You have an old machine that now has to “learn” something new: to run a different material than what it was designed for. Typically, when a machine was initially designed, it had a “speed limit” and a particular material class that it was intended to process. For example, a web coating process was built to apply a coating to a film made of a particular polymer at a specific thickness and width at a given speed.
But markets change, product portfolios change, and raw material supply chains change, especially in times like these where supply disruptions and new market needs have appeared. So how do you assess how far your process and equipment may be pushed before reinvestment is needed? How to tell if that new material is going to be processable in your current machine?
Here are a few suggestions for a process to understand the current state of your process and how well it will adapt to the new reality facing it.
Measure Your Material Properties
Navigating to a new place is easier if you know where you are before you go, in case you have to go back. Measuring the web properties of your current material that are relevant to converting it will create a basis for comparison with the proposed material changes. These properties are dependent on the specific converting process but here is a table of some potentially relevant properties:
| Web material measurements | Process Effects |
| Thickness (caliper) or Basis weight (mass/area) | Wrinkles, wound roll quality, coating interaction |
| Width | Air entrainment, wrinkling tendency |
| Modulus (in longitudinal, transverse and z-directions) | Machine direction registration, tension control, wound roll quality, lateral control, web wrinkling, curl-free lamination |
| Poisson’s Ratios | Web width under tension, wound roll quality |
| Thermal conductivity, heat capacity, thermal expansion/contraction, temperature | Wound roll quality, interaction with coatings, effect on web tension and registration |
| Permeability/porosity | Air entrainment interaction, fluid coating interaction |
| Surface energy | Coating adhesion effect |
| Triboelectric series position | Electrostatic build-up, impact on coating processes and operator safety |
| Surface Roughness | Traction on rollers and lateral control, slippage in wound rolls |
| Optical properties | Interactions with sensors for web guides and other sensors |
| Viscoelastic properties | Time on wound roll quality |
| “Sidedness”: differences in properties from one side to the other | The web may behave different when interacting with different surfaces or coatings depending on which side is presented to the process |
| Material property variation: camber, thickness variation, modulus variation | Wrinkling, lateral control |
Having data on both the existing material and the new material of interest can provide early insights into how different the new material is. Small changes in properties, say within 5-10% may not be an issue, but if the new material is of a very different class, say a new polymer with much lower modulus compared with the previous material, the traditional machine settings could be more likely to fail.
It is time-consuming and expensive to measure a large number of material properties, however. The process engineer has to have an understanding of the requirements of the process, for example the product quality measures for the output of the process, and knowledge about each conversion step from a physics and chemistry standpoint to map out what is critical to be measured from a material property standpoint.
Measure Your Process Parameters
The existing web converting machine may have come with a set of knobs that might or might not now reflect the state of the art in assessing the health of the process or what truly matters to the web. Creating a list if it doesn’t exist of what is currently measured and comparing to what is needed from a physics/chemistry standpoint for each conversion step is helpful. It may be that new sensors will be needed to better understand the state of the process, either physical sensors, or “soft” sensors that use existing sensor data to infer the actual property of interest.
| Process/Equipment Parameters | Process Effects |
| Web tensions | Web breakouts and permanent deformation |
| Driven roller, nip and conveyor velocities | Tension control, lateral and longitudinal control |
| Idler bearing drag & inertia | Tension control during transient conditions |
| Roller alignment | Wrinkling, lateral control |
| Web/roller coefficient of friction | Web traction and lateral and longitudinal control |
| Process control loop parameters | Transient effects during line ramps and splices, web tension control and lateral control |
| Web path span lengths and number of spans | Wrinkling probability, lateral control, registration |
| Coating fluid pressures, flowrates, and temperatures | Interactions with web |
Relate the Process Parameters to the Material Properties and make a prediction
Knowledge about the relationship between the material properties and the process parameters can provide a prediction or assessment of how a new material will convert. This is best done with models or simulations, but models can span from heuristics to algorithms, meaning some relationships might be known at a very loose level, such as Factor X is known to affect Output Y, but how much relative to other factors isn’t known well. Or a predictive computer simulation of the physics of the conversion step might be available, but at any rate, some assessment must be done. In the absence of a model, one approach is to assemble the people with the most knowledge of the process steps and have them map the relationship between inputs and outputs.
Assessment Outcomes
Ideally the assessment will highlight key conversion steps that are either the greatest risk, or where the knowledge is lowest, sparking ideas where focused experiments can be run, or models can be built. It may also provide insights where new measurements are needed, either for the material or the process. And finally, it should highlight if it is obvious the current machine isn’t going to meet the new requirements and where investment is needed to enable the new requirements.