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

Residential Electric Load – Which multiplier w/Tankless/EV? 0.75 or 0.4 or…?

jamesboris | Posted in Mechanicals on

I know how to perform a residential load calculation using either: -(a) the Standard Method, which includes a 0.75 multiplier for fixed appliances (so, a 4500W water heater would be treated as adding 3375W to the residence’s demand), or -(b) the Optional Method, which includes a 0.4 multiplier.

Which multiplier should be used with electric tankless water heaters or EV chargers? The I’ve read that some AHJs require including 100% of the nameplate wattage. One Mike Holt thread reported an AHJ treating these things as continuous loads, with a 125% multiplier. Others have arbitrary additors (in one CA county, you just add 15k… regardless of nameplate wattage!). My nearby AHJs have shrugged.

Stiebel Eltron’s electrical dept said: “A tankless load is no different than any other load. It falls into the 10k+40% category. It’s used very very occasionally.”


Is there an NEC code section on point, or a general rule here? The choice makes extreme differences in load calculations. 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.


  1. charlie_sullivan | | #1

    NEC 625 is on EV charging and makes it explicit that it's a continuous load, not ifs ands or buts. 125%. And that makes sense because if you come home with 0% charge you might charge for 10 hours to recover, and absent an active load management system, that 10 hours could be any 10 hours of the day, including overlaps with any and all other loads.

    The good news is that NEC 625 is also explicit about load management systems being allowed as a means to ensure that your EV charging never leads to overloading your panel or service, such that as long as the other loads don't exceed the capacity. And there are more products than the used to be for that: Emporia and Wallbox being the leading candidates in the US market.

    As for tankless, I don't know what's proper as far as code calculations, but I would recommend being conservative, just because it's such a large load that you can't count on a lot of diversity to average things out. It's also usually simply a really bad idea, chosen by people who have been mislead by proponents who don't understand physics very well. It is not significantly better efficiency than a tank electric water heating system for efficiency, so you are creating a peak demand headache for no benefit, while also wasting money and effort that could be better spent on a heat pump water heater.

    1. jamesboris | | #3

      Thanks so much for the EV charging info Charlie, will start digging into those options.

      I agree that electric tankless is a bad idea in almost all cases... but in a case of low demand (one shower, one washer, one sink), limited space, easy access to re-wiring, and PV capability, I think it can make sense. I think a 40% multiplier is unsafe for a 24kW appliance that can and will, occasionally, actually spike to 24kW. But is 75% safe? 91.73%? It feels like a judgment call that could only be answered forensically, case by case, which is annoying haha.

      1. Expert Member
        Akos | | #4

        You might not have it right now but a tankless could a be used to fill a hot tub or even a larger tub. At point it is a continuous load and would treat it as such.

        A better option if you are tight on space is to use a small tank with a smaller inline tankless unit downstream sized to supply a single shower. This way the tankless only needs to run if the water tank runs out, which should not happen often. Tankless are not zero maintaince especially if you have hard water, lower duty cycle means you don't have to descale them that often.

        1. jamesboris | | #6

          That makes sense. And if I understand correctly... even filling a regular tub, with a ~7 GPM spout, would kick a 24k (or even 29k) tankless to maximum wattage, no?

          1. paul_wiedefeld | | #8

            7gpm x 8.34 lbs/gallon of water x 60 degree rise x 60min/hr = 210,000 btu or 61kw. 255 amps at 240 V. Absolutely massive spike! Tanks are so much better at heating water.

          2. Expert Member
            BILL WICHERS | | #9

            In reply to post #8:

            No, a 24kw tankless won't ever draw more than 24kw. What will happen is that with a flow that exceeds the tankless' unit's capacity to heat the water, or if the entering water is too cold and exceeds the allowable thermal delta, the tankless heater will go to it's maximum RATED power (in this case, that would be 24kw), and the output water coming out of the unit will be BELOW the setpoint of the unit. What is happening is that the heater is putting in the maximum amount of heat that it can, but it's not enough to heat the water far enough, so the resulting water is not as hot as you'd like.

            What your calculation is showing is how much energy would be NEEDED to head 7GPM of water by 60 degrees, and it's more than the 24kw heater can provide. You would need a 61kw heater to be able to get the water at the temperature you wanted at that flow rate.


  2. Cafferatak2 | | #2
  3. Expert Member
    BILL WICHERS | | #5

    If you have local code issues, then you have to size things per your LOCAL code. Municipalities are usually free to amend codes to be more stringent than the national code.

    EV chargers are CONTINUOUS loads. The code defines a "continuous" load as a load of 3 hours or more duration, which certainly includes EV chargers in typical usage. EV chargers typically apply the "125%" (usually known as the "80% rule", which is the same multiplier, but described in the opposite way), so you might have a charger capable of delivering 32A (about 7.7kW on a 240v circuit), but it will be fed from a 40A circuit -- which is necassary due to the 80% rule. In this case, you would size the load on the service based on the "continuous portion" of the overall load, which would be that 32A rating for the charger, not the size of the breaker feeding it.

    For an electric on-demand water heater, I would size that as a continuous load too regardless of code rules that may allow any demand factors. The reason for this is that that one, single, MASSIVE load of the water could easily "click on" and put your entire service over the limit since it's likely to be such a large percentage of your overall service capacity. I would size the service using the full nameplate rating of the water heater in this case.

    Note that sometimes you can tap those large on-demand electric water heaters directly off of your meter using a seperate disconnect in some cases, which can simplify the requirements for the main panel. Code allows up to "six mains" for a structure, so a configuration as I described would only be two, which is OK. I would run this by your local inspector first though since not all will be happy with this arrangement, and you're technically supposed to have lugs in the meter can that are rated for two wires too. The next option is to feed a commercial style main panel from the meter, then use that panel to feed the large loads, so you'd have a heavy feed from the meter to the large panel (maybe a 400A service), and that large panel would have 200A breaker for a main panel for smaller branch circuits, and a 100A (or whatever is needed) for the water heater. You could feed an EV charger from this large panel too. By "large panel" here I mean a panel that takes larger frame size circuit breakers so that you can put in multiple 200+ amp breakers, which isn't possible in normal residential panels.

    BTW, if you go with the "large panel", a common breaker size is 225A, which is a nice fit for 4/0 wire feeding a distribution panel with a 250A busbar. That's what I usually do with commercial projects, because all of the math for the feeders works out nicely, and you end up with a little more wiggle room in that distribution panel.


    1. jamesboris | | #7

      Hi Bill, always a pleasure to read your posts. I've learned a lot! Some followups:

      "I would size that as a continuous load too regardless of code rules that may allow any demand factors... I would size the service using the full nameplate rating of the water heater in this case"

      Continuous loads are sized at 125% of the rating. It sounds like you're suggesting 100% though, right?

      RE: feeding straight from the meter. That's all good to know and will come in handy at times. My projects are really all over the place. I do remodels and repairs, but try not to be a hack. The POCOs around here are all over the place -- very solar-friendly in Austin, super backwards east of it, etc. In most cases, it's easier to just have the utility to upgrade to 400A.

      Just for the hell of it, here's the calc for this project... the WH makes a huge difference as you can see... but you can also see, I make a lot of conservative assumptions:
      In this particular case, there is already 200A service. Going to 400A is impractical for off-topic reasons. So I'm trying to figure out what size tankless makes sense with 200A. This is the Optional method:

      10k general
      + 0.4 x (a+b+c) = 10.12k
      a) (# of small appliance and laundry branch circuits* x 1500) = 12k
      --*Let’s say 8... 2 bath (washer + hair dryer), 2 kitchen (toaster oven + insta-pot), 1 power tool, 1 well pump, 2 misc (notice I'm building in a lot of safety here)
      b) 3W x 500 SF for general lighting/receptacles = 1.5k
      c) 5.2k dryer + 4.4k range + 0.3k hood + 1.9k dishwasher = 11.8k
      +3.5k HVAC
      -RUNNING TOTAL, not including Water Heater: 23620 W.

      Pretty low. Now let's add the WH.

      -w/24k WH load x 100%: 47620 W… / 230 = 207A
      -w/24k x 40%: 33200 W… / 230 = 144A
      -w/24kx 75%: 41620 W… / 230 = 181A

      My gut is "40% can't possibly be safe, no matter what Stiebel says. All someone has to do is fill a bathtub, and it's 100% demand... on the other hand... it's a small house, maybe just don't use your clothes washer or dryer when filling your tub? Residential is just tricky like that, it often really depends on specifics of people's habits. Sometimes it's hard to try to future-proof beyond a certain point with electric with the certain-uncertainties of PV, EV chargers, TX's electric grid collapsing, batteries, etc... to say "hey just go to 400A and run huge wires to your mechanical area" isn't always feasible...

      1. Expert Member
        BILL WICHERS | | #10

        Your assumption here is that all that stuff is going to be running at the same time. That's very unlikely. You anticipate two people drying their hair, while running two toaster ovens and two insta pots, all at the same time, while running the clothes dryer and dishwasher and the range -- and with the range at full capacity, that means all four burners AND the oven going at the SAME time. During that hectic time, the occupants of that house have ALL of their lights on too...

        How likely do you think that scenario is to play out in reality? I'd say *very* low likelihood. You'd be in a sort of berzerk comedy sketch of a house with people madly running about with half dry hair back to the kitchen when some timer beeps, tripping over the laundry hamper on the way and tipping over the insta pot. It's just not a realistic scenario to have all that stuff going on at the same time. That's why you have 'demand factors' to reduce the "add it all up" demand down to something closer to the "here's what will actually be running at any given time" demand. I would say that 40% is probably pretty conservative in practice.

        As a more realistic example, lets say someone just got out of the shower and is drying their hair, and their spouse is in the kitchen with two pots on and something in the toaster oven. Maybe the kids are watching TV, so no one is in the second kitchen. The washer is running, but not the dryer (because the washer isn't done yet), the power tools are put away, that's why someone went to the shower. The insta pot finished earlier, the dishwasher isn't needed because dinner hasn't started yet, and the shower is over so the well pump stopped running. I'll give you that maybe they left all the lights on :-) You're probably well under HALF of that 23.6kW load you added up.

        You size a circuit based on 125% of the continuous part of the load, which is the same as "derating" a circuit by 80%. The reason for this is to keep circuit breakers under 80% loaded in continuous service, which is how they are rated. You size some other things based on 100% of the continuous load, such as loads used for demand factor calculations, since you don't have to derate there. Hopefully that makes sense and answers your question.

        Note that a "400A" service is really only 320A (80% of 400A is 320A) too. I do prefer a commercial-style arrangement with a central distribution panel fed from the meter, and then "big" breakers in that panel feeding branch circuit panels, and the ocassional heavy load like an on-demand water heater or possible an EV charger. Some problems with solar load calcs can also be solved this way. ALL commercial projects are designed like this, unless they are so small they only need a single panel. It's a good idea to plan 400A services like this, and offers a lot more flexibility over the "two 200A panel" layout that is more commonly used.

        BTW, You're not going to make any difference with the Texas grid regardless of how you wire a house, so don't worry about that one :-)


        1. jamesboris | | #11

          All makes sense, thanks Bill -- however, I think your statement here is generally correct, but does not apply to the Stiebel Tempra Plus (and maybe some other fancy ones): "... with a flow that exceeds the tankless' unit's capacity to heat the water, or if the entering water is too cold and exceeds the allowable thermal delta, the tankless heater will go to it's maximum RATED power (in this case, that would be 24kw), and the output water coming out of the unit will be BELOW the setpoint of the unit"

          Stiebel told me that a Tempra Plus unit will never send out water below its setpoint. It will instead modulate and continue to provide only the max it can at its setpoint. So, under ideal conditions (not a crazy high rise needed, etc.), a 4 GPM unit will give your tub the temperature of water it wants... but only 4 GPM. So your tub spout won't be at its (e.g.) 7 GPM capacity, but the water won't be lukewarm. Though I do wonder if 4 GPM would fill an 80-gal tub quickly enough for that first 20 gallons to not have cooled down and... yeah. Limited space (so no room for a big tank) + bathtub is tricky.

          1. charlie_sullivan | | #15

            So it actually has a valve the partially closes to throttle back the flow to 4 GPM? Sounds like an expensive way to reduce reliability, but maybe a preferable mode of operation.

  4. aunsafe2015 | | #12

    On the topic of electric tankless, is there any reliable source explaining whether electric tankless is considered to be "more efficient" than tank heat pump water heaters? For an average family of 4, which would use more energy per year? I assume heat pump < electric tankless < electric tank?

    1. Expert Member
      BILL WICHERS | | #13

      The only advantage any tankless water heaters have in terms of efficiency is that they have zero standby losses, since they don't have a tank that they maintain at temperature. Aside from that, there is no efficiency gain.

      Assuming you mean "efficiency" is how much hot water you get out per unit input energy, a heat pump will generally beat everything, since it is moving heat instead of making it, so it doesn't use as much energy to raise water the same amount as something else would use. Heat pump water heaters do have standby losses, but those losses are low. The only case where I could see an on-demand water heater having an efficiency advantage over a heat pump water heater would be in a seldom occupied structure, such as a remote cabin that is only used a week or two out of the year. Anywhere else, the heat pump water heater is likely to have the overall efficiency advantage.


    2. paul_wiedefeld | | #14

      I mean the Uniform Energy Factor for a Rheem heat pump is 3.75 and for a tankless electric it is .99. A tank electric is .93. So a heat pump outperforms both by a huge margin, while the tankless barely exceeds the tank. Tankless is mostly just marketing fluff unless you're short on space.

      1. charlie_sullivan | | #16

        And not all tankless are that high. They can be 0.94, for example. Not that that matters much since the main story is that heat pump is much better.

        1. Expert Member
          BILL WICHERS | | #18

          The only reason I ever see to use a tankless is if you have a need for very large volumes of hot water, such as lots of people who all want to take showers at around the same time, or if you have a very large bathtub or jacuzzi you need to fill. The only advantage a tankless really offers is that you never run out of hot water with a tankless. For efficiency, a tankless really doesn't gain you much, and tankless units need more maintenance and typically have shorter lifespans. That's all in addition to the downsides of large utility supplies of electricity or gas to power tankless units.

          I'm not much of a fan of tankless water heaters...


          1. paul_wiedefeld | | #19

            For these situations, a high output tank would do just as well, or even better. I agree with you, the hype greatly exceeds the value. Incredible marketing job.

    3. charlie_sullivan | | #17

      If you want a source to confirm that heat pumps use vastly less electricity than electric tankless, and that that tankless and regular tank are pretty similar, you can look up examples on

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |