-prime solutioning
over about 60-80% is desired. This may be possible due to the
increase in melting temperature for the intermediate (about 90 ppm)
boron level. In addition, this 90 ppm level of boron provides a
greater margin against heat treat cracking, and increases the yield
during the solution heat treatment operation.
[0067] Castability experiments have been performed using the
procedure described in U.S. Pat. No. 4,169,742 (herein incorporated
by reference). A master "lean" heat of DSN4 was formed, where B and
Zr were removed, but otherwise the remaining elements (except for C
and Hf) were the same as in SC Rene N4 as described above. A
three-level, four-factor designed experiment (DOE) was then carried
out. Castability was examined using the aforementioned castability
test with the grain boundary strengthening elements (& Ti) at
the following levels (Zr was not varied but kept at the lowest
level), as shown in the Table below:
1 Weight Percent of Elements at the 3 levels Desired for DOE
Experiment Element Low Level Medium Level High Level Carbon 0.06
0.10 0.14 Hafnium 0.25 0.45 0.65 Boron 0.0075 (75 ppm) 0.01 (100
ppm) 0.015 (150 ppm) Titanium 3.37 3.50 3.65
[0068] It has been determined that castability is improved if Hf
and Ti are run at their highest levels, but this also depends upon
the B content. The differences between C and B could not be fully
ascertained because this was not a full factorial experiment (which
would have been 3.times.3.times.3.times.1.times.3 or 81
experiments), and due to the limited ranges of carbon
(0.14%-0.06%=0.08%) and boron (0.015%-0.0075%=0.0075%) versus
ranges for hafnium (0.65%-0.25%=0.45%) and titanium
(3.65%-3.37%=0.28%).
[0069] Hafnium (Hf) is known to cause casting defects known as
"bands", which are transverse linear indications as determined
during FPI examination. It has been determined that 0.75% Hf causes
bands in low or high boron DS Rene N4 (boron 30-50 ppm--or 80-130
ppm), whereas 0.25 weight % Hf and 0.45 weight % Hf resulted in no
bands. From the standpoint of acceptable transverse creep
ductility, the lower level of Hf in production buckets is not
allowed to fall below 0.15 weight %. Thus, for DS Rene N4, Hf is
generally maintained in the range of about 0.15-0.45 weight %.
[0070] Experiments have been carried out using controlled amounts
of boron and hafnium added to a baseline N4 master heat to
determine their effect on castability, expressed as total inches of
crack length. The master heat composition was, by weight, 0.04%C,
9.77% Cr, 7.49% Co, 5.92% W, 1.51% Mo, 4.21% Al, 3.37% Ti, 0.45%
Nb, 4.71% Ta, 0.16% Hf, 0.00% B, less than 0.005% Zr, balance Ni.
The results for thin wall castings (about 60 mils thick) and thick
wall castings (about 120 mils thick) are shown in the chart below.
The least amount of cracking (expressed as inches of crack) is
best.
2 "Inches of Crack Length from Castability Test" 1 2 3 Other heats
made by doping master lean heat
[0071] The chart above shows that thin wall versus thick wall data
are comparable, and that best castability is observed for DS Rene
N4 with 40 ppm (0.004%) boron and no Hf, OR 130 ppm (0.013%) boron
and 0.45% Hf, indicating there is a "saddle point" in the data. No
Hf is not considered to be acceptable as it may decrease transverse
creep ductility. It has been found that castability of the 90 ppm
boron alloy with 0.15% Hf is improved over the castability of 130
ppm boron material with 0.15% Hf. Higher Hf levels may create
transverse "bands" or dross. Banding as noted earlier is a known
casting flaw, and "dross" is a nonmetallic inclusion caused by a
chemical reaction between dissolved oxygen in the metal and free
hafnium in the metal which combine to form a stable oxide such as
HfO.sub.2 (hafnium oxide). In either case, lower Hf (typically
0.15-0.45 weight %) is desirable in creating defect-free
castings.
[0072] The method of the invention includes a ramp heat treatment
up to the solution heat treatment temperature plus the
post-solution heat treatment cooling rate down to room temperature.
Four factors are important to achieving reduced heat treatment
cracking. Each has been investigated at two levels, as discussed
below.
[0073] HIP temperature (2175.degree. F. or 2225.degree. F.);
[0074] solution heat treat temperature (2270.degree. F. or
2290.degree. F.);
[0075] post-solution heat treatment temperature cooling rate (slow
furnace cool at about 35.degree. F./minute, or fast gas fan cool at
about 150.degree. F./minute, both followed by gas fan cooling from
a temperature of about 2050.degree. F.); and
[0076] solution heat treatment atmosphere (vacuum or argon
gas).
