GMP-Talk with Harald Flechl at the LOUNGES 2018 in Karlsruhe
Editor Doris Borchert discussed these (and other) questions with Harald Flechl, senior engineer and author of GMP Publishing, at the LOUNGES 2018 in Karlsruhe.
If we compare the ventilation system with the human circulatory system, we can say that like the circulatory system, the ventilation system doesn’t actually enable life itself. Reducing the cleanroom conditions to the ventilation system alone is an error which is made very often. As with the human body, many factors play a role in keeping the cleanroom alive.
Our bodies are surrounded by clothes - which is equivalent to the shell of the cleanroom.
We can live in different climate zones and are exposed to a variety of environmental influences to which our bodies have to adapt.
The cleanroom is also exposed to a variety of climatic environmental conditions. To ensure that the correct functioning of the cleanroom is maintained under the different environmental conditions, “nutrition” in the form of the provision of energy and the continuous adjustment to the changing conditions are necessary.
We require electrical energy to keep everything operating, heating and cooling media for the correct temperature as well as water or water vapour to maintain a controlled level of air humidity. Like the blood in the human circulatory system, the air is responsible for the transport of oxygen. It must be cleaned – or filtered – in order to fulfil the requirements. A superordinate form of control – the brains of the system, generally called the measurement technology and control engineering, or the building automation system – ensures the correct processes.
Like high blood pressure in the human body, it is also necessary to avoid “high pressure” as a source of everyday stress in a ventilation system. By this I mean inefficiently high volumes of air which are not actually required, but that “decades of practice” have nonetheless made seem normal. To ensure that ventilation systems have a long lifespan, targeted steps should be taken to avoid such high levels of pressure.
Small air ductwork cross-sections increase the pressure losses and/or resistances in the system – com-parable with the arterial calcification in the human body. To be able to transport, at low levels of resistance, air volume flows which are high in comparison with a so-called “comfort ventilation” such as that in an office building, air ductworks with large dimensions are required. The associated requirement for space requires bigger technical installations, however, which also means bigger investments - but people also want to keep their investment costs as low as possible, which at the practical level unfortunately means that the systems are planned in cramped spaces. In this respect, the law of physics is also forgotten: lower levels of resistance reduce the amount of energy required to transport the medium! Or expressed otherwise: small air ductwork cross-sections save on investment costs but result in higher operating costs because of the greater need for energy.
Unfortunately, people often fail to consider the entire life cycle from the investment to the end of the operating life, and only think about the investment costs - although over a 10-year period, the operating costs can account for around 3/4 of the overall costs. This means that every Euro of operating costs saved is equivalent to 3 Euros in investment costs, or expressed otherwise, instead of a 10% price discount upon purchase, it is better to save 3% of the operating costs.
Unfortunately, the bad habit of defining the air cleanliness grades with “air exchange rates” is now commonplace. Starting from the uncertainty and frequently the unawareness of what is going to take place in the room in the future, a cleanliness grade is automatically assigned with a minimum air exchange rate which is normally far too high. It goes so far that this air exchange rate is then used in numerous SOPs for the monitoring of the cleanroom – and this unnecessarily high air change rate is established as being compulsory.
In practice, there are company-specific differences which can be very divergent. For ISO 8 (Grade C), I know of normal air change rates of 20-30-x.
Air exchange rates are often used for determining air cleanliness grades, which is wrong: cleanliness grades are defined on the basis of the maximum permitted number of airborne particles in the ISO standard. In the area of pharmaceutical production, internal company regulations are determined for every grade of cleanroom or type of room. This is for historical reasons and can be attributed to the first “dust-free” rooms of the 1950s, when people used to work with high air flow rates. The FDA Aseptic Guide also contributed to the determination of air exchange rates for the individual cleanliness grades. In this document, an air exchange which is “normally 20-x” is recommended for ISO Class 8 cleanrooms (equivalent to GMP Grade C). Unfortunately, European inspectors often comply with this recommendation.
These recommendations have survived until the present day, although CFD (computated fluid dynamic) simulations are available as a tool for optimising air flows, room airflow patterns and particle dispersion.
A survey that was carried out at more than 100 Grade B to D cleanroom ventilation systems in 2017 actually showed that the authoritative value for the determination of the air volume is based on the level of the thermal load to be dissipated, and the particle values reach a maximum 5% of the per-mitted limit values.
Most cleanrooms are interior rooms which means they have to be kept cool at almost every time of year. At a room temperature of +22°C, to prevent condensation from developing on surfaces, the temperature of the incoming air should not exceed a value of +15°C.
The air flow rate determined results in an air exchange rate of approx. 10-x, or a little more or less depending on the load.
Along with the energy savings, the reduction of the air flow rate to the level which is actually re-quired also results in further advantages, such as lower levels of turbulence and back-flows in the clean areas, a lower burden on the facilities, higher filter service life, and in the case of particle filters, the separation efficiency is also better!
