When an 80gal Heat pump water heater isn’t enough
I’m replacing a 50 gal natural gas water heater and want to “go green”. I’m fairly certain that an 80gal heat pump water heater is in our future, but I’m not satisfied that it will be able to handle our hot water demand – our existing gas heater itself cannot keep up when our family of 6 cycles through showers.
So I’m resolved that, unless our habits change (they won’t), I need to supplement the HPWH…. but with what?
I know there are options out there like the eemax autobooster and homeadvantage II which are electric powered tankless heaters that can be piped serially after my HPWH, but I’m having a hard time figuring out “how much more capacity do I need” to size that correctly. I’m also curious if I should just bite the bullet and pipe a second HPWH in parallel instead? My basement is large enough that should work, but I’m a little concerned it would get too cold in the winter and I’d end up opening a heat vent to the space and wipeout whatever overall efficiency I thought I was gaining.
Has anyone explored this dilema and come to conclusions of their own on how to supplement and willing to share “why” they landed on those conclusions?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Matt Risinger used to advocate for a HPWH that feeds into a Marathon electric resistance water heater. The Marathon would only kick on if the HPWH can't keep up, and it's well insulated and made with that butyl liner that won't corrode like metal tanks. It is resistance, so you take a hit there but it's cheaper than a HPWH and it will only run when you need it and some minor standby losses. On the other hand, this is not unlike a single larger HPWH with electric backup, just know that it will rely on that backup when it needs to.
Using an electric tank in series after a HPWH instead of a tankless is a neat thought... It should hold hot water, having more available on demand for a longer period, but of course it would need to kick on every so often because of the standby losses, and those losses would never not be there vs the tankless which would only ever kick on when water was really in use and also not hot enough. Is it enough of a loss to justify going tankless of the supplemental and probably spending $$ on a panel upgrade? Hard telling (http://waterheatertimer.org/Pages/Calculate-standby-loss-electric-water-heater.html suggests maybe $24/year). This is probably close enough to a 50:50 decision
I probably have a decision... upgrade my electric panel to support a higher draw tankless as supplemental and be ready for an EV, too, someday, or pull the trigger on either a smaller tankless supplemental or a smaller electric tank. I just can't decide which of the three is "best".
If it were me, my preferences are in this order:
1) A single HPWH, work with the family to make the sizing acceptable
2) HPWH feeding into resistance water heater/storage tank
3) Gas tankless
4) Gas tank
5) Electric resistance tank
6) Electric resistance tankless
Please note I put electric resistance at the bottom of the list due to heavy amperage requirements that may require service upgrades but also poor COP of <= 1.0. In most areas of the country natural gas is still pretty cheap and may be cheaper to operate than an electric resistance model.
I'd look at the drain water heat recovery options:
https://www.greenbuildingadvisor.com/green-basics/drain-water-heat-recovery
Raising the temp of the incoming water means you'll be using less hot water and water that needs to be heated will be heated from a higher temp. This would be the option that is simple, durable, relatively cheap, and saves energy.
Lower flow shower heads are also simple, cheap, energy saving, and durable.
Another option would be to store hotter water and use a mixing valve to bring it down to safer levels. This uses more energy, but does expand the capacity of your heater.
You could also have an inline electric tank. The heat pump would heat the water, so ideally the tank wouldn't run much besides overcoming standby loss. Which is generally low.
In general, I think tankless water heaters should rarely be installed in a home and almost never in a green one. Their benefits are extremely limited: mostly, they save space. Their drawbacks are many: poor capacity, poor efficiency, high wiring expense, and huge electric service demands.
"In general, I think tankless water heaters should rarely be installed in a home and almost never in a green one."
Is this statement limited to electric tankless? Or do you feel the same way about gas tankless?
Gas too for sure. A gas tank can be as efficient and have a burner the same size, plus a tank has buffer capacity that a tankless does not.
drain water heat recovery would be a nice "1 and done" solution, and while I remain skeptical about how much it could really help (gotta think long-term savings for sure on this one), drain pipes don't really offer more than 12" of space to wrap around because of connections and the grade I sit at.
Already storing water at a higher temp, but yes - that's definitely a part of the solution.
I guess I wonder for a way to calculate the comparison of the standby loss of using an electric tank heater in series after the HPWH vs using the tankless which only comes on when necessary and not just because the water sat in a tank. I wonder, is the heat loss low enough that the HPWH would do most of the heating over a year?