[0077] HIP or "hot isostatic pressing" is a means by which internal
porosity in the casting can be closed by the application of
external pressure. This is achieved in a HIP vessel. The porosity
is closed by the application of temperatures in the range of
2175.degree. F.-2225.degree. F. and 15,000 psi for an alloy like SC
or DS Rene N4.
[0078] A heat treat temperature of 2290.degree. F. was chosen as
the highest temperature possible for the solution heat treatment.
The temperature of 2290.degree. F. was reached using part of a
RAMP4 cycle to 2290.degree. F., which is set forth in the Table
which follows:
3 Typical RAMP4 Solution Heat Treatment Cycle to 2300 F. Hold
Heating Ramp Rate Temp. Hold Time Rate Purpose/Results 25 F./minute
1400 F. 10 mm. -- Stabilize, and begin introducing 800 microns of
argon gas. Not used if already running in a 100% argon atmosphere.
25 F./minute 2225 F. 8 hour Increase to homogenize 25 F./hour 2250
F. 4 hours Increase to homogenize 30 F./hour 2280 F. 2 hours
Increase to homogenize 10 F./hour 2290 F. 2 hours Increase to
homogenize 10 F./hour 2300 F. 0.5 hours Cool to RT Achieve final
gamma-prime solutioning
[0079] This heating cycle was chosen because there was no evidence
of melting or heat treat cracking using a variety of bucket or
ingot sizes. For the 2290.degree. F. solution cycle, that part of
the RAMP4 cycle above (including up to 2290.degree. F./2 hours) was
chosen. A temperature of 2290.degree. F. was chosen because
previous work by the inventor showed that at 2300.degree. F.,
recrystallized grain (RX) defects could form in DS Rene N4, and to
avoid the RX grains the temperature would have to be lowered. Since
it is only possible to control the temperature to within 10.degree.
F., a temperature of 2290.degree. F. was chosen as the highest
practical heat treatment temperature.
[0080] The second solution heat treatment temperature was
2270.degree. F. This was based upon metallography photographs
showing the percent of gamma-prime solutioning, and was considered
to be the lowest acceptable temperature capable of providing a
full-life bucket.
[0081] The results are set forth in FIG. 1. Heat treating at
2270.degree. F..+-.10.degree. F. was equivalent to heat treating in
the range of 2260-2280.degree. F., and heat treating at
2290F.+-.10.degree. F. was equivalent to heat treating in the range
of 2280-2300.degree. F.
[0082] A reason that it is difficult to determine what causes heat
treat cracking is because the buckets cannot be examined at the
solution heat treatment temperature to see if they are cracked. It
is necessary to cool the buckets down to room temperature for
examination. In addition, the section size of the bucket has some
effect on residual stress, which further complicates the heat treat
cracking issue.
[0083] The HIP temperature was probably not significant because it
is well below the incipient melting temperature. Furthermore, the
HIP cycle is also a thermal cycle and therefore can provide some
homogenization to the DS Rene N4. In this case, the 2225.degree. F.
cycle would provide more homogenization than the 2175.degree. F.
cycle. But based upon the experimental analysis, it was shown the
amount of homogenization provided by either HIP cycle is inadequate
to influence the heat treat cracking.
[0084] In addition to the previous HIP and solution heat treat
cycles, the cooling rate was believed to have an effect on heat
treat cracking. To investigate this, two cooling rates were
employed. The first rate was produced from a gas fan cool in the
range of 100-150.degree. F./minute, which is available on most
vacuum furnaces. The second rate was selected because it was used
during development trials, specifically from Ramp 4 heat treatment
where gas fan cooling was not available--only natural cooling was
available (called furnace cooling). Furnace cooling is achieved by
just turning off the furnace and letting it cool naturally. In this
case, the range was measured to be 35-75.degree. F./minute.
[0085] Finally, the furnace atmosphere was felt to be important.
Two atmospheres are commonly available. The first is a vacuum
atmosphere with some argon backfill, in the range of 400-800
microns. The second atmosphere that is commonly employed (and was
used in RAMP 4 heat treat) was 100% argon (not a vacuum).
[0086] The furnace environment during the heat treat experiment was
determined to be a minor factor. Initially, it was thought a vacuum
or partial vacuum environment could cause heat treat cracking by
volatilizing the grain boundary elements. In this instance, during
a vacuum heat treatment, some elements with a low vapor pressure
can be removed from the alloy, possibly leaving void spaces such as
along a grain boundary (which could be interpreted as a crack).
However, neither atmosphere (vacuum with partial pressure argon or
100% argon) had a significant effect on the heat treat cracking of
the DS Rene N4 buckets.
[0087] FIG. 1 shows that the cooling rate has the greatest
influence on the heat treat cracking, followed closely by the
solution heat treatment temperature (the greater the slope, the
larger the effect). The other two factors--HIP temperature and
furnace atmosphere--are considered to be minor factors. Thus, the
slower cooling rate and the lower solution heat treatment
temperature afforded the best results (least amount of heat treat
cracking).
