Final Design

Description of Final Design

The final design of the W2WGTHPWH consisted of a Nyle Geyser-r Heat Pump Water Heater with a 75.7 liter (20 gal) cold tank and 189.27 liter (50 gal) hot tank connected to the incoming water supply. Each tank had an outgoing water line and was connected to the heat pump itself. The final design utilized the heating element of the 189.27 liter (50 gal) tank to heat up the incoming water to the tank from 20°C to 32°C. The 32°C temperature than triggered the Nyle Geyser-r to turn on and perform the heat exchange from that point on. For purposes of killing possible bacteria that could exist in the incoming water supply the tank had to be raised to 57°C for each test cycle. The heat pump had temperature sensors at the inlet and outlet on both the hot and cold side in order to get the change in temperature that occurred. Temperature sensors were also at installed to measure incoming water going to the tanks and the outgoing water from both tanks. Flow meters were also attached on the outlets going to the hot and cold tank from the heat pump in order to measure flow rates. Lastly, transducers were installed on the hot tank and heat pump electrical cords to measure the current for the total power the system needed to operate. The final design can be seen below in Fig. 1 as well as Fig. 2 showing water flow in the system. Note that in Fig. 2 the water from the cold tank was higher in temperature going to the heat pump and higher in temperature coming out from the heat pump on the hot side.

Figure 1: Completed System with labels indicating components Figure 2: Flow Chart of water flow in the system

Design of Key Components

The key components that needed to be analyzed for this system were the sensors that were installed and the method for gathering the data. The main concern before any sensors could be installed was having a reliable and accurate data gathering technique. For reasons of convenience following a recommendation by Dr. Jan Kleissl, the Onset HOBO® Data Logger was decided upon. The Onset HOBO® Data Logger collected various sensor data from the system including temperature (C), flow (gal), and electrical (A) data from respective temperature sensors, flow meters, and transducers. From there all the sensors needed could be purchased and connected to a single data logger that would put real time measurement data for further analysis. The main area of concern for these components was getting accurate measurement data from the housing of the temperature sensor.

Figure 3: HOBO Data Logger

Functional Requirements of Component

Temperature sensor needs to accurately measure water temperature
System must not leak

Two designs were considered for temperature sensor installation. The first design was to fix the sensor on the outside of the pipe and wrap it in insulation. The second design choice was to fix the sensor into the pipe using a tee pipe and rubber plug.

Comparison of Designs Considered

Justification of the Final Design Choice

The final design choice decided was the rubber plug method because of the assurance of good measurement data. The primary concern of this project was having accurate temperature data to calculate the heat transfer occurring throughout the system. Considering that aspect, the extra work involved in creating a water tight seal was decided over the wrap around method. It was thought that because of the rubber material that the water pressure pushing against the rubber plug would create the seal needed as well as the seal around the temperature sensor and plug.

Summary of the Projects Performance

A MATLAB model to simulate the flow patterns of the system was created to compare to the test results. The MATLAB model made the assumptions that the system was steady state, and each main component (20 gal tank, heat pump, 50 gal tank) were all analyzed as separate control volumes linked together in the end. These assumptions allowed for the creation of an overall COP vs. ∆T graph combining the averages of the different temperatures in a best fit. The validity of the model was verified by its comparison with actual gathered data by performing tests resulting in a relation between COP and ∆T of the Hot Tank. The constant cold tank temperatures utilized were 15°C, 13°C, and 11°C resulting in the plot below (Fig. 4). Another important test conducted was ASHRAE (American Society of Heating, Refrigeration, and Air-Conditioning Engineers) standard test. This test was conducted to discover how efficiently the tank could recover from hourly draws from the hot and cold tank for 6 hours. The ASHRAE test revealed how efficiently the tank recovered lost energy from each successive draw. The results of this test showed that the system works quite efficiently, after the initial increase of temperature from groundwater temperature (usually ~ 20°C) to 57.2°C (135°F), because after each draw the temperature of the hot tank does not fall significantly. The summary of results and recovery time with a simulated MATLAB model is shown in Fig. 5.

Figure 4: Test Results for COP vs. ∆T of the Hot Tank Figure 5: ASHRAE Test Results (3 Tests)

with MATLAB Simulation for Const. Cold Tank Temperatures with MATLAB Simulation