• If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!



Page history last edited by Soren 7 years, 6 months ago




I founded energyconsumption.org and the EC Users' wiki. I own several watt meters and have plugged them into many devices. I haven't written everything down, but these sites are helping me share my finds and encouraging others to do likewise. For links, you can follow me on Twitter as @EnergySoren.


2 October 2016

Picking up where we left off (as I prepare to pay three months of PG&E):

1. After switching the remaining incandescent can lights to LED, I discovered that one bulb would never turn off. The "smart" push-button three-way switch setup evidently drew enough idle current through the LED bulbs to keep one (of three) dimly lit. So I switched to a pair of old-fashioned (mechanical) three-way switches and that bit of phantom load went away. We have three similar switches on the lower stairs (a "four way" circuit). As none of the associated bulbs is glowing (for whatever reason), I have yet to install the mechanical replacements which are now somewhere buried in the garage.


2. After WeMo's software improved, I replaced the mechanical timer + Christmas tree remote attached to the recirculating pump with a WeMo Insight Switch. It is much more convenient on two fronts: a) we use our iPhones/iPads to turn it on/off b) the WeMo software upgrade added an "auto off after N minutes" rule. So we now only interact with the system once, to turn on the pump. It then it turns itself off a few minutes later. It works so well that I gave a (non-Insight) WeMo to my neighbor for his birthday. The WeMo (like the xmas remote) has a small phantom load, but it's worth it to conveniently tame the recirculating pump and have hot water available without first running cold water down the drain.


The only problem I've had is when the pump doesn't run for the full four minutes. That happens if there is warm water in the pipes which recirculates to the pump (which has a thermo-sensor cut-off I have dialed all the way down to what I imagine is "lukewarm"). The trouble is that the WeMo also tries to be smart: it only turns off after it detects that the device under its control has been on for the programmed period of time. The WeMo is programmed to cut off the pump after four minutes of running, which is the time it takes to get hot water to the furthest faucet. But if the pipes already contain warm water, it only takes about two minutes of pumping to bring lukewarm water to the pump and for it to shut itself off. When the pipe near the pump cools down, it runs for another two minutes to bring more lukewarm water and the WeMo's rule never triggers. The net result of the current setup is that WeMo never automatically turns off automatically if pump turns itself off first. One solution might be to turn the thermo-sensor up in an attempt to ensure that the pump always runs for four minutes even if there is lukewarm water in the pipes. Of course that might also just cause it to cycle more quickly. Currently, we have a semi-manual workaround: I've configured the WeMo to notify us any time the pump turns on (also quite handy so one parent knows if the other is preparing for a little one's bath :). When you find a few of those notifications over a few hours, you know a manual switching off is in order.


3. I installed a efergy e2 and stared at it on and off for a year until its receiver batteries died. I never downloaded any data and I only just now downloaded the software (which raised multiple security flags in modern OS X). I still have hope for this tool, but the main thing I used it for was a depressing circuit by circuit analysis of the house's phantom load ... virtually every one of our multi-dozen modern circuits is leaking a few watts. Almost as frustrating, the phantom loads from various circuits never add up, certainly due to out of phase (non-1.0 power factor) loads that the e2 not equipped to accurately decipher.


4. I dressed up as "Energy Man" for Halloween, including getting the energy kit from the San Jose Library so I'd have a temperature "gun" (infrared thermometer) for my tool belt. The kit's documentation was impressive and stepped me through measuring water heater temperature, looking for insulation gaps with the "gun," and installing a new sink aerator which was free with the kit. It included window-sealing foam (which we didn't need) and a few other goodies. Last Christmas, Xinh got me an energy "gun" (the one with laser pointer :) of my own!


5. Probably the most useful thing I've done in the past year or so is to create a cloud-sync'd "Energy Notes" calendar that I can easily access on my phone. Optimizing energy requires a sense of time, and being able to quickly note anomalies lets me go back later to correlate observed aberrations from the norm with energy consumption downloaded from the meter. For example, when I see a small spike in the daily consumption on PG&E's web site which manifests as a higher than normal overnight load, I can now easily confirm that it was due to having left the porch lights. This lets me estimate the electrical costs of laundry, baking, the recirculation pump cycling, and running the (gas) dryer. The slight downside is that rather than try to do reasonably-accuaret live measurements, I often just note what we're doing and hope to sort out the data from the aggregate after the fact.


