Household Electrical Energy Consumption Management



I just bought an electric car, a Chevy Volt. The Volt is known generically as a plug-in hybrid. To lower the costs of operating it beyond what a light hybrid can do, it needs to run for substantial periods on pure electric power, which it can do. Electricity, depending on the cost per kilowatt-hour and the cost of premium gasoline, costs less per mile than gasoline does.

The key to charging an electric car is to do it under conditions where the electrical rates are lowest, that means overnight. This is convenient because most cars spend their nights in a garage or driveway somewhere where power might be available. However, without signing up for a special rate plan or maybe installation of a 2nd power meter (expensive), typical household electrical costs are high enough so that any additional power is charged at a rate that amounts to nearly the cost of gasoline based on how far that power will drive the car.

My electrical power rates are marginally high and in past years I have already made a pass through the house to make changes that reduce our power consumption. I won't be able to make the substantial increase of power consumption due to the car fit in our current baseline, but I can try. Every little bit helps.

The Volt consumes 13.4 kWh per full charge, a pretty good chunk. If I did that every day, it would consume all of my Tier 1 and Tier 2 power allotment by itself. This is also exactly how much power that my whole house consumes in an average day without the car.

The main subject of this page is to describe how to minimize the power consumption of a typical household. Even if an electric car is not involved, a homeowner can save some money by paying attention to the energy consumption of everything in the house.

Electrical Utility Rates

A watt is a unit of power. However, the power company doesn't sell you power, they sell you energy. Energy is power multiplied by time. The typical unit of electrical energy is the kilowatt-hour (kWh). A kWh is 1000 watts consumed for an hour and typically costs between 6 to 35 or more cents. The rate that you pay depends primarily on your geographical location. If you live in the northwest of the US near abundant sources of hydropower, you'll get a low rate. If you live in places far away from cheap sources of power, you'll pay more. If you use a lot of energy, you'll pay much more.

A high power appliance, such as a microwave oven, doesn't consume a lot of energy because it typically only runs for a minute or two at a time. Another high power appliance, such as an air conditioner, can consume a lot of energy because it runs for hours at a time.

Utility rates are based on lots of factors. Geographical location is one, time of use can be another. Costs per kWh tend to increase as you use more energy.

Tier 1 Baseline rates are the lowest rates and are set, in California by the California Public Utilities Commission, so that consumers that don't use a lot of energy can manage their costs by keeping their usage within the baseline. The allocation, also set by the PUC, is expressed in kWh/day varies depending on location. In my area, the energy rate is $0.13 per kWh any time of the day on the standard Southern California Edison (SCE) Schedule D rate plan. The baseline rate is available for usage up to 9.2 kWh/day in the summer (June through September) and 9.6 kWh/day the rest of the year. This is called Tier 1. All the rest of the rates are based on the Tier 1 rate and allocation.

Any particular user's baseline allocation depends strongly on their geographical location, if their home is "all electric" and if they have some special medical condition that requires electrically powered life support equipment. You can find a table of these allocations on your electric utilities' website for your location.

Tier 2 rates are higher. My rate is $0.16/kWh from 100% up to 130% of the baseline allocation.

Tier 3 starts at 130% and goes to 200% of the baseline allocation and costs $0.26/kwh.

Tier 4 ranges from 200% to 300% of the baseline allocation and costs $0.29/kWh.

Tier 5 ranges from 300% to 400% of the baseline allocation and costs $0.33/kWh.

These rates are subject to change, but the trend remains the same. The more energy you consume, the more that energy costs per kWh.

The rates are set this way to provide a disincentive to consume a lot of energy. Folks with whole house A/C, electric heating, and electric water heating get a higher allocation at the baseline rate but the allocation really doesn't go high enough to cover the higher energy costs of electricity vs natural gas. Further, if you have a pool with a power hungry pump, you don't get a baseline allocation increase for that. If you live in a very hot area, there is a large increase in the allocation to cover the expected higher costs to remain more or less comfortable. In Death Valley in the summer, the baseline allocation is more than 4 times my allocation as I do not need A/C.

During summer days when it is hot outside, there are air conditioners running almost everywhere. All these A/C units consume enormous amounts of aggregate energy. The electric utilities have to size their generation, transmission and distribution networks to handle to handle the peak loads. Even then, on some days the system cannot handle the load and brown outs or outright power outages occur. Sometimes those outages are planned and are rotated throughout the service area so that any given customer may see an hour of outage. This is not long enough for refrigerators and freezers to warm to the point of food spoilage, but they sure are long enough to feel the effect of no air conditioning.

The utilities are trying to avoid these peaks by moving as many loads as they can to off hours when the peak air conditioning loads are not as high. Even in the winter, system load is higher in the daytime because business are operating and consuming power. At night, many of those businesses close, turn out their lights and machines and the energy consumption goes down. One of the ways that the utilities try to shift load to the night time is to provide incentives to consume energy at night instead of during the day. Electric car charging is one of those things.

An electric utility can avoid some significant capital equipment costs for power plants and new transmission lines by leveling the consumption. This is financially important to them. One way to level energy consumption is to offer special rates on a Time Of Use (TOU) basis. The newer power meters are smart enough to know when energy is being consumed and can calculate what the rate is and report energy consumption vs time back to the utility so that the utility can bill based on when the energy was used. The rates are lower than a standard rate at night, but also higher during peak periods, sometimes MUCH higher. This a carrot and stick approach.

One method that SCE uses is to simply adjust the rates vs time using the existing power meter. The savings are modest but still important. This plan can reduce the cost of charging a car by 20% to a little more than 50% depending on which Tier the customer is currently in. Another method is to use a 2nd power meter for the car alone. This rate is can be about 66% less costly than a standard Schedule D rate. However, it requires capital expense to install a 2nd breaker/meter panel, which can run $1500 without some special incentives. In my case, the payback at that cost would be 8 to 10 years, too long to make that practical. However, users of full electric cars (that could consume about 1.5x to 3x the energy to charge as compared to a Volt) and need a 220 volt circuit to recharge, the payback period gets reasonable, down to 3 or 4 years for a car driven for most of it's range every workday. The 2nd meter approach provides a 15 hour window in which to charge the car. This will be needed by a full electric car even with a high rate charger.

There are incentives available that make it less expensive to make these capital improvements. Just paying a general electrical contractor to make the changes doesn't work economically.

I cannot charge the Volt from a fully depleted battery to full charge in the 6 hour "super" off peak rate that the SCE plan provides unless I use a high rate charger, but I can charge it for just those 6 hours and get enough charge to drive it the next day for as far as I am likely to go. If I want to go further, then the car burns gasoline but still gets 40+ mpg. Installation of a high rate charger would still cost me about $1500 without some special deal, about $500 to get a 220 volt circuit installed in my garage and $1000 for the 220 volt charger.

The car can be programmed to charge only during the "super" off peak rate. It will start charging at midnight and stop at 6 AM, full filling the "super" off peak rate window. Even at the "super" off peak rate, the energy I need to charge the car costs me $0.17/kWh because my normal household usage consumes most of the baseline "super" off peak rate at $0.11/kWh.

Power Conservation Tips

There are at least four ways to attack the energy consumption of a household.

  1. Turn off the damn lights
  2. Get rid of unnecessary and inefficient appliances
  3. Use higher efficiency lights and appliances
  4. Go totally OCD and track down and measure each and every energy consuming object in the home

A really easy way to consume less energy is to simply turn off the lights when they are not being used. Lighting consumes a lot of energy, easily several kWh/day, out of a baseline allocation of maybe 10 kWh/day. However, this also involves a lot of training of members of the household that are not paying the bills. This can be a hard thing to do. Other power hogs, such as big screen TV's, game consoles and computers, should be shut off or allowed to sleep when they are not in immediate use.

The next step is to jettison a power hungry appliance or two. I looking at that refrigerator or freezer in the garage as a prime example. One refrigerator really is enough for a whole household. A freezer full of food is a target for an extended power outage where all that food can spoil. If you want to stockpile food, do it with non-perishable items. The quantity of frozen food can be managed so that it all fits in a regular freezer compartment in a standard refrigerator even you buy frozen food at Costco. If you want to buy some frozen item that you don't have room for, eat something else first. This also tends to rotate the frozen food so that it doesn't get really old.

