GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter 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-story Victorian desperately needs a cost-efficient radiant floor heating source

ClevelandOhio | Posted in Mechanicals on

Greetings from Cleveland, Ohio, USA

I have been lurking her for a long time, finally, frustrated and running out of time ahead of the heating season I seek your help.

Thank you for allowing to share my heavy burden,



Back story: I was sold a house by a nefarious neighborhood development corporation nonprofit who is now over their heads with indictments and investigations by local and federal authorities. Long story short is that I was forced to rebuild a house or lose it. Originally, I bought this house with new paint, carpets, furnace, fixtures, and then basically was forced to rip everything out and rebuild it. With the new relevations steming from the courts — all this was done in the hopes that I would give up on the house. They did it over and over to many people. I learned everything as I went forward with the project. From digging out and pouring a new foundation to sistering just about every joist.  I decided that I may as well just get the house down to studs and rebuild it my way. The last five years of my life have been dedicated to doing just this — making me poorer, smarter, and meeting some amazing people who helped me along the way — some of which decided to stick around as my good friends.

The house: Circa 19teens crayon-box two story house down to studs with original 2×4 balloon construction furred out by 2×6 on 3 of 4 walls of the house. Ceilings are 10′ & 11′. Floors are 50/50 divided in  a 1.5″ thick glued and screwed sandwich of the original 100 year old oak t&g and plywood (for a later hardwood install) and for the future tiled areas the flooring sandwich consists of two glued and screwed 3/4″ plywood sheets and Bekotec-like panels+pex+screed sandwich.

My mechanicals are in back the second floor walk-in closet. Manual J calculations are asking for 100k heat, 12.5k latent, 50k cooling, 36.5k sensible. I have 6 loops of ~200′ of 1/2 O2 barrier pex per floor. The system is divided 50/50 between a 4-run-per-bay staple-up backed up all the way by omega shaped channel aluminum panels and Schluter Bekotec-type product for tiled areas.

The problem: For the last two months I have been attempting to ascertain which heat source should be driving my radiant heat flooring installation. I have been running from one boiler manufacturer to another and after reading user reviews and running back to hot water tank-type installation. More information I read on the topic and more reviews I read about my options the more scared I become to pull a trigger.

Recently my finances have taken a very dictatorial tone, unallowing me the means of retaining a qualified professional, and due to the aforementioned fears and frustrations — I am just thinking of throwing everything I learned on the topic out of the window, and temporarily throwing a cheap Rheem NG tank in a closed loop with an expansion tank, air eliminator, ‘smart’ circ, and a few valves and calling it a day; on better days I think of buying a HTC Light Duty.

Here is what Ohio Laws say about using a water heater for heating. I am actually not 100% sure that I understand the law completely, but I think I can use a hot water tank to heat my house witout also using it as a domestic hot water source


I have added the photos of my framing detail:


* Vinyl siding
* EPS 1/2″
* Asbestos shingles
* Wood lap siding
* 2×4 uninsulated cavity (old baloon frame)
* Air gap
* 2×6 cavity


I am not sure the correct way to describe my ceiling under the roof
Part of the old 2×4 framing has been widened by 2x6s

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. Jon_R | | #1

    Maybe a Weil-McLain AquaBalance (~$1K, 10:1 turn down ratio)?

  2. MattJF | | #2

    Whoa whoa, 100k btu/hour sounds like a very high heating load for a house that has effectively 2x6 walls. How big is this place? What are the windows/ and how much area?

    Can you post your manual J? Design temp should be 6F.

    I can’t help you with the boiler choice, but I can stumble through a manual J.

  3. MattJF | | #3

    So quickly read about radiant pipe layout...assuming you are at 12" spacing or less would, with 1200 linear feet of pipe per floor, would make the house 2400SF or smaller.

    A 30'x40'x22' tall conditioned space, R11 all six sides, on piers, U1.1 single pane wood framed windows at 15% area, 162 cfm infiltration/ventilation comes in around 70K Btu/H. Your load will be less if your house is any better in any of those categories.

  4. ClevelandOhio | | #4

    Thanks guys :)

    Jon -- I was browsing FurnaceCompare, and I realize that reviews there have a negative bias against all manufacturers, but after reading this -- I decided to steer clear:
    "Aqua Balance"
    "This unit is purchased by WM and rebranded. It's not American made because it's made in Italy. It has a poor reputation in Europe and if you search Ferroli Boilers, you'll understand why. My unit is constantly leaking at the heat exchanger. I've purchased new pieces of copper and gaskets and it doesn't stop leaking. Look elsewhere."

    Matt, I am deepening and finishing and heating my basement in the next six months. This may have thrown my calcs out of wack a bit, but not as much as fresh air. the calculator recomends 420cfm. At 163cfm it would be 10k less.

    1. Jon_R | | #5

      For what little it is worth, the first google link took me to
      which is quite positive, even from people who say they installed hundreds over many years. On the other hand, seems like paying twice as much should have some benefits. And that using a water heater designed for low duty cycles should have some drawbacks.

      I have no direct experience with boiler brands.

      1. ClevelandOhio | | #8

        After taking a closer look I feel like I can assume that 1/5 or 1/4 will have issues.

        Because things are tight -- I must make a choice if $1000 is worth a gamble.

        I apreciate your suggestion. Out of curiousity, do you have other suggestions based on the "paying twice as much should have some benefits" sentiment?

        Thanks Jon!

        1. Jon_R | | #9

          I'd call a dealer that sells both and ask for their pitch about the more expensive one (keeping in mind that they make more from selling it). It's a large numbers question, and for other products, I've seen 1% issues where everyone who has the issue posts about it - leading one to believe that it's a near 100% issue. A stainless steel heat exchanger makes sense to me.

          Maybe this:

          1. ClevelandOhio | | #12

            Thank you so much! I realize that a good boiler is a forgettable appliance -- one would be hard pressed finding many reviews on something that quietly chugs along in the basement, but man -- if it ever breaks... This is why I have stated earlier that I realize that reviews are with a negative bias.


            This guy seems to be a high volume contractor that did his best to be fair to the company -- makes me think if maybe he is worth listening to. Looks like they redesigned gaskets recently, but we wouldn't know if WM has old stock or not. Things like this occupy my thoughts...

            "We have fit around 90-100 27 and 32 C…
            We have fit around 90-100 27 and 32 C HE Modena,s over the past 6 years, They have caused me a lot of headaches !! Lets start with the positives, easy to install, small and great for when fitted above a worktop, and then the problems. At first seemed great, then we come back to service them and we are finding 4 and 5 joints on both heating and domestic water leaking inside. Endless main heat exchanger gaskets leaking condensate which they now have desighned a new gasket to combat this and have seen the evidence in more recent models. We have had a complete boiler replaced as it was so bad, no chemical was added after fitting it resulting in 2 years later another call put in for another fault but the warranty been knocked back and temporally cancelled due to water condition!! This was Ferroli Engineer/Contractor who replaced the boiler, they also left out the wireless receiver that was fitted in the original model by us and said that was nothing to do with us resulting in me having to go out again to re-connect and set up (2 min job).
            It all seams to be one sided, every time an engineer goes out to a job they have an habit of pulling your work to bits in front of the customer resulting in another panicked call by the customer to me. Another positive i need to mention is that the Rep Martin has over the past few years been very helpful and has stepped in on many occasion resolving isues with lost paperwork, un-registered warranties etc and dealing with warranty calls as getting through to Ferroli at one point was a nightmare!!
            This is an honest and fair review and only covers a very small amount of problems and issues and to be fair the company have improved and listened to issues raised by engineers so hopefully can only get better."

            Have you installed many of these without problems? It would make a difference if I heard that from you...

    2. MattJF | | #6

      The load calc came through in low resolution and I can't read it. Try two zoomed in pictures or a .pdf.

      I see some floor and basement numbers. Don't include floor losses if the basement is being treated as conditioned space. Don't include the basement if it is not conditioned.

      The 420cfm infiltration seems too high if you paid any attention to sealing things up as you went.

      1. ClevelandOhio | | #7

        Thanks Matt. I attached the pdf. I am a bit limited with what the program can do as far as flooring options.

        1. Jon_R | | #10

          Do fix the load calculation. 90000 CF in a 1900 sq ft house implies a ceiling height of 47'.
          And use correct insulation values. But with 10:1 modulation range, I wouldn't spend much time on getting load perfectly accurate to select a boiler.

