Aim

• To measure the internal temperature of a bottle undergoing pressurisation in order to determine if the heating due to pressurisation can cause a bottle to fail at a lower pressure.

The strength of PET material used in bottles is affected by temperature. The lowest temperature where this effect begins is known as the glass transition temperature. Wikipedia lists this as both 75 degrees C and 69 degrees C (see table) however, the actual value depends on a number of factors such as additives and the microstructure of PET.

The main question is whether under normal circumstances a rocket made out of PET bottles can reach such a high temperature where the rocket can fail at a lower pressure compared to the same bottle tested hydrostatically. Due to the nature of hydrostatic testing the bottle is cooled by the water and hence does not experience the same sort of heating as a rocket on a launch pad.

A number of factors can affect the temperature inside the rocket and hence the walls of the bottle.

Sources of heat include:

• Sun - With the greenhouse effect the sun can raise the temperature inside the bottle above the outside ambient temperature. It is the same effect as you get in a closed car on a hot day. The colour of the bottle or any dark paint or tape on it can have a significant effect on the internal temperature.

• Pressurisation - By far the largest contribution to the increase in temperature inside the rocket is due to pressurisation. As air is compressed the temperature increases.
• Air source - The air supply type also has a contributing factor to the temperature. When filling from a compressor, the air is heated by the compressor, and warm air comes out of the air hose. When filling from a tank, the air comes out cooler due to expansion. The higher the rate of flow the cooler it is. The length of hose also has an effect on heating or cooling the air from the air source. A long black hose can also heat the air when exposed to the sun.
• Bottle stretching - As the bottle is pressurised and the walls stretch, that process in itself may generate some heat in the walls. (Thanks Christian and Trevor for pointing that out)
• Ambient temperature - This is the outside air temperature. This can be significant if you are launching on a hot day.

Experiment Setup

The thermal tests were performed on a spliced pair of 1.25L bottles. The air was let in through the bottom and the thermocouple was fed in through the top lid and suspended about 15cm below the lid.

The thermocouple was connected to the multimeter set on its thermometer setting. We also placed a watch next to it and a pressure gauge that read the line pressure fairly close to the bottle inlet. We set up a video camera to simultaneously record the temperature, time and pressure relationships.

	2.1L spliced-pair under test. With pressure gauge, temperature gauge and a watch.
	Air is supplied from a SCUBA tank through a pressure regulator.

Results

We filled the bottle at different rates to around 100psi. In some tests, we also filled the bottle with ~800ml of water which would typically be found in a rocket this size. The air bubbled through this column of water as would happen on a launch pad.

Thermal tests

The ambient temperature was 12 degrees C on the day of the tests.

*We only managed to get to 30psi with the little compressor before it decided to splutter and seize up.

** Fills faster because of the reduced volume.

Other Tests

• We measured the temperature of the air coming out of the hose before entering the bottle. During a slow fill rate the air temperature was close to ambient. On maximum flow rate the temperature reached 7 degrees C which was 5 degrees below ambient temperature.
• Three months ago we ran some tests on small compressors to measure the air temperature coming out. The ambient temperature at the time was 17 degrees C. The temp started out okay when the compressor was first turned on at around 18 C. After about 2 minutes of unloaded operations (the compressor wasn't filling a rocket) the air temperature climbed to 29 C. At that point I stopped the test. That was a 12 C increase without any load.
• On one occasion after air had cooled in the pressurised bottle, we let the pressure out rapidly and the air temperature dropped to -1 degree C.

Conclusions / Analysis

• From these few tests it is difficult to reach definitive conclusions about whether it is possible to reach 70 degrees in the bottle. Filling to higher pressures of 130 - 140psi will also have an effect on further increasing the temperature. On a hot day filling to higher pressures would certainly be within the ball park figure of exceeding the glass transition temperature, but we won't be certain until further tests.
  
  We have observed an unusual stretch failure last year when we tried launching a rocket on a very hot day where temperature was around 40C and pressurised to 130psi.
• We were quite surprised that the higher flow rate into the bottle did not increase the air temperature, but this could be partially explained by the fact that at the higher flow rates the air was 5 degrees cooler when it entered the bottle. Faster filling from a compressor could be different.

• We were unable to test even faster flow rates mostly due to our air supply setup in that it has a number of quick release connectors with relatively small holes.

• The other interesting result was that filling the rocket through the water column did virtually nothing to cool the air being compressed above the water.

• Other consequences of filling rapidly and heating the air, especially on bottles sitting behind non-return valves is that the air pressure will drop once the air cools inside the rocket. Depending on various factors this could be as much as 10psi. This is particularly significant if you wait for a while before launching a rocket.

• Because filling a rocket rapidly can raise temperature significantly it is advised to fill a rocket slowly to allow the air to cool.