[0088] When the alloy is DS Rene N4 alloy with 130 ppm boron, the
optimum heat treatment includes a HIP cycle at 15,000 psi for 4
hours in the range of 2175-2225.degree. F. followed by a solution
heat treatment temperature in the range of 2270.degree. F. to
2290.degree. F., followed by a furnace cool of about 35.degree.
F./minute to about 2050.degree. F. and gas fan cooling to less than
1200.degree. F., to prevent heat treat cracking.
[0089] The solution temperature had the largest effect on heat
treat cracking, and is generally 2280.degree. F..+-.10.degree. F.
(i.e. 2270.degree. F.-2290.degree. F.), more usually 2280.degree.
F. This provides for a lower incidence of heat treat cracking while
still achieving adequate gamma-prime precipitate solutioning.
[0090] The cooling rate is generally in the range of 25-45.degree.
F./minute, for example 35.degree. F./minute. The gas fan cooling
may be initiated when the temperature reaches approximately
2050.degree. F..+-.50.degree. F.
[0091] The furnace atmosphere may be 100% argon, or vacuum plus
argon partial pressure (400-800 microns). Vacuum plus argon partial
pressure (400-800 microns) is generally employed. The use of this
small amount of argon helps reduce the vaporization (depletion) of
chromium during the heat treat cycle.
[0092] From this 130 ppm boron group, 1 cracked bucket occurred out
of 19 total, or a 5.2% failure rate, due to heat treat cracking.
Part of the reason for this is the small margin of error between
the heat treat temperature (2280.degree. F.) and the incipient
melting point of this alloy (2301.degree. F.). The temperature
difference between heat treat temperature and melting point is
2301-2280.degree. F.=21.degree. F. This small margin is less than
the error of thermocouples, which would approach 1% of the actual
temperature, or at 2280.degree. F. it would be 22.8.degree. F. This
means the actual heat treat temperature could exceed the true
melting point of the alloy, without the furnace operator's
knowledge. If that happened, it would cause incipient melting,
which in the presence of residual stress may lead to heat treatment
cracks. This is compared to a margin of 54.degree. F. for the 40
ppm boron material between the heat treat temperature and the
potential for incipient melting and heat treat cracking
(2334.degree. F.-2280.degree. F.=54.degree. F.)
[0093] The margin for temperature error with a 2280.degree. F. heat
treatment is shown in the Table below.
4 Incipient Aim Heat Melting Point Treat Margin for Temp.
DSN4/GTD444 (.degree. F., on Temperature Error during Heat Alloys
heating) (.degree. F.) Treatment (.degree. F.) DSN4 w/31 ppm 2346
2280 66 Boron DSN4 w/36 ppm 2344 2280 64 Boron DSN4 w/40 ppm 2334
2280 54 Boron DSN4 w/90 ppm 2311 2280 31 Boron DSN4 w/130 ppm 2301
2280 21 Boron
[0094] The advantage in going to an intermediate level of boron,
such as in the 80-100 ppm range, is in the margin between incipient
melting (when the alloy starts to melt) at the 2280F heat treat
temperature. For example, at 130 ppm B, there is only 21.degree. F.
between the incipient melting point and the 2280.degree. F. heat
treatment. This is not an acceptable range, because the error due
to the thermocouple (TC) alone is 22.8F (1% of 2280F). But at 90
ppm B the range between incipient melting and the heat treat
temperature has increased to 31.degree. F. Therefore, after
accounting for 22.8.degree. F. of TC error, there is still
8.2.degree. F. of temperature margin (31.degree. F.-22.8.degree.
F.) between the incipient melting point and the 2280.degree. F.
heat treat temperature. While 8.2.degree. F. of margin is not a
lot, it is an equivalent margin when compared to other
high-technology SC or DS alloys.
[0095] Buckets from 90 ppm boron heats were successfully heat
treated at 2280.degree. F. with 0% failure rate due to heat treat
cracking. For the 90 ppm boron material, the melting point was
determined to be 2311.degree. F. Thus, with a heat treat
temperature of 2280.degree. F. the temperature difference between
the heat treat temperature and the melting point is
2311-2280.degree. F.=31.degree. F. The temperature difference
between the heat treat temperature and the incipient melting point
is greater than the thermocouple error (1% of 2280.degree. F. or
22.8.degree. F.), so there is less opportunity for unknowingly heat
treating the buckets above their incipient melting point, causing
heat treat cracking.
[0096] It has been found that the amount of boron influences the
incipient melting point of the alloy, i.e. less boron is better.
The amount of boron additionally influences the transverse creep
ductility, i.e. more boron is better (although boron does not
influence the longitudinal creep ductility). Moreover, a higher
solution temperature leads to more gamma prime solutioning, and
more gamma prime solutioning leads to more longitudinal creep life.
