Tankless Electric HWH w/ Battery ‘grid buffer’?
Hi, I have been searching to see if this exists or if this is just a hair-brained idea. I recently came across a stove – the very high end ‘Impulse’ which is induction with an integrated battery backup / buffer. So it can be plugged in to a 120v outlet on a standard circuit and the battery will handle the spikes in electricity as opposed to needing a larger electrical service.
This got me thinking about other high-demand appliances. I had looked into the efficiency of electric tankless hot water heaters in the past but the major draw back is the spikes in demand required from the electrical grid which rules them out as a sustainable option despite their other redeeming qualities.
…so finally to my question, what if you paired an electric tankless HWH with a battery in the fashion of the impulse stove – so the battery trickle charged itself from the grid (or better yet from solar), and the HWH drew power directly from the battery so there were no grid demand spikes.
Does this exist? And any thoughts from the experts on here on this setup? I’ve seen battery backups for tankless HWH but not grid buffering batteries paired with a tankless unit. Can the battery backup systems like ‘Hugo’ or Ecosmart’s Eco27 be configured to function for grid buffering?
It would seem to me you would get all the efficiency of the tankless HWH without the negatives of the grid demand spikes… what am I missing?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
That's what the tank is for.
Nice, yea I see your meaning as far as energy storage...
Tankless water heaters, when sized appropriately are more efficient than even heat pump hot water heaters. So, we could probably go down a rabbit hole of debating the energy loss and efficiency of storing energy thermally in a tank or chemically in a battery, and determine which is more efficient... which would vary based on the size and usage of each system (so a moot exercise in my mind if not designing a specific building system for a project) ...but that's not my question. Maybe I'll start another thread to collect some funny quips on that topic.
Interested in some more thoughtful answers from the expert members that may have knowledge in this area. Thanks.
"Tankless water heaters, when sized appropriately are more efficient than even heat pump hot water heaters." [citation needed]
Yeah, I can't see how that statement could possibly be true. A HPWH with a UEF of 3 would be able to heat the water, cool back down to room temperature, reheat the water, cool down again, and heat the water a third time before coming close to using the same amount of energy as a tankless. Since the water coming into the house is usually quite a bit cooler than room temperature, reheating the water won't take as much energy as it took to heat the first time. So, the HPWH will be able to cool down at least partly and heat it back up for a fourth time. That would be a lot of standby loss.
This is easy: the tank-type water heater is more efficient, period, no question, no exception. So why is this the case? Because it is converting electricity into heat, a process that is 100% efficient. The only losses are in the wiring supplying power to the unit, and those losses are both minimal, and exist in both systems (the battery system will have the same, or more, wire losses).
Why is the battery system less efficient? Because it is converting the energy several times, and every time you do so, you lose a little. Even a 94% efficient system, which is probably overly optimistic, still has 6% more losses than the tank system.
But.... That's for "steady state" operation, where a lot of hot water use is assumed. For infrequent hot water use, things are a lot more complicated, because you have to allow for the standby losses from the hot water heater. This isn't going to be a lot, but may be more than for the battery system, cancelling out some of the advantages the tank-type heater has in terms of operational efficiency. In this case, it becomes more of an economic question, and the tank-type heater still wins.
Batteries can be used to average out short term demand loads (like on demand water heaters) over longer periods of time to avoid the brief load spikes. The battery would supplement the utility supply when the water heater is running (the water heater can use utility AND battery power at the same time, so those systems can be in parallel and still work as you're thinking). The battery ends up doing what the utility world calls "peaking", which means it handles the short duration "peak" loads. That's the most expensive kind of power in the utility world. You'd need to determine if the battery system, which would need to be fairly beefy for this application, would save enough money due to reduced overall system losses over the life of the system, which would likely be 5-10 years or so. I'd be very, very surprised if the battery system would be more economical to operate.
The cooktop is more practical because it has much lower peak loads, allowing for smaller amounts of battery storage to accomplish the task. On demand hot water heaters are VERY large loads, so they need much more substantial battery systems to accomplish the same load leveling.
Bill
Let's not forget that batteries have standby losses as well.
The purpose of the battery in the stove isn't to increase efficiency. It's so you can run an appliance that typically requires 30 to 50 Amps at 240V off of a 120V wall plug, so you can install an electric stove in locations where it might be impractical to increase the electrical capacity. It's a bridge to getting people to switch away from gas.