It is actually the case that compared with the human circulatory system, ventilation systems have a big advantage - with a calculated risk and appropriately derived measures, the air flow rates can be adapted to the actual requirements: that means sometimes, ventilation can be reduced or even deactivated, for example, outside production times, in the “at rest” status or during “provisioning” times. As people are afraid of the questions inspectors ask on how these measures have been assessed and derived, however, this is often avoided, and not - as often claimed - because it is prohibited by the regulations.
Research has shown that if the ventilation system is deactivated and access to it is controlled, almost no changes occur to the status of the cleanroom. This is due to the fact that the individual in the production represents the greatest level of possible impurity, and therefore the highest risk of contamination. Therefore, if no people are present outside of the production times, and none of the production systems are running, there are no emissions of particles in the clean room, and no risk of a contamination. If the system is activated again, with the use of defined qualification steps which are documented in the scope of the validation, it is necessary to prove how long it takes until the production status defined in the user requirement specification, i.e. the “manufacturing” or “in operation” operating status is achieved again. In this respect, it should be possible to convince the internal quality assurance, and subsequently, also the inspectors of the effectiveness. If an explanation of this approach is necessary during the inspection, it must always be explained by a technical expert, and not by a member of the quality assurance team.
By operating systems on an optimised basis according to demand, which is, incidentally, also de-scribed in VDI 2083 sheet 9.2, energy savings of up to 70% are possible. Embarking on such a rethink is made difficult by the fact that energy savings of this kind compare poorly in terms of the costs of discarding one or several batches. When the quality assurance department – which the management regards usually more highly than technical operations management for in-house technology - says that it isn't able to assess the risk, the management is highly likely to decide against the optimisation.
A regular health check on the ventilation system depends on how critically a possible failure is cate-gorised. In this respect, it is necessary to differentiate between legally required checks such as fire safety, hygiene, machine safety etc. and work which is technically necessary for operational reasons. In my experience, it is often the case that ventilation systems are serviced and maintained during planned breaks in the production process (shut downs, campaign manufacturing). This generally results in very high maintenance costs, because shut downs of this kind take place at least once a year. The operating costs of the systems are therefore needlessly increased.
In my experience, there is a considerable potential for making savings in the area of servicing and maintenance. As always in the area of GMP, the risk-based approach is considered to be key, and it isn't possible to make generalisations. I always ask myself the following question, however: what will happen to my product if that or that happens or a component fails.
Let us take the example of an air particle filter: a good air particle filter is indispensable for the cleanliness of the air and the hygiene standard of the system, and is also beneficial for the exchange of energy on the heat exchangers. If operated correctly, however, soiling does not mean the filtering performance of an air particle filter will be ineffective. It means the energy required to overcome the increased level of differential pressure on the filter is increased. By monitoring the life cycle costs, it is possible to determine the right time frame – and by that I don't mean the point in time – for changing the filter. In the case of working according to specific intervals, I have seen particle air filters which were pretty much new but were replaced because of the requirements of the internal maintenance SOP. The differential filter pressure between the clean air and exhaust air side was also recorded - as required by the SOP! This differential pressure is usually far below the recommended terminal resistance of the filters. A longer service life, i.e. the possible use of these filter elements for a longer time, is not considered. It isn’t the maintenance staff who are responsible for this, however. It is usually the internal regulations, or the SOPs. If an SOP requires the replacement and such a replacement is not carried out, this is considered to be an infringement.
In other words: the internal regulations (the SOPs) may be responsible for an inefficiency, that - with the fear of a possible complaint due to the “paranoid” avoidance of risks - stipulate excessive requirements.
The risk-based approach is also applicable to the maintenance, in general terms: what happens to my product if a system, part of a system or a component no longer works or fails?
The ventilation has an impact on the air cleanliness – but what is the impact of a low level of air cleanliness on a closed process in which the product has no contact with its environment? I can then align my maintenance strategy according to this approach and decide whether the system should simply operate until it fails and it is then serviced, whether I simply need to replace an important part, or if it is necessary to install a condition monitoring system to be able to forecast a failure with confidence. In the future, I think approaches of this kind will replace regular technical maintenance. However, it goes without saying that the legally-specified inspections, such as of the fire damper functioning, sprinkler systems or other safety systems and hygiene etc. must continue to be completed and documented in accordance with the national or local regulations.
To maintain the circulation of the air, energy and media are required – comparable with the intake of food by the human body. In this context, “healthy” means using the necessary media in a sensible way, for example, using waste heat and energy flows from other processes. This means, for instance, avoiding emitting waste heat from cooling systems into the environment but rather using it for the necessary subsequent heating in the dehumidification operations, returning the waste heat from air compressors to the heating system, using cooling circuits for heat pumps, operating the ventilators in the ventilation systems at the required operational time with the use of speed control, etc.