However, the solution temperature influences the transverse creep
ductility, whereby a lower temperature is better.
[0097] Thus, optimization of the alloy requires transfer functions
(equations) that describe these features in terms of controllable
factors. Additionally, creep strength and casting yield are not
measured in similar units. Therefore, the transfer function is
expressed as a percentage of the best case for heat treat yield
(100%) and creep strength (100%). The transfer function generation
is described below.
[0098] Heat treat yield is a function of two variables, boron
content and solution heat treatment temperature. If the B content
is too high, incipient melting or heat treat cracking occurs at
segregated areas in the casting, resulting in scrap. If the
solution heat treatment temperature is too high, incipient melting
and recrystallization (RX) limit yield. Recrystallized grains
result from a phase transformation where residual strains in the
material on heating cause the formation of strain-free grains with
little or no strength, i.e. critical defects. The following
spreadsheet shows the data used to generate Heat Treat Yield
Transfer Function Equation 1:
5 Heat Treat Yield Boron (B) (Percent) Temp. (F.) Content (ppm) 100
2280 40 50 2292 130 50 2310 40 90 2280 130 0 2327 40 0 2310 130
[0099] Regression with the data leads to the following regression
equation:
Heat Teat Yield=5448-2.34(Temp)-(0.340)*(Boron content) Eq. 1
[0100] This is the first transfer function for yield.
[0101] A statistical analysis was conducted for the data, resulting
in the following standard tables:
6 Predictor Coef StDev T P VIF Constant 5448.0 671.8 8.11 0.004
Temp -2.3353 0.2907 -8.03 0.004 1.1 B -0.3398 0.1117 -3.04 0.056
1.1
[0102] S=11.59 R-Sq.=95.6% R-Sq. (Adj)=92.6%
[0103] (R-Sq=R.sup.2 or R squared; adj means Adjusted)
[0104] The next transfer function is for longitudinal creep
strength. This is a function of gamma-prime precipitate solutioning
versus the solution heat treatment temperature, as the only way to
get 100% creep strength is to fully solution the material. The
following is data relating the percent of full creep strength
versus the heat treat temperature for DS Rene N4:
7 Creep Heat Treat Strength Temperature (Percent) (F.) 100 2320 90
2300 60 2280 40 2215
[0105] The longitudinal creep strength is in percent of maximum
obtainable, and the heat treatment temperature (t) is the solution
heat treatment temperature in degrees F.
[0106] The data was used to solve for Equation 2 (see the
Regression Plot in FIG. 2). The curve has the correct dependency of
creep strength on solution heat treatment temperature. It will be
noted that as-cast DS Rene N4 has about 40% of the possible creep
strength and that solution heat treatment of DS Rene N4 at
2320.degree. F. gives 100% creep strength. This is the second
transfer function.
[0107] A further important feature of the alloy is creep strength
transverse (transverse creep strength) to the grain boundaries.
This is important in the tip shroud and other areas where loading
is not in a radial direction on he part. The following data was
extracted for transverse creep strength:
8 Percent of Transverse Creep Boron Content Strength. (ppm) 50 40
100 80 80 130 90 100
[0108] This information created a non-linear regression plot as
shown in FIG. 3. Equation 3 is:
Y=-40.7431+2.9113X-1.54E-02X.sup.2
[0109] The three transfer functions (equations) can be solved
simultaneously using an optimization spreadsheet shown below:
9 MULTIPLE RESPONSE OPTIMIZATION 4 5 6 7
[0110] The solutions with respect to Heat Treat Yield, Longitudinal
Creep and Transverse Creep Strength were:
10 Needs Heat Treat Yield 1 1 2 2 3 3 Longitudinal Creep Strength 2
3 1 3 1 2 Transverse Creep Strength 3 2 3 1 2 1 Optimize B ppm 40
40 94.5 94.5 40 94.5 Temp F 2280 2280 2296 2280 2296 2280
[0111] A "1" means optimization on this need first, followed by "2"
and finally "3".
[0112] This results in an optimized alloy with a boron content of
94.5+/-10 ppm and a heat treatment temperature of
2280.+-.20.degree. F.
[0113] FIG. 4 is plot showing creep elongation as a function of
test temperature. FIG. 5 is a plot showing the effect of varying
amounts of boron on incipient melting of SC or DS Rene N4.
[0114] FIG. 6 shows a third and fourth stage bucket fabricated from
an alloy of the invention. FIG. 7 is a gas turbine engine showing
the location where buckets of the invention are used.
[0115] While the invention has been described in connection with
what is presently considered to be the most practical and preferred
embodiment, it is to be understood that the invention is not to be
limited to the disclosed embodiment, but on the contrary, is
intended to cover various modifications and equivalent arrangements
included within the spirit and scope of the appended claims.
* * * * *