We discussed back in this thread: https://www.greenbuildingadvisor.com/question/battery-energy-storage-appliances
H_ZIMM,
It's no doubt my failing, but I don't understand this:
"the major draw back is the spikes in demand required from the electrical grid which rules them out as a sustainable option despite their other redeeming qualities."
Why does that make them unsustainable?
It doesn't, at least not at utility scale. What happens is that a typical on demand electric water heater might need a 100 amp, 240 volt circuit. That's 24kw. For perspective, the AVERAGE load of a typical house is around 1 to 1.5 kw. That means over a day, the house will average somewhere around 1.5 kw, so it will use around 36 kwh per day, or just over 1,000 kwh per month. That's what utilities plan on.
Averages only work out over larger numbers of things though, so the problem is when you get "small and local". A typical house will be connected with several others to a single transformer. 5 houses may be connected to a single 50 KVA transformer, for example. That single transformer is good for 50 KVA, which is the same as 50kw for resistive loads like a hot water heater. That means that transformer can provide 208 amps of 240 volt power all day long, forever. No problem! Those five houses use around 6 amps on average, so 30A for all the houses, plus about 100 amps for the hot water heater is 130A while you're taking your shower as seen by that transformer. That's no problem. Now what happens if you neighbor decides to take a shower too, at the same time, such as before work in the morning? Now that transformer sees 230A, over it's 208A rated capacity. Since the limit is a thermal limit, not an absolute limit like a fuse, the transformer will probably be OK for even a long shower for the two soon neighbors.
Now let's assume ALL FIVE houses decide to take showers before work in the morning. They probably won't all take the shower starting at EXACTLY the same time, and they probably won't all take showers of EXACTLY the same length, but there will be some overlaps. For the time they're all in the shower at the same time though, that transformer sees the 30A average load from the five houses, plus 100A for the on demand water heater in each house, for a total of 530A -- over 127 kw !! -- which is over 2.5 times the rated load of that transformer. The transformer might be able to survive that for a few minutes, but it's doubtful it can do it for an extended period of time. You're also going to see a lot of volt drop during that time, so dim lights and other exciting power anomalies.
That's the problem. The smaller stuff out in the neighborhoods can't handle the SIMULTANEOUS load of lots of huge loads like that. The exact same thing is a problem with EV chargers. Loads way larger than the AVERAGE load of the house can be a problem when MANY such loads run at the SAME APPROXIMATE TIME. The reason is that with these small "sample sizes", the statistics don't work out, and you can have problems. Tank-type water heaters act to average out the load, so a smaller amount of energy can be put into the water over a longer period of time, then the tank handles the short period of heavy load that is represented by the showers in my example here.
So what does the "grid" see? The grid is going to be fine -- up to a point. A typical utility may have (literally) one million residential customers. Do you think they'll all take a shower at the same time? On the same day? Probably not. Chances are some people like showers Tuesday, some Sunday, some every day, some might need one RIGHT NOW because they used acoustical caulk, etc. The PEAK loads don't all "line up" the way they can on the small scale, so they are spread out in time, resulting in a nice, smooth load curve as seen by the utility. The statistics work out in the grid's favor here, with the loads spread out over time so that the average load is less than the possible sum of all the peak loads (this is known as a "demand factor" in the electrical world), and the grid spreads that average load out over a large number of customers and serves it with a large number of very large generators (typical power plants have several generators of 250 megawatts or more each).
What this means is that the on-demand water heaters have nothing to do with "sustainability", and everything to do with system capacity planning. EV chargers face exactly the same problem, but worse, because EVs chargers tend to run for much longer periods of time -- hours -- where on demand hot water heaters are likely to run for tens of minutes at a time. The portion of the grid most local to the houses needs to be significantly upgraded to handle these heavy, brief loads, while the larger grid doesn't have the same issue -- for the larger grid, it's just a generation issue, the peak/average stuff isn't as much of an issue, since the grid itself spreads that out (that's the entire point of the grid, BTW, that and to allow cheaper generation to be used when available to allow for better economies in the system).
But.... Remember Bill's First Law: Physics is a b****. That's the one true law of the universe.... In this case, with the grid needing to be upgraded to handle short term heavy loads, you need to remember that standby losses for things like transformers go up as the device gets bigger, so the tradeoff to support those heavy loads while they're running is higher system losses during periods of lighter loads. No free lunch! Nothing to do with sustainability though.
Bill
Thanks Bill.
Totally doable. But why would you pay the expense to have a device specific battery when the same size battery (Roughly speaking) could serve your whole house? It's an expensive chunk of energy storage for just one energy service. But sure, someone could invent that. For the price point... I'm not sure what the market would be. If it were $3000 would you get one?
The value prop for the cooktop is as a drop in (and a high end one at that) for a high end gas cooktop. One that performs better, allows for cooking by algorithm, and all sorts of fancy stuff. And no electrical 240 run, no utility service upgrade. That's a huge value proposition for a high power, low energy using service.
Conversely, instantaneous water heating is usually high power, high energy (electric tankless wh are upwards of 80% of total home energy consumption in some of my new single family home projects).
On another note-- tankless electric can absolutely be more efficient than hpwh tank! It's a question of how and when. These technologies are treated fundamentally differently, like you might expect when one is a 5 foot tall cylindrical air conditioning unit, and the unit is a quiet cute little box that can hide under the sink so architects can pretend it doesn't exist. And then physics is really a b****.
"Tankless electric can absolutely be more efficient than hpwh tank! It's a question of how and when."
This claim is central so let's dive in a little bit. When is electric resistance going to be more efficient than a heat pump? The heat for the heat pump comes from within the house, when the combined COP of the water heater and whatever heats the house is below 1.0 then just straight resistance heat is more efficient. Let's represent the COP of the water heater with COPw and the COP of the heating system with COPh, the combined COP is given by:
COP = 1/(1/COPw+1/COPh)
Water heaters typically have COP around 3.5, so the combined COP is under 1.0 when COPh is below 1.4. When would you see that? If your heat source is resistance heat, either because that's what you have or it's so cold your heat pump needs supplemental heat. Or if it's just cold and your heat pump is running at a low COP. If you live in a place where electricity is cheap and fossil fuels are expensive it also might make economic sense to use electricity.
For the most part these are rare cases. But if you live some place where electric resistance heating is the most cost-effective home heating, it probably is also the most cost-effective water heating. I say "probably" because you also have to look at cooling in the summer, where the HPWH contributes for free.
So that's resistance vs heat pump. The other part of the equation is tank vs tankless, because in all the circumstances where resistance is more efficient a tank electric water heater would fit the bill as well. Tankless is going to be somewhat more efficient because standby losses are lower, it's hard to put a number on that because it depends on the pattern of use. But tankless with a battery is also going to have standby losses, as well as losses when the energy is converted from electricity to chemical energy and back again.
Now, I'm going to wave my hands a bit here, but based on my experience with both plumbing and batteries there is just no way the standby losses on a tanked water heater are greater than the associated losses with a battery and tankless heater.
To back up my handwaving, the EIA estimates that utility-level electricity storage using batteries returns about 82% of energy stored:
https://www.eia.gov/todayinenergy/detail.php?id=46756
You'd have to have a really terrible storage tank to return that little of the energy stored in it.
Or low utilization?
I'll extend. Draw the boundaries bigger. Here's a very common scenario I see.
New home being built. Hot water options consist of the following:
1. a new tankless electric water heater-- the new home is all electric, and gas is too expensive to provide out to it, and propane is practically impossible for locating the tank per NFPA 58.
2. draw from an existing heat pump water heater located in an existing home on the same lot. And run hot water lines on a timer based recirc pump (yeah... I know) out to the house. length between houses is 100'
You might ask, why can't there be a heat pump at the new house. To which I say... so many REASONS. I know because I bang my friggin' head against with architects about 8 hours a week. It's the first thing I ask about-- 'where can we put a tank water heater?' and the answer is literally nowhere. It's illegal to put in the side and back (setbacks) and it's an architectural non-starter to put it in the front, and it sure as heck ain't going inside because they're loud and stuff. So back to options 1 and 2.
Let's say they use 60 gallons a day, 55 degree inlet water, 120 degree final temp, and COP-1 for the tankless electric. Round numbers to say they use 10 kWh's a day (60 gallons of hot water) for the tankless electric.
How many hours a day of recirc will it take for the HPWH, assuming nominal efficiency, to have an effective COP of 1? R-5 pipe insulation, 120 degree average water temp (between supply and return), 55 degree ground temp, 3/4" line, R-3 insulation.
The heat pump will be less efficient, even operating at nominal efficiency (DUBIOUS) if that recirc operates for ~20 hours a day. Ok... not bad. Can probably accommodate that. If the usage goes down to 30 gallons a day, then it goes down to 10 hours a day.
Assuming that HPWH COP is actually more like 1.5 because the return water is warm, then those numbers go down to 5 hours and 2.5. On-demand? Now you're talking, but some people don't like the push button and go with the occupancy sensor. And some of these houses get a LOT of false triggers given the layout.
What about no recirc line... cool. So occupant has to wait 90 seconds for hot water. For my work, that's often what happens because timer based recirc is effectively illegal where I am. But everyone knows the issue and plumber preps a return line and then they retrofit it after permit when the owner complains.
Obviously, this is a situation where a HPWH is being put at a significant disadvantage. But this is much more realistic than an apples to apples comparison.
Almost nowhere is it a good idea to put the water heater outdoors.
I hear ya
But where would you suggest? If it cant go inside, and no garage or unconditional space, where should you put it?
I just talked to a builder where the design put them inside on 2 projects. He's already changed them out less than a year after final. Owners objected to them so they went electric resistance. Lose lose
If the problem is noise, then some sound proofing in a mechanical room is the way to go, and has been discussed before in these forums. It’s not hard to do.
What you’ve explained in your earlier posts is that the on demand heater may make more sense in certain unique (and very unusual) circumstances, but it’s not “more efficient” on its own. It’s important to be clear about these things so that people can make the right decisions.
Bill
Nyle makes a modular heat pump where the heat pump is separate from the tank. The heat pump itself is a box that is about 30H x 16W x 8D that hangs on a wall.
Making sure there's space for mechanicals is the job of the architect. A tank takes about four square feet of floor space, there's no way your space is so expensive that the additional operating cost of resistance heat over its lifetime isn't going to justify dedicating four square feet to a heat pump.
Where does the nyle hang? I'm unclear what problem it solves if it's going inside.
. $1000 per square foot (closer to $1000 than. 100 at least), 10 square feet ( room 3x3 ) ($10k right there)
Tank plus access plus sound proofing plus ductwork $5k
Or an electric resistance tankless (corrected- typing from my phone)
Plus architect isn't paying energy bills and client just wants a rental unit or granny flat for short/medium term. Substantial myopia on energy costs
I'm not saying your wrong, but I'm just observing that what should be a no brainer is takes a lot of thinking, and convincing and then redesigning (didn't add those costs in)
Welcome to the left coast!!
How do you find room for the refrigerator? The washing machine? The television?
It sounds like the issue has nothing to do with the energy efficiency of the different types of water heaters, but rather the economy of the different water heater types when working around the site specific limitations you’re encountering. Those are not the same thing. It’s like saying a compact hybrid isn’t “efficient” because you can haul sheets of plywood with it.
Anyway, if you’re super tight on space, don’t have natural gas available, and don’t want noise either, then an electric on demand unit may be your best option under those specific circumstances. Sometimes you have to accept lower energy efficiency in those cases, due to other factors that affect the installation.
Bill
DC,
"Making sure there's space for mechanicals is the job of the architect"
I'm amazed it's even presented as an issue. I'd no more leave out space for a water heater than I would for a refrigerator or laundry.
Malcolm: what about the TV?
Thanks for the synthesis Bill
I'll offer an alternative one:
I am absolutely talking about efficiency, but not efficiency of the appliance, rather, the efficiency of the DHW system is in delivering hot water to the fixtures. And with those different appliances, there are cases where the DHW systems themselves are different for a given project using these different technology options, and that effects the Eta(DHWdelivery) metric.
(See all the articles from Gary Klein on this site.)
In the ideal world, architects would appreciate that a DHW tank is as important of an appliance as a television, or washing machine, or other appliance whose size correlates with its capacity to hold stuff we want to keep. But, alas, they've been trained long and hard that DHW is as easy as a little box that hangs on the wall and magically produces endless hot water (itself an ephemeral consumable-- in and out) and super efficiently. And it's just a hard row to hoe right now to fight for that space for a tank back from the architects, and then on top of that have the occupants accept the possible noise and cold air generated when HPWH are integrated inside of conditioned space.
