3. Analysis for the product Crank Case 3000

Collection of the rejection data is essential for defect analysis. This helps us to understand which defect caused the most rejection. As stated earlier, the majority of the defects observed were core shift and sand drop by visual observation. To better understand these results, a bar chart needs to be prepared for the product that has significant rejections due to defects. In this case, from the entire range of products, Crank Case 3000 was used due to its high rejection rate (Table 1).

Furthermore, most of the rejection causes for the product were due to only two defects: sand drop and core shift as shown in the bar chart. Sand drop occurs due to misaligned cope and drag, which in turn also creates core shift. Weak green


#### Table 1.

Defect analysis table for product Crank Case 3000.

Analyzing Tactics and Reduction of Core Shift and Sand Drop for Crank Case… DOI: http://dx.doi.org/10.5772/intechopen.81083

Figure 1. Bar chart of defect analysis for product Crank Case 3000.

compressive strength and incorrect moisture levels also play a huge role in causing these defects (Figure 1).

## 4. Pareto analysis to find critical defects and their causes in product Crank Case 3000

From the preliminary analysis, it was decided that the two defects (core shift and sand drop) were to be studied further with the help of Pareto charts and cause

Figure 2. Pareto chart for product Crank Case 3000 rejection.

effect diagrams to better understand the percentage of rejection and the root cause of the defects.

For this to occur, Pareto analysis was carried out. Table 2 was created in preparation of the Pareto chart. A cumulative percentage of the defects was taken in descending order.

Pareto charts display the 80–20 rule, which means that 80% of the rejections are due to 20% of the causes. Here the causes are confirmed as core shift and sand drop. In Figure 2, it is clear that both core shift and sand drop are above the 80% limit. This shows that according to the 80–20 rule, these two defects are critically affecting the yield and efficiency of the product and foundry.

#### 5. Cause effect diagram to find critical defects and their causes in product Crank Case 3000

Construction of the cause effect diagram was undertaken by considering the significant factors that affect both core shift and sand drop. All these factors are summarized under five basic categories namely Machine, Sand, Mold, Metal and Core. The critical control factors were chosen based on weighted scores obtained from observation of the defect analysis and their causes as shown in Table 2. The level of importance of the factor and the co-relation of the output with input is specified to calculate the weighted score.

After calculating the scores, the factors with maximum scores are highlighted in Table 3.

The score for correlation of the output with input is based on the following scale: 9 = very strong, 7 = strong, 5 = moderate, 3 = poor, 1 = very poor.

Comparing the above causes, one can observe that compactibility, green compressive strength (GCS), moisture and permeability directly affect the defect as they are related to the molding sand. The other causes are non-measurable and need to be solved by brainstorming (Figure 3).

When the GCS is low, the sand cannot be held together, which causes loose sand. Loose sand is a major reason for the sand drop defect. High permeability shows that the sand has many intermediate spaces that allow the gases to escape. Thus high permeability shows that the GCS is comparatively low.


#### Table 2.

Cumulative rejection analysis of product Crank Case 3000.


Analyzing Tactics and Reduction of Core Shift and Sand Drop for Crank Case… DOI: http://dx.doi.org/10.5772/intechopen.81083

#### Table 3.

XY cause effect matrix for identifying control factors.

#### Figure 3.

Cause effect diagram for core shift and sand drop.

Also high compactibility shows low permeability, which can cause gas related defects. High moisture content can decrease the GCS beyond a certain range. Thus to completely prevent core shift and sand drop, a balance in values of these parameters is necessary.

## 6. Why-Why analysis

Any problem always has a cause that points out the core of the problem. The Why-Why analysis is a powerful statistical tool that helps us to find out the root cause of any non-measurable problem. With observation and asking the question "why" five times in a row, the root cause of the problem comes to light. With the help of this tool, the root cause analysis of core shift and sand drop was carried out and the outcome is as shown in Figure 4.

Observing the Why Why analysis, we can confirm that sand composition can be controlled by altering the values of the sand parameters. The change in GCS, moisture, permeability and compactibility is proposed to acquire various data for defects related to these sand parameters.

#### 7. DOE and ANOVA

Core shift and sand drop are primarily molding sand-based defects. Changing the sand properties to optimum properties is one of the main requirements to reduce core shift and sand drop defects (Table 4).

DOE was carried out on compactibility, permeability, moisture content and green compressive strength. By application of DOE, an analysis can be made of how the factors vary with respect to each other.

% Reject ¼ 1; 522; 604–1474 GCS � 39; 782 permeability þ 14; 071 Compactibility


Figure 4. Why-Why analysis for core shift and sand drop.

Analyzing Tactics and Reduction of Core Shift and Sand Drop for Crank Case… DOI: http://dx.doi.org/10.5772/intechopen.81083

The above equation provides an equation that co-relates GCS, permeability, compactibility and moisture with % Rejection. The above equation has an R square value of 0.949, which means 94.9% of the data fits the model. After carrying out DOE, an analysis of variance is required that shows us which factor is the most critical according to the data.

From Table 5, it is seen that the P value of permeability is the least as it is less than 0.05. This shows that moisture plays a critical role in the properties of sand and the defects caused by it.


#### Table 4.

DOE table for sand parameters vs. % rejection.


Table 5. ANOVA table.
