Low-Pain Gain: Performing a Snapshot Audit
The only change in the family lifestyle was learning to turn on the power to an electronic device before using it.
May 06, 2009
This article originally appeared in the May/June 2009 issue of Home Energy Magazine.
I started sizing the PV system for our home by researching our history of electricity usage in past utility bills and found that we consumed an average of about 660 kWh per month. This was well below the national average. But when an estimate for an offsetting PV system came back at around $45,000, I was alarmed.
In order to parse our electricity bill further, I figured our average hourly usage and I was astounded. Our average electrical consumption was almost 900 watts. Where could all this electricity be going?
To get to the bottom of things, I purchased and installed The Energy Detective (TED) whole-house electricity monitor. It worked well, but I was frustrated because it told only how much electricity was being used, not where it was being used. I would watch in fascination as the monitor spiked 100 watts with no apparent cause, or I would sit in a dark house, refrigerator unplugged and everything off, while the monitor glowed to 200 watts.
It was obvious that I needed to find the devices responsible for the observed aggregate flow—something that required measuring current flows with greater resolution. Since the whole-house monitor was measuring current flows at the two main 120V electrical conductors at the home’s main electrical panel, reducing the flow by turning breakers off would limit the flow at the main conductors to just the breakers that were operating. I switched all the breakers off, and then switched them back on one by one, noting the current flow associated with each breaker. (Turning the breakers off one by one and noting the reduction in overall flow is a potentially more accurate method, depending on the contributing devices). I was astonished. With no lights, building systems, or appliances apparently operating, current was flowing at almost half of the breakers!
Cycling breakers was too disruptive for a viable protocol. In addition, the visual resolution on The Energy Detective was too crude to register very low flows (although it logs those flows); also, cycling current in the system might temporarily alter demand to some devices, skewing the results. However, resolution by circuit seemed adequate for an initial investigation, although findings on some circuits would probably require limited additional investigation. I searched for an appropriate tool.
The AEMC SL206 Dual (manual) Range Current Probe was the most compact instrument I could find that would measure the ranges of current I expected to see. It can measure from a few milliamps to over 50 amps. Its size mattered because it would be used in a cramped electrical panel. At the same time, it was reasonably accurate. It was still not ideal. Greater compactness, more accuracy, and lower sensing capabilities would all be better, but it was the best I could find. Since this device is only a probe, which converts current to voltage but has no display, it must be used with a voltage-sensing device, such as a multimeter.
I also picked up a clamp-on meter and line splitter, which made auditing individual plug loads convenient, and a plug load monitor, that allowed limited logging of cycling loads. If you opt to get a clamp-on meter for this application, make sure it has the range capability to work down to 5 milliamps or less, so that it will detect low flows. These low-flow meters are usually referred to as leakage meters.
Developing the Audit Method
To better understand electricity flows in a home, I sought a simple way to get the most information for the effort. However, since the method I am about to describe is not a prolonged monitoring of electricity flows—but rather an instantaneous snapshot—some of the blanks have to be filled in with educated guesses. This exposes both the advantages and the limitations of the method. Its primary advantage is that it is fast and relatively cheap.
|Current versus Power in Nonlinear Devices
In the methodology I describe, I convert current measurements to power (watts) by multiplying by voltage. This assumes that the power factor of the current drawing devices (real power divided by apparent power (volts x amps)) equals one. However, this is often not the case, due to either non-sinusoidal waveforms—typical of electronics, including most CFLs—or to voltage and current waveforms being out of phase—typical of motor-driven loads, including things such as refrigerators and furnaces. This can pose problems, including overestimating potential savings.
This issue can be overcome by always using a true-root-mean-square (TRMS) power meter, which compensates for power factor contributions of both distorted and out-of-phase waveforms. While the Kill-A-Watt plug load monitor and the TED whole house monitor report TRMS power (the Kill-A-Watt can also display power factors), the AEMC probe and the clamp meter I discuss do not. Unfortunately, although I have searched for a compact, affordable TRMS power meter with the range and sensitivity the method requires, I have not yet found one.
Alternatively you can simply be aware of the issue and compensate the best you can. For instance, be suspicious of readings from devices if those readings do not fall within an expected range. (Typical ranges are available from various sources. See, for example www.standby.lbl.gov/summary-table). If looking at lighting, sum the rated wattage of the fixtures for comparison to measured values. Where possible, compare your readings to manufacturer’s claims or validate them with a Kill-A-Watt or other TRMS power meter. And above all, keep in mind the nature of the audit. If an action is indicated with a 50% or more annualized return despite a moderate error in power conversion, it will still be indicated were the error corrected. If, on the other hand, you are parsing the last few percentage points to justify doing something, you should be using a more refined method.
