Energy Efficient Housing

Milestone: Feasibility/Design of Grey Water Heat Recovery System

This milestone will encompass the feasibility and design of a grey water heat recovery system for the energy efficient housing complex. A number of different systems were analyzed. Some are currently offered for sale commercially and others were of our own design which could be constructed from commercially available materials. An analysis of energy savings and economic savings was preformed to determine the feasibility of a grey water heat recovery system for the energy efficient housing complex.

Introduction to Grey Water Heat Recovery

Grey water heat recovery systems utilize the energy contained in hot water exiting the house through the drain. Many household activities, such as showering, dishwashing, and clothes washing, allow hot water to exit down the drain. Despite still being hot enough for use, the water is dirty and therefore cannot be reused. A grey water heat recovery system extracts this heat from the gray water and uses it to preheat the cold water entering the hot water heater. figure 3.1 demonstrates how this process works. The incoming clean water is kept separate from the grey water to prevent contamination, but through the use of a heat exchanger the energy stored in the grey water is transferred to the clean water.

Figure 3.1. Schematic Drawing of a Drain-water Heat Recovery System¹.

There are a number of different types of heat exchangers and grey water heat recovery system arrangements. These arrangements depend upon the specific application and have varying rates of efficiency. The next section will describe some possible arrangements for energy efficient housing complex.

Possible System Arrangements

Throughout researching and brainstorming possible grey water heat recovery system designs a number of possible ideas were considered. Each idea had its own advantages and pitfalls. This sections contains a theoretically and analytical analysis of each of this arrangements, to determine which system would be ideal for this applicatiohl;k;ljk;jkl;xchanger would offer the highest efficiency. Similar to figure 3.1 a coiled copper tube would be used to contain the cold water, the key difference between this system and the system shown in figure 3.1 is that the coiled tube would be contained with in the drain pipe as opposed to being wrapped around it. An example of this system is shown in figure 3.2.

Figure 3.2

Figure 3.2. Shell and Tube Heat Exchanger Schematic.

A particular problem with a system such as this is that even when hot water is in use (i.e. someone is taking a shower) the drain pipe is not completely filled with grey water. The pipe is sized to handle excess flow, thus during times of normal use there will be air and grey water present in the pipe. This presents two problems. One, the maximum amount of heat transfer is not occurring because the coil is not completely surrounded by hot grey water. Two, it is impossible to calculate the heat transfer from the grey water to the cold water.

A simple solution to this problem is to place a throttling valve at the bottom of the drain pipe and a float switch at the top, as shown in figure 3.2. A controller can use the throttling valve to maintain a constant level in the pipe, monitored by the float switch. If the flow down the drain stops the last amounts of grey water will remain in the heat exchanger until more grey water enters the drain. This has the added advantage of allowing cold water to be heated by grey water, even after the flow of grey water has been stopped, as opposed to a system without a throttling valve, where the grey water would have already left the heat exchanger.

A system such as this can be modeled using the following heat exchanger equations, assuming the heat exchanger is perfectly insulated:

Equation 3.1. Both grey water and cold water flow

Equation 3.2. Cold water flow, no grey water flow

Equation 3.3. Grey water flow, no cold water flow

Equation 3.4. No grey water or cold water flow

It would be wise to use these equations in a computer program because small time steps, iterations, and solve functions are required. A computer program was not written to model this system because more practical reasons prevent this system from being feasible.

Placing the coiled copper tubing within the drain pipe is the ideal location for maximum heat transfer, however over time hair, soap and other residues will build-up causing clogging problems and preventing efficient heat transfer. For this reason this system is not recommended for use for this particular application.

Pure Counterflow Heat Exchanger
A single wall shell and tube pure counterflow heat exchanger is shown in Figure 3.2. This heat exchanger operates on the same principles as the shell and tube heat exchanger described above, but has a few defining characteristics that separate it. As can be seen in Figure 3.2, grey water flows down through a number of smaller pipes that are enclosed in a larger pipe which holds the incoming cold water.

Figure 3.3

Figure 3.3. Single Wall Shell and Tube Pure Counterflow Heat Exchanger.

Although this system would be less prone to clogging than the previous, it would still encounter problems, especially at the entrance of the system where a large drain pipe would be split into numerous smaller ones. The major flaw of this system is its lack of sufficient heat transfer area per unit length of pipe. The amount of heat recovered from the grey water is directly dependent upon the amount of heat transfer area (i.e. the area of the pipe containing the grey water that is touching the incoming cold water). Because this design uses straight tubing vs. coiled tubing it takes a significantly longer pipe to achieve the same amount of heat transfer. Thus this option becomes unfeasible and is not further addressed because it requires too much space and adds additional capital cost.

GFX Brand Grey Water Heat Recovery Systems
GFX, an industry leader in grey water heat recovery, has developed the system shown in figure 3.4. GFX has received Depart of Energy Grants to research and develop this technology. After comparing this system to the previous two listed in this report and others currently available on the market we believe this design to offer the best performance, reliability and value.

Figure 3.4

Figure 3.4. GFX Grey Water Heat Recovery System².

The previous analyses found that the best arrangement, based on pure numbers, was a shell and tube heat exchanger. The reason why this method could not be used was because hair and soapy water, as well as whatever else goes down the shower drain, will cause clogging in a shell and tube heat exchanger. One of the best features of the GFX is that there will be no more clogging in it than any normal drain pipe.

As can be seen from figure 3.4, the GFX is also compacted, which eliminated the problems that were encountered with a pure counter flow heat exchanger.

GFX has many models for many different applications and loads. The system design is also simple enough that it could be fabricated by university personnel or students if a particular size was needed or there was an economic benefit.

Particular Considerations

In addition to analyzing the different systems for their own merit, considerations have to be made for how the systems will act within the complete design of the energy efficient housing complex.

The final design will include buildings with multiple showers, but a single hot water heater. This presents a problem because the standard GFX system is meant to fit one shower and one hot water heater. There are a number of possible solutions to this problem.

One option would be to place a GFX system beneath each shower and run the cold water coils either in series or parallel. Both of these options present multiple problems. Running the coils in series will result in a large pressure drop in the incoming water flow because of the length of small diameter tubing the cold water will have to travel through. In addition if the first GFX system is filled with hot grey water (i.e. someone just took a shower), but the next couple of GFX systems are filled with cool grey water (i.e. grey water left over from a shower the previous day), the incoming cold water will be heated by the first system and then cooled in the next couple of systems before entering the water heater. This is obviously not very efficient. Running the systems in parallel presents similar problems. If there are four GFX systems per water heater, but only one of them contains hot grey water, then three quarters of the cold water will be flowing through inactive heat exchangers. Finally, both of these options require an excess of materials, because a system must be placed at every shower drain.

It is clear that if there are multiple GFX systems per water heater a control system will need to be put in place. Control systems, and all the plumbing for this type of operation are an added expense. To avoid as much expense as possible, we believe a single GFX system should be shared by multiple shower drains. As with typical residential design philosophy the bathrooms in a single building will be located adjacent to each other, thus the piping required to share a GFX system will not be an issue.

In the event that the final design contains buildings with more than four showers, it is possible that two GFX systems may need to be used. In this case the GFX systems will be run in parallel. A differential controller will control the flow of the incoming cold water. Thus more cold water will be fed to the GFX system that has hotter grey water, resulting in effectiveness equal to that of a system with one shower and one hot water heater.

Hot Water Load

Hot water demand and use guidelines for apartment buildings are provided by ASHRAE³. The guidelines they give are broken into low, medium and high usage. Medium is defined as the overall average, High is defined by tenants that have a high percentage of children, low income, public assistance, or no occupants working. The high category fits many characteristics of the current University Park, but we believe University Park falls into a slightly different category than what is intended by what ASHRAE describes. For this reason we will take an average between the medium and high categories to determine the hot water load per person for the family housing complex. This data can be found in table 3.1.

Table 3.1. Hot Water Demand (Gallons per Person)³.

Maximum Hourly	Peak 15 Minutes	Maximum Daily	Average Daily
7	2.5	68.7	42

Energy Savings

A particular hurdle in this design process was comparing how much energy could be saved with each system. Complex heat transfer methods were required to determine actual heat transfer from the grey water to the cold water. These heat transfer methods are dependent upon accurate minute by minute or even second by second hot water usage data. Because predicting the minute by minute hot water usage data for a particular family is very difficult, energy savings predictions are largely dependent upon assumptions made regarding their hot water usage.

For this reason it was deemed unfeasible to try and accurately estimate the exact energy savings provided by a GFX system. Instead calculations were made to determine the energy savings seen when the system is under full load (i.e. shower running, water flowing to the hot water heater at the same flow rate as the grey water is flowing down the drain). These numbers will give some quantity to the savings that will be actually seen. Actual savings will depend upon the number of showers taken and length, other hot water demand, and the effectiveness of adding a throttling valve to the GFX system. It would be logical to assume that the actual savings will be greater than the numbers presented here.

Actual performance data on GFX systems is very limited. On their website they do present effectiveness values for the G3-60 model. The G3-60 model is a five foot long 3” drain pipe surround by ½” copper tubing. For a flow rate of two gallons per minute, which would be typical of a low-flow showerhead, the G3-60 is rated at about 62.5% effectiveness.

