The bolt's capping angle and torque monitoring window are generally determined by empirical methods. For example, the torque method's capping angle can be monitored according to 1-720°. The final tightening window of the bolt is generally not determined based on experience, but t

bolts are generally determined by empirical methods. For example, the capping angle of the torque method can be monitored according to 1-720°.

The final tightening window of the bolt is generally not determined based on experience, but the monitoring window is determined based on statistical methods.

Today Screw will chat with you: How to determine the torque and angle monitoring windows using statistical methods?

1

Introduction to the basic principles of SPC statistical process control

Anyone who understands product quality control is familiar with SPC. SPC is a very useful quality management tool that is widely used in manufacturing enterprises.

SPC is a process control tool that relies on mathematical statistics methods. It analyzes and evaluates the production process, promptly discovers signs of systemic factors based on feedback information, and takes measures to eliminate their effects so that the process can be maintained in a controlled state that is only affected by random factors to achieve the purpose of quality control. The purpose of

SPC is to detect whether there are special reasons in the manufacturing process through control charts. If there are special reasons, then they must be analyzed first, and based on the analysis results, it can be decided whether to carry out prevention and improvement.

In order to simplify the discussion, we will discuss based on the original judgment criterion formulated by Shewhart , whether it exceeds the 3σ control limit, and will not consider other criteria for the time being.

When applying SPC, if the data is within the 3σ control limit, it is generally believed that process changes are usually caused by ordinary reasons, and there are no special reasons, and no improvement is needed; if the data falls outside the 3σ control limit, it indicates that there may be an impact in the process. For the special reasons of product quality characteristics , it is necessary to confirm whether there are special reasons, and if so, measures should be taken to improve them.

A simple summary of the role of SPC is to ensure that the characteristic parameters of parts and the parameters required by standards are accurate, discrete and small, just like shooting at the bullseye. This kind of product quality is the most reliable.

For tightening, high reliability and small tightening parameter dispersion are what we are pursuing.

Therefore, it is very appropriate to use the SPC statistical method to control the tightening quality of bolts.



2

SPC application in tightening torque and angle monitoring

SPC The core is a control chart as shown in the figure below. Taking torque as an example, it controls the monitoring range within the range of μ±3σ (where μ is the mean value and σ is standard deviation).



So why is it ±3σ? Not ±2σ, nor ±4σ, nor ±6σ?

is shown in the following table: When



• is controlled by 3σ, the pass rate can be as high as 99.73%; when
• is controlled by 2σ, the pass rate is 95.45%; when
• is controlled by 4σ, the pass rate is 99.99%.

Tighten the rework rhythm based on the actual unqualified products in the workshop:

• When the pass rate is controlled at 95.45% (according to ±2σ), the unqualified rate is slightly higher and the rework workload is large.
• When the pass rate is controlled at 99.99% (according to ±4σ), the pass rate is high and the dispersion is large. The actual bolts are tightened within the monitoring range, but the bolts may not be effectively tightened and the probability of failure is high.
• When the pass rate is controlled at 99.73% (according to ±3σ), the rework proportion is small, the workshop has time to rework, and the torque dispersion deviation is also within the acceptable range.
3

How to determine and adjust the monitoring window?

Through the above analysis, the bolt tightening angle and torque parameters can be monitored using the μ±3σ method.

So, how to implement it?



Screw Jun suggested:

• When the parts are still in the project stage, the torque and angle windows can be opened, or a rough window can be set based on previous experience. It is recommended that the window range should not be set too tight to prevent the failure rate from being too high. Because many uncertainties in the project stage are still being optimized, torque engineers also need to evaluate the tightening curve to determine whether there are any abnormal factors in the process. For individual points with large deviations, if they are caused by various abnormal factors, they will be reviewed later. This data can be excluded from statistics.
• When the project enters sop, it is necessary to conduct μ±3σ statistics on the early tightening data to determine the upper and lower limits of torque and angle monitoring.
• During the mass production process, it is generally not easy to adjust the monitoring window. If there is an alarm of less than 2% or 5% (the specific proportion depends on the location of the connection point), rework can be used to tighten it again. When an alarm rate higher than 5% occurs, the torque engineer needs to analyze the bolt tightening from the perspective of man-machine-material method. Generally, the monitoring window cannot be easily adjusted before the cause is found. If the torque exceeds the window upper limit alarm, it is found through analysis that the friction coefficient of the bolts of this batch is higher than that of the previous batch. If both batches are within the qualified range, the torque window can be adjusted upward appropriately; if the friction coefficient and high If the upper limit of the standard is exceeded, the surface treatment of the bolt needs to be optimized to reduce the friction coefficient, and the torque monitoring window cannot be moved.
• Regarding the torque and angle monitoring windows after mass production, in order to further improve the tightening quality, it is recommended to conduct μ+3σ statistics every six months, one year or after 1000-10000 tightenings to determine the appropriate window for the latest batch of parts.
• In addition, Mr. Screw also suggested that statistical methods can be used to determine the monitoring window (such as 50% of the final tightening torque) during the bolt tightening process to ensure that the tightening process is also monitored.
4

Conclusion

(1) The final tightening window of the bolt is generally obtained by calculating μ±3σ based on the statistical principle of SPC. This can ensure that 99.73% can be within the qualified range without special reasons, while tightening reliability high.

(2). The tightening window determined after the project stage. When the torque and angle exceed the window alarm, the human-machine material method environmental test analysis needs to be performed first. Only when there are no special reasons, the tightening window can be optimized and adjusted.

Today’s topic will be shared here; for more details, please follow the WeChat public account: GAF screw gentleman (GAF-luosijun)