Cast iron radiators and Takagi TH 2

[Beekmansup]

I have a house with very inefficient gas boiler, one 1” one way single loop pipe and 7 cast iron radiators. I wanted to replace this system with a radiant open system for the reason that is the most efficient system. I bought all the supply necessary to build that system including 2 Takagi TH2 high efficiency water heaters, but then I found that the city doesn’t aloud this kind of system for the legionnaire bacteria concern. Those cast iron radiators can work very nice if I have a decent efficient boiler. My question is: can I use this Takagi TH 2 which is 90% efficient with the existing one way T connectors for every ½ “ risers and cast iron boilers and new mechanicals and controllers for single zone? I heard that instant water heater are not like heat boilers to accept hot water for intake but Takagi use them also for heating applications because it can go to 180F.Does anybody has any experience for cast iron radiators and TH 2 water heater?

[jadnashua]

Tankless systems and boilers designed for heating are similar but not exactly the same. Personally, I think you'd get better reliability and easier installation using a boiler for your heating both the house and the hot water. Hooked to an indirect tank, a modern high-efficiency boiler makes great use of the energy for both purposes...a tankless may not. Some tankless systems void their warranty when used for space heating. Again, personal opinion, while some people think they work great, an open system has some pitfalls. One, especially when used with cast iron radiators is that you'd rust them out. An open system is constantly adding oxygen to the system in the incoming water supply - guess what, that will rust out your radiators. A closed system gets (ideally) filled once, and shortly, all the oxygen is used up, then, the water becomes relatively inert.

[Dana]

What Jim said about using potable water in cast-iron radiators- you'll ruin both the Takagi and the radiators going that route.

It's possible to design a system around it using heat exchanger keeping the heating loop isolated from the potable, with a separate bronze-impeller pump in a short loop on the Takagi side and a standard iron hydronic pump on the isolated heating-system water side. You'll have to do the math to get it right, but it's possible to get decent efficiency out of it if you do.

Do you have ANY idea what temperature water you would need to handle the heating load at +7F (the 99th percentile design temp for Wilkes Barre)?

Do you have any idea what the whole house heat load is at design temp? (We can work backwards from fuel-use on the old system, if we have to.)

With some idea as to the load and the amount of radiation you have out there we can probably thumbnail a system that would be close enough to work with some judicious tweaking.

Some Takagi models auto-kill the flame when the return water gets up to about 130F- not sure if that's an issue with this model. No matter what, if you're running temps higher than that for return water you won't be getting anything like 90% efficiency out of it. Yes, you can put 180F out, but if you're constrained to 130F max return you have to design the flows very carefully. But if the TH2 is programmed for 180F output it's useless as a hot water heater, since you'd have to use a tempering valve at the output to mitigate scald risk, and the flow through the tankless at low DHW flows may not be enough to trigger ignition or regulate the output temp well Ideally you'd have sufficient radiator to run the thing with 130F water OUT, which would mean the return water temp would stay in the condensing range, and you won't have to fight to make it work as a hot water heater too.

I used a KD-20 as the "boiler" on my single-temp ~125F combi system utilizing a "reverse indirect" as the heat exchanger for the potable side, running heating-system water through the tankless, but that's a pretty pricey heat exchanger if you don't need the thermal mass for buffering micro-zones (which is why I took that approach.) If you did something similar you'd be well advised to put a filter on the loop to keep iron crud out out of the tankless too- old iron heating systems can have a LOT of crud in them and work fine with a cast-iron beastie boiler, but not so much with a condensing tankless. But I did the math first rather than going with "hack & hope". If you have enough radiation to get your design day water temp requirements down to ~125F or so you might hit 90% with a TH2 and a plate type heat exchanger, but not if you need higher temps than that.

Did you in fact install the radiant? If yes, please describe.

You should also do a room-by-room heat loss calc based on a +7F outside temp, and use some reference like this to see if the radiators in each room can cut it on their own with 120-130F water, or if you would need more radiation.

