Moldmaking of the future: Predictive maintenance
Every product will wear out at some stage: whether it’s your beloved car, your coffee machine or your power drill. If your vehicle goes on strike, your morning wake-up alarm fails to work, or your drill just won’t go into the wall, it’s not only annoying but will also cost you money to have it repaired or to buy a new one.
And a defect on a machine used in industry will generally have meaningful consequences. In the case of plastics injection molding, for example, even the smallest signs of wear in the mold lead to defective end products. A number of parts will already have been produced before the defect becomes apparent, and production will generally come to a standstill until it is fixed. This can cause significant financial loss, reduced customer satisfaction and damage to a company’s reputation.
To prevent this, predictive maintenance is increasingly being adopted in the context of industry 4.0.
Data captured by sensors helps to proactively maintain machines and molds before they fail, thus reducing costly outages. A further positive effect: expensive individual molds last longer, because even the slightest signs of wear can be detected early, while they are still comparatively cheap to repair.
Pressure, sound and optical monitoring
Pressure sensors, for example, make it possible to ongoingly monitor clamping force. They register excessively high or excessive low pressure, as well as uneven mold clamping during the injection molding phase. In this way, impending damage to the mold can be avoided before it becomes expensive. The underlying principle is comparable with the tire pressure sensors on current cars – if the pressure deviates from the optimal values, they issue an alarm before the tire suffers serious damage.
In addition to pressure monitoring, mold manufacturers also recommend the use of temperature, sound and optical sensors. If the measured temperature is outside the specified range, for example, this can be an indication of limescale deposits or corrosion in the cooling channels. Sound-measuring devices detect structure-borne sound during the injection phase. Changes in the sound profile can be an early indicator for cracks developing in the mold, or for the beginning of deformation.
Optical sensors can be used to scan the inside of the mold. They can confirm any damage that has occurred and can detect whether the mold is still the correct shape. This is important because the mold steel is subject to gradual wear – especially when plastics reinforced with abrasive fibers are processed. Continuous monitoring helps to avert even greater damage to the mold at this point.
Optimal sensor positioning
Already at the planning stage for the injection mold, it makes sense to configure the installation of sensors so that they do not collide with the heating/cooling system and ejector system etc. In the course of the CAD design process, it is possible to find the best position alongside the other assemblies of an injection mold. Depending on the particular parameters of the production process that are to be monitored, it is possible to determine the quantity and type of sensors and install these as appropriate. The designer then produces a mold with a long service life right from the start, saving the user considerable repair and maintenance costs.
Planning production outages, lengthening servicing intervals
As with a vehicle, injection molds must also be serviced from time to time. While the machines are being maintained, however, production more or less comes to a standstill. With predictive maintenance, this interruption to production can be planned more flexibly. This is because the sensors notify the person in charge of the machine when the system is not running optimally, requiring maintenance to be performed earlier, in order to prevent damage. If all the parameters are in order, however, the service interval can be extended without any problems.