Large forging manufacturing process

This invention was made with Government support under Contract No. DE-FC21-95MC31176 awarded by the Department of Energy. The Government has certain rights in this invention.

This invention relates to forgings used for large land-based gas turbines, and particularly to large Alloy 718 forgings that are prone to a problem known as abnormal grain growth.

The forging of Alloy 718 involves heating a billet and forging it in one or many steps (also referred to as upsets) to the final required shape. The billet must be reheated before each upset. After forging, the shaped parts are solution treated at a high temperature (1700-1825F), and then aged at a lower temperature (1325-1150F) to develop strength. Under certain process conditions, Alloy 718 forgings develop abnormal grain growth when heated to the solution temperature. This has not been a serious problem for small forgings (as discussed below), but it has been a serious problem for large forgings which, for purposes of this invention, are those over 10,000 pounds in weight.

Abnormal grain growth, also referred to as secondary grain growth or critical grain growth, occurs when a few grains in the material grow to a very large size compared to neighboring grains. This occurrence alters the mechanical properties of the material. Specifically, not only does abnormal grain growth reduce fatigue resistance and yield strength of the material, it also impairs the ability to detect small defects by ultrasonic testing. Abnormal grain growth does however, improve creep resistance at high temperatures, and may therefore be desirable in certain instances.

The possibility of developing abnormal grain growth in Alloy 718 forgings has been known for some time. A prior document which describes abnormal grain growth in Alloy 718 and conditions which promote abnormal grain growth is the “Study of Secondary Grain Growth on 781 Alloy” by J. F. Uginet and B. Pieraggi; The Minerals, Metals and Materials Society, 1997. Abnormal grain growth was not regarded as a serious problem in that process modifications were available that minimized or eliminated its occurrence. These process modifications, however, work well with small forgings but not with large forgings as defined above. More specifically, with small parts, one could:

1. Avoid low strains (amount of upsets in a forging step). There are some difficulties in doing this for large parts because the press capacity may allow only small upsets each time.

2. Use higher strain rates (related to the speed at which the top dies move). This again will not work for very large parts because the higher strain rates require higher press loads that would exceed the capacity of the largest presses in the world.

3. Avoid doing a solution treatment and do a direct age instead. This works out well for small parts because the cooling rate in air after completion of forging is adequate to ensure a fully solutioned structure for a small part. If air cooled after forging, the cooling rate at the center of very large parts will be very slow and will not have a fully solutioned structure. The absence of a fully solutioned structure means that the part will not develop high strength after the aging heat treatment. Therefore, after the solution treatment, the parts must be quenched in oil/water to retain a fully solutioned structure.

This invention involves the identification of a unique processing window for large Alloy 718 forgings which causes abnormal grain growth. By then avoiding this window, abnormal grain growth can be eliminated, thereby permitting large forgings that have a uniform grain structure. Alternatively, the process permits the formation of abnormal grain growth in selected areas when considered desirable.

Initially, a study on the effect of forging parameters was carried out using small specimens, but the process was made to simulate the processing of large forgings. It was observed that abnormal grain growth occurs under specific conditions of:

(a) starting grain size;

(b) forging temperatures;

(c) forging strains;

(d) forging strain rates;

(e) number of upsets; and

(f) solution treatment temperature.

In one embodiment of this invention, abnormal grain growth can be avoided by a forging process which takes into account the above factors, within the parameters disclosed herein.

In an alternative embodiment of the invention, where there is a need to intentionally create abnormal grain growth in any areas of a forging, items (c) and (f) are altered, as also described further herein.

More specifically, the present invention relates generally to a process for forging large components of Alloy 718 material comprising:

a) providing a billet with an average grain size between ASTM 0 and ASTM 3;

b) heating the billet to a temperature of between 1750° F. and 1800° F.;

c) upsetting the billet to obtain a component part with a minimum strain of 0.125 in at least selected areas of the part;

d) reheating the component part to a temperature between 1750° F. and 1800° F.;

e) upsetting the component part to a final configuration such that the selected areas receive no strains between 0.01 and 0.125;

f) solution treating the component part at a temperature of between 1725° F. and 1750° F.; and

g) aging the component part over predetermined times at different temperatures.

When there is a need to intentionally create abnormal grain growth in any areas of a forging, then steps e) and f) are changed only as follows.

e) finish forge the part to intentionally create strains of 0.01 to 0.125 in the required areas.

f) solution treat the component part at 1825-1850° F.

The process in accordance with the invention has advantages over the prior art. Specifically, one can develop a control process which can eliminate abnormal grain growth and have a uniform grain structure specifically for large 718 alloy forgings. Alternatively, one can develop a control process which does produce abnormal grain growth intentionally in specific areas to meet specific property needs. This aspect can be used in both large and small forgings.

FIG. 1 is a photomicrograph showing evidence of abnormal grain growth in an alloy 718 component;

FIG. 2 is an end view illustrating the cross sectional shape of a small scale specimen used in the development of the invention;

FIG. 3 is a model of the specimen showing strain contours and measurements for use as a reference;

FIG. 4 is a model of a component part forged by a conventional process; and

FIG. 5 is a model of a component part forged by the process of the invention.

With reference to FIG. 1, an example of abnormal grain growth is shown in a photomicrograph with a magnification of 200×. Specifically, evidence of abnormal grain growth is shown in the gray areas, one of which is designated by numeral 12 . As indicated above, abnormal grain growth occurs when a few grains in the material grow to a very large scale compared to neighboring grains. Abnormal grain growth reduces fatigue resistance and yield strength of the material. It also impairs the ability to detect small defects by ultrasonic testing. On the other hand, since abnormal grain growth does improve creep resistance at high temperatures, it may be desirable to foster such growth under certain circumstances.

