Published on January 12th, 2017 | by Chris Dragon


Dragon’s Guide To A 100% Renewable Home, Part 2 (Water Heating)

January 12th, 2017 by  

Back in Part 1, we discussed upgrading to LED lights, which saved us more than 19% on our electricity bill. First, I have an update on that.

My father discovered that Philips 65W BR30 reflector bulbs work well with his old X10 dimmer switches. Phillips has a newer 65W BR30 bulb that uses 0.5W less power and probably works with his dimmers, but we haven’t tried it yet. If you need more brightness, Philips 75W BR40 might be your solution, but I think they’re out of production and everyone seems to be out of stock except for eBay, where you can find them by searching for “Philips 431932.” I checked Philips’ web site and found the new version of 65W bulbs, but no mention of a new 75W — hopefully they’ll be released eventually. Note that all these bulbs are rated for damp locations but not for fully-enclosed fixtures.

We also found that these aftermarket reflectors fit perfectly on GE 13909 bulbs and increase their brightness significantly in recessed fixtures, but you’ll be wasting power compared to using a true reflector bulb.

To the right are photos of 3-bulb options pointed at my ceiling.  I don’t own a GE 13909 bulb so I had to use a standard fluorescent that is a similar size and emits light in a similar pattern.  When such a bulb is combined with the aftermarket reflector and left on for a few minutes to attain full brightness, it’s brighter at 16W for ~1600 lumen than the 9W for 650 lumen reflector (Philips 65W BR30), but not as bright as the 12W for 800 lumen reflector (Philips 75W BR40).

I say ~1600 lumen because the CFL I used is rated to start its life at 1700 lumen and dim over time to a mean value of 1400 lumen. I’m only guessing the bulb is 1600 lumen at this point, but it could have degraded below that. If anyone is really wanting an accurate test here, let me know in the comments and I’ll perform the test again next time I have access to a GE 13909 bulb. Even if 13909 is a little brighter than my CFL, I can’t see it being bright enough to warrant the 4 extra watts of power use.

The “bright” measurements in the picture are an average greyscale color value for the entire image where 0% would be a pure black image and 100% pure white.

Step 2: Electric Water Heater

Tanked or Tankless?

A comment in Part 1 said that I shouldn’t have installed a tankless water heater because heat pump water heating (HPWH) is so much more efficient, so I delayed Part 2 to do a bunch of research.

It turns out that it’s probably not possible to use less energy with a tankless water heater vs a HPWH, but the amount HPWH can save you varies from 10% to 75% depending on climate and number of users. We should also take into consideration that HPWHs are expensive up front, unlikely to last over 10 years, and the risk of unexpected repairs costing $200 or more in labor is significant. If HPWH can’t save you enough energy in your particular situation, it makes more financial sense to simply install more solar panels to power a tankless heater.

Most Americans are familiar with water heaters that keep water hot in a large, insulated storage tank. A tankless water heater eliminates that tank by using enormous amounts of electricity to heat water to your preferred temperature as quickly as you need it.

The advantage of tankless is that you waste no energy warming more water than you need or keeping water warm in a tank. This energy loss is known as a “standby loss.” No matter how good the insulation on a tank is, heat will gradually be lost and require extra energy to heat the water again. People have reported using 0.8 to 1.1 kWh of energy each day just to keep the tank warm in a traditional electric resistance water heater. That’s more energy than the 0.75 kWh we recently clocked for a shower in the winter. Clearly a tankless heater saves energy vs a traditional tanked resistance electric heater, but can it save more energy than a HPWH tanked heater?

HPWHs pump heat from the air into the water and they take less electricity to do than a tankless heater uses to directly heat the water. Unfortunately, heat-pump technology is too slow to heat water at the high speeds necessary for tankless-electric heaters.

Calculating the energy saved by a HPWH is surprisingly difficult. First, most HPWHs are “hybrid” which means they use a smaller heat pump with limited water heating speed to gradually heat the tank, but when you want a significant amount of hot water quickly, they fall back on resistance electric heating (just like tankless heaters). Stiebel Eltron’s Accelera line is the only “pure” heat pump water heater I could find: It uses the heat pump about 90% of the time.  Accelera significantly increases efficiency but costs about twice as much up front.