Condensing boilers have the advantage of being able to adjust the output temp automatically with changing outdoor temperature (called "outdoor reset" in the trade), but a tankless does not, so the only way you can get the efficiency out of a tankless is to have enough radiation to run it in condensing mode under all load conditions. You can then design/adjust the loop flows for long low-fire burns that match the design-condition load for output, and the thing won't short-cycle itself to death. But if it turns out you actually NEED 140F or higher water to meet the load at design condition, you'd be far better off with a boiler with outdoor-reset capabilities, and tweak the reset curve to the lowest temp that meets the load.

[Beekmansup]

Thank you for your help guys. I decided to go with a Burnham ES2-4, 77,000 BTU and my cast iron radiators and an Outdoor Reset Module to make this 85% efficient boiler to save more gas and I will line the chimney with 5" ss liner and build a box around the boiler with a duct of filtered outside air. The TH2 I will use it for radiant floors in the joists and hot water with a mixing valve. I'm very impress about your knowledge in heating systems Dana and I appreciate any input from you regarding my project. Thank you.

[Dana]

The Burnham ES2-4, is a bit more than 2x oversized for my whole-house heat load @ design condition (my outside design temp slightly cooler than Scranton/Wilkes-Barre, but comparable.) How did you choose that boiler?

With part of the house heated by the Takagi the load on the cast-iron boiler will be less than the whole house load.

A room-by-room heat load calculation is called for, but short of that, as a starting point, using the fuel use of the previous system it's possible to put some firm stakes in the ground as to the actual size of the whole house load. If you have a winter-season gas bill complete with the end dates of the billing period it's possible to look up the heating degree days (HDD) during the billing period. (Or if the billing has an mean-temperature for the billing period and a daily-use average it's even easier.) From therms/HDD it's simple arithmetic to come up with a source-fuel BTU/hour per degree F below the heating degree-day base (65F is an appropriate base for most homes). Then de-rate by the boilers name-plate efficiency fpr a net BTU/hr-degree, and multiply by the difference between design temp (+7F for Scranton/Wilkes-Barre) and 65F. Unless you leave multiple windows open all winter or have a great deal of single-pane window area (without storms) I'd be surprised if your heat load at +7F is anywhere near 77KBTU/hr.

But the heating fuel bills will tell all. Unlike heat load calculation methodologies based on construction types and presumed U & R values, fuel-use against degree-day data is a heat load MEASUREMENT.

[Beekmansup]

Thank you,Dana.This weekend I will go to that house in PA and measures all the SF of radiators to make a clean calculation of the heat loss. The problem is, I like to change the old boiler and mechanicals with a new more efficient cast iron boiler and mechanicals and keep the old radiators( 7pc.) connected in single one way iron pipe, but so far I didn’t find a satisfactory boiler, a blue print of mechanicals and electrical connections. There are some efficient cast iron boilers up to 90% but they have a confusing data of how to install them especially when it comes to an outdoor reset controller. If anyone have any idea where I can find such information will be much appreciated. I have to postpone the radiant system because I have to renovate the house. The radiant is useless where you have the cabinets and big furniture and king size beds so the old cast iron radiators will be the primary heating system and the radiant I will install later in the exposed no covered joists, mainly in the living and dining rooms. So right now I’m looking for an up to 90% efficient cast iron boiler and find a good diagram for mechanical and electrical connections which includes the outdoor reset device. The only one I found is the Burnham ES2-3 51000BTU and it has an outdoor reset module but the boiler controller doesn’t have an output for the outside circulator pump and have no idea how that circulator will be connected and controlled by the boiler controller board. Is there anyone who can help? Thank you.

[jadnashua]

If a boiler doesn't already have on-board multizone controllers, the most common way to deal with it is to add something like a Taco zone controller box. This OR's the theromstats for the zones into one on/off call for heat, and powers the circulators through on-board relays. The outdoor reset monitors the outside temp, and usually, the return water temp to the boiler, and uses that to decide what temp supply water needs to be which is different than turning the circulators on/off for the zones needing heat.

