mod con tankless combo NG boiler sizing to exisiting cast iron radiators

[jmm]

I am investigating utilizing the above boiler type to replace an oil fired boiler 139K btu output 748 sq ft radiator.
I measured all the radiators sizes and used a radiator btu calculator spreadsheet. i need 161412K btu for the total radiator sq ft is 842 total.
Is this an accurate method for sizing a new mod con tankless combo NG boiler. Thanks in advance.

[Dana]

Sizing the radiation may be a way to size a steam boiler, but a useless way to size mod-con. You size a mod-con by the actual heat load of the house, and adjust the outdoor reset curve of the mod-con to be able to deliver the heat with the lowest temperature at which it will keep up, using outdoor temp as the model of the heat load.

The lower the boiler's output temp, the lower the return-water temp==more condensing efficiency. Ideally you'd be need at most 140F water even at design conditions (and I'm guessing you are probably already there, unless you live in an uninsulated house and leave the windows open, with that much radiation.) If the return water is over 130F it won't be condensing at all. But even if you needed 180F water at 6AM on the coldest day of the winter, most of the time you wouldn't need water anywhere near that temperature for the radiators to keep up.

At +65F outdoor temps most houses have pretty close to zero BTU/hour heat load but at +25F most will have a fairly substantial heat load, but at +45F outdoor temps the BTU/hour requirements will be about half what it is at +25F. The BTU output of the radiators also vary fairly linearly with water temperature, so if you set up the boiler to deliver water hot enough to keep up with the load at 25F, and a lower temp water designed to keep up at a higher outdoor temp, letting the boiler's controls interpolate that line to match a water temp to an outdoor temp works pretty well. It's crude, but effective.

And as long as the mod-con boiler's high fire BTU/hour can keep up with the BTU/hour requirements of your house at the ACCA 99% outside design temperature for your area, it's big enough that you'd never get cold. But the SMALLER the minimum-modulation of the boiler is, the fewer burn cycles there will be, and the closer to the boiler's best-efficiency it will be at any given water temperature.

It's possible to put an upper bound on your boiler sizing by using the name-plate efficiency and the fuel use of the old boiler per base-65F heating degree day, which will also be pretty linear. If your oil vendor stamps a "K-factor" on the billing, that's the same information in a different form.

[jmm]

Thanks great input. cant you calculate hot water ng boiler btu output based on the amount of water that circulates through the burner. In other words if you drained all the water out of the system and measured the total amount in use, lets say 100 gals. So then you would need to calculate how many btu s it takes to heat that volume of water to 180f. This would give you the highest use, (coldest day requirement) anything less than that would be controlled by the PID controller, modulator? Thanks in advance

[Dana]

In a word, no. All of that water is inside the building, and it doesn't matter how long it takes to heat it all up- BTU/hour is the same as heat/time. If it takes 15 years to heat it all up with a 1BTU/hr candle, it doesn't matter, as long as the house only needs 1BTU/hour.

The only thing that matters regarding the boiler output sizing it the maximum rate of heat loss out of the house- the BTU/hour leaving the house at the coldest hours of the the coldest night of the year. The calculating the heat loss at ACCA 99th percentile outside design temperature (based on 25 year binned hourly weather data during the heating season) for your location is more than sufficient for sizing the boiler to keep you warm. Going a whole lot larger than that results in lower efficiency and lower comfort.

The total amount of BTUs stored in the water at 180F (or any other temperature) is irrelevant. It's the rate at which it can release that heat to the house that counts, and that changes with temperature (and the size of your radiators.) The higher the difference in temp between the water and the room, the greater the BTU/hour rate, and the larger your radiators, the greater the BTU/hour rate at any given temperature. But you only need the water temp at which the heat emitted by the radiation matches the heat loss out of the house, and that heat loss also varies with outdoor temperature (and other factors, such as passive solar gain, wind-washing, etc.) Outdoor temperature is a very crude model for tracking the heat load, but it's "good enough" so that using an "outdoor reset" sensing the outdoor temperature to vary the temperature of the boiler output relatively high comfort & efficiency can be gotten out of the boiler.

It's highly unlikely that you would ever need anything close to 180F water, even on the coldest day if your 180F BTU/hour number for the radiators of 161,412 BTU/hour is the right order of magnitude. The graph on page 2 of radiator sizing document I linked to suggests that most radiators put out 170BTU per hour per square foot:



So, if you have 842 square feet of radiator that's 143,140 BTU @ 180F, not 161,412, but it hardly matters- even 143KBTU/hour is literally FOUR TIMES the heat load of my (~2200ft house with ~1500ft of semi-conditioned ~65F) basement at my 99th percentile outdoor temp of 0F.

Unless you're on a high mountaintop in VA your 99th percentile design temperature is going to be higher than +10F, and could be as high as ~+25F if you're along the coast. My 99% design temp is 0F, and if I had your radiators I would never need more than ~120F water: My heat load at 0F is about 35000 BTU/hour, and with 842 feet of radiator that's 35000/842= 41.5BTU/hour per square foot. Looking at the graph, you can expect about 40 BTU/foot out of the radiators with 115F water. And since this is well into the condensing zone, that would yield 95% efficiency or better out of the boiler if sized correctly.

As it happens, I'm currently running my home's system at 130F max water temp, but in fact COULD heat the place comfortably with 115F water at 0F outdoor temps with some improvements to one of my radiant floor zones.) For you to actually need 143K or 161KBTU/hour @ +10F outdoor temps you'd literally have to be living in a tent, or a very drafty 6000 square foot uninsulated barn, or keep some windows open all winter.

With that much radiation you have enough to run at very high condensing efficiency 100% of the time, but for optimum comfort & efficiency, size the boiler to the heat load, not the radiation. The water temp requirements don't change with the size of the boiler but with an oversized boiler it'll turn on/off more often, throwing away a fixed amount of BTUs with every ignition cycle and flue purge, adding up to lower efficiency and more wear on the boiler. With a right-sized mod-con boiler it would burn nearly continuously on the coldest days, but mostly at low to mid fire. It's very likely that even the smallest mod-cons are bigger than your design condition heat load, but as long as it's MINIMUM modulated output is at least 1/3 of your design heat load it'll be fine.

[jmm]

Thanks for that great response. House size 3200 sqft soild brick 50 windows wstorm windows outside temp rating is 17. Thinking outside the box! Would you increase efficiency and cost (ng) by using electric heat wrap on incoming cold water supply thus raising the water temp decreasing the spread ?etween base water temp and max water temp? Richmond va 55 degrees is incoming lets say you raised it to 70 degrees. Also would this reduce the strain on tje boiler?

[jmm]

Hello: I looked back and the spreadsheet that I used; recommended to use 185 btu instead of 165 btu. Looks like the same sizing charts etc. Based on the fact that you know what you're talking about and I am a laymen I am definitely going with your numbers. Thanks again.

[Dana]

Adding any heat to the system with electricity is always more expensive than adding it natural gas, in any lower-48 market.

I don't really know what I'm talking about, I just make it up as I go along!

With the oil use numbers (or a K-factor) it's a lot easier to estimate the design condition heat load than by trying to do a half-baked calculation based on the number of windows, taking a WAG as to the wall area and attic R, etc. A real heat loss calc (using that freebie Taco tool or similar) will get you much closer.

BTW: Is it solid brick, 3-4 wythes thick, or is it a cavity wall (with a few inches of air between the outer wythe and the interior brick)?

[jmm]

Adding any heat to the system with electricity is always more expensive than adding it natural gas, in any lower-48 market.

I don't really know what I'm talking about, I just make it up as I go along!

With the oil use numbers (or a K-factor) it's a lot easier to estimate the design condition heat load than by trying to do a half-baked calculation based on the number of windows, taking a WAG as to the wall area and attic R, etc. A real heat loss calc (using that freebie Taco tool or similar) will get you much closer.

BTW: Is it solid brick, 3-4 wythes thick, or is it a cavity wall (with a few inches of air between the outer wythe and the interior brick)?

solid one foot thick brick with interior plaster walls built
1925

[Dana]

At a foot thick you're looking at ~R3 total (counting plaster + air-films etc.), which is way better than nothing, but does make for ~17-20BTU/hr per square foot of wall area to the heat load @ +17F, with an interior temp of 70F. With storm windows the window losses @ +17F are about 25 BTU/hr per square foot.

With 50 windows @ ~10 square feet per you have 500 square feet x 25BTU/ft for 12,500BTU/hr of window loss.

Assuming it's a 40x40' two story and 10' ceilings/1' joists for a total of 22' height the wall area is (160 x 22=) 3520, less 500' of window for ~3000' of wall area. Assuming 20BTU/foot that's a wall area loss of 60,000 BTU/hr @ 17F.

Assuming R25-ish whole assembly-R (R30 batts or blown between joists) you're looking at 2.1-2.2BTU/hr per foot at the top of the house, so the loss out the ceiling is another (1600 x 2.2=) ~3500BTU/hr.

Total conducted loss is then on the order of 12,500 + 60,000 + 3,500= 76,000 BTU/hr plus air leakage losses.

Brick walls tend to be pretty tight, but attic floor/top-floor ceilings and basements aren't air sealed there can still be a considerable stack effect. But brick also has a substantial thermal mass, which tempers the peak loading by quite a bit at a foot thick, so knocking 15% of the boiler sizing it would still keep up. Design conditions only persist for a few pre-dawn hours at a time on the coldest nights of the year, and the stored heat in the thermal mass brick averages out the out the load presented to the heating system over many hours- the interior surface wall temp doesn't rise or fall nearly as quickly with outdoor air temp as low-mass walls would.

Without seeing the place or measuring anything a WAG based on the minimal info presented would be that your true heat load at design temp would be somewhere between 55-90 KBTU/hr, which is a pretty big range, and it would be good to get a better handle on it before picking the boiler. What's the oil use per year, and your zip code (for weather data), or if available, the K-factor on the oil bills?

Assuming the high end of the estimate of 90KBTU/hr, with 842 feet of radiator, at design condition you need to get (90,000/842=) ~110BTU/foot out of them. Using the graph that means you'd need at most 150F water during those coldest hours, and lower during the rest of time, so you'd still be in the condensing zone for return water temps most of the time, if not quite at the design conditions. If the true heat load is more like 75K you only need (75000/842=) ~90BTU/foot out of the rads, which (per the graph) means ~140F average water temps, and the return water temps will be low enough to hit the ~90% efficiency range even then, and lower temp/higher efficiency when it's not as cold out. Either way you have more than enough radiation to get quite a bit of efficiency advantage out of a condensing boiler using an outdoor reset control strategy that raises & lowers the boiler output temp in response to outside air temperatures, rather than a (low, but) fixed output temperature type combi-boiler. But if your heat loads are low enough that your peak water temps requirements are ~130F or less, there may be an argument for using a condensing tank hot water heater (like a Polaris or Vertex etc.) rather than a more expensive & complicated big-burner tankless type combi.