In an NREL study of 14 HPWHs in a cold climate, HPWHs averaged out to be 1.9 times more efficient than electric-resistance heaters. The ability to create 1.9 times more heat than the electricity they use is often referred to as “1.9 COP” (aka 1.9 Coefficient of Performance).

The problem with the NREL study is the wide variability of results. First, 12 HPWHs were hybrids (A.O. Smith Voltex and GE GeoSpring), and 2 were Accelera. Whatever efficiency benefit those double-priced Accelera units have is lost in that 1.9 average. Worse, one unit installed in a small, unventilated closet was reported to have 1.0 COP (no better than tankless heating) while another in a warm basement averaged 2.6 COP (and likely benefited from air warmed by a whole-house furnace, which means its true energy saving was not actually 2.6 COP).

NRDC and Ecotope created a software package to model HPWHs under various conditions, including climate, ground water temperature, number of users, hybrid vs pure heat pump, and installation location. Although I don’t trust a fully-software model quite as much as real-life measurements, this data should get us much closer to accurate HPWH energy estimates for particular situations. You can download Excel data for California, or for the entire USA. They even released source code for their software model if you’re feeling particularly adventurous, but you’ll need to supply things like your own climate and groundwater temperature data.

Most importantly, the lowest HPWH efficiency predicted by Ecotope’s model is 1.1 COP in North Dakota with 1 person using it. There is no case in their estimates where HPWH would be less efficient than a tankless heater. Notice that they model electric resistance tanked heaters at 0.91 to 0.98 COP because of standby losses, but there are no standby losses in a tankless heater so it should be considered to be 1.0 COP.

Estimating hot water energy use

First, estimate your average groundwater temperature. You can find it on the map to your right, or you can measure it with a thermometer after letting your water run for a couple minutes. The problem with measuring it is you won’t get an average temperature unless you measure in the middle of spring or fall. The problem with the map is it doesn’t account for unusual topography in an area, like living up a mountain.

Decide how warm you want your water to be (105F is preferred by most people), then plug those values into this equation:

EPG = ([preferred temp] – [groundwater temp]) * 0.00242kWh

In our case, this results in EPG = (105 – 54) * 0.00242 = 0.123kWh/gallon. EPG is Energy Per Gallon: How much energy it takes to heat a gallon of water to your preferred temperature.

See how much water your showers use by running them into a jug for exactly 1 minute. Measure the amount of water in the jug using something like the measurement lines on a blender. I think our shower head is rated at 1.5gpm but years of lime buildup have reduced it to 0.726gpm. I don’t really want to clean it and start using more water, though I’ve noticed it takes longer to rinse hair. If you don’t have a low-flow shower head, please get one.

Use a timer to judge how many minutes per month each shower is in use, then multiply by the gpm of each shower head to get gallons per month. Multiply the final gallons by EPG to get monthly electricity use for heating water with a tankless water heater.

If you take baths, estimate gallons used per month and multiply that total times EPG.

I highly recommend switching to cold water clothing detergent. If not, add as many gallons as your washing machine uses times number of loads per month. Multiply total gallons times EPG.

Estimate how many gallons of hot water you use for dishes per month, then multiply times EPG.

Estimate how many gallons of warm water you use for hand washing and how many months per year you use warm water for hand washing (we only use warm water in the winter). This is probably too difficult to measure considering how little water hand washing should take, but I’ll leave the decision up to you. If you come up with a number, multiply it by the number of months you use warm water, then divide by 12 to get extra power use averaged across all months of a year.

Do you use much hot water for anything else?

  • Hot tub?
  • Car washing?
  • Obsessive daily floor mopping?

Estimate kWh per month for anything else you can think of.

Add all your kWh/month figures together and that’s about how much energy a tankless water heater will use per average month.

Look up the expected COP value for your particular climate and situation in Ecotope’s California or USA data.  The data you need will be in the “CA Data by Location and HH Size” or “US Data by Location and HH Size” tab which should appear at the bottom of most viewers.  If you live outside the USA, find a COP in a US location with a climate that reasonably matches your own.

Notice that they provide values for “Pure heat pump” (ie Accelera) and “Hybrid heat pump”, so it matters which HPWH model you choose (we’ll discuss that choice soon).

Divide your kWh/month total by the COP you decide upon to get your average water heating electricity use per month.

