5.0 Appendices

5.1 Preliminary findings memo

To: Fred Smothers and Glen Starkey
From: Mary Ann Piette and Peter Rumsey
Cc: Kris Kinney, Chris Shockman, Tony Sebald, Satkartar Khalsa
Date: Oct. 28, 1998

Re: Findings from the Information Monitoring and Diagnostic System (IMDS)

As part of our research project with your building we have been examining the energy performance data collected to identify energy saving opportunities at the Hong Kong Bank Building at 160 Sansome Street. This memo outlines our key findings to date, along with recommended actions to reduce energy use. We have put these items in writing for you to review and would like to discuss them with you.

For each finding we have included a description of the problem, a conservative estimate of the energy saving and other benefits, and recommendations to achieve the benefit. Energy cost savings are based on a simple 10 cents per kilowatt hour. Your actual savings may be substantially higher. We have not included any potential peak demand savings.

Please note that these are PRELIMINARY ROUGH ESTIMATES and we look forward to discussing these energy saving opportunities with you in more detail. This is not an exhaustive list of all possible O&M and retrofit energy savings. We are still examining the overall energy performance of the building.

The building currently uses about 1,838 million watts per year (MWh/year), not including steam. We estimate O&M savings of about 275 MWh/year, which is about 15 percent of total energy use, as summarized in the table on the last page of the memo. These savings are similar to the level we predict can be found in most buildings with the type of information we have collected from the IMDS. In addition to the low-cost O&M savings, we have identified another 15 percent savings that can be achieved with more aggressive retrofits.

We hope that you are able to implement some of these recommendations while there is some cooling season left. When we discuss these savings possibilities with you, we will show you the data from the IMDS that were used to develop these energy savings estimates.

1. Lights, plug loads, and other

Finding 1.1: High nighttime plug loads
Description: The plug loads at the building are about 0.7 watts per square foot (W/sqft). While this is not high compared to other buildings, the nighttime load remains fairly high, at about 0.3 W/sqft.

Implications: Nighttime energy use is higher than necessary, and could be cut by about 0.2 W/sqft. These savings are equivalent to about:

0.2 W/sqft * 98,000 sqft * 3500 hours/year * $0.10/kWh) = $6860/year

Not only will this save energy, it will reduce cooling loads.

Costs to implement such a change are negligible with tenant cooperation, which we've been able to affect in other buildings.

Recommendation: Develop an information campaign for tenants to turn off unneeded equipment, such as computers, monitors, copiers and printers. Inform tenants about savings from enabling auto-power down features of Energy Star office equipment. This could be done with flyers, email, or a memo explaining the opportunity for savings. We are happy to help in this area. (See http://eetd.lbl.gov/EA/Reports/39466 and http://www.epa.gov/office.html).

Finding 1.2: High early-evening lighting
Description: As observed by the chief engineer, interior lights are on longer hours than needed.

Implications: Early evening lighting energy use is higher than necessary, and could be cut by about 0.3 W/sqft during early evenings. If this level of lighting power were cut during work evenings by four hours, these savings are equivalent to about:

0.3 W/sqft * 98,000 sqft * 1000 hours/year * $0.10/kWh) = $2940/year

Not only will this save energy, it will reduce cooling loads.

Costs to implement this change are negligible assuming janitorial assistance.

Recommendation: Develop an information campaign for tenants or janitors to turn off unneeded lighting.

2. Chillers

Finding 2.1: Retrofit chillers and include a 100-ton chiller
 
Description: Both chillers one and two operate at about 0.8 kW/ton and spend the majority of hours under 100 tons.

Implications: 120-ton screw chillers are available that operate at about 0.67 kW/ton. A rough estimate of energy savings is then:

100 tons * 2000 hours * 0.13 kW/ton * $0.10/kWh = $2600/year

There will also be significant pumping and cooling tower savings with a smaller chiller.

A 120-ton screw chiller is about $30,000, but this purchase may be warranted given interest in replacing the existing chillers which are reaching the end of their life.

Recommendation: Explore retrofit options more thoroughly using measured cooling load data.

3. Cooling tower

Finding 3.1: Lower condenser water temperature
Description: The condenser receives water between 72 and 75 degrees F from the cooling towers. During high-load days, the towers run full out, but during low-load days they cycle frequently. A lower condenser water temperature would reduce chiller energy use and reduce cycling. The existing chillers can operate with condenser water temperatures as low as 55 degrees F (this has been verified with the manufacturer).

Implications: The rule of thumb is: there is a 1.2-percent degradation in chiller efficiency with every one degree of condenser water temperature. A 10 degree F change in condenser water translates to a 12-percent improvement in chiller efficiency. We verified this relation with actual data from Chiller Two, identifying a one-percent degradation in efficiency with each change in condenser temperature.

100 tons * 2000 hours * 0.1 kW/ton * $0.10/kWh = $2000/year

Costs to implement are negligible.

Recommendation: Reduce the condenser set point to 62 degrees F. This may require the VFD retrofit described in Finding 3.3, but could be experimented with to improve the chiller efficiency.

Finding 3.2: Run cooling tower water without fans
Description: The cooling towers could be used to provide some natural cooling from convection without the fans. This strategy would be useful on mild days when the towers cycle frequently under current operation.

Implications: A rough energy savings estimate is as follows:

20 hp * 0.746 kW/hp * 750 hours/year * $0.10/kWh = $1119/year

Not only will this reduce fan energy, but extend equipment life.

Recommendation Experiment with running the water in both tower cells without the fans. Examine the cooling provided and the resulting condenser supply temperature.

Finding 3.3: Install VFD?s on cooling towers
Description: The cooling towers are cycling on load days and only one cell is used at a time. The use of variable frequency drives would greatly improve the ability to fit the tower fan use to the load. This would reduce cycling, which is not good for the fans or the chiller, and significantly reduce fan energy use.

Implications: The use of VFD?s on the cooling tower has two effects. First, it could cut tower fan energy. Currently there is one 20-horsepower (hp) fan. With a VFD, average operation would involve using both fans with the VFD?s, or: 2 * 20 hp * (0.5)3 = 5 hp. Annual savings are thus about :

15 hp * 0.746 kW/hp * 2000 hours/year * $0.10/sqft = $2238/year

Part of the savings would come from running both cooling tower fans, and providing a lower condenser set point than is currently available. The VFD may be a necessary element to achieve the energy savings outlined in Finding: 3.1.

A VFD retrofit will require about $2500 for each VFD and another $2500 for the installation of each, or $10,000.

Recommendation: Explore possibility of VFD retrofit in relation to other cooling tower strategies presented.

Finding 3.4: Reduce condenser-side pressure drop
Description: The rated flow for the condenser is 615 gallons per minute (gpm). Chiller Two condenser flow is about 500 gpm and Chiller One, 530 gpm.

Implications: The reduced flow reduces the condenser heat transfer. Increasing flow would improve the condenser performance and chiller efficiency. We estimate that the chiller performance suffers by about 0.05 kW/ton, or

100 tons * 3000 hours * 0.05 kW/ton * $0.10/kWh = $1000/year

Recommendation: Check to ensure that all balancing valves are open and check valves move freely. Check that condenser tubes are clean.

4. Pumps

Finding 4.1: Reduce use of two chilled water supply pumps together
Description: Two chilled water pumps are occasionally used with one chiller.

Implications: This operation dilutes the chilled water supply temperature and results in higher than needed pump and fan energy. We found about 45 hours of two pumps with one chiller on during the last few months of monitoring. Annual occurrence is probably 50 percent greater or so. While this condition does not appear often, it does result in significant energy losses when in use. These losses can be estimated as follows:

67.5 hours * 20 hp * kW/hp * $0.10/kWh = $101

Fan and chiller energy is also impacted since these equipment need to run longer hours than they should during such conditions. Thus, additional savings are:

25 hours * 175 hp * 0.746 kW/hp * $0.10/kWh = $326

80 tons * 1 kW/ton * 25 hours * $0.10/kWh = $200

Recommendation: Do not run two pumps with one chiller

Finding 4.2: Trim chilled water pump impellers
Description: The chillers are rated for 450 gpm. Chillers one and two currently operate at about 580 gpm. The pump impellers should be trimmed to reduce the flow and energy use.

Implications: The pumps consume about 7.5 kW each. The energy savings from trimming the impellers is as follows:

(450/580)3 * 7.5 kW/pump * 2000 hours/year * $0.10/kWh = $701/year

Cost is about $500 per pump.

Recommendation: Trim pump impellers to achieve rated flow through chillers.

5. Fans

Finding 5.1: Modify ducting to change return fan to be used as exhaust fan
Description: The current fan system uses a return fan instead of an exhaust fan. Significant energy savings could be available if the system was modified to allow the return fan to be used as an exhaust fan. This analysis should also include examining the opportunity to change the filters and silencers used in the supply fan area.

Implications: A ballpark estimate for fan energy savings from the modification of the return air fan is as follows:

50 hp * 0.746 kW/hp * 6000 hours/year * $0.10/kWh = $22,380/year

Recommendation: A comprehensive retrofit should be evaluated, which could also include changing the filter system on the supply fan to include a pre-filter to catch any large debris, followed by cartridge filters. This would lower the pressure drop and reduce fan energy. The silencers in the air distribution system may be more extensive than necessary. The pressure drop from the silencers should be measured and evaluated.

Finding 5.2: Tune VFD operation on supply fans and return fans
Description: Glen and Fred have been examining the use of the VFD?s on the supply and return fans and have reported that the VFD?s are not being used in an optimal fashion. We are happy to assist in debugging the operational problems that appear to be defeating the energy savings from the VFD?s.

Implications: A ballpark estimate of savings from using the VFD?s to achieve a 10 percent reduction in flow suggests power savings related to the savings in flow, cubed, or:

(0.9)3 * 75 hp * 0.746 kW/hp * 3000 hours/year * $0.10/kWh = $12,236/year

Recommendation: Needs further review

        Summary of findings
        Type*
        One-time cost to
        implement
        ($)
        Annual energy savings
        (kWh/yr)
        Annual cost savings
        ($/yr)
        1. Lighting and plug loads        
        1.1 High nighttime plug loads
        O&M
        0
        68,600
        6860
        1.2 High early evening lighting
        O&M
        0
        29,400
        2940
        2. Chillers        
        2.1 Retrofit chillers w/ 100-ton chiller
        R
        30,000
        26,000 
        2600
        3. Cooling tower        
        3.1 Lower condenser water temperature
        O&M
        0
        20,000
        2000
        3.2 Run cooling tower water without fans
        O&M
        0
        11,190
        1119
        3.3 Install VFD?s on cooling towers
        R
        10,000
        22,380
        2238
        3.4 Reduce condenser side pressure drop
        O&M
        0
        10,000
        1000
        4. Pumps        
        4.1 Run one chilled water supply pump 
        O&M
        0
        6270
        627
        4.2 Trim chilled water pump impellers
        O&M
        1000
        7010
        701
        5. Fans        
        5.1 Modify duct/return to exhaust fan
        R
        50,000
        223,800
        22,380
        5.2 Tune VFD on supply & return fans
        O&M
        0
        122,360
        12,236
        Totals         
        *O&M - Low cost change & O&M
        O&M
        1000
        274,830
        27,483
        *R- Retrofits
        R
        90,000
        272,180
        27,218

5.2 Whole-building energy use analysis

The IMDS began collecting data on May 1, 1998. The only energy consumption data available for the building before that date are from utility bills, so we used these bills to evaluate whether there have been energy savings since the IMDS was installed. Data from January 1997 through June 1999 were used. Data are available from 1991; however, for comparison, only the previous year was used, to reduce the effect of significant changes in occupancy, the conversion to VAV, and changes in the use of the building. The number of days in each billing cycle is not known, so data are not normalized, which might have smoothed some of the variation. Although the system was not used much for several months after the installation, we used the installation date (May 1, 1998) as a change point. Temperature data was averaged from University of Dayton?s daily temperature archive located at http://www.engr.udayton.edu/faculty/jkissock/weather/ for the city of San Francisco.

Figures 5-1 through 5-3 show regressions for the two time periods for total monthly electricity (kBtu; thousand British thermal units) and steam (kBtu) versus average outside air temperature (degree F). These regressions were performed in Emodel©, an analysis software that also calculates energy savings for retrofits.

5.2.1 Electricity

Figure 5-1. Monthly electricity (kBtu) versus average outside temperature (degree F)

The baseline period shows a modest correlation with temperature, while the post-installation period shows none. See the Phase Two report (Piette, et al., 1998) for further discussion of the baseline data. The savings calculation indicates that 7547 kBtu were saved every month between June 1998 and June 1999; however, the uncertainty is so large, it is impossible to say if there has been any change at all.

5.2.2 Steam

Figure 5-2. Monthly steam (kBtu) versus average outside temperature (degree F)

The lesser quality of the fit of the regression line in the post-installation case compared to the pre-installation period is due to the single extreme outlier, which also affects the energy savings. This could be due to non-normalized data.

The savings calculation for steam is negative, indicating that more steam energy has been used since the IMDS was installed; however, once again, the uncertainty is greater than 100 percent, making it impossible to determine if there was any change. Although steam is not measured by the IMDS, there is still an opportunity to save energy on steam. Steam data may be incorporated at a later date or in future implementations.

5.2.3 Total building energy

Figure 5-3. Monthly total energy (kBtu) versus outside temperature (degree F)

As seen above, both the pre-installation and post-installation models seem reasonably good; however, the savings calculations have such great uncertainty. We conclude that the IMDS has not resulted in any significant energy savings to date.

Total energy use from 1991 through 1998 is shown in Figure 5-4. This shows that energy use has been changing significantly over the years. Electricity use was slightly down in 1998, but steam increased. This increase was primarily related to weather. As shown above, however, there is little statistically significant difference in energy use from 1997 to 1998.

Figure 5-4. Multi-year whole-building energy use

Start of Report
Legal Notice
Acknowledgements
Preface
Executive Summary

Section 1. Introduction
Section 2. Conclusions and Recommendations
Section 3. Discussion
Section 4. References

Project Home  |  Commercial Building Systems  |  Building Technologies
This page last updated 12/20/99 by SKhalsa