Step 3 is to replace inefficient power hogs with more efficient substitutes. If your refrigerator is more than 10 years old, it is likely a power hog. Not all of them are, you need to measure it to be sure, see below for the method. However, a 30 year old refrigerator that still keeps food cold may be doing it an an awful cost, perhaps 4x what a new one would cost to run. Powering such a behemoth can cost you $250 a year or even more depending on your marginal energy rate. In such a case, a new refrigerator can pay itself back in 2 years to maybe 4 years mostly depending on the excess energy consumption of the old one and how expensive the new one is. Inexpensive models with high efficiency can be found for $500, but don't expect to get stainless steel or a built in ice maker at that cost.

Your big screen plasma TV can draw 700 watts when running, costing you more than 10 cents an hour. A newer big LCD could run half that or maybe you could get along with a smaller one that consumes less power yet, around 100 watts.

Another big energy waster is regular incandescent lighting. Get rid of as many of those as you can. New Compact Florescent Light (CFL) bulbs work very well, will usually last much longer, and consume a little less than 25% of the energy of a standard bulb with the same "wattage" rating. Look for the instant rebate supported pricing at Costco or other big box stores where the bulbs cost about $1 each although you can get them at in multipacks for about $1.25 each any time. CFL's have downsides too. Most don't fit in lamps where the shade clips over the bulb, but some do. You have to look for these. CFL's tend to run dim at startup where their light output is about 1/4 of what they will do after a 2 minute warm up. Most CFL's and dimmers do not get along. Any kind of a electronic controller will tend to blow them up. A few models of CFL's are rated to be dimmable, but they tend to make a buzzing sound when dimmed. Dimmable CFLs also confuse X-10 lamp modules and often do not completely go out when turned off with an X-10 module. CFL's are generally not applicable to dimmers, motion sensing lights, X-10 modules, or completely enclosed fixtures. You might get away with a "60 watt" CFL (which only draws about 13 watts of real power) in an enclosed fixture, then again, you might not. The heat build up in the fixture may kill the bulb. If that lamp fixture is used only for short periods, less than a hour at at time, then there will be likely little problem. If that bulb is on for extended periods, then don't use a CFL in that fixture.

A newer form of lighting is the Light Emitting Diode (LED). These bulbs are somewhat more efficient than CFL's but cost a lot more. Some also come in some very odd shape factors. For the same amount of light, a good LED bulb will consume 2/3 of the power of a CFL but might also cost 10x to 20x that of a CFL (depending mostly on the CFL cost). An LED bulb does not have as much of a heat problem as CFLs as the LED bulbs are more tolerant of the heat that they do make. Some LED bulbs will also work with the dimmers or other circuitry that are not tolerated by CFLs. Dimmers generally won't blow them up, however they also tend to buzz annoyingly and will also confuse X-10 modules.

One particular type of LED lamp stands out from the rest. These are the bulbs from CREE Research, now available at Home Depot for $10 for a "40 watt" bulb (6 watts real power) or $13 for a 60 watt version (9.5 watts real power). According to my light meter the 40 watt bulb makes as much light in a direction perpendicular to the axis of the bulb, as an incandescent 60 watt bulb. The 60 watt version is brighter than a 60 watt incandescent. Further, the Cree bulbs are dimmable and don't make the buzzing sound that some other LED bulbs make while running on a dimmer or electronic controller such as an X-10 lamp module or a motion detector light switch. When used with X-10 modules, some other LED bulbs do not actually go off, they just get really dim when the module is commanded off. The Cree bulb goes off and doesn't appear to freak out the module and turn itself back on. These LED bulbs are expensive, but they do cost about half of what other LED bulbs cost. The CREE bulbs replicate the shape of an incandescent lamp so that they work in cases where a lamp shade clips over bulb.

