Ideas around home electrification & decarbonation

Hi folks, here’s some general thoughts on how Sense could better support those of us who are converting their home to be fully electric, a.k.a. de carbonizing their home. Some of this has been replacing gas consumers with electric, some has been replacing older electric appliances with modern, more efficient ones. I’m just coming to the end of my process, which has looked like this:

Electric baseboard heaters -> ductless heat pump mini-split system
Natural gas water heater tank -> electric on-demand tankless water heater
Old clothing washer > High Efficiency clothing washer
3 x gas vehicles -> 3 x electric vehicles + 240V level 2 EVSE & 120V Level 1 EVSE
Natural gas fireplace insert -> discontinued use, might replace with electric insert in the future
Gas range -> Electric induction range (using existing 240V 40A circuit)

Also installed grid-tied solar PV, 9.22 kW total system size, inverter maxes out at 7.5 kW

All of the above happened within a ~2.5 year period, most of it in the past year, and I’ve had Sense for nearly 1 year at this point. My only regret is that I should have gone for a heat pump water heater instead of on-demand, but oh well. By the way, all of that cost less than a single Tesla!

So now my house is fully decarbonized, I’m cancelling my gas service today! Of course this means that my electrical load has increased significantly.

Sense sends notifications about our increasing electrical consumption, which certainly is a good thing as it could indicate an appliance going bad and using more electricity than it should. But in my case, the increased load is something to be celebrated, as it means I’m weaning off gas. That leads me to my first suggestion:

What if Sense allowed you to specify if you have a fully electric house, versus a mixed fuel house? That would allow Sense to customize the ‘compare’ section, so that fully electric houses could be compared to other fully electric houses, instead of comparing fully electric houses to mixed fuel houses (and discounting the role of fossil fuels).

Here’s my more ambitious suggestion:

What if Sense allowed a user to input details from their fossil fuel consumption (natural gas, propane, heating oil, gasoline, diesel, etc.) and helped the user make projections about switching this consumption over to electricity?

There are tools out there to help you make these projections about specific items, such as switching from gas to electric vehicles, but I haven’t seen anything that helps you convert your entire lifestyle off fossil fuels. It would certainly take some work, but I think Sense is in a good position to help someone evaluate this proposition.

It could look something like this:
User has a natural gas water heater tank, and is thinking of switching to an electric heat pump water heater. User can input details from their natural gas bill into Sense, and specify the gas appliances that are currently in use (say, water heater, gas range, and gas fireplace in my example). Sense can make some predictions about how much energy the water heater is using (perhaps just based on regional averages, since it would have no way of splitting up the gas bill into consumption by different appliances, or perhaps the user could input info from the EnergyStar sticker on the appliance). So Sense helps the user estimate their current cost of heating water, then project out their new cost of heating water by switching to electric, perhaps even calculating the return on investment. Repeat this same process for all gas appliances and consumption, to come up with a total cost/savings of electrification.

What do you think?

I hear you, and therein lies the rub!

I think a lot of people are thinking about this a lot because it’s a super difficult calculus. Read: hard.

A few thoughts (which I’ve outlined elsewhere to a certain extent), channeling the devil’s advocate:

  • Most big ticket appliances and especially vehicles have a ton of embodied energy so are best kept in service or up-cycled vs recycled. Of course, mileage will vary, so to speak (a lot!)

  • Operating a single-family home beyond it’s solar capacity is something that requires a LOT of thought. Spend money to install a heat pump or spend money to add solar? Add solar now or when perovskite takes over the world?

  • Given the above, do the EVs run on the sun (alone) or do they need the grid? Can the usage patterns be modified to improve efficiency?

  • And don’t even get me started on batteries/PowerWalls and so on!

Short answer: Sense is doing ML on the electrical signal and the decision making you are talking about feels more like ML^2 to me. Did I mention: hard?

1 Like

Thanks for the thoughtful reply! I get that this is hard, and that’s why I think it would be enough for Sense to just offer prediction/projection tools based on averages, instead of actually using machine learning to attack the problem. It is the same type of thing someone could do in Excel, Sense would just provide a more user-friendly interface.

