With forming process automation a must for glass containers to survive, the latest series of articles from Paul Schreuders will dig into this step-by-step and share his vision on effectively reducing the weight of glass, at first and within this article, in general terms. In subsequent contributions, he will explore the different (individual) steps to be taken towards reducing the weight of glass.

Following the Paris Climate Change Conference (COP21) and taking their social responsibility seriously, many if not all food and beverage packing companies are actively working to reduce their carbon footprint. Since packaging is a substantial part of this carbon footprint, supply chain collaboration is a key for success. Knowing the competitive field of aluminium, plastics and bio-based packaging, for glass the keys to survival are to recycle and reduce weight (improving the content-to-glass ratio).

Looking at the characteristics of glass container forming today, the following statements should be concluded:

  • Overall efficiency is too low at 85%-90%.
  • Except for those having hot end infrared camera systems installed, the overall (quality) feedback loop is at best very slow (cold end and hot end are still acting very independently, where the hot end focus is production and the cold end focus is quality).
  • Hot end forming is highly human-dependent (experience-based instead of fact-based).
  • The workforce is ageing and it is becoming more difficult to hire and retain good hot end operators and/or specialists.
  • Health/safety remains a concern.
  • Containers are (designed to be) too heavy to remain competitive as preferred packaging.
Figure 1: Because of the relative glass thickness fluctuations horizontally, bottles are in general 40% too heavy (Source: Prof Dr-Ing H Hessenkemper, Glas- und Emailtechnik, TU Bergakademie Freiberg).
Figure 1: Because of the relative glass thickness fluctuations horizontally, bottles are in general 40% too heavy (Source: Prof Dr-Ing H Hessenkemper, Glas- und Emailtechnik, TU Bergakademie Freiberg).

Containers are designed to be too heavy

Following the completion of research on this topic, because of the relative glass thickness fluctuations horizontally, bottles in general are 40% too heavy (see figure 1). Of course, the actual number depends on the production process (blow-blow versus narrow-neckpress-blow) and may even be the colour of the bottle (green, amber, flint). It should be realised that in addition to horizontal glass thickness fluctuations, there are also vertical glass thickness variations. A rough estimate is that bottles are in total at least 50% too heavy. In general, bottles are designed to be this heavy: Due to glass thickness fluctuations and in order to meet the customer’s specification of 1mm, a design specification will range from 1.7mm up to sometimes more than 2.5mm.

This oversizing is performed on purpose and in order to avoid the risk of producing faulty bottles; in earlier days, when they were unable to make strength calculations, this was the way to construct buildings that would last for centuries. This oversizing compensates for the low level of forming process control and is generally accepted by glass container manufacturers and their customers.

On many occasions, glass container manufacturers say that their customers want heavy bottles. Everybody knows though that food and beverage packing companies want to avoid the risk of breakage on their filling lines, which is absolutely logical but different from simply wanting heavy bottles.

Level of forming process control is too low

The reason behind the low level of forming process control involves the many process disturbances within the glass container forming process and its surroundings. Due to unpredictable changes in cullet quality, viscosity, temperature, glass homogeneity, ambient temperature, deterioration and wear of material and even swabbing and/or stop/starts of sections, the glass forming process itself is unstable and the outcome is often unpredictable. This leads to variations in gob condition (weight, temperature, shape), gob loading (speed, length, time of arrival,position), temperatures (parison, moulds, plunger, neckring) and consequently, to low efficiency and bottles with glass wall thickness variations and critical defects. Specifically, the deterioration and wear of material (neckring, deflector, trough, moulds, plunger) is mostly cavity-related, highly unpredictable and cannot be avoided.

In addition to these many sources of forming process variation, the industry relies mostly on the experience of operators and/or specialists to observe, analyse and control the process. Due to its complexity (many variables), lack of tools, time constraints and last but not least IS machines that have become bigger over years and are up to 48 cavities nowadays, manual forming process control is ‘one hell of a job’ or even impossible! This is without mentioning the harsh environmental conditions (health and safety) in which operators must work. And manual process control can easily be counter-productive: Instead of reducing process variation, process variation increases. This can be seen easily by manual swabbing, where every operator swabs differently, with different results.

Controlling the forming process up to a higher level than is the case today and stepping away from today’s efficiency and weight standards requires a different approach. This approach is needed in order to improve competitiveness with other packaging materials, as well as giving the industry a truly ‘green’ image and ensuring the industry will sustain, despite the ageing workforce.