Is solar plus tankless cheaper for your household?

Say you decided you need 60 kWh/mo for tankless water heating but that an HPWH could get the job done at 1.6 COP.

60 kWh/mo / 1.6 COP = 37.5 kWh/mo for HPWH water heating.

60 – 37.5 = 22.5 kWh/month / 30 = 0.75 kWh/day extra for tankless heating.

Instead of installing a HPWH, you could supply that extra 0.75 kWh/day with additional solar panels. Our solar system produces an average of 30.68 kWh per day using 21 panels. 30.68 kWh / 21 = 1.46 kWh per panel per day. So we could eliminate the need for a HPWH by adding just half a solar panel. Not only would that be much cheaper than a HPWH, but it qualifies for a 30% federal tax rebate and it carries much less risk of needing expensive repairs. Plus, solar panels are expected to last 25 years instead of 7 to 10 for a HPWH.

I’ll go into detail about estimating solar system energy output, solar system cost, and so on in Part 5. For now, just write down how much extra energy you expect a tankless heater to use and add that extra when you total everything a solar system needs to cover. I’ll also remind you in Part 5.

HPWH models

50 and 80 gallon HPWHs

Say you believe HPWH will save you so much energy with your climate and water-use patterns that it’s better to install one instead of additional solar panels.

Or maybe you don’t have room for more solar panels.

Or maybe you just want to learn what HPWHs are out there.

If so, let’s look at models and think about cost and reliability.

NREL shows a timeline on page 3 where we can see the first mainstream HPWH was released in 1950. The number of models available has gone up and down, roughly corresponding to electricity prices. Around 2010 there was another new push as a few manufacturers brought HPWHs to America (in some cases, descendants of models that were created decades ago). Around 2012, HPWHs developed quite a buzz in the US when GE moved its GeoSpring HPWH production to the USA. In fact, it was the first new product to roll off the assembly line at GE’s Kentucky plant in 50 years. Yet, at the end of 2016, GE ceased producing GeoSpring, citing low sales.

Buyers cite a number of problems with GeoSpring:

  • GeoSpring is 2-3 times more expensive than resistance electric heaters. Reduced electricity use may pay for that in a few years if you use a lot of hot water.
  • GeoSpring is the quietest of the HPWHs but some owners were still bothered by the noise. Most HPWHs are about as loud as a traditional air conditioner, yet more likely to start randomly at night which may disturb some sleepers.
  • Reliability problems. Some claim reliability problems were addressed after the first year or so, but Amazon reviewers talk about problems in units purchased as late as 2015. GeoSpring’s overall rating of 2.6 stars is the worst amongst HPWH brands.
  • Difficult to service. Plumbers don’t know heat pumps and HVAC techs don’t know plumbing. Now that HPWHs have been prevalent for 5+ years, I suspect there are enough techs that can deal with them in most areas.
  • Extra service visits eat up potential savings on electricity bill. Manufacturers rarely pay for labor even if warranty covers parts.
  • In many situations, you can cover the estimated extra electricity used by a tankless heater by buying a larger solar system. The cost of a solar size increase will often be less than the extra cost of HPWH, and solar is expected to last ~3 times as long with much less maintenance.

Around the same time GeoSpring was released, Rheem came out with a competing HPWH. Rheem has been making water heaters since around 1930 and is one of the largest manufacturers in America. Unfortunately, Rheem’s heater earned only 2.9 stars in 17 Amazon reviews. The company offers a 10 year warranty, but it does not include labor and one lady says they refused to replace her unit under warranty because they are no longer making that particular model anymore (they are making an updated version). 3 of the 17 reviews complain of failures within the first year or even the first day. Home Depot gives Rheem 4.2 stars in 46 reviews but almost all are initial impressions. 3 reviews say it stopped heating water within the first 10 months and a 4th negative review is a duplicate of one on Amazon.

Ruud also came out with a HPWH, but I couldn’t find reviews on Amazon or anywhere else obvious. It turns out that’s because Ruud and Rheem are the same company.

As the largest water heater manufacturer in North America, it’s no surprise that A.O. Smith also released a HPWH. Like Rheem, A.O. is an old American company that has also made water heaters since the 1930s. Despite being so large, I’d never heard of them because they seem to mostly label their heaters with other brand names including State, American, Reliance, Lochinvar, Takagi, and U.S. Craftmaster.