Or maybe a recirc pump between the two tanks would avoid that? Doing so probably adds unnecessary complication, though.
Ah if you can't fit the DWHR unit, you're out of luck. That'd be a very high ROI change if you could route most of the shower water through it.
A standby loss of an electric tank is low, <1kwh/day is probably reasonable, so if the HPWH was feeding it water above its set point, you'd be paying $20-30ish dollars per year.
I have found that low-flow shower heads make a dramatic difference in our hot water usage. In particular, the High Sierra 1.25 GPM gives a good shower:
https://www.amazon.com/gp/product/B09VD22H8W
With those on all showers I'm able to survive with a 66 gallon water heater and a family of five.
The biggest drawback is that if you don't have a circulator it takes a while to get hot water. And a circulator isn't recommended with a HPWH.
If someone has a bathtub faucet with their shower it often has a higher flow rate than the shower head. We can run the tub faucet for ~10 seconds and then have warm water to the low flow shower head.
I wonder what the hot water gallon per hour rating of your gas water heater is, what temp it's set at, and if it's still performing optimally if it can't keep up. I would think an 80 gallon heat pump water heater set to store water at 140F+ with a mixing valve could do fairly well, particularly with an efficient shower nozzle. If you need over 100 gallons per hour it might be worth getting a Sanco2 83 gallon hpwh vs 2 HPWHs.
The gas heater I'm replacing has a 42.56gph recovery rating, which I think is just about double the recovery rate of the HPWH even when it fires all its extra heating elements. It does only have an 81 gallon first hour rating though as compared to the 96 gallon first hour rating of the A.O. Smith HPTS-80 I'm considering. Other than a tank leak it's developed, it seems to be running fine and "catching up" in a very reasonable amount of time after a series of long showers depleted it. We do already keep the heater hot in the tank: 160.
So... all things considered, maybe they're a close match (w/heating elements also firing) when thinking about the 45-90 minute timeframe? But I'm really trying to figure out the 10-15 year total cost of ownership. If I let the HPWH's internal heating elements supplement the heat pump component every/most times the water heater kicks on and probably would stay on until the tank is back up to full temp even if it wouldn't be needed for a half a day, I'm losing out on the efficiency the HP could offer. If I add an external tankless, then it only fires when water is being used and isn't hot enough... vs using the heating elements to heat a full tank and let it sit there when the HP could have done that itself, just over a longer time period.
The only "gotchas" I see are
1) potential need to upgrade my electric service depending on the kw rating of the supplemental device. The eemax autobooster only requires 7.5KW and a 30amp breaker, not bad. If I go up from there, though, to the 18KW HomeAdvantage II for example, which could probably handle even the HPWH going out of service, then I might need some panel work and/or service upgrade (working with an electrician on that). Neither are really expensive coming in at just under/over $350.
2) It means I'm not getting that nice energy efficiency of the HPWH for anything beyond the first tank's worth can put out before it would need to kick on again.
The Sanco2 83 is tempting, but we live in a climate with extended sub zero temps in the winter, and it seems like they haven't quite perfected the installs for those yet... reviews are suggesting I'd need to worry about heat tape, power outages, etc, to prevent piping from bursting. Hopefully they make some more advancements and that can be the replacement for this current round in another 10-15 years.
A 7.5kw inline heater can't do much. Even combined with 4.5kw from the HPWH, that's just 40,000 btu/hr, so if you have run out of preheated water, you'll only get 1 GPM at 80F lift. 18kw can only do 1.5GPM at 80F lift.
Storage seems a much easier path here.
True - but when most people would say a tank is "out", it's not really out-out... temp probably normalizes somewhere in the middle when we start complaining, right? When I consider what's probably coming out of the tank at that point, I'd likely be raising the temp by 30-40F to reach "shower" temp.
Storage (with an electric resistance element) being fed by the HPWH, I think, is definitely becoming the foggy winner - especially something with exceptionally low standby loss. The one last hump I cannot get over yet, is the cost of that storage compared to the smaller tankless. If that one can't cut it though, and I need to bump my electric service, then the choice is clear - storage.
While an 80-gallon HPWH is a greener option compared to a traditional gas water heater, it may still struggle to meet the high demand of a large family, particularly during peak times.
Department of Energy - How to Choose a Water Heater: https://www.energystar.gov/productfinder/product/certified-water-heaters/results a small world cup
+1 on raising storage temperature. I have read that Canada requires 175F to prevent Legionella. That temperature requires a thermostatic mixing valve downstream for safety.
