GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Two DHW tanks in parallel, series or on two separate loops?

maxwell_mcgee | Posted in Mechanicals on

Hi folks,

Curious if there’s an optimal design for hot water distribution.

My house has two clusters of wet areas — one on the South side of the house covering the Primary Bathroom, a second bedroom ensuite and a basement guest room. There’s a second cluster of wet areas on the North side of the house (third and fourth bedroom en suites, laundry room, powder room). Plus there’s also a kitchen which isn’t particularly close to anything… 

I haven’t done the exact math, but approximately speaking, Gary Klein’s hot water rectangle (https://www.greenbuildingadvisor.com/article/the-hot-water-rectangle) would be something like 20% if I encapsulated the wet areas on just the North side or just the South side of the house, but would be around 70% if I tried to to build a common distribution loop for all the wet areas.

I’m committed to the idea of having two Heat Pump Hot Water tanks — both for capacity and resiliency reasons (want to still be able to shower even if one tank breaks down). So let’s take that for granted.

If we focus on the two wet zones, is there a preferred approach between:

1) A single distribution loop with the two HPHW tanks connected in series
2) A single distribution loop with the two HPHW tanks connected in parallel
3) Two separate distribution loops — one North, one South.

And does the answer change at all if we decide to utilize a hot water recirculation loop?

And then for the kitchen which is far from everything, does it make any sense to use something like a small point of use hot water tank ( https://www.homedepot.ca/product/rheem-mini-tank-120-volt-4-gallon-compact-point-of-use-electric-water-heater/100168341 ) fed with a hot water line from one of the other tanks? This way we can have hot water pretty quickly when washing hands, but also get continuous hot water if/when needed (e.g., a long session of manual dishwashing)

Thanks!

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.

Replies

  1. Expert Member
    DCcontrarian | | #1

    I've never seen a heat pump where the manufacturer doesn't recommend against recirculation.

    I'd go with two separate loops. For resiliency put in valves that allow you to connect the two loops and isolate either heater.

    I had parallel tanks in an old house. Although the plumber had made a valiant effort to plumb then with exactly symmetrical piping, one always had slightly higher flow than the other. In periods of high use the higher-flow one would run out first and when it did it would supply cold water that would mix with the hot water from the other tank. Since the now-cold tank had higher flow the water would be mostly cold mixed with hot and decidedly lukewarm. It did not at all satisfy the goal of having a longer hot water supply. The higher-flow tank also wore out quite a bit sooner.

    I've also lived in houses with serial tanks and that does seem to provide double the hot water supply.

    1. maxwell_mcgee | | #2

      Thanks, very helpful. I like the series with valves to isolate into two separate loops idea. Potentially the best of both worlds.

      What's the issue with using a recirc loop with a HPHW? I haven't heard about this -- would love to learn more.

      1. Expert Member
        DCcontrarian | | #4

        I'm not sure, but I'll guess. Recirc causes higher standby losses, which means the water heater has to cycle on and off more often. Heat pumps don't like to cycle.

      2. graham78 | | #5

        With Rheem, the incoming water temp from the recirc pump can be too high activate a heating cycle. Tech bulletin ins attached

        1. maxwell_mcgee | | #6

          Super helpful! Thanks for sharing.

          I wonder then...if I'm planning on generating my own electricity via rooftop PV, should I just stick to simpler electric resistance hot-water tanks instead?

          Any thoughts?

        2. Expert Member
          DCcontrarian | | #7

          Thanks for that, it's interesting. The language is guarded but the impression I get is that the tank on a HPWH is more temperature-stratified than normal, and they rely on that stratification to have a good temperature delta, delta is everything with heat pumps. A circulator pump mixes the water and ruins the stratification.

        3. andyfrog | | #12

          Does this also apply to two HWPH in series?

  2. jamesboris | | #3

    For your kitchen, a small point of use is cheap and DIY-able for almost anyone. I've got one of those small Stiebels and the build quality seems good. Only had it 2 years though.

  3. Expert Member
    BILL WICHERS | | #8

    The usual way to extend capacity is to connect the two water heaters in series. As DC mentioned though, that may be an issue with heat pumps. You might want to check with the manufacturer before trying that. Obviously two seperate hot water circuits avoids this issue.

    The usual way to deal with standby losses for hot water recirculating loops is to use a small ("coupla gallon" range) electric tank-type water heater, and plumb the recirculating loop in and out of that, then supply the "cold" water to that little water heater from the big water heater. The end result is that the little water heater deals with ALL of the losses from the recirculating loop, but any use of hot water is sourced through, and is primarily heated by, the main "big" water heater. This setup works well, and is also used with on demand water heaters when there is a need for a recirculating loop.

    bill

    1. Expert Member
      DCcontrarian | | #10

      I didn't think about this, but with heat pump heaters if you put two tanks in series the downstream one is going to have the same problem as it would with a circulator, where the incoming water is warm but not necessarily hot, and efficiency suffers because there isn't much of a delta.

      1. Expert Member
        BILL WICHERS | | #11

        Yep, you'd have that issue, with the downstream heater not really kicking in until the first one ran out of hot water. I'd mitigate that issue by setting the downstream water heater's temperature setpoint slightly below that of the first unit. In that way, the downstream heater will mostly run only to deal with it's own standby losses, until the upstream unit runs out of hot water when demand is high. Once the upstream unit runs out of hot water, the second one will start to run more to keep up with demand. This might not be optimal overall, but it's probably the best you can do while keeping things simple.

        Bill

  4. Expert Member
    DCcontrarian | | #9

    With two heaters in series, the upstream one does most of the work, the downstream one only kicks on when the upstream one is exhausted and starts providing cold water to the downstream one. So one approach would be to put a heat pump upstream and a resistance heater downstream. You would then be safe to run a circulator so long as it only included water from the downstream heater. All of the standby losses from the circulator would be borne by the downstream heater.

    There's no reason the tanks have to be the same size, you could even make the downstream tank a small point-of-use sized tank.

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |