8. Commissioning (multi-cavity analysis) - Validation procedure for injection molds
The eight step in validating a injection mold with the overall process shown in injection mold validation flow chart is Commissioning (multi-cavity analysis). The steps before:
1. Mold certification
2. Dry cycle mold
3. Process stability test
4. Gage repeatability & reproducibility (R&R) test
5. Mold viscosity test
6. Balance of fill analysis
7. Gate Freeze Test
Purpose:
The purpose of commissioning is to ensure that all cavities in the mold deliver the same quality, i.e., there is no difference between cavities in the mold on the critical dimensions. The time required to perform this analysis is a function of the number of cavities and number of critical dimensions. The time on the injection molding machine is minimal compared to the time required to measure the parts. However, by performing this analysis it will significantly reduce the number of parts required to test and measure for future experiments on the multi-cavity tool. A solid understanding of creating and interpreting statistical control charts is necessary to perform the multi-cavity analysis.
All the steps during the procedure that involve intimate contact with the injection molding machine are to be done by a qualified injection molding machine operator.
Procedure:
1. Set melt temperature at resin manufacturer's recommended mid-range.
2. Set mold temperature at resin manufacturer's recommended mid-range.
3. Set fill rate and transfer method (position, time or pressure) based on result of the 5. Mold viscosity test.
4. Set hold time based on result of the 7. Gate Freeze Test
5. Set cooling time long enough so that parts eject without being distorted.
6. Allow the mold to run at least 1 hour, which should be long enough for the system to reach thermal equilibrium and for the process to stabilize. Longer stabilization time may be necessary for higher cavitation tooling like 128 cavity stack molds.
7. Ensure the measurement technique is stable and measurement sigma is known. The measurement technique must be accurate and precise enough to capture small amounts of variation in part dimensions and attributes.
8. Take one all-cavity shot every 15 minutes for an hour (n=5) with each cavity identified. (Base your choice of time on capturing most of the variation that is present.)
Caution: For destructive test methods, additional all-cavity shots will be required - to sample every 15 minutes. For every destructive test method, collect one additional shot.
9. Group all the parts from a particular cavity together to get samples of 5 observations each. The number of samples corresponds to the number of cavities.
10. Condition parts for 24 hours at 23C (73.4F) and a relative humidity of 50% with a standard tolerance of 2.0C (3.6F) and 5% relative humidity.
11. Measure and record all critical dimensions, attributes and variables for all parts.
12. Construct an and R Chart where the Range Chart will capture the variation within a cavity over an hour and the Chart will capture the variation between cavities. Calculate the Cp, CpK and target Z. This information will be used as a flag if the mold is considered off specification. For constructing an and R Chart use a spreadsheet Commissioning (multi-cavity analysis) which should be modified for your specific requiment.
13. Interpret the Range Chart.
• If the Range Chart is in control (pass C, P, and R), all the cavities deliver the same variability which can be calculated by , where d2=2.326 for n=5 (See Factors for constructing variables control chart on page 45), and is the average of the range measured within each cavity.
• If the Range Chart fails, and #7 was done, there is a particular cavity that is different from the others relative to the variability it produces. When the upper control limit (UCL) and the lower control limit (LCL) are not statistically meaningful, in comparison to the upper specification limit (USL) and lower specification limit (LSL), continue with the validation process.
14. Interpret the Chart (if the Range chart is in control ONLY)
• If the Chart is in control (pass C, P, and R), all the cavity averages are statistically not different. All the variability in the mold occurs within one cavity and all cavities are statistically not different.
• If the Chart fails, the special cause cavity must be investigated and corrected, i.e., is it because of mold, material, process, or people? When the upper control limit (UCL) and the lower control limit (LCL) are not statistically meaningful, in comparison to the upper specification limit (USL) and lower specification limit (LSL), continue with the validation process.
15. Upon completion of the molding run for the multi-cavity analysis, explore the molds processing window to determine key parameters (factors) to be varied during a design of experiments and the amount of variation (levels) of each.
A multi-cavity analysis curve shown in figure.
1. The range chart is in control (pass C, P, R): The variation (range) within any cavity is not significantly different than the mold average-range of 0.0027" (.0068 cm).
2. The average chart fails C (cavities 2, 9, 13): Cavity 2 is consistently bigger than the rest of the mold while cavities 9 and 13 are consistently smaller than the rest of the mold. All other cavities are not significantly different than the average inside diameter of 2.1728" (5.52 cm).
The cause for these out-of-control cavities should be investigated and identified. Root causes may be measurement errors, different steel dimensions, imbalanced runner system (small/large gates, different probe tip temperatures, etc.), different cooling conditions, etc. Use your investigative skills to identify the special cause. When the upper control limit (UCL) and the lower control limit (LCL) are not statistically meaningful, in comparison to the upper specification limit (USL) and lower specification limit (LSL), continue with the validation process.
A multi-cavity analysis curve shown in figure.
1. The range chart is in control (pass C, P, R): The variation (range) in weight within any cavity is not significantly different than the mold average-range of 0.121 grams.
2. The average chart is also in control (pass C, P, R): All weights are not significantly different than the average preform weight of 29.024 grams, i.e., any one cavity is representative of the quality (weight) of the 16 cavity tool.
Since there is no significant weight difference between cavities, on-going monitoring of the weight (as an overall process stability indicator) can be achieved by sampling ONLY one cavity randomly from the 16 cavities in the mold. Remember to spread the observations within samples to capture the "true" process variation.
