Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

gravity system upgrade

No problem with the mistake--we ALL make them--but now there's a different quandry.

How much radiation is installed (preferably in sq.ft. of EDR)?

It now sounds as if the heat loss in the "newly insulated" house is very close to the 180F output of the radiation. That sounds <I>very</I> odd for a gravity conversion unless a lot of radiation has been removed.

At 130 mbh load using standard assumptions, that's 13 gpm required flow and you'd need at least 1 1/4" main piping.

While your heating balance on the ground floor may be just fine now, it's impossible to calculate what will happen when you replace the mains. Unless there is some utterly compelling reason to replace the mains, I'd leave them be. The savings by not replacing the mains would more than pay for TRVs on the ground floor rads as well as the 2nd floor.


  • Eric_8
    Eric_8 Member Posts: 66
    circulator sizing for gravity system upgrade

    I'm upgrading a system that was piped as a gravity system but has been operating well with a single Series 100 1.B B&G pump for years. We have about 110 MBTU/HR of radiation (at 180, not including pickup losses. cast iron radiators), and an estimated 200 gallons of water in the existing system.

    Specs call for a fully condensing (Monitor Products MZ40) boiler, and complete replacement of headers in basement to implement a first and second floor zone. Once this is done, the only old piping that will exist is risers going directly to the radiators, that is there will be no horizontal runs of the old gravity system pipe. The existing risers are either 1.5 inch or 1.25 inch.

    Heat loss of the newly insulated structure is about 50% of the existing radiation (measured at 180), so we expect to be able to run this system at lower temperatures.

    A debate has developed as to whether circulators and piping should be sized to the expected heat load, or whether they should be sized in view of the fact that the original system was gravity feed and thus has larger pipes. The Hot Tech Topics (id 125 and id 22) give differing advice--id 125 says to up-size circulator on a gravity system, id 22 says to size it to the expected heat loss of the structure.

    Which approach seems best for the job as described? We would like to keep pipe sizes down, and avoid using higher flow than necessary as the condensing boiler will work best with larger delta T (i.e., lower return water temps).


  • I would highly recommend you look into TRVs (thermostatic radiator valves) for your radiators.

    Not only will this eliminate the possibility of "circulation problems" when you remove those huge mains but maintain the still relatively huge branch piping, but you'll get room-by-room "zoning". No wiring required and you'll only need a single circulator for the radiators.

    You say there is 110 mbh of radiation (@ 180F) but the heat loss of the "newly insulated" structure is about half.

    That means a heat loss of about 55 mbh.

    The MZ40 (with two 71 mbh stages) woud [seem] to be ridiculously oversized! By those numbers you'd never need the 2nd stage--EVER!

    The MZ condensing boiler line was one of the first to come to the U.S. and it has an exceptional reputation for reliability among contractors who have installed. Unfortunately though, the MZ series is perhaps the only residential condensing boiler that does not also offer input modulation.

    With a gravity conversion system, I would highly recommend using a condensing and modulating boiler. If TRVs are used, I would say use only a condensing/modulating boiler as the system will utterly thrive and you'll be truly amazed at the reduction in fuel consumption.

    If you want a truly first-class condensing/modulating boiler look into the Vitodens 200 by Viessmann. If the price is scary, there any many other choices in a great range of prices. But remember that with generous--not gigantic--main piping and TRVs a Vitodens 200 can often drive a gravity conversion system without any form of primary-secondary using only its built-in variable-speed circulator and the only required "accessories" for the system will be venting (must be Viessmann) and a differential pressure bypass valve.

    p.s. TRVs are also about the only reasonable way to get flow rates down to reasonable levels in a gravity conversion system. As you already know, this allows a higher delta-t which in turn translates to higher efficiency through a mod-con. If you drive the entire system via a single circulator (no primary-secondary) and use a reasonable reset curve, you are guaranteed conditions in the system that will maximize efficiency.

    Nearly all mod-cons can drive a TRVd system with a single circulator, but you do have to pay very careful attention and do some calculation to verify that flow will stay in the acceptable range for the boiler.
  • Brad White
    Brad White Member Posts: 2,393
    Lower Flow

    Pipe size and system volume has nothing to do with required pump capacity. Size the flow for the heat loss at a specified Delta-T. Actually, so long as you average water temperature is the same, (180-140 and 170-150 each have a 160F average) it is all the same output really.

    If as you say you have 200% of the required radiation, you are in an enviable position you well know. You can run your water at about 135 to 140 degrees average water temperature versus maybe the conventional 170 F. average of most hot water systems.

    If your heat loss is really 55 MBH (half of the 110 MBH worth of radiation), you can easily heat that with 5.5 GPM if properly distributed and that is at a 20 degree delta-T. A one-inch main or two 3/4" mains will do it nicely. Larger pipes only add volume really, they are not heat loss.