6. I've labeled the circuit breakers so that I can turn off most of the house when we go on vacation. I've recently spied some interesting behaviors, including the difference between leaving just the fridge turned on vs. fridge+garage (which includes the water heater and a big phantom load in the door opener) and what I'm pretty sure is the fridge defrosting itself every two days. I'm looking forward to posting some graphs, perhaps on a new Awareness page.



Current Summary and Future Direction

We consume 120kWh-140kWh/mo depending on season (and vacations), an average of ~175W. This is slightly below the consumption of our "efficient" neighbors. We also consume ~100 therms of gas each year, roughly half in the cooler months. This is significantly above our "efficient" neighbors, especially in the cooler months.


Our nighttime idle load remains ~120W (fridge + base phantom + required porch/garage lights). Daytime isn't much better (~95W if no one is home). Per the below, the fridge is at ~35W, which means we still have a 60W problem I haven't done much about in nearly three years. Interestingly, those 60W are a full third of our total energy consumption.


While 300 days of accumulation seems to have overflowed the kWh total on the Watts Up Pro attached to the fridge (:P), my memory from multiple observations is that it was averaging ~26 kWh/month for much of the year. Seasonal variation is much less than in the old house (especially compared to before marriage :) thanks to insulation, an east-facing unit, and active window management to chill the house each night during the warmer months. 26 kWh/mo comes out to ~36W, which is only marginally better than the old fridge (~40W?) despite being nearly 20 years newer and the lowest Energy Guide number I could find for a bottom-freezer model (400 kWh/year ~ 60W average in the relatively harsh Energy Guide test). The old unit was a top-freezer and those now sport Energy Guide numbers under 300kWh/year (~47W) average.


To optimize my fridge, I have a crazy plan to turn some solid foam insulation my uncle sent me (wrapped around excellent audio speakers :) into magnetic panels that can add exterior insulation to any fridge. Our fridge has some air space between the cabinetry above and around it, but it will be interesting to see how impeding that air flow vs. adding more insulation affects energy consumption. Hopefully the insulation we won't throw off the fridge's internal fridge/freezer balance more than the controls can overcome. Note: high-efficiency top-freezer units no longer spit hot air out the back, opting to emit it out one side of the front on the floor. :P


After playing with a thermal camera at Josue's house, future phantom energy battle plans include a FLIR ONE thermal imager for iPhone, which I hope to use to spy warm electronics (like the smart switches) hiding in plain sight.


On the light bulb front, I've asked Belinda at SuperGreen Solutions to order me some LED replacements for the (now burned out) T12 fluorescents under the kitchen cabinets and for the 40W incandescent small-base bulbs that shine from the microwave onto the cooking range.


Our gas consumption is also fairly high compared to "efficient homes" on the PG&E web site. Plans for addressing that include a furnace tune-up (I've noticed it "short cycling" :P) and someday a heat pump water heater. Need to keep an eye on the garage temperature this winter. Pumping heat out of the garage and into the hot water tank will be great in the summer.

25 Jan 2014

The days of paying PG&E every six months are over. With a 100W+ base load (including the new fridge), our bills are running about $100/month which means PG&E gets grumpy a lot sooner if you don't pay your bill. :P


There are lots of inefficiencies, from that base load to many incandescent lights. There are also appliances we never had in the old place: dishwasher, laundry machines, etc. The first control system I installed was for the recirculating pump on the water heater. A keychain fob controls a plug-in switch intended for Christmas tree lights. I've also been swapping in BR30 LED bulbs (TCP brand from the local hardware store) as the incandescent bulbs in the can fixtures burn out. Hint: if the bulb changes color (to orange) when you dim it, it's a replacement opportunity.


I will try to document what I find once I have access to live readings. I haven't yet asked about access to the (smart) electric meters, but they are far enough away that I plan to order an efergy e2 whole house meter for easy household energy visibility & optimization.