          1. ClevelandOhio | | #11

            Yikes! Haha....

            Yes, companies definitelly figured out how to combat the oversizing problems with the turn down ratios...

        2. Expert Member
          Dana Dorsett | | #21

          So, 18.6K of basement wall losses? Really? Are you really putting R30 between the floor joists rather than insulating the foundation walls?

          Where does the 14.3K of infiltration come from?

          And 28.7K of "fresh air" load? Planning on sleeping with the windows open or something? Just huge load attibuted "fresh air" is probably close to the total REAL design load of a tight 1900' house with an insulated foundation.

          It's time to stop screwing around and either get a REAL Manual-J tool and training on how to use it, or to just hire that out to a competent P.E.. The last thing you should be doing is trying to spec the radiant floor when it's clear that the load and water temp requirements are effectively a complete unknown. If the basement is as HORRIFICALLY lossy as the LoadCalc indicates, the radiant floor money will be better spent on air sealing and insulating the foundation walls.

          I'm not sure if Nate Adams does his own Manual-Js or pays pros to do it, but he's reasonably local and can at least advise you on where to get those services.

  5. Expert Member
    Akos | | #13

    I've had good luck with condensing tankless (Takagi or Rannai) heaters for heat and DHW. Most have a minimum fire around 15000Btu, check that your zones are sized to prevent cycling.

    Provided your water is not too hard and if your codes allow for open setup, it is the simplest setup. The heaters are pretty high pressure drop, just have to size your pump accordingly.

    One side benefit of the open system is that you can tack on your hot water recirc as a dummy zone.

    1. ClevelandOhio | | #17

      Before I dove into my research couple of months ago, Takagi was my #1 choice. As I started reading customer reviews I realized that if anything fails in the boiler, Takagi will make you retain an HVAC expert who then must take the unit COMPLETELY apart and shoot the photos of the completely disassembled unit. They do not reimburse you for the service calls.

      But if you can sweet-talk Rinnai customer service they will send a guy out to you free of charge, I think, for the first year. Rinnai customer service is apparently really bad, but it seems like, if you are patient, you can get a resolution. I am a patient person.

      I am pretty set on the closed loop system because I want to lengthen the life of equipment. I will do 1:3 glycol mixture. If I buy a boiler or tankless I will mate them with a 30 gal buffer tank. This is why I have been looking at HTC Phoenix Light Duty.

      1. Expert Member
        Akos | | #19

        Tankless units are cheap enough that if something does go wrong, you just replace it. Once the all the piping is in, replacing the unit is very quick. Maybe replace a blower or ignitor, but that is about it.

        I hate this idea, but I've taken one complete apart before to weld up a burst pipe and they are definitely not worth while fixing if you are paying an hourly rate. Having said that, for the complexity inside they are surprisingly reliable.

        If you are going for a closed system, I would stick to plain H2O unless you are doing driveway melt. Glycol systems need maintenance to keep the mixture from turning acidic.

        Floor heat systems have enough water volume in the field to not need any buffering even if the zone is slightly undersized. I would check your smallest zone and the min fire of the unit.

  6. MattJF | | #14

    So your infiltration and ventilation numbers were super high and I think you were double counting your basement and floors.

    For venitaltion use the formula here:
    (I did 7.5*4+.01*1900= 49 CFM)

    I did a quick run with for no basement and a finished basement. See attached.

    I am at 42.6K BTU/h in the worst case. This likely gives you some different options than at 100k BTU/h.

    Do you really have R27+ center cavity walls with R25 in an attic and R11 cathedral ceiling? That is just not a common combination, but can happen with retrofits.

    Finished basement will need R15 walls and you will want some floor insulation for comfort.

    If your floors are really uninsulated it would be very worthwhile getting them insulated with your radiant heat flooring. This load calc doesn't factor the uninsulated higher temp floor.

    Are you sure your windows are not low e coated?

    If you really want to understand this, build a spreadsheet. Heat load = Area * (68-(outside design temp)*U value. Add all the components.

    Let me know if I missed something in the load calcs.

    1. ClevelandOhio | | #16

      Thanks Matt, I'm going to take some photos in a bit.

      Do new double pane windows come standard with low e coating?

      I havent gotten around to windows just yet...

      1. MattJF | | #18

        Most modern windows have some sort label. Mine are located above the upper sash.

      2. ClevelandOhio | | #25

        Matt, I just attached framing detail to the original post.


  7. Deleted | | #15


  8. Jon_R | | #20

    My experience with using a tankless water heater as a very low duty cycle boiler was poor. Had to add a bigger pump and even then flow was too low and so water temperature was too high. It didn't condense, so efficiency was lower.

  9. Expert Member
    Dana Dorsett | | #22

    I'm still stunned at just how far off the LoadCalc is from anything "reasonable". Why 420 cfm of ventilation air? As Matt F alludes to, ASHRAE 62.2 for a 1900 foot 3 bedroom house (presumptively 4 occupants) would be only 87 cfm. Building Science Corp's recommended ventilation rate would come in at 49 cfm. With heat recovery ventilation the ventilation's heat load would be in the calculation error noise, not 28,681 BTU/hr. Even without heat recovery, at BSC ventilation rates the ventilation hit would be less than 3500 BTU/hr.

    Even with the adjustments made in the subsequent attempt it's showing more than 3400 BTU/hr of infiltration loss, in addition to the 3450 BTU/hr of no-HRV ventilation air.

    In that second pass it's still showing 11,647 BTU/hr of floor losses (to a basement that is probably not even losing that much heat if allowed to drop to 40F.) Is some of that floor over a ventilated crawlspace or pier foundation or something?

    With a radiant floor system the losses through the warmer-than-room-temp floor and tubing/heat spreaders to unconditioned space below are significant, and the LoadCalc tool isn't going to factor any of that in.

    >"Do new double pane windows come standard with low e coating?"

    IRC 2018 code minimum windows for zone 5 would be U0.30 or lower, which would usually be a single low-E coating and argon fill double pane. That type of glass is manufactured in high enough volume that it's pretty cheap compared to what that would have cost 35 years ago.

    LoadCalc's window selection options are pretty limited, so there isn't any good way to get there using LoadCalc.

    Even when I've optimized the hell out of every thing when using that tool it's been higher than a pro-tool's load numbers by more than 20%, and often more. For a quick & dirty sanity check or for sizing a condensing hot air furnace it's OK, but inadequate for designing a radiant floor or specifying a heat pump.

  10. MattJF | | #23 definitely has some issues. Its infiltration calc is is a bit of a mystery to me, but I haven't read through that part of Manual J to compare it.

    As with all of these, garbage in = garbage out.

    The original 420cfm was from listing the volume of the house at 90K CF.

    I was just trying to punch out a more realistic upper bound. I think that is all loadcalc is good for. Your real world load won't end up higher.

    I carried his original uninsulated floor, which is in the category of crazy stuff considering the radiant floors. I suspect there are few other things that are incorrect, or I hope are incorrect.

    As for infiltration on my redo the 3400BTU works out to basically 49 CFM.

    With an N-Factor of 14.8 that works out to (49 CFM*14.8*60)/20,790 CF = 2.07 ACH
    Now there is fudge factor on fudge factor there, but it seems not totally crazy. I really have not done enough of this to know what real infiltration is like.

  11. MattJF | | #24

    For the floor load of 11,647 BTU, it seems like loadcalc is using something like a 50F crawl space temp and an R value of 2.5 (U0.4) over 1500SF.

    Doesn't seem too out of line based on the inputs we were given, but I am hopefully punching in garbage.

  12. ClevelandOhio | | #26

    Matt & Dana

    I have attached framing detal. As you see, I havent insulated yet :)

    I am going to read through your posts carefully to try to understand. I am not insulating the floors over basement of course, but I had no option in to NOT INSULATE!

    Please be patient with me and I am going to get through this.

    My guess is that my actual heating BTUs are closer to 35k and not 100k

    1. Expert Member
      Dana Dorsett | | #27

      >"My guess is that my actual heating BTUs are closer to 35k and not 100k"

      My guess is that your actual heating load is between 25K-35K, and might even come in under 25K if the foundation walls are insulated to current IRC code minimums (=R15 continuous insulation, for zone 5, Cleveland.)

