PETpla.net Insider 10 / 2020

BOTTLE MAKING 38 PETplanet Insider Vol. 21 No. 10/20 www.petpla.net a higher temperature. In order to under- stand how this can indeed be accom- plished, we need to understand how the oven lamps heat up the PET preforms. 6.3 Types of heat transfer There are three ways of conveying heat and all three are present in a reheat stretch blow machine:  conduction  convection  radiation Conduction occurs when two parts touch each other and heat flows from the warmer to the colder part. An example of this is a warm mandrel transferring heat into the preform neck. Conduction is not suitable for use as a means of reheating because PET like most plastics is a poor conductor and it would take too long to reheat preforms this way. Convection is heat transfer with par- ticipation of air. Heating up of air in the ovens of the blow machine takes place and depending on the temperature of this air it can either heat or cool the outside of the preforms. Air heating is difficult to control, highly dependent on environmental conditions and further- more heats the outside of the preform more than the inside; just the opposite of what is required. The output of the oven lamps radi- ates to the preforms in the form of waves. These waves can be readily absorbed by the PET or penetrate with little absorption. When surface absorp- tion takes place, a large amount of heat is transferred directly to the outside sur- face of the preform. From there it would then travel by conduction to the inside. This behaviour is undesirable since the outside wall would overheat with the inside wall staying colder. Therefore a high degree of surface absorption is not suited for optimal reheating. 6.4 Light absorption characteristics of PET Oven lamps can be adjusted in a voltage range of 0–220V. On some machines a percentage setting of 0–99% indicates this. One less under- stood feature of this control is that voltage output affects both temperature and wavelength. We measure wave- length for this part of the electromag- netic spectrum in either micrometers (one millionth of a meter) or nanometers (one billionth of a meter) and 1000nm make up 1 μ m. While these lamps all work in the infrared spectrum, this spectrum ranges from 0.7 to 100 μ m and it depends on the voltage setting, which wavelength the lamp is emitting. It should be noted that while the lamps work mostly in the infrared spectrum, they also emit wavelengths on either side of this range. Shorter waves of 400–700nm are in the visible spectrum (Fig. 6.3). Fig. 6.3 Typical range of emitted wave- lengths of infrared lamps. Diagram cour- tesy of Philips. Fig. 6.4 shows how PET absorbs the output of the infrared lamps depend- ing on the wavelength of the emitted radiation. It was recorded by emitting the different wavelengths shown as the horizontal axis through a strip of PET and measuring the wavelength after it bounced off a reflector. To the left of the horizontal axis are the short waves. Here PET absorbs up to 50% of the emitted radiation. Between about 1,000 and 2,200nm absorption is about 20% with 80% of the heat being lost. At longer waves to the right of the axis absorption increases again. Fig. 6.4 Lamp settings change both heat flow and wavelength. Optimal wave- length is indicated between the two black arrows. High absorption rates would lead to an overheating of the outside skin of the preform, which is detrimental to an optimum heat profile. Therefore the majority of the lamp output should be between 1,000 and 2,200nm. This is the output the lamp emits at a voltage of 220–110V or 100–50%. While this is admittedly a wasteful process, it allows the heat waves to penetrate the preform walls evenly, heating inside and outside walls to roughly the same degree (Figs. 6.5–6.7). Fig. 6.5 Summary graph of empirical temperature data of preform inside and outside wall at 40% lamp output. Inside temperature lags. Graph: Mr. Bonnebat Fig. 6.6 Summary graph of empirical temperature data of preform inside and outside wall at 70% lamp output. High absorption at outside wall leads to great difference. Graph: Mr. Bonnebat Fig. 6.7 Actual temperature data of pre- form inside and outside wall at 100% lamp output in one position. Inside tem- perature increases faster due to ideal radiation wavelength and convective cooling of the outside. Graph: Bonnebat Now that the preforms are reheat- ing evenly, the next step is to cool down the outside wall of the preform. This is accomplished by blowing air into the ovens at a rate high enough to lower the temperature to below blowing tem- perature. This air will absorb heat more from the outside of the preform wall than the inside since it comes more into contact with it. An oven temperature of 85°C (185°F) has been proven to be sufficiently low to facilitate this effect. However, oven temperature readings are highly dependent on where the