[Dana]

Thank you,Dana.This weekend I will go to that house in PA and measures all the SF of radiators to make a clean calculation of the heat loss. The problem is, I like to change the old boiler and mechanicals with a new more efficient cast iron boiler and mechanicals and keep the old radiators( 7pc.) connected in single one way iron pipe, but so far I didn’t find a satisfactory boiler, a blue print of mechanicals and electrical connections. There are some efficient cast iron boilers up to 90% but they have a confusing data of how to install them especially when it comes to an outdoor reset controller. If anyone have any idea where I can find such information will be much appreciated. I have to postpone the radiant system because I have to renovate the house. The radiant is useless where you have the cabinets and big furniture and king size beds so the old cast iron radiators will be the primary heating system and the radiant I will install later in the exposed no covered joists, mainly in the living and dining rooms. So right now I’m looking for an up to 90% efficient cast iron boiler and find a good diagram for mechanical and electrical connections which includes the outdoor reset device. The only one I found is the Burnham ES2-3 51000BTU and it has an outdoor reset module but the boiler controller doesn’t have an output for the outside circulator pump and have no idea how that circulator will be connected and controlled by the boiler controller board. Is there anyone who can help? Thank you.

The size/surface area of the radiation is of absolutely no use in determining the heat loss of the house (or room) at design temperature- NONE!

Fuel use over a particular known period where the weather data can be correlated is a HUGE help in determining the maximum BTU output that would be required of the boiler, and the smallest boiler that meets that load will run the most efficiently, all else being equal.

If fuel use data isn't available, a Manual-J or IBR type of heat loss calculation can get you there, and that is standard GOOD heating professionals would use. Those methods work by calculating the steady-state heat loss out of all exterior surfaces based on insulation & construction type, window type & are at a fixed difference in interior & outside temperatures. Typically 70F is used for the interior temp, and the local 99th percentile outside temp should be the lowest used for outside temp. Even though the 25 year peak lows might be 5-10F lower, it doesn't dwell there long enough to matter- the thermal mass of the house carries it through without a loss of comfort to the occupants.

Doing the heat loss calc on a room-by-room basis, and correlating the size of the radiation in each room to that room's heat loss is what determines the water temp requirements for meeting the load. Most existing hydronic systems with a 180F nominal water temp are at least 2x over-designed for the true heat load, and will in fact meet the true heat load at design temp with 140F water. If that's true of your system you would be able to run a condensing boiler in condensing mode nearly 100% of the time if you set up the outdoor reset curves correctly.

Outdoor reset is somewhat overrated for use with cast-iron boilers if the boiler is sized correctly for the load, and even more so when you have excess radiation that can deliver design-day heat at 140F or less. Since gas-fired cast iron boilers will need return water entering the boiler to be over 130F to avoid destructive condensation, simply setting it up with a thermostatic mixing valve on a system or boiler bypass plumbing to keep the return water to the boiler at 130-135F and letting it fire until the thermostat is satisfied it'll meet it's AFUE numbers, and the temp at the radiators may never exceed 140F in any cycle. Only if the radiation truly NEEDS 180F or something at design condition will outdoor reset deliver much in the way of enhanced efficiency or comfort. If the boiler is way oversized for the load there can be efficiency gains with outdoor reset, but in those cases a heat-purging smart controller would usually do slightly better.

Outdoor reset mixing valves etc can make for a more even radiator & room temp though, which is more of a comfort issue than a system efficiency issue.

If you're not up to doing your own Manual-J type heat loss calc, there are contractors out there capable of doing that, as well as specifying the hydronic boiler/system approach that makes sense for the amount & type of radiation you have. Spending some money up front for a real design can save you from buying unnecessary or inappropriate equipment that ends up costing you even more in reduced efficiency over it's lifespan. Without first calculating the heat loss of the house at design temp (which is NOT equal to the heat output of the radiators with 180F water, or at least not very often) it's all just a shot in the dark.

[Beekmansup]

I did contact a friend plumber (he doesn't have college degree in plumbing) for a loss heat calculation and he told me that an accurate +/- 3% LHC will be $150 to find the correct size of a boiler but he appoint me to this BTU calculator which have a +/- 10% accuracy who give me 40,342 BTUs. The monster I have right now has a 90,000 BTUs output (calculated and installed by UGY gas Co. in 1988) and the open flue chimney efficiency. You don't have to be a boiler scientist to see that I was spending a lot of $ heating the air around the house. I know you have more experience (or education?) then more of us but please Dana let me know what do you think about this. Thank you.