Can you further insulate your tank to make up for increased heat losses?
I love the idea of drain water heat recovery, but it might be hard to retrofit.
The last option is tankless. We have Rinnai ( gas ) installed by a previous owner and no complaints, except that we would prefer to go all-electric.
bsawers,
Just to clarify:
THE ONTARIO BUILDING CODE | HOT WATER TEMPERATURE
9.31.6.1. Hot water temperature
(2) An electric storage-type service water heater shall have a minimum set storage temperature of 60°C.
Thanks for the clarification. 60C is only 140F so I’m wrong on the temperature. Do you happen to know if the rationale for 60C is Legionella?
Yes. The deliver temperature has a range:
(1) Where a hot water supply is required by Article 9.31.4.3., equipment shall be installed to provide to every dwelling unit an adequate supply of service hot water with a temperature range from 45°C to 60°C.
Bruckner,
If I were you, I'd be going through every faucet and shower head in the house to make sure they're all low flow. It's almost impossible to buy a new one today that isn't, but some have been "modified" by homeowners to deliver higher GPM.
Then I'd have a family meeting addressing the issue. Of course, some people (kids!) tend to take very long showers ad it's hard to change their ways. A 10 minute shower is 12.5 gallons. Times 6 is 75 gallons. Maybe some education on the environmental impact of the home using 75 gallons of HOT water a day just for showers, or 1.25 gallons PER MINUTE might help them understand better.
Having said all that, the Marathon Electric Tank as storage is the best idea IMO.
I may or may not be one of those homeowners looking to compensate for an unwillingness to fully go that route, but isn't fully ignorant of efficiency. A "meet in the middle" approach over a "all the way or no way" argument will go further for most people not already on board.
I'll check out the marathon - you're the second person to bring it up.
The efficiency of heat pumps is directly proportional to the temperature change. So increasing the thermostat setting affects efficiency rather dramatically. If your incoming water is 60F, 120F is 60 degrees of lift and 140F is 80 degrees, a 33% increase and correspondingly a 33% decrease in efficiency.
Considering that the whole point of a heat pump is the improved efficiency that seems counterproductive.
Since a water heater typically only runs a couple hours a day, it seems that what you want is more storage capacity and longer run times.
A higher lift is less efficient no doubt but you are excluding the fact that less water is being heated. To get 2 GPM of mixed 100F using incoming 60F and a 140F tank, you need 1gpm of hot. To get the same 2 GPM of mixed 100F using a 120F tank, you need 1.34 GPM.
But if you can make and store more water made at a COP of say, 4.0, that is overall more efficient than cranking the temp up and reducing COP to say, 2.7 for a smaller volume. The standby losses are so small that it does not drag down the efficiency by much. Storage is the answer to your problem, as many of us have already suggested above.
Agreed but storage costs money. If an 80G tank at 140F is sufficient than 80G + 40G at 120F seems frivolous. Making 140F vs 120F doesn’t take that much more energy.
But at the end of the day you need the same number of BTU's regardless of the temperature of the tank. To get 2GPM at 100F from 60F water, you need 40,000 BTU/hr; if you need it for 15 minutes you need 10,000 BTU. If you're heating the water to 140F first, you use 33% more energy than if you heat it to 120F first, to get those 10,000 BTU.
I think you’re still missing the mark. Say you need 100G per day of water at 100F. That’s the total demand.
Now you can do it at 120F:
Incoming 60F, you need 2 120F gallons for every 1 60F gallon. So the daily needs are 66 gallons heated from 60F to 120F, 9.7kwh.
At 140F, you need 1 140F gallon to 1 60F gallon. At 50 total G, that’s 9.8 kWh. Same 100 G at 100F.
I will admit that absolutely a heat pump is less efficient getting to 140F vs 120F, because they all are less efficient as the gradient increases. But it’s a smaller difference. Standby losses will also be higher.
Your math is a little off, probably rounding error, I get the same 9.76 kWh whether the water is at 120F or 140F, we both seem to agree on that. That's not even a factor in my calculation.
All of the 33% drop in efficiency I'm claiming is coming from the fact that the COP of the heat pump is lower at 140F than it is at 120F. I claim the COP is 33% lower. The reason I claim 33% is that the temperature delta is 33% higher, and the heat pumps I've looked at have a linear COP vs delta curve -- it's a straight line.
Interesting - thank you for sharing. I don't think I've come across anyone explaining the efficiency calculations would be different considering different target temps/lifts. Can you provide a source where I can read more about this?
