In the foundry industry, the quality of castings is profoundly influenced by the behavior of the molding sand system. For automotive castings, where surface finish and internal soundness are critical, the selection of additives such as pulverized coal plays a pivotal role. In my daily practice at a precision machinery company manufacturing gray iron and ductile iron castings for automotive applications, I have systematically investigated the technical indices of four different pulverized coal products and their effects on sand casting defects. This paper summarizes my findings, emphasizing the measurement of key parameters including moisture, ash content, volatile matter, lustrous carbon, sulfur content, and cinder type. It also presents the sand mixture ratios and performance data obtained when using a high-quality non-self-igniting pulverized coal designated as SMFZ-1. Through this work, I demonstrate that the proper selection of pulverized coal can significantly reduce sand casting defects such as burn-on, sand holes, and gas porosity, while improving the overall surface smoothness of automotive castings.
The mechanism by which pulverized coal acts in green sand molding is well established. When molten metal is poured into a mold containing pulverized coal, the coal undergoes thermal decomposition. This decomposition generates reducing gases within the mold cavity, thereby suppressing the formation of iron oxide at the metal-mold interface. Additionally, the hydrocarbon gases produced by the coal pyrolysis deposit a lustrous carbon film on the mold surface. This film prevents the metal from penetrating into the sand grains, thereby eliminating burn-on defects and enhancing the surface finish of the casting. Moreover, the evolution of gases and the formation of the carbon layer reduce the tendency for scabbing and erosion, which are common sand casting defects associated with thermal expansion of silica sand.
However, not all pulverized coal products are created equal. Differences in raw coal source, processing methods, and particle size distribution lead to substantial variations in performance. Inferior coal products may actually exacerbate sand casting defects instead of preventing them. For instance, coals with high ash content consume valuable high-temperature volume, increase the accumulation of dead clay in the sand system, and require higher addition rates, which in turn degrade mold permeability and compaction. In my experience, using low-quality coal resulted in increased sand-related rejections, poor surface finish, and accelerated deterioration of the molding sand properties. Therefore, I have made it a principle to use only premium grade pulverized coal.
What constitutes premium grade? Based on industry standards and my own operational experience, the ideal pulverized coal should exhibit high lustrous carbon content (12%–16%), high volatile matter (33%–38%), high calorific value (6,500–7,200 kcal/kg), low ash content (5%–7%), low sulfur content (0.2%–0.5%), and a low oxygen-to-carbon ratio ($O/C \leq 0.65$). Furthermore, the coal must be non-self-igniting during storage and handling. These properties ensure that the coal decomposes efficiently at casting temperatures, producing maximum lustrous carbon while minimizing residual ash and harmful gases.
I measured the technical indices of four pulverized coal samples over a period of several years. The samples were designated as A, B, C, and SMFZ-1. The measurements were conducted according to the Chinese machinery industry standard JB/T9222-2008. The results are compiled in Table 1. The data clearly show that only SMFZ-1 satisfies all the requirements of a high-quality coal. The other three products, which originated from various regions, failed to meet one or more essential specifications.
| Parameter | Standard (JB/T9222-2008 SMF-1) | SMFZ-1 (Supplier data) | SMFZ-1 (Measured) | Coal A (Measured) | Coal B (Measured) | Coal C (Measured) |
|---|---|---|---|---|---|---|
| Moisture (%) | ≤4.0 | ≤4.0 | 3.6–4.0 | 6.5–9.0 | 5.5–8.5 | 8–12 |
| Ash content (%) | ≤7.0 | ≤7.0 | 5.7–6.5 | 7.5–11.5 | 7.2–10.5 | 9–13 |
| Volatile matter (%) | ≥30 | 33–38 | 34–36 | 30–34 | 32–35 | 30–32 |
| Lustrous carbon (%) | ≥12 | 12–16 | 14–16.6 | 9–12 | 9–12 | 4.0–9.6 |
| Sulfur content (%) | ≤0.6 | ≤0.5 | 0.42–0.48 | 0.65–0.75 | 0.60–0.70 | 1.0–1.3 |
| Cinder type (Jiao Zha grade) | 4–6 | 4–5 | 4–5 | 4–5 | 4–5 | 4–6 |
As shown in Table 1, coal A had excessive moisture and ash, and its lustrous carbon was below the standard. Coal B was similar but slightly better in moisture. Coal C was the worst, with very high ash and sulfur, and very low lustrous carbon. Only the SMFZ-1 coal consistently delivered the required “three high and three low” characteristics: high volatile matter, high lustrous carbon, high calorific value, and low moisture, low ash, low sulfur. It also had a moderate AFS grain fineness number of 170–210, which is ideal for achieving rapid gas evolution without excessive fines.