The further steps are required in validating a injection mold according to injection mold validation flow chart is dry cycle mold:
9. Design of experiments
10. Qualification (process capability study)
11. Mold metal Adjustments - centering process
12. Verification (30-day run)
1. Mold certification
2. Dry cycle mold
3. Process stability test
4. Gage repeatability & reproducibility (R&R) test
5. Mold viscosity test
6. Balance of fill analysis
7. Gate Freeze Test
Purpose:
The purpose of commissioning is to ensure that all cavities in the mold deliver the same quality, i.e., there is no difference between cavities in the mold on the critical dimensions. The time required to perform this analysis is a function of the number of cavities and number of critical dimensions. The time on the injection molding machine is minimal compared to the time required to measure the parts. However, by performing this analysis it will significantly reduce the number of parts required to test and measure for future experiments on the multi-cavity tool. A solid understanding of creating and interpreting statistical control charts is necessary to perform the multi-cavity analysis.
Procedure:
1. Set melt temperature at resin manufacturer's recommended mid-range.
2. Set mold temperature at resin manufacturer's recommended mid-range.
3. Set fill rate and transfer method (position, time or pressure) based on result of the 5. Mold viscosity test.
4. Set hold time based on result of the 7. Gate Freeze Test
5. Set cooling time long enough so that parts eject without being distorted.
6. Allow the mold to run at least 1 hour, which should be long enough for the system to reach thermal equilibrium and for the process to stabilize. Longer stabilization time may be necessary for higher cavitation tooling like 128 cavity stack molds.
7. Ensure the measurement technique is stable and measurement sigma is known. The measurement technique must be accurate and precise enough to capture small amounts of variation in part dimensions and attributes.
8. Take one all-cavity shot every 15 minutes for an hour (n=5) with each cavity identified. (Base your choice of time on capturing most of the variation that is present.)
Caution: For destructive test methods, additional all-cavity shots will be required - to sample every 15 minutes. For every destructive test method, collect one additional shot.
9. Group all the parts from a particular cavity together to get samples of 5 observations each. The number of samples corresponds to the number of cavities.
10. Condition parts for 24 hours at 23C (73.4F) and a relative humidity of 50% with a standard tolerance of 2.0C (3.6F) and 5% relative humidity.
11. Measure and record all critical dimensions, attributes and variables for all parts.
12. Construct an and R Chart where the Range Chart will capture the variation within a cavity over an hour and the Chart will capture the variation between cavities. Calculate the Cp, CpK and target Z. This information will be used as a flag if the mold is considered off specification. For constructing an and R Chart use a spreadsheet Commissioning (multi-cavity analysis) which should be modified for your specific requiment.
13. Interpret the Range Chart.
• If the Range Chart is in control (pass C, P, and R), all the cavities deliver the same variability which can be calculated by , where d2=2.326 for n=5 (See Factors for constructing variables control chart on page 45), and is the average of the range measured within each cavity.
• If the Range Chart fails, and #7 was done, there is a particular cavity that is different from the others relative to the variability it produces. When the upper control limit (UCL) and the lower control limit (LCL) are not statistically meaningful, in comparison to the upper specification limit (USL) and lower specification limit (LSL), continue with the validation process.
14. Interpret the Chart (if the Range chart is in control ONLY)
• If the Chart is in control (pass C, P, and R), all the cavity averages are statistically not different. All the variability in the mold occurs within one cavity and all cavities are statistically not different.
• If the Chart fails, the special cause cavity must be investigated and corrected, i.e., is it because of mold, material, process, or people? When the upper control limit (UCL) and the lower control limit (LCL) are not statistically meaningful, in comparison to the upper specification limit (USL) and lower specification limit (LSL), continue with the validation process.
15. Upon completion of the molding run for the multi-cavity analysis, explore the molds processing window to determine key parameters (factors) to be varied during a design of experiments and the amount of variation (levels) of each.
A multi-cavity analysis curve shown in figure.
1. The range chart is in control (pass C, P, R): The variation (range) within any cavity is not significantly different than the mold average-range of 0.0027" (.0068 cm).
2. The average chart fails C (cavities 2, 9, 13): Cavity 2 is consistently bigger than the rest of the mold while cavities 9 and 13 are consistently smaller than the rest of the mold. All other cavities are not significantly different than the average inside diameter of 2.1728" (5.52 cm).
The cause for these out-of-control cavities should be investigated and identified. Root causes may be measurement errors, different steel dimensions, imbalanced runner system (small/large gates, different probe tip temperatures, etc.), different cooling conditions, etc. Use your investigative skills to identify the special cause. When the upper control limit (UCL) and the lower control limit (LCL) are not statistically meaningful, in comparison to the upper specification limit (USL) and lower specification limit (LSL), continue with the validation process.
A multi-cavity analysis curve shown in figure.
1. The range chart is in control (pass C, P, R): The variation (range) in weight within any cavity is not significantly different than the mold average-range of 0.121 grams.
2. The average chart is also in control (pass C, P, R): All weights are not significantly different than the average preform weight of 29.024 grams, i.e., any one cavity is representative of the quality (weight) of the 16 cavity tool.
Since there is no significant weight difference between cavities, on-going monitoring of the weight (as an overall process stability indicator) can be achieved by sampling ONLY one cavity randomly from the 16 cavities in the mold. Remember to spread the observations within samples to capture the "true" process variation.
The further steps are required in validating a injection mold according to injection mold validation flow chart is dry cycle mold:
9. Design of experiments
10. Qualification (process capability study)
11. Mold metal Adjustments - centering process
12. Verification (30-day run)
Comments
I have one question related to this I made R chart for my product that have 32 cavity the thing is more than cavity is out of UCL(16.86179) and LCL (16.84782) put in relation with USL(17.40) and LSL(16.38) there is huge difference, so i think there is no significant difference between the cavities that will require me to do CAP but how can i reflect this in my report in order to continue validation
please help