Some Tips for ensuring consistent quality centrifugal castings
from short air-cooled dies
Introduction:
For Ni-Resist groove insert manufacture the ideal system is to have carousel type centrifugal casting machines with dies as long as possible, and equipped with automatic control of rpm, temperature and cycle time. The large volume reputed manufacturers generally use 2-meter dies. Timed water-cooling is adopted to ensure uniform temperature of the die from end to end.
In producing captive castings or moving to low cost suppliers, the centrifugal casting machines generally used have short dies of 300mm length without water-cooling. The control on microstructure becomes more difficult, particularly at the rear end of the pots due to more rapid solidification. The front of the pot stays hot longer, being closer to the hot metal poured in at this end.
In order to secure high quality and consistency of microstructure of the inserts produced from such pots, from one end to the other, some simple principles need to be rigidly followed in the production process.
Melt Preparation:
Input raw materials:
The high percentage of Nickel in this material requires and efficient melting and mixing furnace of adequate capacity. The best suited for this purpose is a Mains frequency Induction furnace of minimum 500-Kg capacity and preferably of 1000-Kg capacity.
The input raw materials need to be of reasonable purity. Carbide forming elements like Mo, Va, Ti and W should be kept to traces. Caution: Many commercial pig irons contain high Mo and Ti. It is desirable that in the final insert these elements should be controlled to within the following limits:
Mo > 0.05%
Ti > 0.04%
W > 0.02%
Va > 0.02%
Steel scrap is a source of tramp elements that can cause unpredictable problems in Ni-resist materials. The steel scrap should be limited to max 10% of charge as far as possible and of a known source and composition.
Chips and borings should also be kept as low as possible, if they cannot be avoided altogether. It is desirable to limit these to a max of 60% and should be dry and clean. Mixing chips uniformly in the melt is difficult. The losses of certain elements from the chips are high and there is a risk of air and hydrogen trapping in the chips, both of which promote bad graphite and coarse/clustered carbides.
Sulfur in the melt is necessary in a narrow range. It should be kept around 0.06% for getting the desired hardness. Higher Sulfur causes problems to the microstructure.
Carbon Equivalent:
The Carbon Equivalent ( CE ) is known to be a critical parameter for cast irons. This factor plays a significant role in the type of microstructure developed in the casting. It is closely inter-related to the rate of solidification, which in turn is influenced by the cross-sections of the castings and temperatures.
The general convention is to compute the CE value according to the Volume/Surface (V/S) ratio of a casting. For 3-dimensional castings, which are of varying cross-sections, this is very difficult. The centrifugal pots, though 3-dimensional, are however of symmetrical section throughout the length and hence V/S can be calculated for these as shown in the following formula:
Volume / Surface Area Ratio V/S
V/S = ( D2 - d2 ) H / 4 ( DH + dH + ( D2 - d2 ) / 2 )
Where, D = Outer Diameter ( OD ) of the Centrifugal Cast Pot
d = Inner Diameter ( I.D. ) of the Centrifugal Cast Pot
H = Height of the Centrifugal Cast Pot
A better factor for the centrifugal cast pot would be the Degree of Saturation, which is given by the formula:
Degree of Saturation Sc :
;
where, C, Si and Ni are the actual values in weight-%
For our applications the desirable value of Sc = 0,80 - 0,95
Installing a CE meter at the furnace will enable a tighter and more immediate control over the Carbon and Silicon content in the metal before casting the pots.
Recommended range of C, Si & Mn after inoculation:
Carbon: 2.70 – 2.80%
Silicon: 2.10 – 2.20%
Manganese: 1.20 – 1.30%
Melting:
It is important to stratify the charge loading into the furnace for optimum mixing of the large quantity of the Nickel and copper in Ni-Resist material. The input material charging sequence desirable is to first change 50% Pig iron + Scrap, and when it just melts add Nickel, Copper, Ferro alloys, carbon, and finally top up with the balance 50% Pig Iron.
Carbon should be maintained at start in bath at 2.75 - 2.80%, Ni at 14.5 -15%, Silicon at 1.8 – 1.9%, Chromium at 1.15 –1.2%, Mn at 1.2 – 1.25%
The role of Silicon Carbide:
For ensuring adequate nucleation sites in the melt, SiC plays a significant role. Therefore SiC should form a part of the original charge make-up calculation. And further SiC additions are needed just before super-heating the melt, once the chemistry is checked and approved; and also after every few ladles are tapped out, so that carbon depletion is compensated. This will help ensure close limits for the carbon content and hence a more consistent CE and microstructure.
The recommended quantity of SiC in the charge make-up is 0.25% of total charge along with the alloying elements added to the charge. This addition is calculated into the charge make-up for carbon and silicon. After the melt reaches a temperature of 1400C, the chemistry is checked and corrections applied if required, and rechecked. Once the chemistry is approved, a further 0.1% SiC is added and immediately the furnace is brought to full setting and the melt is super-heated to the working temperature (generally 1480 –1500 C, depending on drop of temperature expected by the time the pouring into die is completed from each ladle), and tapping out into a properly pre-heated ladle starts.
Addition of SiC must be done also at regular intervals during the pouring cycle for replenishing the carbon losses. In order to assess the exact requirement of quantity and timing of SiC additions for a given melting set-up and operating conditions, it is best to run several closely monitored heats for carbon drop with time. In a continuing production mode, checking the carbon content in each ladle tapped will give sufficient data to determine when the carbon drops off more than 0.05% at which stage a calculated quantity of SiC must be added to the furnace and allowed to ingest a few minutes before tapping out gain.
Ladle Filling and inoculation:
Before any ladle is deployed to the production area it should be thoroughly heated so that the lining is fully dry and as hot as possible with the torch flame used. The final heating at start of shift and after a break is to be done by reladling with the molten metal from the furnace. More than one reladling may be necessary to bring the ladle to the minimum required temperature of 750C before metal is tapped for use. A contact pyrometer should be used to ensure that the temperature is right.
Theoretically, high Ni materials are self-inoculating. However for our sections of centrifugal castings and the rigid controls on microstructure, it is desirable and necessary to make an inoculation in the ladle as close to the time of pouring as possible. The operating range for inoculation with High Grade Ferro Silicon ( 75 % ) is calculated for bringing in 0.20 to 0.25 % Si pick- up. It is advisable not to use the conventional cast iron inoculants like superseed in this material due to interference by Strontium.
Introducing rare earths will help our microstructure. The possibility of using a better inoculant RESEED, with cerium, can be explored.
Also controlled quantity of Nickel-Mag introduced with the inoculant can have very good effect on avoiding chilled graphite.
Dies:
Design:
Controlled water-cooling is ideal for ensuring a uniform temperature of the die from one end to the other. However if bringing in water cooling is difficult try feasibility of air cooling of the die under a suitable jacket with inlet at front and hot air exiting at rear.
Since machinability problem is mostly confined to the rear end inserts it is advisable to study the microstructures of the rear end inserts and determine the minimum cut-off needed to ensure that only good microstructure inserts get into production. For naturally air-cooled short dies of 300mm length generally a cut-off of at least 40-mm is required.
In the long term the length of the pots should be increased by 30 – 35mm by making new rear end plugs. This will enable even a larger cut-off without compromising on the number of inserts cut from each pot.
A more optimum solution lies in re-designing these dies to take a 600 mm long pot whereby productive is increased even with a cut-off of 50mm at the rear end.
Insulation:
Many short dies use dry insulting materials in ID which act as insulation and a separator. There are superior wet insulation materials, which would give a better protection against chilling.
It is also desirable to insulate the front and, if possible, the rear flanges of the dies.
The ID surface of the die should not be below 300C at anytime during useful production. A contact pyrometer should be handy at each die for ensuring this.
At the beginning of each shift and after every break, 1 or more dummy pots should be poured as needed to ensure that the temperature of the die at the ID is above 300C.
As production continues, the temperature of the die will increase progressively and stabilize at 450 – 500 C. This is an optimum temperature for the short, naturally cooled dies.
Records:
It is essential to monitor continuously all parameters and maintain proper and accurate records.
In the development stages it is prudent also to maintain the carbon and silicon records for each ladle and develop the SiC addition program therefrom.
The control of microstructure should be done as frequently as possible/necessary on the last useful insert cut from the rear end of a pot, till the parameters for the casting and cut-off are fully stabilized.