Process stability is key to optimisation

There is a lot to do and great potential lies ahead. The focus should be on dealing effectively with unpredictable changes in cullet quality, viscosity, temperature, glass homogeneity, ambient temperature, deterioration and wear of material and even swabbing and/or stop/start sections, as well as minimising the effect of these changes on gob condition (weight, temperature, shape), gob loading (speed, length, time of arrival, position), temperatures (parison, moulds, plunger, neckring) and consequently efficiency and the quality of bottles. Key for dealing with these unpredictable changes are timely and consistent information about gob condition, gob loading, temperatures and the quality of bottles. For the past 15 years, (hot end) sensors have become available for monitoring and analysing bottle cavity variations, loading variations, temperature variations and forming variations. The use of these sensors leads to consistent information and thus, the same observations for all operators. Critical defects can be eliminated at the hot end and thus, quality to the customer improves. Any process deviation (including gob forming, gob loading and temperature deviations) are seen in real-time and allow for fast remedial actions. Instead of sharing opinions about ‘what they think they see’, operators and specialists can focus on finding solutions. Furthermore, production improvements with regard to swabbing, job changes, section stop/starts, gob forming and gob loading lead to improved stability, resulting basically in higher efficiency and less defects produced. Experience learns that all different sensors once applied basically have the same positive effect; the process variation is reduced and the output more stable and predictable. This effect is visualised in figure 2. Despite the enormous amount of benefits to be achieved, the (manual) use of these sensors requires strongly motivated management and shop floor operators/specialists, which seem not always to be easy to organise. In addition, cavity-related variations especially (deterioration and wear of material, swabbing), remain difficult to control to the highest level, even for the strongly motivated.

Figure 2: The application of hot end sensors and control loops (blue = ‘on’) leads to more narrow working ranges, more stability and predictability. This results in higher efficiency, less glass wall thickness fluctuationsand less defects produced.
Figure 2: The application of hot end sensors and control loops (blue = ‘on’) leads to more narrow working ranges, more stability and predictability. This results in higher efficiency, less glass wall thickness fluctuationsand less defects produced.

Process stability requires automation

In order to create the best circumstances for the most effective forming process control, automation is required. In this context, the sensors observe, analyse and through feedback to the IS timing, the necessary adjustments are made. As the feedback is computer feedback, adjustments can be made within milliseconds. Even the very best operator/specialist will never be able to do this. Acting upon variations within milliseconds is extremely effective in minimising the negative effect of the unpredictable changes in cullet quality, viscosity, temperature, glass homogeneity, ambient temperature, deterioration and wear of material, swabbing and stop/start of sections. As a result, huge benefits can be achieved.

Within the past five years, (hot end) automated control loops have become available for controlling gob weight, ware spacing, mould temperature, plunger process and vertical glass distribution. It is to be expected that in the near future, more control loops will become available. Experience shows that all different control loops once applied basically have the same positive effect; the process variation is reduced and the output is more stable and predictable. This effect is shown in figure 2.

Vertical glass distribution

Glass distribution is the distribution of glass within the bottle. As indicated previously, there are basically two angles of glass distribution; a horizontal one and a vertical one. Glass distribution tends to drift over time, for example due to fluctuating ambient temperatures (day and night rhythms) and/or changes in the cooling capacity of IS machines due to weather changes.

In addition, unwanted changes in the forehearth, changes in positions of loading and deterioration and wear of materials are common causes of changes in glass distribution. Less variations in glass distribution mean less quality problems related to glass distribution, as there are thick/thin bases, thin spots, thin necks etc. Also, the IS machine will run more smoothly, as the number of outliers reduces due to more stable glass distribution.

A well developed and tested control loop relates to this vertical glass distribution. It basically works as follows: For every single bottle, the actual vertical glass distribution is measured (by means of an infrared camera system at the hot end). Any change in comparison to a defined setpoint (= optimal vertical glass distribution) is compensated for by an automated algorithm. This automated algorithm makes changes in the IS timing for contact and cooling time, as such changing the actual vertical glass distribution back to the setpoint. This automated algorithm works continuously. The results of this control loop are impressive (see figure 3).

The aim of controlling vertical glass distribution is to achieve stability across all cavities and to make glass distribution independent from ambient temperature fluctuations (day/night) and from feeder (glass) temperature fluctuations, whereby human interaction is excluded. The result is hardly any fluctuation in vertical glass distribution, which directly allows for weight decreases. Automatically controlling vertical glass distribution goes beyond the capability of current operators and specialists and brings the forming process as a whole to a higher level of control. As such, this and also other control loops are key towards the future of operating an IS machine.

Figure 3: Example of actual changes in vertical glass distribution (blue, red and green lines), without and with automated control (automated control = green area). The variations left are due mainly to manual swabbing.
Figure 3: Example of actual changes in vertical glass distribution (blue, red and green lines), without and with automated control (automated control = green area). The variations left are due mainly to manual swabbing.

The industry can do 25% better

Serving customers is what everyone aims to do. Knowing that all food and beverage packing companies are actively working on reducing carbon footprint and realising that packaging is a substantial part of this carbon footprint, glass container manufacturers and their suppliers need to act. Stepping away from the traditional way of working and introducing hot end sensors and automated control loops will allow the industry to make step-wise changes in performance. The author believes strongly and experience proves that with the right sensors and control loops in place and rightly applied, industry performance in terms of higher efficiency, less glass wall thickness fluctuations (which allows for reducing weight) and less defects produced can be 25% better or more.

Having said that, the glass container industry should make its own plans for COP21 and take its social responsibility. Applying different sensors and loops individually but especially in combination brings huge savings within reach. As such, it is possible to produce much lighter and stronger containers with (almost) zero defects at higher speed, with minimal human dependency.

In future issues of Glass Worldwide, the author explores the different (individual) steps to be taken towards optimising process stability and reducing the weight of glass.