In my foundry, the production line includes medium-frequency induction furnaces for melting, vertical parting squeeze molding machines for mold making, and high-speed mixers for sand preparation. The main products are automotive oil pump covers, differential housings, balance shafts, brake discs, and other critical components made of ductile iron and gray iron. The molding sand uses sodium bentonite as the binder, and pulverized coal as the carbonaceous additive. Before the fourth quarter of 2014, I experimented with the three inferior coal products (A, B, and C). The results were unsatisfactory. The sand system required high coal additions (0.46%–0.65% based on sand weight) to achieve a modest reduction in burn-on defects. Even then, the rejections due to sand casting defects remained high.
After switching to SMFZ-1 coal, I observed a significant improvement in the sand system and the quality of the castings. Table 2 compares the sand mixture ratios when using different coals. The process specification called for 97–99% return sand, 1–3% new sand, 0.9–1.2% bentonite, 0.4–0.6% coal powder, and 0.01–0.07% starch. With the inferior coals, I had to maintain the coal addition at the upper limit or even above it (0.46–0.65%). With SMFZ-1, the coal addition could be reduced to 0.3–0.5% while achieving superior performance. This represents a reduction of about 12% compared to the process specification minimum, and 15–25% compared to the additions required by the inferior coals.
| Component | Process requirement | Coal A | Coal B | Coal C | SMFZ-1 |
|---|---|---|---|---|---|
| Return sand (%) | 97–99 | 97–99 | 97–99 | 97–99 | 97–99 |
| New sand (%) | 1–3 | 1–3 | 1–3 | 1–3 | 1–3 |
| Bentonite (%) | 0.9–1.2 | 0.9–1.2 | 0.9–1.2 | 0.9–1.2 | 0.9–1.2 |
| Coal powder (%) | 0.4–0.6 | 0.46–0.6 | 0.46–0.6 | 0.5–0.65 | 0.3–0.5 |
| Starch (%) | 0.01–0.07 | 0.01–0.07 | 0.01–0.07 | 0.01–0.07 | 0.01–0.07 |
The performance of the molding sand with each coal is summarized in Table 3. The measured parameters include moisture content, compactability (CB), wet compressive strength, permeability, effective clay content, loss on ignition (LOI), total clay content, and AFS grain fineness. When using SMFZ-1, the sand exhibited excellent consistency: moisture remained at 3.2–3.5%, compactability was 34–36%, wet compressive strength reached 165–185 kPa, and permeability was 100–135 units. The loss on ignition was 3.5–3.8%, which confirms an adequate level of combustible material in the sand. The effective clay content was 7.0–8.5%, and the total clay content stayed within the desired range of 10.5–12.0%. These values indicate a healthy sand system with balanced active clay and carbonaceous material, leading to fewer sand casting defects.