The Single Family House

By choosing efficient equipment and controlling it aggressively, our energy costs used to look like this:

In short, January 2013 was the first time in 8 years that we owed more than $10 in one month for electricity. Our heating is all gas and that does spike in the cooler months. We've eliminated much of the warm-season gas consumption by turning off the furnace's pilot light and taking showers exclusively from the water heater's pilot (this works just fine with a high-quality 1.5 gpm shower head).


After moving out of this house and zeroing out the bill, I learned that not only does it cost ~$0.15/day to be connected to the electric grid, but it costs ~$.10/day to have a gas connection.


Notes: no solar; see amusing "similar homes" comparison graphs at the very bottom of this page.

24 August 2013

 I owe PG&E more than $100, time to pay the piper. Here's the latest graphs:

Overall, the numbers are similar to last year, though a bit more varied. We were away from home in December and March, which left us home for the entirety of the coldest month. The space heaters and heat lamps added up in January while in March it was still cool enough for the fridge not to have to do much work (so electricity use plunged). Yes, January 2013 was the first month since moving to this house that we owed more than $10 for electricity.


Today, I happened to examine an Energy Star refrigerator (a fairly cheap/generic one from a few years ago) in a confined space. There was no heat build-up behind or around the fridge! This appears to be due to the fact that room air is drawn in at floor level via the left side of the front grill while warmer air (containing heat removed from the fridge) is expelled out the right side of the same grill (both at angles so as to avoid a loop). My 1996 fridge (and many others) seem to draw air in from the floor but then expel it out the back. Modern kitchen designs often enclose refrigerators making the older design part of a wasteful setup where warmth around the fridge makes it work harder. Previously, the only fridges I'd seen expelling warmth out the front were high-end ones designed to be built-in.


Other changes (in no particular order ;)

  • In late October, we'll be moving to a newer (2008), bigger place about a mile away.
  • In December, we're expecting a child! [an opportunity to continue the low-energy lifestyle :]
  • We're experimenting with just X-10 for the Internet: substituting self-control for the timer.


15 January 2013

Belkin, of all companies, is the first company I have found to produce the trivial three-prong version of the classic cutoff switch! I'll need to find someplace local that carries it (bike > UPS > USPS?).


Also, I poked around the newer parts of pge.com/myenergy recently and let it run my solar demand calculations. Even with a slight over-estimation of 75 kWh/mo (see graphs below), http://pge.cleanpowerestimator.com/ says it only takes ~500W (AC) of solar panel to supply us our current ~900 kWh/year. The cost estimate of ~$2500 after rebates is probably a bit low, but perhaps we can keep it under $5k extra if we do it as part of getting a new roof.


Of course if we add a pump for solar hot water and some sort of heat pump, we'll be moving a lot of heating from gas to (fairly efficient) electric and our electricity consumption is likely to go up. That's fine as long as we keep moving towards my goal of "net zero carbon:" use gas where it makes sense medium-term, but put enough excess electricity into the grid to offset a similar amount of gas used elsewhere.


As I ponder the carbon equivalence of local solar vs. remote gas (on a grid with other types of generation), I will post the size of the solar array necessary to offset our current gas usage under a variety of scenarios.


27 November 2012

Alfredo from Elements Zero Energy is coming Thursday morning to do an energy audit! This is the first step in the Energy Upgrade California program, but more importantly it is the first step in creating a much more comfortable and efficient house. Hopefully Alfredo will continue to be patient as I ask about ductless mini-splits, solar thermal + electric heat pump(?) space heating, combining heat appliances, metal roofs, etc. :)


The initial audit report, is good. I was confused about the air flow recommendations, but Alfredo explained that while they test at 50 Pascals, the final "ACH50" number is calculated using expected wind speeds in San Jose, height of my house, etc. If our house was in a place with howling winds, it might achieve 15 air changes per hour ("ACH"), but since it is not, it's only about twice as leaky as a minimally non-mechinically-ventilated house.


