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

Is a 350,000 BTU heater too big?

slateandall | Posted in Energy Efficiency and Durability on

This is our first winter in old house < 4000 sq ft.

Several people have commented that our 350,000 BTU heater (water heat) is crazy big for the house.

Is there any drawback to a big heater (other than the cost of initial install)?  Anything to watch out for going forward?

Souther New England, cold winters but not too cold.

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. Expert Member
    Dana Dorsett | | #1

    Yes, it's crazy big. How crazy?

    The heat load of reasonably tight 4000' insulated 2x4 framed house with clear glass storms and R20 fluff in the attic will come in about 80,000 BTU/hr @ 0F even without foundation insulation, about 60,000 BTU/hr with basement wall insulation. With no wall insulation and a lot of air leakage it may come in at or above 100,000 BTU/hr, but if it's that bad it's very cost effective to start fixing the building envelope. But even a 4000' tent doesn't have a heat load of 350,000 BTU/hr without leaving some windows or doors open.

    So let's say it's a fairly leaky sucker with almost no insulation and has a heat load of 100,000 BTU/hr at 70F indoors / 0F outdoors. A boiler with 300,000 BTU/hr would be able to heat the place at an outdoor temp of -140F.

    Are we expecting hell to freeze over or something?

    Or is it the idea that everybody sleeps with the windows open on the coldest night of the year?

    There's a double digit efficiency hit for a cast iron boiler for being 5x oversized for the heat load, even if there's enough radiation to emit 100% of the boiler's output on every zone. Even at 3x oversizing it's about a 10% hit. (See Table 3, and compare the steady state efficiency column to the 3x oversize factor column: ) If it's broken up into multiple zones the number of burn cycles soars and corresponding efficiency hit becomes even worse.

    This is why ASHRAE recommends an oversize factor of no more than 1.4x from the 99th percentile temperature heat load when specifying non-modulating heating equipment.

    With a modulating condensing boiler some of that is relieved but the modulation ranges aren't infinite, and gross oversizing can still bite in several ways. At condensing temperatures the radiation output rate is lower, and oversizing leads to the risk of short-cycling the boiler into lower efficiency (and shorter lifecycle). Some napkin math to run on that lives here:

    Since you don't have a heating history to compare fuel-use against heating degree-day data you can't put a stake in the ground on boiler output requirements, HIRE a third party such as a RESNET rater or professional engineer (and NOT an HVAC contractor) to run an Manual-J heat load calculation using aggressive assumptions (per the Manual) on things like R-values & air tightness.

    For a sanity check while shopping for a pro, try some online Manual-J tools such as or Those will usually oversize reality by a double-digit percentage, but not triple-digits. If sizing a cast iron boiler using those tools, don't oversize by more than 1.2x (if possible).

    As part of the analysis for the heating system run a zone-by-zone tally of the radiation. If much of the radiation is cast iron, use this as a guide:

    1. slateandall | | #7

      I tried those two websites for Manual-J—way above my technical ability. The Columbia Heating document is very interesting. I will calculate EDR for the house. That seems like a fun doable project. How do I know if I have gravity hot water or a pump?

      Oh, and there’s only one zone but radiators in some rooms barely heat up for some reason.

      1. Expert Member
        Dana Dorsett | | #8

        > How do I know if I have gravity hot water or a pump?

        Gravity hot water systems have very large pipes, usually 4-6" in diameter. If the system has a pump it's usually pretty obvious. It might look something like this:

        Is it possible that your system is steam?

        1. slateandall | | #9

          I found a flow control valve and a pump. There’s also an expansion tank. Pressure gauge seems to be a constant.

          Yes, two of the original cast iron pipes are 4”. They connect together to a single smaller new copper pipe that goes into the FCV/pump.

          There are 4 other cast iron pipes (smaller than 4”) that connect to a single newer copper pipe that goes to the boiler.

          I’m thinking this was converted from gravity to pump? The 2 thick cast iron pipes supply the radiators and the 4 smaller ones are returns?