The A.O. Smith Voltex is almost as expensive as Stiebel Eltron’s Accelera line. I found 7 reviews of the 80-gallon Voltex, 6 of them positive and the last one complains only about the noise level. I found 3 reviews of the 50-gallon Voltex, one complaining of a leak on day one and another of a rotten egg smell caused by the anode rod. Unfortunately I could find no other reviews on the internet or at hardware stores. Reviews of other A.O. Smith heaters on Amazon are somewhat low and all had 5 reviews or fewer.

I noticed that nobody ever mentions Whirlpool when comparing the four big HPWH makers even though Whirlpool Corp. is huge. I’m pretty sure that’s because A.O. Smith secretly makes the Whirlpool and Kenmore branded HPWHs.  The pictures of Whirlpool and Voltex are identical except for the label, and the Kenmore just has the Voltex rectangular covers replaced with oval ones. I also found a reviewer saying they contacted A.O. Smith for warranty service on a Whirlpool-brand HPWH.

I found one lonely 4-star review of Kenmore’s HPWH on the Sears site, and 7 reviews of Whirlpool’s HPWH at Lowe’s. One review says the heat pump failed in two months. Another said the pressure release valve leaked on the first day and replacing it did not help. His plumber told him he needed an expansion tank to relieve the pressure, which is simple physics in some homes, but the reviewer blamed the HPWH due to ignorance.

EnergyStar lists a number of other HPWHs, but none of the remaining manufacturers have any Amazon reviews and most of them are A.O. Smith brands. I searched for every heat pump manufacturer on the list up to page 10 when they switched from heat pumps to tankless gas.

Anything that involves water also involves corrosion. Adding the complexity of a heat-pump compressor to transfer heat into that water makes for some tricky engineering. The NREL report says:

By 1984, ~17 HPWHs were on the market (Calm 1984). While sales reached over 10,000 units per year in the 80s, the HPWH market soon collapsed because “these early machines suffered from high purchase prices, high maintenance costs, excessive noise, poor longevity, and limited installation options.”

With the discontinuation of GeoSpring, I’m worried we’re heading into another period of collapse as renewables drive down electricity prices and word of mouth spreads stories of unreliability in at least the early versions of rebooted HPWH models. Nevertheless, if anyone is going to get an HPWH right, it might be Stiebel Eltron.

Stiebel Eltron Accelera

Stiebel is a German company started in 1924 and it released its first HPWH in 1976 in Europe during the oil crisis. As I mentioned earlier, Stiebel’s “Accelera” HPWH is the only “pure” HPWH I found, meaning it’s sized to use the heat pump about 90% of the time. It uses a design that eliminates the sacrificial anode rod and uses no water pump, two common points of failure. Unfortunately, Stiebel’s design comes with a price (around $2500 for its 80 gallon version) that is almost twice as high as Rheem’s HPWH.

Stiebel’s price likely explains why the unit has only 3 Amazon reviews, two good and one complaining it’s too big for their space (good idea to measure before ordering!). Home Depot gives it 4.4 stars in 19 reviews. One reviewer says the thermostat failed in the first year. Another claims he’s on his second unit in two years, that both leak refrigerant, and that he knows two others with the same problem (although I’m suspicious that one guy encountered the same problem 4 times).

This Australian site has 75 reviews, but the average is a disappointing 3.3 stars. Stiebel’s HPWH was released there around 2008 and many reviewers mention failures after the 5-year warranty ran out. I noticed Stiebel’s warranty has since increased to 10 years, suggesting they have improved reliability, and its reps claim the same thing. On the other hand, at least 3 reviews complain about units installed in 2012 or later.

Stiebel has a total of 5 out of 94 reviews that complain of failures in units purchased in the last 4 years. If I tally similar failure reviews for each manufacturer, I get these results:

  • Stiebel Eltron: 5/94 = 5.3%
  • Rheem: 6/63 = 9.5%
  • A.O. Smith: 2/18 = 11.1%

Note that there are far fewer total reviews of all the A.O. Smith related brands, so I’m less confident in that result.

Many Stiebel reviewers mention that their warranty service experience is good, though one says they’re on their 4th replacement unit and Stiebel doesn’t pay for labor. One review mentioned they prefer EcoSpring HPWHs but I don’t think they’re available outside Australia.

A few reviews mention it can take hours for the tank to re-heat after it’s depleted in winter and some say water never gets warm enough in winter. In fact, local Stiebel reps have admitted the design can simply freeze up in winter temperatures and that they need to install external resistance heating elements to compensate. NREL discovered that most HPWHs use pure resistance heating below 45F, but Stiebel limited the size of its resistance heating element to the point of being unable to keep up in low temperatures. Stiebel’s own manual only shows COP data down to 41F which I’ll bet implies it doesn’t work below that temperature. So if you’re installing in a space that stays below 45F when you need hot water, I don’t recommend Steibel’s design.

A number of reviews also complain about the noise being much higher than air conditioning heat pumps, so be sure to consider that if you have no place to put it that is away from bedrooms or if you have “thin walls.” I get the impression that a large amount of noise or ringing was actually a defect in certain units, but reviewers said Stiebel refused to consider it a valid reason for warranty service. Stiebel’s specs say the unit emits 60db, which is roughly equivalent to conversational speech or a traditional air conditioner.

BTW, I have to give Australians props for coming back to review Stiebel’s HPWH after 6+ years. You don’t find many American reviews beyond initial or first-year impressions of a product.

HPWHs are taller and heavier than traditional heaters since the heat pump sits on top. At the same time, larger tanks are better for houses that use enough hot water use to justify the cost of HPWH. The extra tank capacity lets you draw out dozens of gallons during high-use periods, then gives the heat pump time to re-heat the lost water without using resistance heating. Stiebel’s 80 gallon HPWH is 6.27 feet (1.913 m) tall, which may not fit in some basements. It’s also 956 lb (433.9 kg) when filled, so make sure your flooring is up to the weight.

Overall, among the few choices in HPWH manufacturers, Stiebel would be my choice in warmer climates if it could be installed away from bedrooms and price was not an issue. My impression is that Stiebel has fixed problems in its initial design and pushed the failure rate lower than other manufacturers, though I admit I don’t have enough data to be completely sure of that. I also appreciate Stiebel’s long history in HPWH design, higher efficiency than competitors, and high ratings in other products it manufactures. It still concerns me that with such a long HPWH history, there are so many reports of Australian Accelera units dying in under 6 years. Make sure you get one with at least a 10 year warranty.

In colder climates, I would go with Rheem because its resistant heating elements will work well below 45F. Rheem has slightly fewer early failure complaints than A.O. Smith and slightly more praise in water heater repair forums that I read (of course, that’s highly subjective). Rheem is also much cheaper.

If you want the quietest HPWH, you might pick up a discontinued GE GeoSpring before they run out of stock. GE says it will continue to provide GeoSpring parts and warranty service, but double check that still applies to units purchased after production ended.

Another cold-weather option is the Sandeen CO2 HPWH. It claims to work down to -15F and uses CO2 instead of traditional refrigerants that cause enormous greenhouse warming when leaked into the atmosphere. Even the latest refrigerants that do less damage to ozone still contribute to climate change. CO2 as refrigerant is more energy efficient but also requires higher pressures, making components more expensive. Unfortunately, that puts Sandeen at $3500 which is probably beyond the sensible price range for most users, so I didn’t research it.

If anyone’s had a good, long-term, cold-climate experience with an HPWH, let us know in the comments. You might also choose to install a Stiebel in a heated space, but then you’re paying to heat the air that heats your water and that’s not saving as much energy as it might appear. Even if you don’t directly heat your basement, it benefits from radiant warmth from floors above, and I’ve read people can feel the cold in the floors above a basement after installing an HPWH. Those cold floors will drive your house heater to work harder.

Hot water tanks may grow Legionnaires’ disease

COP vs tank temp on 5 different HPWHs

Despite complaints and lawsuits from water that can burn in 5 seconds, hotels often keep their water heaters at 140F to kill Legionnaires’ disease. The bacteria that causes LD grows best in 90-105F water but is not killed until water rises above 122F. It infects the lungs when water vapor is breathed or when drinking it allows a bit of water into the lungs, and infection causes a form of pneumonia.

Yet OSHA says the risks are small. About 95% of people are resistant to LD bacteria and home water systems are less likely to be infected than larger systems. Even so, OSHA says if you don’t have young children or elderly in the house who are likely to be scalded, keeping your water heater at 140F is “a good idea.”

The problem with 140F is it reduces the efficiency of an HPWH by about 0.5 COP vs a 122F setting. If you reduce to 105F, you save another 0.5 COP, though your hot water won’t last as long. Those COP losses aren’t quite as serious as they look, because you only get the lowered efficiency while heating the last 35 degrees from 105 to 140, but they’re still significant. You can save a lot of that power by only turning it up to 140F once a week, but few have the discipline to do that in the long run. On the other hand, a tankless heater never leaves warm water in a tank, so LD isn’t something you have to worry about.

Direct solar water heating

10-15 years ago, I read that using panels with water running through them to heat the water was the cheapest form of hot water. I even grew up in a house that used such “solar thermal” panels with a gas heater as backup. However, due to the high upfront cost of plumbing and panels for those systems, maintenance issues with pumps and valves, and the possibility of damage in below-freezing climates, it’s now become cheaper to use solar PV to power an electric water heater. Solar thermal water heating has gotten more expensive as it’s fallen into disfavor, solar PV has gotten cheaper, and you need the most hot water in the winter when solar thermal heating is at its lowest level.

We went tankless

Electricity cut-offs on left, water heater on right

We live in a semi-cold climate, use little hot water, and four years ago, heat-pump water heaters were prohibitively expensive (they’re not much cheaper now). Being into energy efficiency, we’re also into saving hot water. We wash clothes in cold water and rarely run the dish washer (although hand dish washing may use more water, depending how you do it). Having low hot water use makes it difficult to justify a HPWH vs adding additional solar panels. I’ve now measured the heater as using just 6% of our electricity in the winter, and it will use less in the summer. For comparison, an average household uses 13% of their energy heating hot water.

We installed a Stiebel Eltron Tempra 15 tankless heater around Oct 2012. The Tempra 15 is enough to heat water for one shower at a time. It’s rated to increase water temperature up to 65F degrees with 1.5 gallon per minute water flow (which combines nicely with a low-flow shower head) or 43F degrees with 2.3 gpm flow. In our area, daytime temperatures can stay around 32F for a few days at a time but we’ve not had any trouble getting warm water since ground water stays well above freezing. Even in severely-cold areas, groundwater lines are run deep enough to stay above freezing, and as long as it stays above 40F, this unit can get water up to our preferred 105F.

Tankless brands

I spent a lot of time looking at different brands before choosing Stiebel Eltron. However, my second choice would have been one of the EcoSmart line. EcoSmart tankless heaters get almost as good ratings and cost significantly less. EcoSmart supposedly uses parts that are easier to come by replacements for in local hardware stores, but Stiebel uses parts that are likely to last longer. For example, lime tends to precipitate out of water and crystallize on hot surfaces, so the heating chambers in Tempra are much larger, which makes it nearly impossible for lime buildup around the heating element to short to the side of the heating chamber and destroy the element or the chamber.

Few other tankless heaters had many reviews except for Rheem. Rheem’s star average is pretty good, but one recurring complaint I found (at least four years ago) was Rheem heaters would get stuck on, heating water when it wasn’t actually flowing through the heater. Not only is that a huge waste of power, but it could destroy the heater and cause PVC plumbing pipes to swell up and burst. Problems like that are another reason I chose Stiebel Eltron, seeing as they have a long history of making tankless heaters and working out the bugs.

Cleaning tankless lime buildup

Many complain that tankless heaters fail because lime deposits build up inside. Stiebel Eltron’s larger pods protect against that, but lime buildup still reduces the efficiency of the unit. If you get a lot of lime building up around your shower head where hot water flows, you need to do a yearly cleaning of your tankless water heater using vinegar. Honestly, we get so little lime buildup in the shower that I waited four years to do our first cleaning, and Stiebel’s installation manual says no de-liming is necessary, but it is something that should be done periodically to keep peak efficiency.

Here’s a video on how to de-lime a tankless heater. The only tricky part is you need a 1/6HP submersible pump. You can get a whole kit with a pump, hoses, and bucket, but they charge over $150 shipped. A better idea is to buy the pump for $80 and use your own hoses and bucket. Amazon also has the hoses and bucket, but I found them cheaper at a local Home Depot (hoses and bucket). You can also get a pump for $40 but I’m not sure it will be durable enough for acidic vinegar. A grit guard at the bottom of your bucket is suggested in the video, and it’s also helpful for things like hand washing your car.

The video says your tankless unit must have a “flush kit” installed which is basically two valves that let you easily connect garden hoses to pump in vinegar that will only go through the tankless heater piping, not through your house piping. If you don’t have those two valves, and you feel comfortable with DIY maintenance, you can do this:

Buy two 3/4″ male GHT to female NPT adapters plus plumbing teflon tape. Again, they were cheaper at Home Depot (they use different names for the same items: 3/4″ MH x 3/4″ FIP and thread seal tape). I think all tankless brands use the same standard pipes, but you might want to double check your tankless has two 3/4″ NPT male threaded pipes coming out of it to allow water to enter and exit. Learn more about NPT-threaded pipes here.

Turn off the power and water. Unscrew the two water hoses connected to the NPT pipes coming out of the heater and hold a bowl under them to catch draining water. In the case of Tempra heaters, you actually need to remove the left hose in order to clean a small sediment filter, so even if you had a flush kit with garden hose connections you’d need to disconnect the left NPT pipe.

Clean off any old teflon tape on the NPT pipes, then tightly wrap new teflon tape 3 times clockwise around the threads on each pipe, leaving the two threads nearest the end of the pipe bare (if you wrap counterclockwise, the tape will try to loosen and bunch up as you screw connections over it). The teflon tape acts as a sealer to make the tiny spaces between threads water tight. Screw the GHT to NPT adapters on over the teflon tape and hand tighten them 4.5 full turns, then 3 more full turns with a wrench. Don’t use too much force and risk distorting the threads, as that will make it harder to keep the joint water tight. Be ready to tighten a bit more if you notice leaks (wrap the joint in rags before starting the pump if you want to avoid a vinegar mess).

Screw your garden hoses on to the GHT to NHT adapters, again, not too tight as the rubber in them will take care of sealing without much pressure. Follow the video instructions at this point.

Once you’re done with the video instructions, connect your original water hoses to the NPT-threaded pipes on the water heater using a new wrap of teflon tape and the same tightening procedure. Be ready with your bowl and rags because you may need to tighten them a bit more if they leak after turning the water back on.

After cleaning, before you turn the power back on, run water through the heater until you hear no more air gurgling. Doing this is necessary to remove air from the heating chambers because trying to heat air will burn out the heating elements.


There are an enormous number of rebates offered by federal, state, and even city agencies. To check your area, click here, then use the Apply Filter button to narrow down results to your state. You can then add a filter for Coverage Area > County or City or Zip Code, although the way they force you to choose a zip code from a list is really frustratingly slow in California due to the number of zip codes. Next, add a filter for Technology > Energy Efficiency > Appliances > Tankless Water Heater (or Water Heaters to look for HPWH rebates).

I didn’t find any rebates for tankless heaters in my area, but I did find a low-interest loan program. For HPWHs, our electric company offers a rebate “up to $200.” There was also a federal tax rebate for HPWHs that expired at the end of 2016 and will probably not be renewed, sadly.

Leak detection

When it comes to detecting leaks from any kind of water heater, I highly recommend the Honeywell RWD41. I have one below the Tempra 15 (which also puts it below the washer/dryer), below all my sinks, dishwasher, and toilets.

They’ve saved me multiple times from dripping leaks I wouldn’t have noticed for who knows how long without them. My one complaint is they don’t chirp for long when they’re low on battery so I don’t always hear them crying before they’re dead. Instead, I set an alarm to check their voltage every 6 months and recharge the Tenergy 9Vs in them when necessary.

To be continued…

Whether you go tanked or tankless, remember the most important thing is to power it all with 100% solar electricity.

In part 3, we’ll talk about eliminating gasoline. Onward!

Check out our new 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , ,

About the Author

I grew up programming on an Apple IIe and now run a software company. Despite a hatred of family camping trips, I've always had a deep concern for the creatures we share this planet with. In respect of those creatures, I spent a number of years working at an environmental non-profit, but these days I just try to live with sustainability and encourage others to do the same. I've even come to tolerate camping trips!

Back to Top ↑