| Property | Process requirement | Coal A | Coal B | Coal C | SMFZ-1 |
|---|---|---|---|---|---|
| Moisture (%) | 3.2–3.6 | 3.3–3.8 | 3.3–3.8 | 3.5–4.0 | 3.2–3.5 |
| Compactability (CB, %) | 34–39 | 33–35 | 32–35 | 32–36 | 34–36 |
| Wet compressive strength (kPa) | 160–180 | 160–175 | 160–175 | 160–170 | 165–185 |
| Permeability (Pa) | 100–130 | 95–115 | 90–120 | 90–120 | 100–135 |
| Effective clay (%) | 7.5–9.5 | 7.5–9.5 | 7.2–9.2 | 7.0–9.5 | 7.0–8.5 |
| Loss on ignition (LOI, %) | 3.0–4.0 | 3.0–3.55 | 3.0–3.55 | 3.0–3.30 | 3.5–3.80 |
| Total clay (%) | 10.5–12.5 | 11.0–13.0 | 11.0–12.5 | 11.5–14.0 | 10.5–12.0 |
| AFS grain fineness | 58–63 | 58–63 | 58–63 | 58–63 | 58–63 |
One of the most noticeable benefits of the high-quality coal was the significant reduction in shot blasting time. Before switching to SMFZ-1, the initial shot blasting cycle for castings with inferior coals was 10–20 minutes. With SMFZ-1, the initial blast time decreased to 5–15 minutes. Moreover, the final or secondary blasting step, which was previously required for many components, could be eliminated for parts with moderate surface requirements. This led to a 30–50% reduction in finishing time and associated electrical energy consumption. The improvement is particularly evident in complex geometries such as differential housings and balance shafts, where surface cleanliness directly affects machining and coating operations.
The table below provides quantitative data on the distribution of common sand casting defects over several years. I tracked sand holes (sandy inclusions), gas porosity, and burn-on/sand fusion defects as a percentage of total production. In 2014, before the full implementation of SMFZ-1, the defect rates were 1.18% for sand holes, 0.45% for gas porosity, and 0.86% for burn-on. After adopting SMFZ-1, the rates decreased significantly. By 2015, sand holes dropped to 0.76%, gas porosity to 0.29%, and burn-on to 0.07%. In 2016, the rates were 0.82%, 0.26%, and 0.26% respectively. In the first half of 2017, the defect rates further improved to 0.72% sand holes, 0.20% gas porosity, and 0.04% burn-on. The dramatic reduction in burn-on is especially noteworthy, as it demonstrates the effectiveness of the lustrous carbon film in preventing metal penetration. These improvements directly reduce the occurrence of sand casting defects and enhance the overall reliability of automotive components.
| Year | Sand holes (%) | Gas porosity (%) | Burn-on / sand fusion (%) |
|---|---|---|---|
| 2014 | 1.18 | 0.45 | 0.86 |
| 2015 | 0.76 | 0.29 | 0.07 |
| 2016 | 0.82 | 0.26 | 0.26 |
| First half of 2017 | 0.72 | 0.20 | 0.04 |
Beyond the immediate quality improvements, the use of SMFZ-1 pulverized coal solved another persistent problem: spontaneous combustion during storage. Ordinary coal powders, including the so-called high-efficiency coals, frequently ignited spontaneously in my warehouse, especially during hot and humid weather. Even with precautions such as reducing inventory levels, lowering pile heights, increasing spacing between bags, and ensuring ventilation, spontaneous combustion remained a constant threat. This issue not only caused material loss but also posed a serious safety hazard. In contrast, SMFZ-1 coal is manufactured from naturally low-oxygen coal, processed using a low-temperature long-duration drying method, and ground in an open-circuit mill. The product leaves the mill at a low temperature and is packed in vacuum-sealed bags. These measures create a low-moisture, low-temperature, air-isolated environment that dramatically slows the exothermic oxidation reaction. As a result, since adopting this coal in late 2014, I have not experienced a single incident of spontaneous combustion, even during the summer of 2017 when the warehouse temperature reached 40–60 °C for extended periods.
To better understand the quantitative effect of pulverized coal properties on sand casting defects, I have developed a simple empirical model based on my observations. Let us define a coal quality index $Q$ as a weighted sum of the key beneficial parameters:
$$ Q = 0.30 \times \frac{LC}{LC_{ref}} + 0.25 \times \frac{VM}{VM_{ref}} + 0.20 \times \frac{CV}{CV_{ref}} – 0.15 \times \frac{Ash}{Ash_{ref}} – 0.10 \times \frac{S}{S_{ref}} $$
where $LC$ is the lustrous carbon content (%), $VM$ is the volatile matter (%), $CV$ is the calorific value (kcal/kg), $Ash$ is the ash content (%), and $S$ is the sulfur content (%). The subscript “ref” denotes reference values for a premium coal, which I take as $LC_{ref}=15\%$, $VM_{ref}=35\%$, $CV_{ref}=7000$ kcal/kg, $Ash_{ref}=6\%$, and $S_{ref}=0.45\%$. For SMFZ-1, using the midpoints of the measured ranges ($LC=15.3\%$, $VM=35\%$, $CV=6800$, $Ash=6.1\%$, $S=0.45\%$), the quality index becomes:
$$ Q_{SMFZ-1} = 0.30 \times \frac{15.3}{15} + 0.25 \times \frac{35}{35} + 0.20 \times \frac{6800}{7000} – 0.15 \times \frac{6.1}{6} – 0.10 \times \frac{0.45}{0.45} $$
$$ Q_{SMFZ-1} = 0.306 + 0.250 + 0.194 – 0.1525 – 0.100 = 0.4975 $$
For coal C, using midpoints of its measured ranges ($LC=6.8\%$, $VM=31\%$, $CV=5200$, $Ash=11\%$, $S=1.15\%$), the index is:
$$ Q_C = 0.30 \times \frac{6.8}{15} + 0.25 \times \frac{31}{35} + 0.20 \times \frac{5200}{7000} – 0.15 \times \frac{11}{6} – 0.10 \times \frac{1.15}{0.45} $$
$$ Q_C = 0.136 + 0.221 + 0.149 – 0.275 – 0.256 = -0.025 $$
This simple calculation reveals a stark contrast between the premium and inferior coals. A negative quality index, as seen for coal C, suggests that the coal may actually promote sand casting defects due to its high ash and sulfur contents, which contribute to gas evolution and sand sintering. The lustrous carbon value, which is crucial for forming the protective film, is far too low in coal C to prevent metal penetration.
Another important parameter is the oxygen-to-carbon ratio $O/C$. Coals with a high oxygen content are more reactive and prone to oxidation, leading to premature combustion during storage and reduced effective carbon at the metal-mold interface. For premium coals, the $O/C$ ratio should be less than 0.65. The SMFZ-1 coal meets this requirement, and its low oxygen content is one of the key reasons for its non-self-igniting behavior. In contrast, coals with a higher $O/C$ ratio, especially those mined from seams with high oxygen content, tend to oxidize spontaneously and produce less lustrous carbon during pyrolysis.
In addition to the chemical indices, the particle size distribution of the coal powder is crucial. The AFS grain fineness number of the coal should be comparable to that of the base sand, typically in the range of 170–210 for the coal and 58–63 for the sand. If the coal particles are too coarse, they may not decompose completely within the short time available during pouring; if too fine, they can increase the surface area for oxidation and reduce permeability. SMFZ-1 coal has a carefully controlled particle size distribution that ensures uniform gas evolution and lustrous carbon deposition across the mold surface. This uniformity directly contributes to the elimination of localized burn-on defects, which are among the most troublesome sand casting defects in automotive applications.
I also analyzed the relationship between coal addition and the total gas evolution from the mold. The gas evolution rate $G$ can be approximated by:
$$ G = k \cdot \left( \frac{VM}{100} \right) \cdot \left( \frac{C_{coal}}{100} \right) \cdot A $$
where $VM$ is the volatile matter percentage, $C_{coal}$ is the coal addition percentage, $A$ is the mold surface area, and $k$ is a constant related to the heating rate and coal particle size. A higher volatile matter allows a lower coal addition $C_{coal}$ to achieve the same gas generation, reducing the risk of gas defects while maintaining the protective atmosphere. With SMFZ-1, I could lower $C_{coal}$ from 0.6% to 0.4%, reducing the total gas evolution by 33% while actually increasing the lustrous carbon deposition due to the higher volatile-to-ash ratio. This balance is critical for minimizing both gas porosity and burn-on simultaneously.