September 2012

I'm not sure if it's the amount I have due or the amount of time since my last payment, but I'm trying to avoid delinquency notices by paying PG&E about before it's been six months. I figure it is better to post new graphs now even though there is some overlap with the previous set:

Our new Energy Star cordless phone + answering machine (circa February?) dropped our minimum below 10W which oddly causes PG&E's smart meter reads (and charge for!) 0.000kWh. Since Feb 2012, we have been using a few less kWh each month. We'll see what happens next: I make my own graphs (overlaying all of this information and/or my day-to-day) or insulate the fridge.

Peak monthly amounts owed, by energy type: $20 for gas in December, $9 for electricity in November.


November 2011

Time to pay the PG&E bill again.  As the following (new, pretty PG&E + Opower) graph of my gas usage shows, average outdoor temperatures near 70 allowed us not only to turn off the (largish) pilot light on the furnace, but also to stop using the burner on our gas water heater.  That's right, we took hot showers using only warmer incoming water, ambient gains (despite the water heater blanket?), and heat from the water heater's pilot light. For three months, we averaged only one therm a week!



Since firing up the heaters again in October and November, the daily graph (not shown) makes it clear we now burn through a therm day or two.  It will be interesting to see what this graph looks after we've managed to get some sealing and insulation done (I finally have contractors lined up to visit for estimates :).


For the curious, our electrical usage remains steady between 60-70 kWh/mo, with winter space heaters eating into the savings realized by the fridge (which varies from 25-35 kWh/mo depending on ambient temperature):

[for those who measure their bills in monetary instead of energy units, these add up to < $20/mo]



April 2011

After PG&E mailed me a 15-day warning notice, I logged in to pay my bill.  It was a dime short of $150 -- and covered gas and electricity (and ClimateSmart charges) for 7 months.  That's still more than $20/mo and our house still isn't properly sealed or insulated. :P


The big news from the pge.com login is that our SmartMeterâ„¢ is finally hooked up!  In fact, it was hooked up in early February, but I missed out on that special gift because I was too lazy to pay our bill regularly.


The web site says our effective daytime idle on medium-warm days is 40W.  The floor as read by the meter itself is has lately been 11W (--sleepingPPCMachines).  The fridge should be the only dynamic load when we're not around.  And that 30W-40W depending on temperature matches the 20-30 kWh/mo I've measured for the past several years.  To double-check a single day, I suppose I could record the meter's kWh count and zero the kWh counter on the fridge before leaving for work.  In the evening, I'd check compare the fridge's consumption with the day's delta on the meter.  11W * 8 hours + fridgeEnergy should come out to ~40W (current web site value) * 8 hours on the meter.


January 2011

I've had several questions about my "Internet Timer."  The timer is primarily a social tool for our household's ability to focus, but it started as an energy-saving project and it highlights the energy savings opportunities of sleep and away times.  Thanks to Keith for the prod to post the writeup.


We have two shutoff mechanisms (both mentioned on Projects).


One is a simple X-10 switch which we use to remotely turn the Internet access equipment on and off.  This adds overhead to having Internet access and forces us to be conscious of when we're using it vs. not.  In our practice, this mostly serves to save energy on networking equipment during the day when we're not home.


The other mechanism is a 7-day programmable timer.  It limits when the X-10 switch is capable of being turned on.  It generally allows Internet access during the day but it (currently) cuts off Internet access from 10:30pm to 5:30am.  It also limits Internet access during the day on Saturdays and in the mornings to ensure I take a shower even if I get up early and start working online.  In our practice, the timer primarily keeps us the off Internet at night (while also saving energy if we forget to use the X-10 switch to turn it off).


After I replaced the ~12W of DSL modem and AirPort base station with a ~6W combined unit from NetGear, the control equipment is starting to consume a significant fraction of the power being switched.  The original inspiration for the timer was to reduce the leakage on the X-10 receiver, but the timer is warmer than the X-10 receiver.  In the end, keeping the Internet equipment off when we're not using it is a win, but it's important to be aware of the power costs of control equipment.  For example, putting X-10 in every outlet would let you control everything, but you'd also have a high constant power drain due to the X-10 equipment's phantom draw.


Early 2010

As noted below, we now have a Smart Meter with digital readout.  Here is my initial guess at where the 13W it reads at "house idle" are going:

- 2W to the X-10 controller used to switch the DSL+802.11 Internet device on and off (only saves 6W? :P)

- 2W to the "Internet Timer" that cuts power to the X-10 switch / limits when we use the Internet

- 1W to the doorbell transformer

- 1W to the motion/daylight sensor on the garage light

- 3W to the answering machine/cordless phone combination (older Energy Star model)

- 4W to two sleeping ppc laptops?

- I forgot the watt meter on the fridge ... it will be fun to figure out the real values (and amusing when they don't add up because of the weird power factors :).


9 March 2010

Good news ... I may end up with efficient ventilation and efficient heating after all!


Here's what happened this morning.

The friendly folks at the nearby Built-In Center have all their stuff set up to be plugged in.  So it was easy for me to plug my watt meter (in this case a Kill-a-Watt EZ with the required "liberator" for easier reading) into the heat/light/fan unit I had reconciled myself to buying.  Unfortunately, the NuTone 9427 (supposedly identical to the Broan 164 mentioned yesterday) had an exhaust fan that consumed 50W with 123VA of "slosh" (i.e. not at all correctd for power factor). Even the tiny "cooling fan" for the heat lamps drew 17W. :P


I explained how sad I was that I couldn't get a heat lamp + fan with an Energy Star motor.  I asked whether the one that came without a ventilator could be hooked up to an external fan.  That one didn't have a vent, but they were happy to sell me the vented fixture without its "crummy old motor" (saleman's words :). The guy didn't think they had any Energy Star "external/inline" fans but when I mentioned FanTech, he showed me the display and noticed the Energy Star logo.  He immediately started asking his partner which model fan had compatible-sized ductwork.


Off went the salesman to get the boxes from the back room.  Even while looking up prices for us, his partner was more than happy to answer my questions about controls.  Out came the boxes, out came the various parts, and off came the "crummy old motor."  I only paid for what I took: they'll order another external fan part (and bracket) to put that kit back together and they'll sell the motor I didn't want as a spare part.  "Everything's here' to sell," the salesman said, "so we're happy to give you what you want."  I was more than happy to pay the contractor price (since my contractor had sent me over :).


The electricians may charge extra to install it or I may have to do it myself ... but either way, I'm hoping this all works out.

8 March 2010

(need to move some of this data to a new Ventilation page; the rest to Heating)

The electricians are now waiting on lighting circuit(s?) for me to procure a heat/light/fan for the bathroom. I would like to get one that is Energy Star rated but my desire for a heat lamp (250W vs. 1000-1500W heater) has been complicating the effort.  Energy Star's ratings demand 1.4 cfm/W for smaller fans and 2.8cfm/W for larger fans.  They exclude heat lamp units (?) -- perhaps because installing one incorrectly could leave you getting all your light from a very inefficient bulb?  Or maybe they're not such a good idea.


Fan efficiency/power data thus far:


  • 55W for my existing uncleaned, un-oiled fan (no idea it's CFM rating; guessing ~70)
  • what appears to be 50W extra for the fan portion of the Broan 162 (70 cfm => 1.4 cfm/W)
  • what appears to be 45W extra for the fan portion of the Broan 164 (70 cfm => 1.55 cfm/W); it also delivers 85% of its throughput all the way up to .4 "in-wg" of static pressure
  • The Energy Star rated Fantech 100 delivers 100 cfm for only 19W (5.26 cfm/W)!
  • Panasonic has a variety of fans that deliver 3.5-4.5 cfm/W (FV-08VQ3 looks about right)
  • Broan's Energy Star fans provide no specs on their fan energy consumption :P 


Assumptions / Calculations


  • the heat function would operate 20 minutes / day, 4 months of the year (40 hrs/yr; 10 hrs/cool mo)
  • the ventilation function would average one hour / day, 12 months a year (365 hrs/yr)





  1. forced-air heat: 40 hrs * 1500W = 60 kWh/year or 15kWh/mo during cooler months
  2. radiant heat: 40 hrs * 250W = 10 kWh/yr or 2.5 kWh/cool mo 
  3. double-radiant heat: 40 hrs * 500W = 20 kWh/yr 
  4. inefficient ventilation: 365 hrs *  45W = 16.5 kWh/yr or 1.4 kWh/mo
  5. efficient ventilation: 365 hrs * 25W =  9.1 kWh/yr or .76 kWh/mo
  6. #1 + #4 =  ~70 kWh/yr
  7. #2 + #3 = < ~30 kWh/yr 




  • Inefficient ventilation wastes less energy than inefficient heat :P
  • But why can't I have both? 




Should we go for a wall-mounted fan in the kitchen? 


  • Panasonic WhisperWall FV-08WQ1 delivers 70 cfm for 




Have you thought about an energy recovery ventilator?  This recovers the heat from the exhaust.  The problem being, the CA climate doesn't make them particularly useful, especially moreso when you already use as little energy as you do, and when you have relatively wide comfort standards.



I was pleased to find several ERV's in the catalogs and web pages for various vents and such.  I have yet to hear of someone resealing an older house so well that it would require an ERV.  But sucking air out of the bathroom will suck it in somewhere else and the outside morning air is sometimes (but not usually!) the wrong temperature.


Could you post something on the Ventilation page about how ERVs work and what climate is best and why?  In Northern California's Mediterranean climate, it is damp and cool at the same time.  Does the ERV like moist air or dry?  Obviously moist air takes more energy to heat.  Answer there, not here. :)


Because we've had no timer on the fan, we've generally just left the window open after showering.  During the heating season, we'll probably leave the window closed and fan off until afterwards.  The rest of the year, we'll just have the window open (and the dry air will do most of the hard work anyway :P).



6 March 2010

I'm pretty excited to have a smart meter on the side of my house.  I haven't even looked to see what sort of feedback PG&E is offering online, but it cycles through 6 readouts every 18 seconds and today I used it to measure the consumption of the fan in the bathroom.  Just after the fridge turned off, I took the bulb out (there's a single switch for fan+bulb; that will be fixed soon! :) and then went outside to establish baseline (18W w/Internet on!) followed by having the fan running (~73W).  That's 55W for my current bathroom fan.  Definitely worth having on its own switch/timer and probably worth beating when I replace it with what I hope will be a fan/heat lamp/light unit this week.


Why this week?  Because I'm finally be getting the house rewired.  Mostly, it paves the way to getting wall and under-house insulation which I have needed for years.  It is also an opportunity to install a whole house meter, ideally one with multiple channels so I can monitor several different areas of the home independently.  I feel like some of the meters (Brultech? the new TED) offer some amount of multi-channel functionality, but I don't have details yet and need to place my order soon!  As a bonus, the owner of Voltz Electric (yes, with a 'z' :) even seems interested in energy awareness and conservation technology as a way to grow his business.  I need to put together an Electricans page to give him and others some ideas about how to compete as the most effective at reducing energy costs!


27 December 2009

I noticed that we're paying PG&E 11.531 cents/kWh which is, I believe, slightly higher than before (this is before the nominal -- 0.254 cents/kWh -- ClimateSmart charges but includes generation, distribution, taxes, etc).  I'd noticed the top-tier rates going up (beyond 40 cents/kWh), but not so much the rates at the baseline.  This is even as my church (a business customer) got a letter saying it would receive a one-time credit on its electric bill because gas has gotten so cheap (and indeed, my same December bill shows gas at a seasonally-low $1/therm).


Last weekend (11-12 August, 2007) I finished my microwave hack (mentioned on Projects). Today (20 August 2007) I am powering off a laptop that has been asleep in my house for almost two years:


ansel.local:/Users/slimer=> uptime

17:20 up 506 days, 4:55, 4 users, load averages: 0.40 0.37 0.48

I may bring this machine back up to the sleep state if I have need for it, but in the meantime, it's finally getting shut off.




For those who want to see them, 28 Cecil's "similar homes comparison" graphs from PG&E/OPower:

Each highlights our highest month's consumption for 2013. I am proud to be dragging the "efficient homes" category down a bit. We still have quite a bit to do on winter gas usage. These have been like this for years now. I don't know how similar the similar homes are (ours is quite small).

Comments (0)

You don't have permission to comment on this page.