BTU Calculator
1. Which best describes your heating system?

High Temperature Heat (e.g. Baseboard, Fan Coils, Radiators)

2. What is the square footage of your home (the area to be heated)?
815 Sq.ft.

3. Location: Please select the city closest to your home.
• Pennsylvania


• Scranton



4. How old is your home?


More than 50 years old

5. Renovations: Have you made significant renovations to your home?


No, my house has not had major renovations since the original construction.



Your estimated BTU requirement is: 40,342 BTUs

[Dana]

40KBTU/hr seems sky high (insanely high?) for 815 square feet of conditioned space at Scranton's 99th percentile design temp +7F. That's 50BTU per square foot of living space- there are tents out there that can do that well. For comparison purposes, it's more than the whole-house heat load of my circa 1923 mostly one-story bungalow @ +7F, and that's ~2400' of fully conditioned ~68-70F space, plus 1500' of semi-conditioned (~65F) basement. When I first moved in it had storm windows, but almost insulation to speak of, yet was still under 50KBTU/hr @ +5F as measured by the fuel use.

If the house has literally NO insulation, and leaky single pane double hungs without storm windows your place might hit 50K @ +7F, if some of those windows are missing glass. :-)

If the house you're living in has only single-pane glass and no insulation, you'd be far better off spending the money on air-sealing, insulation, and storm windows (preferably low-E storms on the east, north, and west sides, standard clear storms on the south side. (There's a D.O.E. program subsidizing low-E storm windows, and you can order them through a box-store from at least one vendor.)

Let's assume for argument that the 40K number is remotely accurate and that this is an uninsulated house with 8 leaky single-pane double hungs. Adding 8 high quality storm windows would cut the heat load by at least 10%, but probably more like 20%, bringing the heat load down to ~32K, for a cost of about $1500, maybe less. Adding R38 of blown cellulose over 815 square feet of attic would also run ~$1500, cutting the heat load by another 20%, bringing the total heat load down to ~26KBTU/hr. Estimating wall insulation & foundation insulation costs are a bit trickier as it depends on construction type, but you should be able to get the heat load under 20KBTU/hr (which would be 25BTU/foot) without breaking the bank, if you're judicious about how you go about it (and it may already be that low.) When heat loads are that low you have many options that aren't available at 50KBTU/hr+. With a heat load of 20KBTU/hr at +7F your radiators would almost certainly be able to deliver the full load with water no hotter than domestic hot water temp, and it would be pretty easy to set it up to heat with one of the Takagi TH2s. There are also ductless air source heat pumps ("mini-splits") that would be able to deliver that amount of heat @ +7F at a system cost of well-under $10K, with an operating cost comparable to heating with natural gas. (A mini-split would have the additional benefit of high-efficiency air-conditioning.)

If you don't have fuel use data to work from, it's still possible to put a reasonable upper bound on it with the crudest-of-crude heat loss calcs based on construction type. (But if you DO have fuel use numbers, lets have them, eh?)

How many windows (and type), how tall are your ceilings, and what is the rough outside dimension of this house?

Walls are studwall, masonry or...???

Attic & roof type?

Slab on grade, crawlspace, or full basement?

[Beekmansup]

The house is a Cape Cod type - brick walls on the outside studs with gypsy walls in the inside (lime, sand, gypsum over wire mesh) no insulation in between except air. The 2 rooms in the second floor (attic???) are very hot in the summer and cold in the winter which means no insulation or a very low R type. Have no access to inside attic to see what they used. I did change the windows, 4 vinyl upstairs, double pane low E and 6- brake in aluminum double pane low E downstairs not a smart choice but easy to install. The radiators are inside the walls about 3", no insulation in the basement joists. Last week I measured everything again and it looks like I have 1100 SF downstairs - 5 rooms 8f h ceilings, 6 radiators (why 2 in the kitchen???) Upstairs 2 rooms no radiators, basement has 4 radiators, never use them (I will cut them off). The boiler (Burnham) has 149,000BTU input 99,000 output the regular and most common inefficient chimney boiler. I do not have data for gas usage and it will not help too much because the DHW 80 gallons tank with a pilot light spends also a lot of gas. So at this point I guess I have to drill 1/4" holes in between the outside bricks and inject foaming insulation in the empty space, 1" -2" holes in the upstairs rooms to blow some fiber glass or to put some reflective radiant barrier on walls and ceilings or both. So that is my situation now and the priority is to replace the boiler with an efficient one. That is why I choose ES2 series cast iron from Burnham 85% energy star with little modification to mechanicals and a new ALPHA 15-55 F/LC Cast Iron Circulator Pump. Any suggestions ????

[Dana]

How deep is the cavity space between the metal-lath plaster and the brick? If it's more than an inch you can get a SUBSTANTIAL reduction in heat loss by having a contractor install non-expanding injection foam. Do not use expanding polyurethane foams in this application- there are too many expensive ways to screw it up, and the glue-nature of SPF makes removing/repairing any mishaps even more difficult. If it's only a half inch or so, it's probably not going to make much of a difference. With a narrow cavity you might be able to hit R6 for wall-R, but cheaper/better approach would be to add 2" of rigid iso on the interior, giving up ~3" of room dimension on the interior walls. This is a major undertaking, but if done all at once to avoid living in a construction zone, it would pay off in comfort & fuel savings in the long run.

With a 2-wythe brick wall and a ventilation cavity the wall's R value is going to be something like R2.5. If the windows are lower-efficiency double-pane (or have storm windows), they too are about R2.5, which makes the Manual-J type heat loss calc a lot simpler: Assuming a 67F interior temp and a +7F outdoor temp at design condition (6AM on the coldest days of the year), every square foot of exterior wall + window is losing about (67F-7F) x (1/R2.5)=24 BTU/hr.

So assuming a ~35' square exterior footprint with 9 feet between floors that's 35x4x9 = 1260 square feet of wall area. At 24 BTU/ft that give you:

24 x 1260= 30,240 BTU/hr.

Assuming you can easily get the roof/attic up to at least R20, the loss through the roof gives you

1100' x (1/20) x (67F-7F)=3300 BTU/hr

Add them together and you're still under 35K, but if the ceiling, windows & doors are leaky you could be at 40-45K or higher, maybe even 50K if it's super-leaky, but still nowhere near the 77K output of the 4-plate Burnham ES2-4. With that boiler you'd be good down to about -50F, a temp not seen since the last ice age. The 3-plate ES2-3 would be a better choice, and run more efficiently after you've insulated & air sealed, brining the heat load down to ~30K or less (this IS do-able.)

But even the ES2-4 is such a large step down in input BTUs and up in efficiency that you will probably need a flue liner to narrow it down in order to avoid flue condensation rotting out your chimney from the inside over time. What it the cross sectional area of the flue?

If you intend to insulate the basement, insulate the basement walls, not the first-floor. If you insulate the floors it puts an R-value between the conditioned space and the boiler, so the standby and distribution losses of the boiler are all lost, rather than accruing to conditioned space. If you insulate the walls the basement stays warmer & drier, and the standby losses aren't fully lost.

To figure out the best way to insulate it, what type of foundation?

In the attic, blowing in cellulose will be more effective than fiberglass, due to the high convection losses of blow fiberglass in a warm side down configuration. The buoyancy of the air warmed by the ceiling moves relatively freely through even a foot of low density fiberglass into the cold attic air, making low-density fiberglass perform way below spec when it's 25F or less in the attic. By contrast, cellulose is more air retardent, and has a fairly stable R value across temperature. Use only "sulfate free" or "borate only" goods though- cheap sulfate fire retardents are corrosive when wet (and stinks when wet too), whereas borates are benign to metals & humans, but toxic to the gut-flora of wood boring ants/bees/wasps/termites, rendering them unable to digest wood.

Simply blowing insulation blindly into the attic isn't a good idea, since you can't tell where the insulation is going or how much there is. If there isn't an access hatch, make one, even if it's in a closet or hall ceiling. Insulation that blocks soffit ventilation and jams up against the roof deck can cause localized rot areas on the roof deck & rafters. More. If the wall cavities vent into the attic and you block them off with fiber insulation blown into the attic you may end up with moisture issues bleeding through causing the paint, then the plaster to fail on in your cavity-wall. (With non expanding injection foam this is less of an issue, but it's important to use a high-quality silane/siloxane masonry sealer on the exterior if you fill the cavity. Non expanding injection foam is vapor permeable, but waterproof, and the high water vapor drives of summer sun on rain or dew wetted brick can also cause paint failure in that configuration, even through the foam.)