The Rheem HPWH is 4200 BTU/hr. That equates to 134 gallons of hot water per day at a 90F rise. This should be reasonable for 6 people doing 1 shower/day for normal situations. I think the problem you have now is that the short-term demand of 6 people showering at the same time drains the storage capacity. In short, the daily recovery should be enough if you can smooth over the demand. Therefore it seems like adding a storage/electric resistance tank would make the most sense.
If you have extensive hot water consumption above this, such as baths, 30 minute showers, rain spa showers, handwashing dishes for 30 mins a day etc. - then I would consider getting a second HPWH for maximum savings.
We have an 80 gallon hpwh that serves our family of 7. I had a mixer valve installed on the hot side of the water tank to allow me to set the output temperature by mixing in cold water. The water heater is set to 135F and the mixer valve is set to about 113F delivery temperature. I have power monitoring on the water heater so I know exactly what it is doing. It is set to hybrid mode so the heat pump works almost all the time, only during periods of high demand do the elements kick on. The elements only work on Saturday nights. That's our whole family bath/shower night before church on Sunday. Everyone is on a different shower schedule throughout the week. There are 2 hot baths that get taken and 4 showers. The hot water is noticeably cooler for my shower since I'm the last, but it's fine and not cold. If the young children took showers instead of baths it would probably be better for me. I also have a 50 gallon electric water heater as a preheater (set to 120F) to the 80 gallon that I can turn on if I want to be certain that I'm going to have a hot shower, or if my wife wants to indulge in a bath in the 110 gallon master tub. This system works very well for us. The house is on well water. The incoming water temp is about 55F.
Oh and on Saturday night there's the 2 dishwashers running, my wife or I washing a bunch of large dishes by hand, and an inevitable load of laundry in the washer which is putting demand on the hot water. The water heaters are in our finished and heated basement closed off from the rest of the space by a pair of louvered doors. We have 2.5 gpm shower heads installed but the flow meter reads 2.2gpm when they are in use.
Thank you for sharing. Interesting - you're the first person to suggest a "preheater". Why did you choose that order?
Yeah, I'd go the other order. Use the conventional electric as a "post-heater" where the elements cover the standby losses and when the HPWH can't keep up.
It also allows you to have a circulator.
How would you introduce a circulator? Between the two tanks? How does it help/how is it controlled?
The circulator piping would go between the most distant hot water-using fixture and the intake of the conventional tank. The easiest way to control a pump is to have it run continuously. There are various energy-saving ways of having the pump run less but none is completely satisfactory.
Also -- The reason you can't use a circulator with a heat pump water heater is probably also a reason that putting a conventional tank in front isn't a good idea.
The efficiency of heat pumps depends in great measure upon the temperature difference between the heat source and the heat sink. Heat pump water heaters stratify the water more than regular heat pumps, so that the water being heated is as cold as possible. Introducing hot water into the supply pipe of a HPWH kills the efficiency.
My house is 100% off grid so we have to be pretty conscious of our power usage during the winter. Having it be a preheater gives me the most control over our power usage and keeps the HPWH and thermostatic valve in total control of the outgoing temperature. I wanted to give my wife a normal lifestyle despite being off grid so this lets me manage things without any bother to her, and if I'm not around the system always works fine.
If you go the other way you get the maximum out of the HPWH, the thermostat on the conventional only kicks on when the HPWH can't keep up and only to the extent it can't keep up. If there are times when you'd rather have cold water than use the power to heat it resistively, you can just shut off the breaker on the conventional.
Deleted
Reply to #37.
I agree that COP is lower at 140F vs 120F, because with an ambient air temp of 70F (not incoming water), the delta is higher. But I think it’s different than space heating: with water heating you’re heating the whole thing. So the COPs should be identical from incoming temp to 120F. Only from 120F to 140F would there be a difference in efficiency. So 75% the same efficiency, 25% would be 70% as efficient or whatever.
You're absolutely right that the cold side of the heat pump is the air in the room, not the water in the tank. Which actually makes the decline steeper, the difference between 120F to 70F and 140F to 70F is 70/50 or 140%.
Most HPWH's only run when the water is close to set point. If they're in hybrid mode, if the water gets too far below setpoint they switch into resistance mode. So when the heat pump is running it just about always sees almost the full delta. I'd also argue that a higher setpoint means the tank recovers more slowly -- the BTU output of the heat falls as the temperature delta increases -- which means more time spent in resistance mode.