          1. Expert Member
            Dana Dorsett | | #10

            It sounds very much like a converted gravity system. The large amount of thermal mass in the system water means it will never short-cycle, but the gross oversizing still isn't your friend.

            One of the guys in my office lives in a ~5000 square foot 19th century antique house that used to have a similar gravity feed boiler later converted to pumped. About a decade ago it was replaced with a (now unavailble) Freewatt system that had a 1kw co-generator married to a (still oversized for th heating load) 190,000 BTU/hr (output) modulating condensing boiler and an indirect water heater. Fuel use dropped by literally half, and in his case the cogenerator ran 24/7 for about 4-5 months out of the year knocking a huge chunk off the power bill too.

            At some point it's going to be worth getting rid of the beastie boiler and going with a modulating condensing boiler that's appropriately sized for the heating load. With an ancient converted gravity feed the standby and distribution losses are huge, but a fuel-use based load calculation would put a very firm upper bound on the actual heat load (probably still a way oversized number). See:


            Start saving your gas bills to track this.

            In addition to the fuel use load calculation it will be worth paying a third party who it NOT an HVAC contractor to run an aggressive Manual-J heat load calculation on the place. If there are any insulating/air-sealing/window upgrades planned, the Manual-J should be for the "after upgrades" condition of the house. In my area a proper Manual-J performed by a certified professional engineer or RESNET rater runs $500-1000, but getting this right will save you more than the fee on upfront system costs and efficiency down the road. I can't count how many times I've seen systems proposed by HVAC contractors that are still more than 2x oversized for the actual loads. Modulation ranges aren't infinite- right sizing the boiler is still important even on high mass single zone systems.

            Some pictures of the boiler & plumbing might provide other insights.

  2. Jon_R | | #2

    Consider adding a water tank to reduce short-cycling/efficiency loss.

    I'd improve insulation and air sealing now and then collect some data.

    1. Expert Member
      Dana Dorsett | | #4

      >Consider adding a water tank to reduce short-cycling/efficiency loss.

      With high mass radiators it probably wouldn't be short-cycling- don't go there unless it's ACTUALLY short cycling, which can be determined by observing the boiler's behavior, logging the lengths of burn times, and numbers of burns per day. If the burns are always 10 minutes or longer there is no further degradation attributable to short cycling. If they're ~5 minutes it can be about a 1% hit, but that grows rapidly with further reduction in burn times.

      A retrofit heat purging boiler control would pull back some of the efficiency lost to oversizing by reducing the average standby & distribution temperatures. System #3 in Table 3 of the Brookhaven National Lab took only a ~1% hit at 3x oversizing, primarily due to the heat purging control:

      A heat purging boiler control also lengthens the burn times but reduces the total accumulated burn time per day/year, and reduces the number of burn cycles.

  3. slateandall | | #3

    Thank you! I will take your advice and gather data and hire an independent pro, and look into water tank.

    The only data I have so far is from my Nest thermostat. For example, one morning when it was 49F the heater ran for 2 hours to raise temp from 63F to 70F (Likely overshot by a bit). The rest of the day it ran twice for 30 min each time to keep temp at 70F. The high that day was 58F.

    During those three hours of use I think something line 8 therms were consumed.

    1. Expert Member
      Dana Dorsett | | #5

      >The only data I have so far is from my Nest thermostat.

      A call for heat from the Nest isn't exactly the same as the burner actually running. The Nest only calls for heat from the boiler- it doesn't control the burner directly. During a 2 hour call for heat the burner may have been cycling on/off as the boiler.

      The rate of temperature recovery is more a function of the radiation size than the burner size. A million-BTU/hr boiler putting heat into radiation that can only emit 50,000 BTU/hr doesn't heat the place up any faster than a boiler that puts out 50,000 BTU/hr.

      How many zones, how much radiation per zone?

      Got a ZIP code?

  4. Deleted | | #6


Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |