US2913337A

US2913337A - Shell molding

Info

Publication number: US2913337A
Application number: US524293A
Authority: US
Inventors: Herman C Kretz
Original assignee: Cooper Alloy Corp
Current assignee: Cooper Alloy Corp
Priority date: 1955-07-25
Filing date: 1955-07-25
Publication date: 1959-11-17
Anticipated expiration: 1976-11-17

Description

SHELL MOLDING Herman C. Kretz, Livingston, N.J., assignor to Cooper Alloy Corporation, Hillside, N..I., a corporation of New Jersey No Drawing. Application July 25, 1955 Serial No. 524,293

12 Claims. (Cl. 75129) This invention relates to shell molding, and more particularly to the shell molding of alloys which are high in nickel, especially stainless steel alloys.

Alloys which are high in nickel are in wide use, particularly in the form of stainless steel, most commonly the 18 and 8 family of chromium nickel stainless steel alloys. These have long been cast successfully in sand molds, and in recent years have been cast in shell molds. The change to shell molding introduced new problems difiicult to overcome. Specifically, castings have been found by X-ray examination to contain gas. Many things have been tried in an efiort to overcome this difficulty, including experiments comparing the use of an arc furnace with an induction furnace for melting the steel; variations in temperature of the melt and in the temperature of pouring, with different delay times between melting and pouring; heating compared with cooling of the shell molds before pouring; overcuring and undercuring the molds; washing the molds with various washes and drying the same before pouring; special precautions for venting; and changes in the resin content of the sand.

The primary object of the present invention is to overcome the foregoing difiiculties. I have found that one important step in that direction is the introduction of an oxidizing agent in the melt. This is contrary to usual practice, which includes the addition of a reducing agent.

I have further found that a reduction in the silicon content reduces the gas in the casting, but that unfortunately the reduction in silicon content leads to surface pitting, which is itself undesirable. I have further found, however, that surface pitting seems to depend on the relation of the silicon content to the manganese content, and that if the manganese content is kept substantially smaller than the silicon content, surface pitting may be avoided while limiting the silicon content to an amount which helps eliminate gas in the casting.

The invention accordingly comprehends the addition of an oxidizing agent; a control of the silicon content, which control allows for the presence of the oxidizing agent; and limitation of the manganese content, all as are hereinafter described in greater detail in the following specification.

The invention applies particularly to steels having a high nickel content, for the nickel appears to have an aflinity for gases. It therefore is particularly important in the casting of chromium nickel stainless steel of the 18 and 8 type.

The difiiculty arises when using shell molds. The invention may be considered an empirical discovery, regardless of the underlying theory, but as a matter of possible explanation, and not in limitation of the invention, I may explain that I believe that the resin content in shell molds breaks down at high temperatures when the molten steel is poured into the mold. It seems probable that gases are given off, and that these gases are absorbed by the molten metal when in'a reducing state. Such a reducing state has always heretofore been favored for casting purposes.

2,913,337 Patented Nov. 17, 1959 ice In experimental work I have lowered the resin content of the sand mix, and found no reduction in the gaseous condition when dropping from, say, 7 or 8% to 5% resin content in the sand mix. A further reduction to, say, 4% resulted in some reduction in gas, but by this time the resin content was too low for practical casting operations, that is, the shell mold lacked strength.

It is my theory that by changing the melt from a reducing state to an oxidizing state the gases combine chemically or for some reason the tendency to absorb gas is reduced, and such has indeed proved to be the case. This makes it possible to use the full desired resin content. In my present work the shell mold mix consists of 87% of #160 New Jersey silica sand, 5% zircon flour, and 8% thermosetting resin.

A convenient oxidizing agent to add to the melt is nickel'oxide. This may be added in a range of form, say, 1.3 to 2% of the melt, and I prefer to use 1.33% of the melt. Thus with a standard charge of 1,500 lbs. I add 20 lbs. of nickel oxide.

The invention as so far described is helpful, but does not wholly eliminate gas, as revealed when the castings are subjected to close X-ray scrutiny. I have found also that less gas is present in castings made from heats having a comparatively low silicon content. However, in

working in this direction with lower and lower amounts of silicon, I found a growing problem caused by severe surface pitting, leading to generally poor appearance. Conversely, again raising the silicon content improved the surface, but reintroduced the problem of gas. The usual silicon content of 1 to 1.25% heretofore used in the art, with ordinary sand molds, is out of question for use in shell molds. However, an intermediate value of silicon appears to be feasible and desirable, provided that the manganese content is also reduced, so as to keep the manganese content substantially lower than the silicon content.