At $0.15/kWh, a regular 60 watt bulb costs $0.21/day if it ran all the time. At that cost, the $10 "40 watt" CREE bulb, drawing 6 watts, would pay for itself in about 55 days while producing almost the same total light output. This makes the CREE bulb not expensive at all. The 40 watt CREE bulb draws about half the power as a 60 watt CFL which costs about $0.05 per day to run. Even compared to a CFL, the CREE bulb pays for itself in 220 days. If the duty factor of the lamp is lower than 100%, the payback period is correspondingly longer. The 60 watt CREE bulb draws 9.5 watts and is actually a bit brighter in directions perpendicular to the axis of the bulb and draws about 3/4 as much power as a 60 watt CFL.

The decision to use a CFL or LED in place of an incandescent is clear. If you can do it, then do it. The trade between a CFL or and LED is less clear and really depends on how much time that particular lamp is used. If the lamp is on several hours a day, every day, then it ought to use an LED. Lamps that are on only intermittently are more cost effectively served by a CFL.

Regular fluorescent lighting (4 or 8 foot bulbs) is good to go the way that it is. They are already as efficient as CFLs. There are a couple of types of florescent lamp, T8 and T12. Each type has to be used in a fixture designed for that kind of bulb although the pin outs are the same. If the lamp and fixture mismatch, there may be starting problems. The T8 is somewhat more efficient than the T12 which is not common anymore anyway.

A CFL is a fluorescent lamp configured to work in a standard bulb socket.

This leads to Step 4. Only the most dedicated energy hunters will go this far, but it is amazing what you can find. Read on to try this.

Measurement and Estimation Methods

To really know what this electrical stuff is costing you, you have to either estimate it's actual consumption or measure it. Estimation works well in many cases and is a lot easier than measurement.

To measure the energy consumption of any given load, you need some sort of instrument to do it. It doesn't have to be elaborate or hard to use. A good instrument is called a Kill-A-Watt. These can be found at electronics retailers or at for between $15 and $20. They are well worth the investment and are really easy to use. The least significant digit for power is 1 watt. This is good enough for the work that you will put it to. If something draws less than a watt, you can probably ignore it. There is also a slightly more expensive version that does some of the easy arithmetic for you. Go ahead and order one. By the time it arrives, you will be done with the the estimation work.


Many loads can be characterized accurately enough by estimation. Further, many of them are wired to your house and don't have a plug or run at 240 volts so you can't use the Kill-A-Watt to measure them. For example, a 100 light bulb draws almost exactly 100 watts. If the bulb is really old, it might draw a little less as it dims with age. There is no need to measure it. The CFL equivalent draws 23 watts, or 23%. 60 watt bulbs are similar and the ratio is the same. A 4' florescent bulb typically draws 40 watts, the rating is written on the bulb, just look at them to get the rating. Many appliances have a sticker on them somewhere that specifies the power consumption. These ratings are usually close enough.

Just walk around your house and count lamps and types. Then estimate how many hours per day that they run. Multiply the watts and hours per day and you get watt-hours. Divide by 1000 and you have kWh. Add the kWh up for everything and you have your total house consumption for the estimated items.


Loads that do not run all the time, such as air conditioners, refrigerators, washers, dryers, dishwashers, furnace blowers, microwave ovens, toaster ovens, griddles and such can draw a lot of power, but they don't run at a specified duty factor. You have to measure these and it takes some time to get an accurate measurement.

A washer or dryer can be characterized by the load. Make a measurement of the energy consumed by one load and then multiply that by the number of loads run per day or week or whatever time period you want to consider. Car charging works the same way, measure it once for a full charge and you have the power for each day that it gets a full charge.

Wall warts are a special case. These are the power adaptors that come with many small electric gadgets such as disk drives, routers, cell phone chargers, and such. Their power rating written on the label is the loaded rating which may or may not tell you anything. Many are generic and have a rating of 12 volts at 1 amp, for example, but are actually used at a half amp. You have to measure these. Most of them also draw a little power when they aren't even being used but are still plugged in. These should be measured without the load attached too. Most will draw only a watt or maybe 2 when unloaded which isn't a lot, but you might have 50 of the damn things plugged in. They can add up.

Another way to tell if a particular wall wart is a serious parasite without measuring it is by evaluating it's temperature by feel. If you can feel that it is warm when it isn't supposed to be doing anything, then it is drawing too much power. A 1 watt parasitic load will typically not heat a wall wart enough to make it feel warm. One that is perceptibly warm is probably drawing 3 watts or more.

A device that draws 1 watt of power and runs 24/7 will cost you about $2.20 a year assuming a $0.25/kWh energy rate. To estimate the cost of any other device once you know the power rating on the device, just multiply it out. A 100 watt bulb burning all the time costs $220/year to operate although you'll got through 4 or more of them in a year.

To make measurements with a Kill-A-Watt, just plug it in. It should read between 110 and 120 volts at the default setting. Then press the "Watt" button, the reading should go to zero with nothing plugged into the Kill-A-Watt. Then plug in a load and read off the instantaneous power. If you know the duty factor of that load, say 100% for something that runs all the time, you can get the kWh by dividing by 1000 and then multiplying by the number of hours in the period that you want to use. If you are recording in kWh/day, multiply by 24. If you want kWh/average month, multiply by 722 (4.3 weeks) or 8760 for kWh/year.

For loads that have an unknown duty factor, such as a refrigerator, plug the load in to the Kill-A-Watt and let it sit for a week. At the end of the week, press the kWh button and write down the total kWh consumed. Then press the same button (the pink one) again to get the number of hours that the test ran. Divide the kWh by the hours and this gives you kWh/hour. Then multiply by the period that you want to use.

If your refrigerator draws more than 600 kWh/year, consider getting a new one. A new Energy Star 18 cubic foot refrigerator will consume a rated 383 kWh/year but can consume as little as 250 kWh/year. Larger ones will consume more. Side by side and freezer below models are less efficient that freezer above models. Smaller units tend to draw less power than larger units. The cheapest ones to buy typically cost more to run. Buy the smallest one that you can use and then in that size range, pick the one with the lowest usage per year on the mandated energy label that the store MUST post with the unit. If you don't see the yellow label at all, don't buy that one.

Computers and game consoles also change their energy consumption quite often. A powerful desktop computer can draw more power as a refrigerator if you leave it running all the time. Set the things to sleep soon after you stop using them. A typical Macintosh desktop computer will draw 50 to 150 watts depending on what it is doing, but only 1 or 2 watts when sleeping. Laptop computers draw less, but can still pull 40+ watts when working hard with a bright screen and up to 85 watts while charging the battery too.

Don't forget your cable/DSL/FiOS modem and router. These typically run 24/7 and can draw several watts apiece. An Apple Airport Extreme draws about 8 watts, a Time Capsule draws more. A modem might draw 5 watts.

Printers can also draw a lot of power. Some HP printers draw 5 or 6 watts even when turned off. A laser printer can draw 1000+ watts while printing.

Those that live in all electric homes will have a huge load to deal with. Electric heat can consume several kilowatts to heat a house. Air conditioning is another huge draw. Wall or window units can be measured with a Kill-A-Watt. Whole house units will have to be estimated based on their ratings and a guess of duty factor. I have neither electric heat or any kind of A/C so I have not had to deal with determining their actual consumption. However, my forced air furnaces do draw considerable power to run the blowers. This can be measured.

Test Data

I have been in the process of measuring and estimating loads for everything that I can measure. I didn't account for the electrical kitchen appliances (not including the refrigerator) and I was missing about 3 kWh hour of consumption. Once the kitchen appliances were added in, even with guessed duty factors, 2 kWh of that was accounted for, but maybe not accurately. I added a line item to cover the aggregate error based on the average of my daily consumption as reported on my power bill for the last year. This is the state of the current measurements and estimates.

spreadsheet of power consumption

I'll be refining this Mac OS Numbers spreadsheet as time goes on.

This page has been accessed hit counter times since 20 Jan 13

© 2013 George Schreyer
Created 20 Jan 13
Last Updated April 14, 2013