You brought up some great points, let’s dig into them!

It is always best to start with your priorities. My first priority is to reduce my carbon footprint to the point where my lifestyle is actually carbon negative. I’m attempting to accomplish this by retrofitting my house (built in 1967) to be as close to net zero as possible, and by maintaining about 1/3 of an acre of forest on my property. I also spearhead community solar and materials reuse projects, as well as teaching sustainability topics. My second priority is to save money and reduce my cost of living as much as possible. I’m in my mid 30s now, and I’m hoping to retire in my mid 50s, so my goal is to get my cost of living to nearly zero by that point, if not sooner.

  1. It is true that anything manufactured has embodied energy/carbon associated with it. I’ll focus on cars, since that’s where I’ve put most of my research efforts. 10-20 years ago, it was definitely true that replacing a standard car that gets, say, 20 mpg with a hybrid that gets, say, 40 mpg probably wasn’t worth it. However, now that used EVs are affordable and available (not globally, but at least in many places) that is no longer true! Replacing a gas or diesel powered car with an EV is definitely the right environmental move, and the embodied carbon from manufacturing will be covered within the first 2-5 years of driving. This video dives into more detail. There’s also some manufacturing facilities that are trying to be net zero, such as the BMW i3 plant in Germany that is powered entirely by onsite renewables. Of course there are other sources of embodied carbon in paints, plastics, etc., but those are present in all vehicles, not just EVs.
    Another point on this, nobody switching to an EV just throws away their old car (unless it is truly at end of life), most people are going to sell the ICE car to someone else. So the ICE vehicle is still likely to be driven until the end of its useful life. The point of EVs, in my opinion, is to stem and eventually stop the manufacture of new ICE vehicles. I sold my 2008 Toyota RAV4 (24 MPG) for $7500, and bought a used 2015 Nissan Leaf (112 MPGe) for $9000, and started saving about $100/month in fuel costs. So financially, the Leaf pays for itself in about 15 months. And it is a much nicer car than the RAV4!
    ICE vehicles can also be converted to EV, but currently that is more of a hobbyist/niche thing, it isn’t yet economically feasible on a large scale. But certainly a converted EV has a much lower carbon footprint than a newly manufactured one, even if it is less efficient while driving. I’m starting a program with my local community college to do just this. A few of the major European manufacturers are starting to release retrofit EV conversion kits for their classic vehicles, which is great to see.

  2. I would argue that multi-dwelling units (a.k.a. middle housing) are the future for the vast majority of the world’s population, and that’s where most of this energy efficiency tech scales the best (for example, one heat pump system servicing multiple units). Most of us on this forum probably live in single family homes, though, otherwise why buy Sense? In my case, I do operate my house beyond my solar capacity for about 9 months of the year, so I’m definitely reliant on the grid. I’ve opted into my utilities’ carbon offset program, so the electricity I do import from the grid is carbon neutral (at least in theory). I would love to be self-sufficient year-round, but I’m not quite there yet, and I don’t think I have enough roof space to achieve that (at least until all my appliances are extremely efficient). Solar will always get cheaper and better over time, but it is already a good investment. My solar PV ROI will be between 10 and 14 years, better than the stock market! Considering the pace of climate change, I don’t believe there’s time to wait.

  3. EVs can run 100% on solar, in my case I drive two EVs on 100% onsite solar in the summer months. If I lived in a sunny climate, I could easily do it year round. I do also set my charging timers to not charge during peak hours, to reduce the likelihood that I’m charging from fossil fuel peaker plants in the winter. There are fancy EVSEs that do this automatically, and take a look at what Wattime is doing with their EVSE integration…they go beyond charging timers and actually monitor the current mix of energy in the grid and automatically charge your car only from the ‘green’ end of the energy mix. Currently a brand new tech, but hopefully it will filter out in the future. Vehicle to Grid (V2G) also holds a lot of promise for helping promote renewables by smoothing out the demand curve, thus reducing reliance on peaker plants. It also reduces the need for home batteries, thus reducing the amount of battery storage any home needs significantly.