I have found, and I recommend that the silicon content in the final analysis of the steel be kept in a range of 0.40% to 0.65%, and that the manganese be limited to an amount not exceeding 0.30%. However, the silicon content as calculated theoretically in the initial charge preferably should go up to 0.85%, because in the final analysis the silicon content goes down to a value less than 0.65%. I believe this is because some of the silicon is oxidized by the oxidizing agent put into the melt.

If the silicon content in the final analysis goes below 0.40% there is great likelihood of surface pitting or poor surface finish, and on the other hand, if the silicon content in the final analysis goes above 0.65% there is strong likelihood of gas porosity being found in the castings.

' Gas dissolved in the hot metal may be released when the metal cools, causing porosity, according to one theory.

These values apply to stainless steel of the 18 and 8 type. I refer particularly to stainless steels known under the following type numbers of the Alloy Casting Institute (ACI), with the equivalent type numbers of the American Iron and Steel Institute (AISI):

AC1: AISI CF-ZO 302 CF-S 304 CF-SM 316 CF-8C 347 HH 309 I-IK 310 It should be noted that there are minor variations from these analyses which may be designated by a different AISI or AC1 number, but which in essence belong to the same generic alloy grouping, as the abovelistt For example, by reducing the carbon content of type Percent Carbon "Maximum" .07 Chromium 19 to 22 Nickel 27.5 to 30.5 Molybdenum 1.75 to 2.5 Copper Minimum 3 The above analysis does not include the silicon and manganese, which are to be proportioned as set forth above. It also does not include selenium, which may be considered a residual, as explained later.

Something may be said in respect to the melting prac tice used to make up the shell mold heats. The silicon must be added in the form of ferrosilicon and not as calcium silicon. It has been customary to add calcium silicon as a reducing agent, whereas here the melt is to be in an oxidizing condition. Calcium silicon therefore should be avoided, and instead the silicon is brought to desired amount by the addition of ferrosilicon. The ferrosilicon is added to the charge in solid form, and the components are all melted down together. It appears to be undesirable to add the 'ferrosilicon after melting down. The nickel oxide also preferably is included in the original charge. If, as is usually the case, revert scrap is used, allowance must be made for the percentage of silicon in the revert scrap. Indeed the charge may be entirely revert scrap.

Ferro selenium is commonly added as a de-gasiiier, and in the present invention it is also desirable to add ferro selenium. The amount is small, it being 6 oz, for a 1,500 lb. charge, and it helps counteract hydrogen porosity. The ferro selenium is added after melting down. The form selenium may be approximately 55% selenium, as is now the standard practice in the industry.

The theory of this invention is that under the prior melting practice, the metal is in a high state of deoxidation, due to the addition of various deoxldizers, and also due to the presence in the metal of around 1% silicon. In my method I increase the amount of oxygen which is dissolved in the metal at the time of pouring. This I do by starting out in the melt with a reduced value of silicon of about 0.85%, and then adding nickel oxide which has the effect of both reducing the silicon content of the metal by oxidation, and of dissolving oxygen in the metal through the decomposition of the nickel oxide. At this point, due to the low value of silicon in the metal (0.40 to 0.65%) the metal can support a relatively large amount of oxygen in solution. Thus when the metal is poured into the shell mould, the reducing gases formed by resin breakdown, either combine with the dissolved oxygen harmlessly, or are repelled from entering the metal because the oxygen content of the metal is high.

To summarize the charge can contain up to 100% revert scrap. The manganese in the charge is kept below 0.30%. The silicon in the charge is calculated at 0.85%, including the addition of ferrosilicon to the extent required to reach 0.85% silicon. Twenty lbs. of nickel oxide is included in each 1,500 lbs. charge. No deoxidizer is used. Six oz. of ferro selenium is used per 1,500 lbs. charge to counteract hydrogen porosity. The term selenium is approximately 55% selenium.

All melting is done in an induction furnace, designed for a normal charge of 1,500 lbs. The nickel oxide and any requiredjerrosilicon are added to the cold charge. The metal is melted and brought up to tapping temperature. Just prior to tap the ferro selenium is added to the metal, in accordance with the usual practice with stainless steel induction furnace heats. On analysis of the final product, the manganese comes out under 0.30%, usually 0.15% to 0.25%. The silicon averages 0.40% to 0.50%.

It is believed that the method of my invention, as well as the advantages thereof, will be apparent from the foregoing detailed description. It will also be apparent that changes may be made, without departing from the scope of the invention, as sought to be defined in the following claims.

I claim:

1. In the art of casting steel having a nickel content of not less than about 8%, to desired shape in shell molds having adequate resin content for strength during casting, the method of avoiding gas and pitting, which includes adding ferro silicon to a cold charge before melting the charge, and thereby adjusting the silicon content to be in a range of from 0.40% to 0.65%, while limiting the manganese content to an amount not exceeding 0.30%, and subsequently casting the molten metal to desired shape in the said shell molds.

2. In the art of casting steel having a nickel content of not less than about 8%, to desired shape in shell molds having adequate resin content for strength during casting, the method of avoiding gas and pitting, which includes adding an oxidizing agent to the cold charge before melting the charge, adding ferro silicon before melting the charge, and thereby adjusting the silicon content to be in a range of from 0.40% to 0.65 and limiting the manganese content to an amount not exceeding 0.30%, and subsequently casting the molten metal to desired shape in the said shell molds.

3. In the art of casting steel having a nickel content of not less than about 8%, to desired shape in shell molds having adequate resin content for strength during casting, the method of avoiding gas and pitting, which includes adding nickel oxide as an oxidizing agent to a cold charge in an amount of approximately 1.33% of the charge before melting the charge, adjusting the silicon content in the initial cold charge not to exceed 0.85% and in the final analysis of the steel to be in a range of from 0.40% to 0.65 and limiting the manganese content to an amount not exceeding 0.30%, and subsequently casting the molten metal to desired shape in the said shell molds.

4. The method defined in claim 1, which includes the step of melting the specified cold charge in an induction furnace.

5. In the art of casting steel having a nickel content of not less than about 8%, and a carbon content not more than about 0.08%, to desired shape in shell molds having adequate resin content for strength during casting, the method of avoiding gas and pitting, which includes adding nickel oxide as an oxidizing agent to a cold charge before melting the charge, adjusting the silicon content to be in a range of from 0.40 to 0.65%, controlling the manganese content to be less than the silicon content, and subsequently casting the molten metal to desired shape in the said shell molds.

6. In the art of casting steel having a nickel content of not less than about 8%, and a carbon content not more than about 0.08%, to desired shape in shell molds having adequate resin content for strength during casting, the method of avoiding gas and pitting, which includes adding nickel oxide as an oxidizing agent to a cold charge in an amount approximately 1.33% by weight of the charge before melting the charge, adjusting the silicon content to be in a range of from 0.40 to 0.65%, controlling the manganese content to be less than the silicon content, and subsequently casting the molten metal to desired shape in the said shell molds.

7. The method defined in claim 5, which includes the step of melting the specified cold charge in an induction furnace.

8. The method defined in claim 2, which includes the step of melting the specified cold charge in an induction furnace.

9. The method defined in claim 1, which also includes the preliminary step of making the shell molds out of sand with some flour and a resin content of 7 to 8%.

10. The method defined in claim 5, which also includes the preliminary step of making the shell molds out of sand with some flour and a resin content of 7 to 8%.

11. The method defined in claim 2, which also includes the preliminary step of making the shell molds out of sand with some flour and a resin content of 7 to 8%.

12. In the art of casting steel having a nickel content of not less than about 8%, and a carbon content not more than about 0.08%, to desired shape in shell molds having adequate resin content for strength during casting, the method of avoiding gas and pitting, which includes adding an oxidizing agent to a cold charge before melting the charge, adjusting the silicon content to be in a range of from 0.40 to 0.65%, controlling the manganese content to be less than the silicon content, and subsequently casting the molten metal to desired shape in the said shell molds.

References Cited in the file of this patent UNITED STATES PATENTS 1,059,709 Byrnes Apr. 22, 1913 1,982,421 Emmons Nov. 27, 1934 2,193,088 Charlton Mar. 12, 1940 2,374,396 Urban Apr. 24, 1945 2,499,306 Forster Feb. 28, 1950 2,542,266 Talbot et a1 Feb. 20, 1951 FOREIGN PATENTS 658,110 Great Britain Oct. 3, 1951 OTHER REFERENCES

Claims

1. IN THE ART OF CASTING STEEL HAVING A NICKEL CONTENT OF NOT LESS THAN ABOUT 8%, TO DESIRED SHAPE IN SHELL MOLDS HAVING ADEQUATE RESIN CONTENT FOR STRENGTH DURING CASTING, THE METHOD OF AVOIDING GAS AND PITTING, WHICH INCLUDES ADDING FERRO SOLUTION TO A COLD CHARGE BEFORE MELTING THE CHARGE, AND THEREBY ADJUSTING THE SILICON CONTENT TO BE IN A RANGE OF FROM 0.40% TO 0.65%, WHILE LIMITING THE MANGANESE CONTENT TO AN AMOUNT NOT EXCEEDING 0.30%, AND SUBSEQUENTLY CASTING THE MOLTEN METAL TO DESIRED SHAPE IN THE SAID SHELL MOLDS.