Serial 1.5D models
6.3 Test mock-up
6.3.5 Assessment of measurements including accuracy and stability
Initially, an assessment of the accuracy of the measurements will be made based on one of the measurement series. For this, the steady-state measurement series 1 will be used. In this series there is only cooling in the upper deck. Before starting the measurement series, the upper deck has been running at the same temperature for several days to ensure stable conditions.
Figure 6.31 shows the heat load in the room, which is calculated based on the control signal as shown in Eq. (6.6). The control is based on a PID-control algorithm, which means that initially it starts by quickly rising to full signal – equivalent to 88W/m² - before dropping down to the desired signal, which in this case is around 58W/m². The control can also be seen in Figure 6.32, where the mean air temperature of the measurement positions 0.7m and 1.1m above the floor level are used to control the room air temperature. In this case it can be seen, that the temperature and heat loads both becomes fairly stable.
0 4 8 12 16
50 55 60 65 70 75 80 85 90
Time [hours]
Heat load [W/m²]
Heat load in room from radiators
Figure 6.31 Heat load to room during measurement period. Data shown for series 1 in Table 6.6.
0 4 8 12 16 22.5
23 23.5 24 24.5 25 25.5
Time [hours]
Temperature [°C]
Controlled air temperature
Mean temperature of Tair @ 0.7m and 1.1m
Figure 6.32 Controlled temperature in room found as the mean value of the measured air temperatures at a height of 0.7m and 1.1m. Data shown for series 1 in Table 6.6.
The heat removed through the pipe to the flow unit is shown in Figure 6.33. In this series, the supply temperature is 15°C, the temperature difference from supply to return is approximately 1.7K and the flow is 0.2kg/s.
0 4 8 12 16
50 52 54 56 58 60 62 64 66 68 70
Time [hours]
Heat flow [W/m²]
Heat removed through pipes in upper deck
Figure 6.33 Heat removed through the pipe in the upper deck. Data shown for series 1 in Table 6.6.
Since the guard wall has not been installed for these measurements, the temperature in the guard cannot be controlled, and therefore the unwanted heat loss through the inner walls will be larger than when this wall is installed. Figure 6.34 shows the guard temperature in the six measurement positions outside each of the six room surfaces. As it can be seen here, the guard temperature is generally around 1K lower than inside the room, the only exception being the temperature below the deck, where the air is colder than outside the walls and ceiling, where there is a larger air circulation.
0 4 8 12 16 20
20.5 21 21.5 22 22.5 23 23.5 24 24.5 25
Temperature [°C]
Time [hours]
Guard temperature
Tguard,ceiling Tguard,floor Tguard,wall1 Tguard,wall2 Tguard,wall3 Tguard,wall4
Figure 6.34 Temperatures around the room. The temperature above and below the decks are shown in bold. Data shown for series 1 in Table 6.6.
The temperature difference across the 95mm of insulation in the four walls is shown in Figure 6.35. This difference is between 0.5K and 1K, which also coincides with the fact that the guard (laboratory) temperature is around 1K lower than the room temperature. In this case this corresponds to an unwanted heat loss of around 21W for the 69m² of wall assuming that the average temperature difference in this case is 0.8K. This corresponds to around 1W/m² of heat load in the room.
0 4 8 12 16
−1.5
−1
−0.5 0 0.5 1 1.5
Time [hours]
∆Twall [K]
Temperature difference across insulation in walls
Wall 1 Wall 2 Wall 3 Wall 4
Figure 6.35 Temperature difference across insulation in walls. A 1K difference corresponds to approximately 0.38W/m². Data shown for series 1 in Table 6.6.
The room surface temperature on the walls, floor and ceiling is shown in Figure 6.36 for a situation where the wall temperatures are measured in the middle of the surface. As it can be seen from, the floor and ceiling surface temperatures are very different from the wall surfaces.
The wall surface temperatures are, on the other hand very similar, differing by around 0.25K-0.5K.
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 20
21 22 23 24 25
Time [hours]
Temperature [°C]
Surface temperatures
Floor Wall 1 Wall 2 Wall 3 Wall 4 Ceiling
Figure 6.36 Internal surface temperatures in room with one thermocouple placed centrally on each wall.
Data shown for series 2 in Table 6.6.
Based on the results for the wall surface temperatures, it was decided to move the
measurement position to the same wall surface and place them as shown in Figure 6.23. This is shown in Figure 6.37 for measurement series 5. Again, the temperatures can be seen to be almost identical, differing by around 0.5K. The lowest surface temperature is 0.5K above the wall and the highest is 2.5m above the floor, while the cooling of the wall from the ceiling surface is apparent for the upper part of the wall.
0 6 12 18 24
21 21.5 22 22.5 23 23.5 24
Time [hours]
Temperature [°C]
Surface temperature in different height of the wall
0.5m above floor 1.5m above floor 2.5m above floor 3.5m above floor
Figure 6.37 Internal surface temperatures placed on the same wall in different height of the wall. Data shown for series 5 in Table 6.6.
The temperature difference between the side of the deck and guard has been measured as shown in the middle drawing in Figure 6.24. This is shown in Figure 6.38. This temperature difference is approximately 4K, with a temperature difference between pipe and guard of approximately 8K. As a comparison to this, the Heat2 calculation shown in Figure 6.28 has a temperature difference across the insulation layer besides the deck is approximately 4.5K in a situation where the temperature difference between pipe and guard is 9K. This seems to confirm the findings in the assessment of the accuracy, which predicted an unwanted heat loss of around 10W for the sides of the deck.
0 4 8 12 16 0
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Temperature difference between side of deck and guard
Time [hours]
∆T deck−guard [K]
Figure 6.38 Temperature difference between side of deck and guard. Data shown for series 1 in Table 6.6.
Figure 6.39 shows the surface temperature of the deck ends in two positions; in the level of the pipes and in the middle of the deck. Here a direct comparison between the measurements and calculation made in Figure 6.27 is more difficult, since the Heat2 model assumes that the pipe temperature is the same throughout the pipe level. However, the temperatures seem to confirm the simulation results with temperatures that are more influenced by the deck temperature than the guard temperature.
0 4 8 12 16
15 15.5 16 16.5 17 17.5 18 18.5 19 19.5 20
Time [hours]
Temperature [°C]
Surface temperature at the end of the deck [K]
Pipe level Middle of deck
Figure 6.39 Surface temperatures at the end of the deck. Data shown for series 1 in Table 6.6.
Finally the conditions in the two decks are compared. Generally, the flow in the lower deck is only half that of the upper deck, which is due to less pump capacity in the flow unit. This of course influences the conditions. However, conversely to this is the fact that the floor surface in the laboratory building is heavily influenced by the lower “ceiling” surface of the lower deck. This means that less heat is absorbed by the lower deck and therefore a smaller
temperature difference than for the upper deck. In most cases, the heat flow from the deck to the fluid in the lower deck is approximately half that for the upper deck.
This is also the case for the temperature difference across the floor covering as shown in Figure 6.40.
0 5 10 15 20
−2
−1.8
−1.6
−1.4
−1.2
−1
−0.8
−0.6
−0.4
−0.2 0
Temperature difference across floor covering for both decks
Time [hours]
∆Tfloor covering [K]
Lower deck
Upper deck
Figure 6.40 Temperature difference across floor covering for the two decks. In both cases, the difference is measured in three positions according to Figure 6.22. Data shown for series 5 in Table 6.6.
This temperature will also lead to different heat flows, which are higher in the lower deck than in the upper. Since the guard wall has not been finished, the upper deck sees a much higher radiant temperature, because the deck is placed directly under the ceiling radiant heating panels in the laboratory which have a surface temperature of approximately 45°C.
This is estimated to give a radiant temperature, which is 8K higher than the air temperature above the deck, and therefore an environmental temperature which is around 4K higher than the measured air temperature. Another, but opposing reason is that the lower deck is equipped with a 4mm parquet floor in the room, which of course leads to an extra but small thermal resistance. This problem is expected to be solved once the guard box is finished.
Concluding on the data presented in this section, it has been shown that the test mock-up can be used to measure the thermal conditions and energy flows in the room equipped with thermo active components, and that the measurement conditions have been shown to be satisfactory in spite of the fact that the test mock-up has not been finished at the time of the measurements.
6.3.6 Results from measurements