  4. I’m interested to hear your thoughts on batteries! I’m particularly keen to play with Lithium Iron Phosphate batteries for stationary storage applications.

Thanks for the discussion!

Congrats on all your changes. You say you replaced 3 cars for less than buying a single Tesla? I know I paid more for my Tesla than any car before, but all cars are expensive these days. I am incredulous.

I too struggle with that comparison as if using energy was BAD and you have to move the needle to be good. The same goes for the power company charging us tiered rates, implying that the more you use the worse you are. You suggest tying in our natural gas bill, as it goes down we are saving energy that is otherwise hidden from Sense. We would also need to tie in our gasoline bill and maintenance stops for oil changes and brake pads that we would be replacing w/o using electric.

And does Sense give us kudos for the energy we generate? I haven’t figured out if that comparison with others includes the energy we generate or just that we use.

Here’s the rough financials of my electrification process (all in USD):

Heat pump mini split system - $8,500 (after installation & $800 incentive)
Electric on demand water heater - $2,330 (after installation, $2k went to plumbers and electrician)
HE Clothing Washer - $350 (bought used, self installed)
Discontinued use of gas fireplace - Free, but made possible by heat pump install
Electric induction range - $1040 (bought new, free delivery, self installed)
240V Level 2 EVSE - $450 (self installed, including running a new dedicated circuit)
Solar PV - $13,000 (received $3500 incentive, before that the cost was $16,500. I did most of the design work and bought all the parts, but had to have it professionally installed to get the incentive. This cost doesn’t take the US Federal tax credit into account since I’m a 1099 contractor, but I’m hoping to claim some of that over the next few years, which will bring the price down even more. If I were able to claim the full fed tax credit, the final price would be around $7k.)

My ICE vehicles were:
1995 GMC Suburban (sold for $1500)
2001 Toyota Prius (1st gen!, sold for $2000)
2008 Toyota RAV4 V6 (sold for $7500)
Avg. fuel cost was ~$130/month

My EVs are:
2015 Nissan Leaf (bought used, $9000 after $2500 incentive)
2011 Think City (bought used, $3800, selling this one soon)
2017 Chevy Bolt (bought used, $17,500 after incentive, just bought this one a few weeks ago)
Avg. fuel cost is ~$30/month

If you add all of that up (subtracting the cars that I sold), it comes out to $44,970. So perhaps I was too flippant in saying that it costs less than a Tesla, since you can now get a Model 3 for less than that, but it is still less than every other type of Tesla! And before buying the Bolt last month it was sub-$30k. As you can see, I’ve also saved money by doing a lot of DIY installs and design work.

The other component that is harder to quantify is the recurring cost of living, since that has changed dramatically over the past few years. Here’s a graph showing my family’s monthly cost of living since we bought our house in late January 2017 (note that this just shows energy/fuel consumed, it doesn’t account for food, water, phone/internet, and other household goods):

The Oct 2019 number is incomplete/inaccurate, but you can see in Sept 2019 I spent less than $20 thanks to solar credits earned over the summer! It has been less than a year since we installed solar, so I’ll have to wait for a while before I can calculate an actual monthly average now that the electrification process is complete. I’m guessing my average will be somewhere around $50/month year round.

Another factor is that my EVs haven’t fully replaced the old functionality of my ICE vehicles. We used the Suburban and RAV4 to move furniture to fill our house, and we’ve basically given up that ability now that we have only EVs. The once or twice per year that we need to move a piece of furniture, we would have to rent or borrow a truck. The RAV4 also used to be my roadtrip vehicle, and the Bolt now takes that space. Obviously an EV roadtrip is fairly different than an ICE vehicle roadtrip, so there is some bit of compromise there. An EV roadtrip will be less expensive, but will also require more time and move at a slower pace due to the charging stops.

1 Like

Very nice.

What rebate did you get that was because you hired a company to install? If they installed the panels that you bought and you designed the layout for, what does them drilling holes in your roof net you?

I thought your costs would be more than my Tesla. Why didn’t you get an electric RAV4? Ok, they would move furniture, but I guess would NOT be your road trip machine.

And don’t generalize about electric road trips like that. I get in the car and say where I am going and it plans the trip and mostly drives me there. It is almost priceless and it’s not really a penalty of time going electric. Used to be my best time was 4.3 hours getting from SF to LA, now it takes over 5 but it “feels” much less. And I am not waiting for the car to charge, it is waiting for me to nosh and pee. Charging is usually less than 15 minutes if I make two stops (as my old man bladder demands) or closer to 50 minutes if I stop once. At a place to eat nicer than McDonalds the car is waiting for me then too. 50 minutes because it takes longer to charge the battery all the way up once than charging it over half way up twice. [/TeslaTirade]

1 Like

I live in Oregon. There’s a non-profit here called The Energy Trust of Oregon (ETO). They get funding from local utilities and other sources, and dole it out to individuals and businesses through a variety of rebates, incentives, discounts on products, etc. They support large community infrastructure too. They are a fantastic organization:

The solar incentive that I got was for $3600 towards my rooftop PV, but it required that I use an installer from their list of approved installers, if I had done the install myself I wouldn’t qualify for the incentive (the incentive is actually paid to the installer directly). It took a bit, but I shopped around and found an installer that would do a ‘labor only’ install for $4300. Most installers rejected me because they make most of their profit from the design of a PV system, not from the installation, so mine was a low margin job. So my out-of-pocket expense for the professional install was $700, that seemed like a no-brainier to me, as I was planning to spend at least a few hundred on hired labor to help me with the install anyway. I think I got a great deal, though obviously this scenario isn’t very common, and did require a lot of time and effort on my end.

I truly love the electric RAV4 (the ‘Toyesla’ as I like to call it), but I decided against it because it is about the same price as the Bolt, and the Bolt has more than double the range. The safety and tech on the Bolt is also better, though I’m looking forward to seeing Toyota put out a new RAV4 EV in the near future (not just a compliance car like the previous ones).

I definitely hear what you’re saying about road tripping in a Tesla, but I think Tesla is in a class of its own in this regard. No other cars really come close at this point. So I guess what I meant earlier is that road tripping in any EV besides a Tesla has some compromises at this point. But that state of affairs is changing rapidly. I’m hoping to be able to afford a Tesla some day, but for now the Bolt gets me there for about half the price.

The other aspect of road tripping that changes is that certain long drives can actually be off-limits in an EV, due to a lack of charging infrastructure. I have family in Reno, NV, and to drive there in an ICE vehicle there is a faster route that takes 18 hrs round trip (Portland to Klamath Falls to Reno). The stretch between Klamath Falls and Reno is beyond the range of every EV on the market today except the Tesla Model 3 Long Range. Even with that car, the trip absolutely relies on the Klamath Falls supercharger being operational. To do the same road trip in my Bolt, I have to take I-5 all the way to Sacramento (so Portland to Sacramento to Reno). That drive takes nearly 23 hrs round trip, not including charging times. That’s an extra 5 hrs of driving time alone, with the added inefficiency of driving more total miles, and not accounting for the charging time. So it is a significant compromise, in my opinion. However, the environmental benefits of decarbonization and the money saved make it worth it for me. This is also a temporary problem, as I’m sure more Level 3 chargers will be added to the faster route at some point in the future.

Good info. You made me worry I had screwed myself by doing my own install. I know my Tesla still cannot get to Minot, ND near where the factory for my motorhome is, but they are building the Superchargers in North Dakota now, so soon. Back to topic:

I still need a way to decarbonize keeping the house warm at sunrise. I am burning gas in the mornings running the fireplace or the furnace. I am thinking about solar water heater panels that warm a tank in the crawlspace that would radiate during the night, but I think it would still be too cold in the AM when there isn’t any sun yet. And to even be slightly warm in the morning it would probably have to be too warm at night. Still, as it gets colder I’ll have to run the fireplace at night too, so having the tank of rainwater under the house that I can pump to the roof to warm might be useful.

Decarbonizing heating of both air and water is definitely a challenge, especially in colder climates. Here’s some approaches:

  1. Heat pumps, heat pumps, heat pumps. They are very efficient at heating air and water (the water ones are often called ‘hybrid water heaters’ as they are a hybrid between heat pumps and traditional electric resistance heaters). Even if you are powering them from the grid, they require less power to run than other heating systems, so it reduces strain on the grid (the peaks are lower), which reduces the need for fossil fuel peaker plants.

  2. Hydronic heating, as you mentioned. A well insulated water tank can keep water hot for days, so you could ‘charge’ the tank the previous day with solar, then use the water for heating in the early morning. A heat pump could be used for water heating here as well, just be aware that you generally can’t mix water used for space heating with potable water, so you would need two separate systems. I would suggest a system on a timer, so that the hot water only starts flowing through your radiators from 4a-6a, or something like that. No need to heat the house all night, just heat it a bit before you wake up.

  3. Batteries, not as efficient as the ‘thermal battery’ suggestion above, but still a good option. My mini split heat pump system consumes 1500W per head unit, and I have two of those (one for upstairs, one for downstairs). So 3 kW if I want to heat the whole house (~2300 sq ft), which takes about an hour, so we’ll call that 3 kWh. If you are handy with electronics and have plenty of spare time, you can build a DIY home energy battery system with a 3 kWh battery capacity for $1000-$1500. Or you could buy a prebuilt one for $5000-$10,000, which will probably have a larger capacity. Again, charge it up from solar during the day, then discharge it in the evening and in the early morning. Personally, I would rather use Vehicle to Grid (V2G) for this since I already have several massive batteries in my EVs, but currently this isn’t an option.

As for heating during the night, I’ve found electric blankets to be very economical. They consume between 100W and 200W, and generally cycle on and off, so they use very little energy overall. Use them for sleeping, sitting on the couch, etc. It will always be cheaper to heat your body directly than trying to heat all the air in a house. In my office, I also sometimes use a seat heater on my office chair (it is a 12V seat heater meant for cars, but I plug it in through a 120V AC to 12 V DC inverter). Electric mattress pads are also an option, though I prefer the electric blankets.

Another thing I’m considering for my house are infrared ceiling tiles. They are radiant infrared heaters, so they feel great (just like a fireplace or hydronic system), but they don’t emit any visible light like old school heat lamps. There are both hardwired 240V options as well as plug in 120V ones. Pretty easy to install, and I think the prices are reasonable:

I’m thinking about using the above for certain hard to heat areas of my house, small rooms that don’t get great airflow and therefore don’t get heated by my heat pump very effectively. They can also be put on timers or certain thermostats (the 240V ones are wired the same way as electric baseboard heaters, so they can use thermostats meant for baseboard heaters).

Finally, the easiest first step is to sign up for your power utilities’ carbon offset program, if they have one. That way you can use electric heating from the grid while you figure out better long-term solutions, and still be at least carbon neutral.


First, thank you for the incredibly thorough and constructive write-up. I definitely think there are ways for us to support what you describe going forward, but it’s definitely a small userbase — though, hopefully a userbase that will grow in time. I’ll pass all of this along to the Product team.

There’s a lot to unpack in your exhaustive breakdown. Here are a few somewhat random points:

  1. Without knowing much about these I would say “Definitely not!”. Logic says heat rises so any IR re-irradiating suface is no more efficient in raw terms than electric radiant. I would put effort (and money) into increasing thermal mass and insulation and where necessary generating heat at floor level. Passive solutions always win in the long term. Sunlight on a dark floor!

  2. Small “dead” spaces should be connected with controlled ventilation rather than dedicated heating/cooling. e.g. A multi-zone mini-split system is essentially an energy distribution system … the basement can serve to keep the upper floors cooler if the basement has a “zone”. If small unoccupied and out-of-zone spaces need to be kept cool/warm then they should be included in the overall airmass by leaving doors open or having active ventilation (room-to-room fans) rather building dedicated heating/cooling systems (imho) … this speaks to the complexity of balancing centralized/localized systems.

  3. Back to your initial solar & decarbonizing … the largest “loss”, if you can call it that, with any energy generating system, and particularly solar vs carbon, is the lost production during failure (this is where Sense can shine) and, more importantly, the lost production prior to activation. In general, solar procrastination is its greatest loss! What I mean here is if, let’s say, you spent the $20K you intend to spend on an ICE-to-EV conversion to get additional solar production, where does that put you in terms of overall carbon? The beauty of having solar is the potential constant generation over a significant lifecycle. This is a far-from-easy calculus which gets much more complicated when you include battery storage and V2G. Solar installs may last more than 20-30 years but how long do batteries last and what is the realworld carbon debt for recycling them (or not!).

  4. You are probably well aware of the Northern Euro approach to home heating … but check out the tanks!

  1. And of course, Kotatsu!

Thanks ixu, I appreciate your replies.

Regarding infrared heating tiles, they can actually be wall mounted and placed near the ground (similar to baseboard heaters), but the folks I’ve talked to told me they work best when ceiling mounted. The idea is that they simulate warmth from the sun, so that you FEEL warm even if the air temperature isn’t that high. It is a different approach to heating, creating the perception of warmth without actually trying to heat the air (though they will heat the furniture a bit). They would be used for zone heating, so you just turn it on when you enter the room and turn it off when you leave, same as an overhead light. I wonder if they could even be put on motion sensors…

I agree with focusing on passive solutions, but in retrofits this is quite a challenge. I have easy attic access so I’ll be adding another layer of insulation there (on top of the fiberglass batting that’s already up there), but my walls are much more difficult to work with. I either have to deconstruct them from the exterior or interior, and then either make them thicker or use blown-in insulation and hide the plug holes. All of that is expensive and makes my house into a construction zone for weeks or months. Also, my house has no existing ducting (since it was designed with electric baseboard heaters), and there’s really no room to add ducting without it being pretty ugly. So ducting the smaller rooms just isn’t an option in most cases. Sunlight exposure in these rooms is also very slight or non-existent, some of these rooms are underground in the basement level with no windows. Currently we just open and close doors to control airflow as you suggested. I have thought about adding room-to-room fans, that is a reasonable solution, especially if I can automate them somehow. In my basement, I do use a ceiling fan (in winter mode) to help spread the heat pump’s effect around to the smaller rooms. It helps, but it takes hours and hours to be effective. Don’t get me wrong, I love my heat pumps, but they are somewhat of an all-or-nothing approach to heating (I have just two zones, upstairs and downstairs) as compared to the granularity of heating that individual room heaters can offer.

I’m imagining the IR ceiling tiles being useful in rooms like my laundry room, where we only want it to feel warm when we are in there doing laundry, at no other time would that room need to be heated. The IR tiles seem like the fastest and cheapest way that I’ve found to achieve that, but I’m open to other ideas!

I’ve looked into radiant floor heating as well (hydronic), but that also poses a challenge. My house is slab on grade, so I would need to add a few inches of insulation on top of the slab, then add the radiant floor layer. So it would make my 8ft ceiling height shrink by 5 or more inches. Wish I had higher ceilings to work with…

I do drool over the Northern Europeon solutions to these problems, but unfortunately a lot of those products aren’t available in the US, or are prohibitively expensive. It is neat to see how they retrofit buildings that are centuries old.

The Kotatsu is neat, I do something like that in my office with blankets and an IR space heater. Keeps my legs warm, but the only problem is that my hands on the mouse/keyboard stay pretty chilly! That’s why I’m thinking an overheat IR panel would work well in that room as well. Obviously there is a risk of fire anytime you mix heaters and blankets, so it has to be actively managed.