Reflective paints on interior rooms have almost no effect on heating bills, but slipping in aluminized radiant barriers between the radiators and walls will be worthwhile (or even cut up sections foil-faced 1" iso), at least until you've figured out how to better insulate the walls fully.

[Beekmansup]

Man, you are a gold mine in knowledge of the heating systems calculation and I really appreciate your help. I wish to send you something besides my thanks to show you how much your advice means to me but I don't know how or where. My background involves in dally basis the mechanical components of HVAC, plumbing, electrical and building automations to the maintenance level and I know to distinguish a bright mind when i see one. Right now I cleared up my mind and thanks to you I will know what to do. The only thing which still keeps me off the project is the fact that I am on the budget and I have to make a decision regarding the boiler. Like I said the Burnham ES2-3 was my choice because I had to go with the rest of the old system but when it comes to $ I will end up with (boiler 85% EF, full IQ control system 3 modules and a 5" liner inside the flue) for a total $ 3305.85 (cheapest price I could find) the cheapest alternative will be a 90% EF Weil Mclain GV90-3 with a Tekmar 256 Boiler Control for $2569.90 and a much easier installation of the direct vents with PVC. So far I didn't find too much info regarding the quality of those units and I don't want to end up with more problems trying to save a couple hundred bucks. So, again, any idea?

[Dana]

Going with the direct-vent condensing Weil McLain has several advantages beyond the modest improvement in AFUE. Since combustion air is piped directly to the unit rather than drawing from conditioned space, it doesn't induce air infiltration, and cannot backdraft when bath-fan/kitchen-fan/clothes-dryers etc. are operating, no matter how tight you make the place.

Also, with direct vented boilers you can seal up the old flue penetration in the basement, which stops the stack affect of the flue from inducing air infiltration 24/365. If you're not going to remove the old chimney, you can seal up any boiler/HW heater penetrations in a number of ways. I've found cutting in a piece of fiber-cement tile backer to fit and mortaring it in place, flush with the rest of the brick/stone face to be quicker & easier than bricking it in, and makes opening it for for inspection or re-commissioning the flue easier, should that ever be needed.

A condensing boiler also doesn't need the extra plumbing & design to protect it from return water that is too cool. If you take the time to set up and tweak the Tekmar to the lowest temperatures that actually keep up, and you may be able to beat the AFUE numbers, especially if you re-tweak it every time you make a major insulation or air-sealing improvement.

As a very rough WAG, if the radiators are as old as the house, the odds are they were at least enough radiation to keep up with the load at design temperature with 180F water, when the load is at least 2x what it is/will-be after windows are tightened & storm windows added, the attic gets insulated, and the overall air leakage gets cut in half with a lot of DIY detailing. With 2x the radiation it means the peak water temp required after the place gets tightened up will likely be 140F or less, with return water to the boiler in the 120F range. At 120F return water it's running ~90%, minimally condensing- it's likely you could ALWAYS be in condensing mode. The AFUE test is done with 140F out, 120F return water and a ~30% duty cycle. But combustion efficiency rises with lower temps- if your seasonal average return water temp ends up at 110F or lower you'd be over 90% for as-used efficiency.

I assume you'll be heating the domestic hot water with the Takagi? If not, an indirect-fired tank running as a separate zone off the boiler makes sense. To be sure, don't vent any water heaters into the (now way-oversized) old flue or it'll have a significant risk of condensation & backdrafting problems.

[Beekmansup]

Very interesting, I did learn something new but I still have a big Q: If the boiler has a sealed fresh air intake and if you clean properly the chimney, why you can't take the intake air from there? The reason for such idea is that air from 30' has less dust than 1' from ground and if you seal it properly the wall and the pipes you will not have any problem with the back draft when you have a negative pressure in the basement. In the winter time the cold air is more dens and will travel down and I don't think it will create a problem for the intake fan. Did anybody try it so far? What do you think?