    I do have to ask though, while I like the 2-stage feature of the MZ-40 at 140 MBH input it is also grossly over-sized.

    The issue I see is, you really only need call it 6.0 GPM and your boiler can crank out about 12.0 GPM worth when both stages are firing.

    One stage at 70 MBH is less than a single MZ-25 at 94.5 MBH and I would submit that you will never need that second stage unless concurrent DHW is called for. Even with the MZ-25 you would not likely ever notice a loss of heating during that time, especially with high mass radiators.

    Why not a smaller mod-con rather than two over-sized stages? Curious.

    EDIT: Mike T. posted as I was lolly-gagging. What he said.
    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"

    -Ernie White, my Dad
  • Eric_8
    Eric_8 Member Posts: 66

    Hi Mike--thanks for your thoughts. I made one major mistake--the heat loss calculation is just shy of 120,000, not 55,000. I read off the wrong set of notes earlier..

    we're planning to use TRV on the top floor, but have found the lower floors to heat fairly evenly at all times of the day and thus weren't convinced that the effort and expense would be worth it--the thinking was that a good piping plan and a reset control would get us similar savings as the mod-con and TRVs used everywhere. It also seemed that the 71,000 stage 1 for the MZ40 was a good match for 80% or more of the heating season, so we fell on the side of the MZs great service history in lieu of other condensing units.

    Anyway, that's all history, and a little academics!

    The two main zones, when re-piped, will be just shy of 50,000 load each. Convention would suggest a 1" pipe (about 80,000 BTU/HR with a 20 degree delta-T--and we seek to have a larger delta T). If the only part of the "gravity" piping that is left are the risers, I wonder if going over 1" is necessary. (the flr 1 and flr 2 zones are, and will, both be piped direct, with very similar pipe runs to each radiator (i.e., the new headers run the length of the building, equal distances from all the radiators).

    I'm not sure I understand the theory calling for larger piping for the new basement headers in this case if 1" pipes can carry the heat load needed. What do you think?

    thanks again,
  • Eric_8
    Eric_8 Member Posts: 66
    revised heat loss number

    Brad, you're thinking more the way I am. Check out my second message--I made a mistake, heat loss is just under 120,000. Does that change any of your thinking?

  • Eric_8
    Eric_8 Member Posts: 66
    getting to the details..

    Wow, it's always so much more interesting when someone is paying attention!

    Yes, existing number seems low--I can't really explain it. If we use 170 BTU per sq ft of radiation, I guess it's about 650 sq ft. Sorry, I don't have the individual EDR per radiator to double check this number, but starting from the BTU output that I believe is correct, that's what it comes to.

    Two factors here:
    (1) I made my second mistake--the new heat loss calculation includes about 40,000 of new space (basement)--so "apples to apples", the newly insulated space, that has the existing 110,000 BTU of radiation, has about 80,000 BTU of expected heat loss.

    (2) Even so, one has to imagine that the original system must have run very hot on some days...that's all I can think of.

    So with these numbers, and zoning, the two zones that will feed the remaining parts of the existing system are about 40,000 each. Seems like a 1" pipe would do it. Should the existence of the old risers and cast iron radiators (see my second message, this being my third!) scare me into increasing the pipe size for these two basement headers do you think?

    best regards,

    PS: the headers are hung very low-they are coming out to make way for a basement renovation.
  • Brad White
    Brad White Member Posts: 2,393
    Hi Dave

    With 120 MBH heat loss, that does make the boiler more appropriately sized. I would still opt for a modulating versus a staged boiler but I will take the MZ-40 over a single-stage boiler any day.

    Pump away-

    By the way, back when it was gravity, the system had outdoor reset by default, did you know? On mild days the boiler fired less often and on colder days more often. You can see how you would get the appropriate inverted temperatures and the correct gentle flow depending on the outdoor temperature.

    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"

    -Ernie White, my Dad

  • Something sounds wrong with the heat loss estimates, especially the basement.

    If this is a full two-floor house (as opposed to 1 1/2 floors) there's almost no way for the 2nd floor to have the same loss as the 1st. More importantly it's virtually impossible for the basement to have the same heat loss as each of the floors above unless it's substantially above grade and has a number of leaky windows.

    I have no idea of your climate, but 650 sq.ft. EDR of radiation in a gravity system (presuming it's original) makes me think you have about 2,000 sq.ft. on the two main floors.

    All this leads me to believe that the smallest condensing/modulating boiler in most lines will be more than adequate for all three floors.

    If you don't use TRVs on the ground floor, make CERTAIN that the main piping serving it has effectively zero head loss at a HIGH flow rate of say 10 gpm. Add head loss in the mains and the zero head loss in the branches will almost certainly result in new heat balance problems.
This discussion has been closed.