Using Equation 3.5 for this case with cold inlet water at 40° F and grey water at 95° F, we find an energy savings of 570.6 Btu/min. For a typical eight minute shower this would be 4564.8 Btu or 1.34 kWh.

Equation 3.5. Heat Exchanger Effectiveness

In reality what these savings would represent for the energy efficient housing complex is hard to quantify. If say there are 450 people living in the complex and they take 1.25 showers per day, this would amount to a savings of 753.75 kWh. However on the electrical meter this would be an even greater value, because conventional water heaters are not perfectly efficient. For instance, take a typical electric water heater with an efficiency of 90%, it would take 837.5 kWh of electricity to add the same amount of heat as the grey water heat recovery system.

Other factors will change the results seen with this system in actual use. As mentioned before, because of the large size of the housing complex multiple showers will drain through the same GFX system. Although there will not be a large change in performance, it is difficult to predict the actual performance without any experimental data.

The addition of a throttling valve to hold grey water in the drain tank at all times will certainly improve performance. Again without experimental data, it is hard to predict the actual effect of adding a throttling valve.

Finally, the largest change in actual savings vs. the predicted values will come from the hot water usage habits of the occupants using the system. The predictions above were made for equal grey water and cold water flow (i.e. only the shower is running, nothing else). In reality if someone is using hot water (anywhere in the building) while the shower is running, additional savings will be seen. For a normal house this is usually not a concern, because most families will not wash dishes while someone is taking a shower. But in a building with four or eight different apartments, it will be very common for someone to be washing dishes while another apartment is using the shower. It is hard to quantify this savings without knowing exactly when people will use hot water.

Economic Savings

The prediction of economic savings largely depends upon the energy savings predictions. As stated before it is very difficult to predict accurate energy savings, but we will use our previous predictions as a base for this economic analysis. It can be assumed that while these numbers are not exact, they are at least with in a ballpark figure of actual savings.

For this analysis, because we do not know the exact total occupancy of the housing complex, we will calculate savings on a per person basis. Using the same assumptions as above, we know we have a 4564.8 Btu/shower. Say a person takes an average of 1.25 showers per day and we have a savings of 5706 Btu/day per person.

To translate this into a monetary value we will use the current University Park as an example. University Park currently uses natural gas water heaters. These typically have an efficiency of about 60%. So 9510 Btu/day would need to be used by the current water heater to meet the output of the grey water heat recovery system. The average cost of natural gas for the University of Maine is $0.75/therm or $0.000008/Btu. This translates to the per person savings shown in table 3.2.

Table 3.2. Per Person Savings.

Daily	$0.071325
Monthly	$2.17
Yearly	$26.03

Assuming four showers are connected to one drain, thus an average of fifteen people are utilizing the grey water heat recovery system, there will be a monthly savings of $32.55 per system. The GFX system with an installed cost of $439 will pay for itself within 14 months.

In this case allowing multiple families to share one GFX system makes this option particularly attractive. Other factors will change the actual economic savings. Particularly worth noting are rising natural gas prices. The more expensive natural gas becomes the greater the savings that will be gained by the GFX system.

This example compares the savings vs. the current University Park, which relies on natural gas for water heating. The energy efficient housing complex will rely primarily on solar energy for water heating and only use natural gas or another fuel for peak demand needs. This will drastically change the economic savings with the GFX system, since solar power is essentially free after the initial cost.

Conclusion

The GFX system offers large savings and a very low initial cost. Final design sizing, economics, and implementation will depend primarily upon its interaction with other water heating systems (i.e. solar water heating). For this reason the final design of the grey water heat recovery system will have to wait until the solar water system is designed. However at this time we know the system will be a GFX model, insulated on the outside, possible containing a throttling valve, and will be shared by multiple showering units. In addition a similar GFX system will be placed in the laundry facility.

¹EERE Consumer’s Guide: Drain-Water Heat Recovery. 12 Sep. 2005. United States Department of Energy, Office of Energy Efficiency and Renewable Energy. 22 Nov. 2005. <http://www.eere.energy.gov/consumer/your_home/water_heating/index.cfm/mytopic=13040>

²Graywater Heat Recovery (DHR) System: GFX. 2005. Waterfilm Energy Inc. 22 Nov. 2005. <http://gfxtechnology.com/contents.html>

³American Society of Heating, Refrigerating and Air-Conditioning Engineers. 2003 ASHRAE Handbook - Heating, Ventilating, and Air-Conditioning Applications - SI edition. New York: ASHRAE, 2003.