The method can be summarized as follows: (1) identify all the current flows in a home as either active or passive; (2) isolate and audit the passive flows, noting contributing devices; (3) inventory the active devices and estimate their consumption; and (4) compare the findings to the historical consumption as shown by utility billings. The resulting data will give a rough picture of overall system flows, which can be used to inform a conservation strategy and predict the results of implementing it.
The passive flows are the base load flows—the near-constant background electricity usage that is not associated with actively operating appliances, lights, or building systems. The passive flows are associated mainly with idling electronics, such as unused furnace controls, TVs, or touch pad appliances. The active flows are everything else—the flows that are associated with devices that are switched or cycled on and off by occupants, timers, thermostats, and so forth. (Note that many devices, such as appliances and electronics, fall into both categories.) Because this method is static, it is best suited to evaluate conservation opportunities presented by examining passive flows. If the audit indicates large active flows, additional investigation using a different method, would be called for.
To isolate passive flows, switch off all dynamic and transitory loads, so that the instrumentation does not sense them. All lights and appliances are turned off; thermostats are set back so no demand is placed on conditioning units; and timed loads are disabled, as though the house were empty. Note that nothing should be disconnected. Many devices produce both active and passive flows, and disconnecting them during the audit will skew the results.
Next, I read the current flows at each circuit by removing the panel cover and inserting the current probe. I enter each reading into a spreadsheet, along with the circuit location. I assign the noted flows to one of six broad categories: lighting; entertainment; outlets (plug loads not part of any other category); building services (furnaces, water heaters, and so on); appliances; and miscellaneous (primarily mixed circuits containing more than one of the other categories). This makes it easier to examine usage patterns and pinpoints areas where electricity is being wasted. Table 1 shows the audited pre-conservation current values for our home.
I use the passive flow audit results to survey the system so that individual devices, or groups of devices, can be assigned to the circuit flows noted. The more accurately the home’s electrical system is mapped, the less detective work this requires. A circuit designation as noted in the panel box may accurately describe the connected devices. If not, the system will have to be mapped. Even if the system is already well mapped, when considering certain conservation strategies it will be necessary to individually audit some devices or groups of devices if they are connected to a single circuit. This would be done to ascertain their contributions to noted aggregate flow so the conservation potential of altering their operation can be investigated. For plug loads I do this with a line splitter and a clamp-on meter. For integrally wired loads, I find somewhere to interrupt the circuit that isolates the device or device group and use the AEMC probe and multimeter or a clamp-on meter, whichever is appropriate.
For active flows, I usually inventory the home and then interview the occupants to see how the contributing devices are operated. How much effort I put into doing this depends on how much energy a given device uses, and how willing (or able) the occupants are to reduce that usage. For example, lighting systems are often a source of significant active consumption, and can often be modified relatively easily to reduce that consumption. For this reason, they often warrant investigation. The active-component consumption of a microwave that is used only occasionally probably isn’t worth considering, since not only is cumulative consumption low, but the occupant would probably be unwilling to give up the convenience. The passive component of the same microwave, which in all likelihood exceeds the active component, is probably worth investigating.
Refrigerators present a special case. Since the length and frequency of their cycling affects overall consumption, a snapshot audit is insufficient. To get a rough estimate of how much a refrigerator is consuming, I monitor it over the course of the audit with a plug load monitor and insert the average rating in the spreadsheet. However, if the refrigerator is being considered for replacement, a longer monitoring is required, at least three or four days, to allow cycling through the defrost cycle. If the refrigerator is not on a dedicated circuit, I disconnect it during the circuit audit, so that I can measure the other flows on the circuit.
To estimate total system performance, I sum the audited passive flows in the spreadsheet in each category and then convert to watts by multiplying by the voltage and then add active flow estimates (see Table 1). These estimates can be derived in several ways. For lighting, I like to sum the wattages of lamps that are frequently used and multiply the total by the number of hours of daily use. Then I divide the product by 24 to find average hourly usage.
For other active flows, I usually simply enter an “Active Adjustment” factor—little more than an educated guess, which can get slightly more refined in the next step—as a multiplier of the passive flows for the same category. The product is then added to the Passive Watts and multiplied by the number of hours in a month to obtain the “Projected Monthly kWh” by category. The projected monthly kWh in all categories are summed to give the “Total Projected kWh/month.”
Next I examine historical utility data and compare these data to the projection. For this, I like to average a year’s worth of utility bills. Active-flow assumptions can then be adjusted in order to make the consumption predicted by the audit approximate that shown by historical utility data. Manipulating active flows like this essentially takes the portion of the historically observed consumption that cannot be accounted for in the audited passive flows and spreads it across the consumption categories. As I explained above, how accurate you choose to be in treating active flows will depend on the emphasis of the audit, and on whether a particular active flow is a good conservation target. Keep in mind that the purpose of the audit is not to predict overall usage, but to identify conservation targets. Truing up the audit predictions with the historical data by manipulating the active flows simply estimates the overall flows as closely as practicable, given the data we have.
Plugging the Leaks: Using an Audit to Green a Home
With the current flows in our house identified and quantified, I could see much more clearly where our electricity was going, and what steps we could take to reduce. I studied the spreadsheet and made a list of the contributors to overall consumption that were primarily operating passively. As applicable, I tested individual devices or groups of devices to ascertain their contribution to the aggregate flows noted. Through this process I discovered, for example
- Unused X-10 automated switches and receptacles consumed between 2 and 3 watts each. This accounted for some of the noted flows at lighting circuits when no lights were turned on.
- The warm-air furnaces consumed between 6 and 11 watts each continuously, and the tankless water heater consumed about 4 watts.
- Entertainment electronics stacks consumed between 10 and 60 watts each continuously.
- An idle computer and its peripherals consumed over 100 watts continuously.
- Garage door openers consumed about 6 watts each continuously.
- The telephone answering system consumed about 5 watts continuously.
In all, the passive flows accounted for well over half of our electricity consumption. A few simple steps could save a lot of energy. We considered doing the following things:
- Remove X-10 components from the system.
- Put electronics stacks on switched power strips.
- Cycle current to the furnaces off in non-heating months by switching the circuit off at the breaker. (This can alternatively be done at the furnace, if the unit does not have a dedicated circuit.)
- Install a switched power strip at the microwave oven.
- Install switched circuits to the garage door openers so they could be turned off when a car was in the garage.
- Replace the incandescent lamps of all commonly used light fixtures with CFLs or LEDs.
The effects of implementing these measures can be roughly quantified. I calculated the potential electricity savings using the data I had measured in the audit and estimated the cost of having an electrician or a skilled handyman perform each task. Savings are calculated using a blended tiered rate of approximately $0.22/kWh, applicable in my utility area between approximately 350 and 550 kWh consumption per month. The costs of implementing these measures are low, and returns on investment are favorable (see Table 2).
When we implemented all of the conservation measures I listed, I was shocked, but in a good way. Comparing July through December 2006 to the same period in 2007, average monthly consumption falls from 660 kWh to less than 360 kWh, a reduction of more than 45% and very close to the predicted savings. Since the reduction moved us to a lower rate tier, the result was a cost savings of almost 65%.
Our savings are even more impressive when you consider that because we now needed less electricity to run our home, we were able to install a smaller PV system. We shaved our average consumption by just over 400 watts. In our climate zone, offsetting baseload with PV solar capacity costs about $50 installed cost per baseload watt. This means that we saved approximately $20,000 in PV system costs.
These simple measures have had minimal impact on our lifestyle. The most intrusive change was having to power up a computer to use the Internet, although I found that I had simply gotten used to hopping online, and not having instant access cut down on my tendency to do so. Other than that, the only change in our lifestyle was learning to turn on power to a unit before using it.
A Final Note
The most startling realization that came out of developing this audit process is the high energy cost of having so many electronic devices idling in our electrical systems. This is something I try to impress on my clients, especially those who are considering installing a PV system. A forgotten 5W transformer equals a 500W light bulb burning 15 minutes a day. An unused entertainment stack pulling 25 watts is like 15 10W CFLs left on for four hours. This seems insane, especially when we consider that the utility derived from these idling devices is almost zero. The folly is multiplied when we buy PV systems to offset this consumption, essentially paying for all that waste up front.
It is as though we cannot turn our homes off. As an architect, I look to specify electrical devices that don’t add to a home’s baseload, but I often find it difficult. Furnaces, water heaters, phone systems, intercoms, sprinkler systems, lighting controls, gate operators, computers, entertainment systems, and countless other items shed energy continuously. Depending on how a device is used, this can go on for months, or even years, without resulting in any convenience or utility. Although some of this is changing as new regulations limit the baseloads of some electronics, the existing installed base of energy-wasting devices is huge, and many items are still unregulated. Users are largely unaware of this constant energy drain, and even well intentioned consumers, who shut off the proverbial lights as a nod toward conservation, often have no idea what is actually happening at the electric meter. We need to sound the alarm.
Frank Bergamaschi is a California licensed architect and currently practices in San Francisco.
>> For more information:
The method of measuring the current flow for each breaker by turning the breakers on one by one is described in Parker, Danny, et al. “How Much Energy Are We Using?” available on the web at: www.fscc.edu/en/publications/pdf/FSEC-CR-1665-06-pdf
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)