For purposes of developing the process in accordance with this invention, small scale specimens were used. The specimens initially as supplied for testing had a cross sectional shape as indicated in FIG. 2. A side elevation of the specimen is shown generally in FIG. 3 . The specimen in FIG. 3 is also shown to include typical strain contours, with strains in each labeled area indicated adjacent the figure.

The process developed in accordance with the invention is based on the test regimen described below.

The small scale specimen or billet 14 is an Alloy 718 forge material with grain size of ASTM 4-5 and ASTM 8-10. The specific geometry of the specimen as shown in FIGS. 2 and 3 allowed strains of different levels to be generated in the same specimen, thus minimizing the number of specimens.

The test methodology included small scale upsets done in a servo-hydraulic testing machine. The specimen 14 and forging dies were both heated and maintained at the temperature of testing, i.e., it was an isothermal process. Finite element modeling of the forging process was done using a commercial code DEFORM.

The following parameters were initially considered as factors in the occurrence or prevention occurrence of abnormal grain growth:

(1) starting grain size;

(2) forging strain;

(3) number of upsets;

(4) reheat time during upsets;

(5) solution treatment time;

(6) cooling rate;

(7) forging temperature;

(8) forging strain rate;

(9) solution treatment temperature.

A preliminary study of the test results showed the following:

(1) Abnormal grain growth does not occur if the grain size is larger than ASTM 8.

(2) The number of upsets did not have a significant effect.

(3) The reheat time was controlled by the size of the forging. Large parts need longer reheat times.

(4) The solution treatment time was also controlled by the size of the part.

(5) The cooling rate from solution treatment temperature was also controlled by the size of the part.

A detailed Design of Experiments (DOE) study was done with the remaining factors, using eight sub-scale test specimens as shown in FIG. 2, based on the following matrix.

Specimens were cut up after the upset experiments for microstructure analysis. It was observed that the abnormal grain growth was located in the low strain region, but when strain reached a certain level, the abnormal grain growth disappears. The locations of abnormal grain growth were recorded and strain level at the certain location was then calculated by commercial forge modeling software DEFORM 2D. The highest strain value (Hstrain) of each specimen represents the amount of abnormal grain growth in the particular specimen. By running statistic software Minitab 12, it was determined that lowering forging temperature and lowering the solution heat treatment temperature could reduce Hstrain and thus the possibility of abnormal grain growth, but strain rate has little effect on Hstrain and thus the amount of abnormal grain growth generated.

Extensive modeling of the finish forging steps of the full size forging was made. Forgings which had evidence of abnormal grain growth were cut up and locations correlated with the temperature/strain/strain rate history of the small scale specimens. After this correlation was established it was possible to design a process which would avoid abnormal grain growth. In addition, the solution treatment temperature was reduced to 1725F.

A comparison of the forged samples 22 , 24 in FIGS. 4 and 5 illustrate the reduced area of low strain and thus reduced abnormal grain growth achieved by the process of this invention. With reference to FIGS. 4 and 5, samples 22 and 24 each include a double cone-shaped geometry, with a notch at 16 and annular steps or shoulders formed at 18 and 20 . The samples 22 and 24 are components made by a conventional process, and by a process in accordance with this invention, respectively. Sample 22 in FIG. 4 exhibit a relatively large area in the low strain range, which has a tendency for abnormal grain growth. On the other hand, sample 24 in FIG. 5 shows a very limited low strain region. This low strain region will be removed by subsequent machining and the possibility of abnormal grain growth is thus eliminated.

Based on the above described test results, it was determined that in order to avoid abnormal grain growth, the following process should be employed:

a) Start from a billet of average grain size ASTM 0 to ASTM 3.

b) Heat the part to a temperature between 1750° F. and 1800° F.

c) Upset forge the part in to get a minimum strain of 0.125 in all areas of the part for each upset; this will recrystallize the part to a fine grain size of ASTM 6-8.

d) Reheat the part at a temperature of 1750° F. to 1800° F.

e) Upset forge again (if necessary) to get a minimum strain of 0.125 in all areas of the part for each upset.

f) Reheat the part (if step e) is performed) at a temperature of 1750° F. to 1800° F.

g) Final forge the part so that no areas of the forging receives a strain of between 0.01 and 0.125. Do not reheat the forging or re-strike it at this stage as it is very likely to cause abnormal grain growth.

h) Solution treat the part at a low temperature of 1725° F.-1750° F.

i) Age the part at 1325° F. for 8 hours followed by another 8 hours at 1150° F. (This is a standard practice for Alloy 718).

With the above process, abnormal grain growth can be eliminated from all areas of the component part.

When there is a need to intentionally create abnormal grain growth in any areas of a forging, then steps g) and h) are changed only as follows.

g) Finish forge the part to intentionally create strains of 0.01 to 0.125 in the required areas.

h) Solution treat the part at 1825-1850° F.

In this way, the component part can be selectively forged to create areas with no abnormal grain growth as well as areas where abnormal grain growth occurs but where creep resistance at high temperatures is improved.

In still another alternative, if the start-up grain size of the billet in step a) above is ASTM8-10, then steps b) and c) can be eliminated, and the process can continue with step d).

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.