Troubleshooting Processes Using Cavity Pressure Sensors

Troubleshooting can take on many meanings for injection molders. All too often they find themselves running to a machine, turning knobs and pushing buttons, hoping with fingers crossed that the next parts out of the mold will be better than the last. Obviously, this practice typically results in wasted production time and escalated material costs and usually doesn’t address the root cause of the real problem. It is very easy to fall into a trap of making multiple adjustments that are not truly related to the problem to be solved, stacking more problems on top of the original problem. Short-term improvements to part quality will often times only mask the actual issue.

Most molders would agree that before any effective troubleshooting can take place, one must first determine what changed in the process. The ability to determine what changed to cause the problem at hand relies on the quality of data available. If the troubleshooter has access to accurate data related to the original or target process, shooting or better yet, killing the trouble can be simple. The key is having that crucial information available so when shots are fired, the trouble is killed and not just wounded from sporadic shooting. Trouble that is merely shot but not killed, is guaranteed to return.

In past articles of this Cavity Pressure Series, we’ve discussed utilizing data from cavity pressure sensors to create an electronic template or blueprint of the process. This template is a representation of the desired cycle from data captured inside the mold cavity where the parts are being molded. We’ve discussed using templates and cavity pressure data to create machine and location independent molding operations. This strategy allows the mold to be moved to another machine or to a totally different facility and the cavity pressure template matched, producing the same parts regardless of geography.

As we discussed last month, containment strategies using cavity pressure technology allow the molder to immediately contain any suspect parts based on in-cavity conditions. The same events that cause suspect parts to be contained, also illustrate process variations that can significantly streamline Troublekilling practices. When process variation occurs that will affect part quality, the graphical profiles or curves from the cavity pressure sensor(s) will clearly show changes happening inside the mold.

The two types of data that any high quality data acquisition system such as the eDART System™ should provide are: a picture of each molding cycle in real time and data over time to detect and display trends. Both types of cavity pressure information prove priceless to the troubleshooter. The real time cycle data is plotted over time as the part is being molded. With the template visible, the current cycle curves overlay the template. Any variations in the cavity pressure profiles are immediately evident as a variance from the template curves.

Figure 1 shows a Cycle Graph with a Template and the current cycle drawn on top.

Defects such as dimensions, flash, shorts, and part weight are directly correlated to the total volume of plastic delivered to the mold cavity, which is directly correlated to the amount of cavity pressure in the cavity. If the cavity pressure curve reaches a peak that is lower than the template or ideal cycle, there was less plastic delivered. If the peak cavity pressure climbs higher than the template peak, this would indicate more plastic being packed into the cavity. Troublekilling dimensional inconsistencies may require recording data over time and correlating trends in cavity pressure data to specific cycles or to specific parts. All the cavity pressure data is saved to a file that can be replayed for evaluation. Specific production runs or even single cycles can be retrieved by recalling the date and time of the cycles to be analyzed. This ability to go back in time and analyze cavity pressure information is very crucial since some molding defects may not come to light immediately.

Cosmetic concerns such as gloss level and texture level are related to the rate of packing. The slope of the cavity pressure curve during pack (pressure rise over time) illustrates the pack rate. If the cavity pressure curves reach their peak sooner or later that the template, a variation in gloss or texture may result.

If there is a cavity pressure sensor located near the gate area inside the cavity, it can be utilized to determine if the gate is sealed. If the gate is not sealed at the time that the holding pressure is released, plastic will discharge from the cavity back into the runner. The gate end cavity pressure curve will suddenly drop concurrent with the end of holding time. Figure 2 shows two cycles overlaid on each other. One showing gate seal and the other shows discharge.

Thermal variations such as heater band cycles or hot runner heater cycles will many times show up on cavity pressure data. This type of variation may be subtle and require more data to detect. Thermal cycles will typically be repeating cycles within the data. For example, as the hot runner heaters cycle on and off, the viscosity of the material is affected. Hotter plastic will have somewhat less resistance to flow and therefore render a higher peak cavity pressure. So the peak cavity pressure will trend up and back down with repeatable frequency from thermal cycling.

By now the benefits of having cavity pressure data are clear. Data from inside the cavity is the only data that can clearly and without doubt open a window for the molder to view all aspects of the molded part being created. Troublekilling is obviously streamlined by the ability to “see” what is happening inside the cavity. With the lights on in the cavity, any changes to the ideal conditions in the mold will be illuminated.