      As a different freebie/cheapie tool (one that also often oversizes for balloon framed houses with full-dimension or other non-standard construction types compared to current methods) would be It has it's own set of drawbacks, but it's at least a recognized ACCA Manual-J tool (unlike Loadcalc, which is Manual-J-ish and not terrible compared to some I=B=R type online calculators, but also not great.)

      As a point of reference, I live in a location with a design temp of +5F (comparable to Cleveland's +6F) in a full-dimension 2x4 framed (not balloon framed) 1920s antique 1.5 story bungalow with mostly clear-glass (not low-E ) storms over the antique wood double hungs, ~2400' of fully above grade fully conditioned space, and 1600' of insulated but not actively heated basement, and an insulated ~150' of crawlspace. My design heating load is between 35-40,000 BTU/hr as calculated by fuel use or calculated by either Manual-J or I=B=R methods. (The numbers differ a bit depending on method, but that's the range.)

      Simply scaling your house to mine by the square feet of fully conditioned space your house would come in at about 28-32K. But if you're installing low-E windows and have furred out the studs for deeper insulation your house should come in less than that. My house is also an extremely inefficient shape, a footprint that has 14 corners, and the framing fraction of the wall is much higher than balloon framed houses, which is why I believe you could be under 25K if the basement walls were insulated.

      With code-min U0.30 windows your window losses will be closer to 5000 BTU/hr rather than the LoadCalc's 9302 BTU/hr number, a difference of over 4000 BTU/hr. With 3400 BTU/hr here (overstated infiltration) , 4000 BTU/hr there (windows) and probably another 1000 BTU/hr of overstated wall (the lower framing fraction of balloon framing) it doesn't take much other overstatement reduction before that 42.5K load number drops to under 30K.

      1. ClevelandOhio | | #31

        I have finally figured out how to use Coolcalc last year. I think my load values were half the ones I posted here. Now there is a new version -- I will try to dive into it when I have time.

        My basement walls have 2" of XPS around two of the foundation walls that I have dug up. The other walls have nothing.

        Dana, do you have a system recomendation for 25-35k btu range?

        Thank you!

        1. Expert Member
          Dana Dorsett | | #36

          >"My basement walls have 2" of XPS around two of the foundation walls that I have dug up. The other walls have nothing."

          I suspect from the description that the 2" XPS (R8.4, derated) stops at or slightly below grade, with NO insulation of the foundation wall above grade(?). If yes, it's leaving the greatest heat loss segement, the above grade foundation wall completely uninsulated. Code min requires R15 continuous insulation, or < U0.050 (which is R20 "whole assembly"). To get there takes something like 1.5" of EPS on the interior side from the slab level all the way to the top of the foundation, trapped to the foundation with a 2x4/R13 studwall.

          Alternatively 2.75"- 3" of fiber faced roofing polyiso from a few inches above slab level (or above the high-water mark, if the basement has a history of flooding) strapped to the foundation with 1x4 furring through-screwed to the foundation with masonry screws, mounting the code required thermal barrier against ignition (half-inch wallboard is fine) on the furring. The reclaimed roofing foam approach can be cheaper than thinner foam + batts if using reclaimed roofing foam, which is available from multiple reclaimers in NE OH. Keep running this search every week or so and you'll find most of them:

          >"Dana, do you have a system recomendation for 25-35k btu range?"

          That requires a more detailed analysis of the radiation and water temperature requirements. A (relatively) cheap mod-con boiler like HTP's UFT-080W is pretty flexible, but may or may not be appropriate for your application. HTPs Phoenix Light Duty condensing water heater (76K burner) might be a less expensive total heat & hot water solution, but not if you need higher water temps to meet the design day load with the existing or planned radiation.

          Bottom line- simply hacking at a radiant heating design is fraught with risk of it...

          A: costing too much

          B: not working adequately

          C: not working efficiently

          D: all of the above.

  13. Deleted | | #28


  14. MattJF | | #29

    I just saw that the photos were in the original post. I can't really do anything with those unless you tell us how you are going to insulate. Design the heating system for the end result and supplement if needed unless it is going to be a long time.

    It seems like you have effectively a double stud wall. What what is going in there and what is going in the original wall?