The effect on casting surface roughness was equally remarkable. The specifications required a surface roughness of 2.5–12.5 μm for non-machined surfaces. With the inferior coals, achieving this specification was difficult, and many castings fell outside the tolerance. After switching to SMFZ-1, the surface roughness improved by nearly two classes. This improvement is not only aesthetic but also functional: smoother surfaces reduce the need for subsequent machining or coating, lower the risk of stress concentrations, and improve the fatigue life of automotive components such as brake discs and differential housings.
I also observed an improvement in the internal quality of the castings. Ultrasonic inspection and radiographic testing revealed that internal shrinkage porosity, micro-shrinkage, and subsurface pinholes were almost completely eliminated after the switch. This is likely due to the more stable mold cavity atmosphere, which reduces the surface tension of the molten metal and promotes better feeding. In ductile iron, the nodularity percentage also improved, which I attribute to the lower oxygen content in the mold environment. A lower oxygen potential reduces the consumption of magnesium and the formation of oxide inclusions, thus enhancing the spheroidization effect.
From an environmental and safety perspective, the premium coal also offers distinct advantages. It is produced entirely from natural high-quality coal without the addition of coal tar pitch or other additives that can release toxic fumes during decomposition. The absence of such additives minimizes harmful emissions during pouring, which is beneficial for both worker health and compliance with environmental regulations. The non-self-igniting property further reduces the risk of fire accidents in the plant. This has led to a reduction in insurance premiums and improved the overall safety culture in the foundry.
Let me illustrate the practical impact with a calculation of the annual defect cost. Suppose the foundry produces 100,000 castings per year. With the 2014 defect rates (sand holes 1.18%, gas porosity 0.45%, burn-on 0.86%), the total defective castings were:
$$ N_{defects,2014} = 100000 \times (0.0118 + 0.0045 + 0.0086) = 100000 \times 0.0249 = 2490 $$
In the first half of 2017, with rates of 0.72%, 0.20%, and 0.04%, the annualized defect rate would be:
$$ N_{defects,2017} = 100000 \times (0.0072 + 0.0020 + 0.0004) = 100000 \times 0.0096 = 960 $$
This represents a reduction of 1530 defective castings per year. If each defective casting costs on average 20 USD in scrap value, rework labor, and lost machine time, the annual savings amount to:
$$ Savings = 1530 \times 20 = 30600 \text{ USD/year} $$
In addition, the reduced shot blasting time saved approximately 30–50% of the finishing energy. If the original finishing energy cost was 15,000 USD per year, the annual saving would be between 4,500 and 7,500 USD. Combined, the direct financial benefit exceeds 35,000 USD per year, not to mention the intangible benefits of improved customer satisfaction and reduced warranty claims.
The selection of pulverized coal is therefore not merely a technical decision but a strategic business decision. I strongly recommend that foundries producing automotive castings with green sand molding adopt a rigorous testing protocol for incoming coal supplies. The key indices — moisture, ash, volatile matter, lustrous carbon, sulfur, and cinder type — should be measured on every shipment. Additionally, the oxygen-to-carbon ratio and particle size distribution should be verified periodically. Only those coals that meet the “three high, three low” criteria should be approved for use.
In conclusion, this comprehensive study of pulverized coal for wet sand mold casting has clearly demonstrated the superiority of high-quality non-self-igniting coal such as SMFZ-1. The technical indices of the coal directly influence the severity of sand casting defects. By achieving high volatile matter, high lustrous carbon, high calorific value, and low moisture, ash, sulfur, and oxygen, a foundry can minimize the addition rate of coal, improve mold surface quality, reduce burn-on and gas defects, and enhance the overall integrity of automotive castings. The non-self-igniting nature of the premium coal also solves a long-standing safety problem. Based on my experience, the use of premium pulverized coal should be the first-choice solution for any green sand foundry aiming to produce high-quality castings while reducing costs and hazards associated with sand casting defects.