Installation of HPWH inside conditioned space has been shown to be generally the most efficient place to put them, both for the unit, and for the conditioned space of the home generally, as the HPWH is effectively a cogeneration system for much of the year-- DHW production AND space cooling.
Here's an example Floorplan
https://snapadu.com/adu-plans/snap-adu-floor-plan-1br-1ba-493-sqft-30x16-long/
Luke,
Lazy floor plan. Even in small units there is enough room. Move the door to the washroom next to the bath. Lose the WIC and replace it with a conventional closet running the whole wide of the room with the tank water heater at one end.
Good commentw malcolm. You can say that, but when I say stuff like that, I get fired. Ask me how I know...
Lol
And climates where it would make sense to put a water heater outdoors are the ones where a heat pump is most compelling -- mild weather year-round. It's when you get to places that are seriously cold for long periods that you have to worry about the effect of temperature on COP.
HPWHs have their downsides as well, noise being one of them. There are obviously situations where it makes sense to use a tankless heater instead, but the reasons usually don't have anything to do with efficiency.
You can certainly come up with specific scenarios such as a little-used fixture located far from the main water heater. But, specific scenarios don't support a general claim that tankless water heaters are more efficient than HPWH.
"$1000 per square foot"
You must spec some very high-end finishes for your mechanical spaces!
Deleted
@frank
I claimed that these cases are common, and common enough to put asterisk on the general claim. forest for trees.
Of course I accept the general claim. I fight to make it real every day. But sometimes the math and physics say otherwise...
I wanted to create a solution space that was generous to the original post. And there is one, and it seems more useful of a response to present a model of thinking than to say 'nope terrible idea '
$1000 per square.ft.
It's costs all in. Not just construction. And yeah , it's ludicrous. Low income housing cost story this year reported $600k-$900k per unit in my county.
In my experience with affordable housing in the NY metro area, there's always room for an electric resistance WH. Many people forget that there are shorty water heaters that fit under a countertop. If there's an L shaped kitchen, the dead (or difficult or expensive to access) space in the inside corner is almost perfectly sized to fit a shorty. I've seen them on a strong shelf above the washer/dryer. In lieu of a stackable W/D, there are some pretty nice all-in-one units that would leave space above for a shorty. In two story homes, under the stairs has space for either a tall one or a shorty. And the issue of HPWH noise is really not all that bad. A well installed HPWH doesn't make any more noise than a typical central forced air furnace, and designers are still specifying those. Attics can sometimes be a good location for a HPWH, though I hate the potential for leaks up there. For a rental unit or granny flat, any of these solutions would work well. I agree with Malcolm that the architects are just being lazy. Space should be assigned for the WH just like all of the other appliances.
And with the affordable housing use case, I think the risks of simultaneous operation (as Bill described in #9) could rise considerably. If these floor plans are picked up for use in an affordable housing community, then it is pretty likely that people take showers at nearly the same time community wide. The local utility would HATE that.
The point is that there are so few use cases where the integrated battery option is a good one that it is unlikely anyone would mass-produce such a system. Unfortunately, the best option is probably still to continue to try to educate the architects as to why tankless WH is a bad idea. Heck, just the cost of the panel and service cable upgrades would pay for a HPWH.
Good thoughts Peter
There's a lot of pressure locally to bring back acceptance of those lo boys. Currently they're very difficult to get past code compliance, and so option is effectively off the table.
Luke,
Can you expand on why they are not seen as code compliant where you are?
California energy code. Prescriptive water heating requirements in sections 150.1 and 170.0 for SF and MF respectively. Require a minimum uef.2.0 water heater
For a large affordable housing complex, a central hot water system with a building-wide recirculating loop in the hallways would be ideal. This type of system would only work in multi story tower type structures, but would allow for the hot water heating system to be in a remote mechanical room. That would bring up “who pays” questions, but there are ways to meter hot water use if needed, and an HOA type organization could handle maintenance.
Bill
Thank you Bill and DCcontrarian (and others) for the detailed and thoughtful responses. I understand the full picture now and I will stick with an appropriately sized heat pump water heater.