    Is the roof going to be vented or unvented? If vented, it would be a vented cathedral with a small attic, pretty standard. You would just use the actual surface areas of each component.

    Is this whole place stripped our like you show?
    Did you put together a set of plans you are working from?
    Is this place going to be inspected?

    Do you have defined locations for the thermal and air barriers? An air barrier seems to require particular attention for this assembly, I see a lot of daylight and "creative" construction.

    Are your radiant floors installed? If they are not, I would strongly encourage going a different more cost effective route. You have a lot of work ahead of you if all 1900sf looks like what are in the photos.

  15. ClevelandOhio | | #30

    I included the photos of an exterior wall side of the wall "sandwich" of an exposed area. My wall system, so far, looks like this:

    * Vinyl siding
    * EPS 1/2"
    * Asbestos shingles
    * Wood lap siding
    * 2x4 uninsulated cavity
    * OSB
    * Air gap
    * 2x6 cavity

    Q: It seems like you have effectively a double stud wall. What what is going in there and what is going in the original wall?
    A: New wall (2x6) is going to have XPS in it -- this is all i know at the moment -- input is appreciated. Old wall cavity (2x4) will remain empty.

    Q: Is the roof going to be vented or unvented? If vented, it would be a vented cathedral with a small attic, pretty standard. You would just use the actual surface areas of each component.
    A: Vented; understood.

    Q: Is this whole place stripped our like you show?
    A: Yes

    Q: Did you put together a set of plans you are working from?
    A: That depends on what you are asking. I am using Chief Architect -- and adjust things as I go.

    Q: Is this place going to be inspected?
    A: Yes

    Q: Do you have defined locations for the thermal and air barriers? An air barrier seems to require particular attention for this assembly, I see a lot of daylight and "creative" construction.
    A: No. I honestly havent gotten to the insulation yet -- so I do not have a plan for it yet. Do you have any advice? The only daylight that is coming in is from the windows and from the old attic vent.

    Q: Are your radiant floors installed? If they are not, I would strongly encourage going a different more cost effective route.
    A: Yes and no -- I am drilling joists for the stapleup now. I already bought most of the materials. Staple-up is only going to make up for half the surface area of the 1st & 2nd floors -- the other half is Bekotec under tile. When ready, basement will be pex-in-slab. May I ask what you had in mind as far as cost effectiveness?

    "You have a lot of work ahead of you if all 1900sf looks like what are in the photos."
    Yes it does look like that. I have this caught-with-my-pants-down feeling right now -- I am feeling pretty intimidated after all your questions.

    I really apreciate your help.

  16. MattJF | | #32

    Sorry to make you feel intimidated, I’ve been there.

    I’m going to suggest you start another thread specifically addressing the best way to insulate and air seal your place after thinking through a plan.

    Air sealing is really important, you want to be able to draw a continuous airtight layer completely surrounding your house. You have to decide which surfaces are airtight and then do the best you can to make those airtight as possible. In your case I would plan on at least an airtight drywall approach. Read up on that. You may try to air seal the osb layer and framing with tape and caulk, this could be a backup approach.

    Do you have basement under the entire place, just thinking about air sealing the floor.

    For wall insulation I would plan high density fiberglass batts or dense pack cellulose. Get cellulose quotes. I don’t see the need for foam insulation unless you can slide more than 1.5” behind the interior framing.

    How deep are the roof rafters? Read up on insulating cathedral ceilings.

    Did you pull the windows? If you did, review how to flash windows.

    You can do this, you just need to make a detailed plan as lot of things are easier to build when they work together. A huge amount of the important detail in design plans is not in the picture, but in the notes and material specs. Your heat load can only be accurate after you define these things.

  17. ClevelandOhio | | #33

    You are right -- this thread went off-topic for the right reasons, but it is getting claustrophobic here now. I wonder if there is a moderator that can split-off some of the responses from this thread to make a new thread: Two-Story Victorian Insulation Plan

    1. GBA Editor
      Brian Pontolilo | | #34

      Hi Brian.

      Unfortunately, I can't migrate replies from this thread into a new post. The best thing to do is start a new thread for one of the discussions and then post a link here.

  18. ClevelandOhio | | #35



Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |