Boiler Water Temperature Calculation

[Flat_Twin]

Just as an exercise I'm trying to find a way to calculate required hot water boiler temperature with known EDR and heat loss values. I want to compare what I've observed over the winter with our new modcon boiler to what an established way of calculating boiler water temperatures says.

Emitters are 10 cast iron radiators with 164 total sections, 23 in tall. 164 x 3.50 sq ft = EDR 574
EDR 574 x 170 for hot water hydronic = 97,580 btu output.

Our house is a 1874 sq ft two story federal style built in 1850 with modern windows, doors, siding insulation. I tried to be as accurate as possible room by room with wall construction, insulation type etc. Design temperature for this area is 5 degrees. Heat loss per slant fin app is 70,400 btu

Thermostat is at 72, nice even heat throughout. ODR is set at max water temp 130 @ -5 outside. Min water temp 85 @ 60 outside temp.

The boiler runs almost non stop. I notice some cycling on mild sunny days or when we're using the kitchen oven for long periods. On the other hand, on very windy days the indoor temperature may sag a degree or two but catches up eventually. Today with light winds and 35 degrees outside the boiler is heating with 100 degree water. The highest observed boiler temperature was 125 at around -3 outside late in December.

You may think I need something better to do but I was curious if there was a way to project what the boiler temperature would be with known EDR and heat loss figures.

 

[Dana]

If you know the heat load in BTU/hr and the total EDR, reduce that to the ratio of load per square foot EDR.

The SlantFin app is typically 25% or more above reality as measured by fuel use against heating degree day data, but for purposes of illustration let's use the 70,400BTU/hr in to the 574' EDR, for a ratio of 123 BTU/hr per square foot EDR.

The output of radiation varies fairly linearly with temperature over a wide range, which allows you to use a simple nomograph such as the one on p2 of this document to estimate the water temperature at which it will deliver 123 BTU/hr per square foot, which is something like 155-160F.

Note, that at 130F it would only be delivering about 70BTU/hr per foot EDR, so if the system is actually keeping up, it implies the heat load at -5F is really only about (70BTU/hr x 574'=) 40,180 BTU/hr, which is WAY below what the SlantFin app was indicating even at a design temp that's 10F warmer (!), but could be close to what you would get using fuel-use against heating degree-days.

What boiler do you have?

 

[Sponsor]

 

[Flat_Twin]

The boiler is a WM Eco 110 installed early November. Since then we've used 480 ccf NG . Last winter which was also a mild winter, we spent $1760 on fuel oil at around $2.20 gal

What you pointed out is part of my curiosity about why it works as well as it does at such low boiler temps. Most of the time it's running at 20% fire until outdoor temps drop into the 20s or lower, then its 20-38%.
This does seem to indicate the actual btu loss is less than what the calculators estimate. The old oil boiler was rated at 100k btu. I suspect the Eco 70 would have worked even better with 14k btu minimum fire over the 22k min fire of the Eco 110. Still, short cycling isn't a problem. The distribution piping is large 2.5 inch to 1.25 as it nears each radiator. From what Ive read, that much mass helps buffer excess btu and decrease cycling.

 

[Dana]

You're correct that the ECO 70 would have been the more appropriate choice, since with the ECO 110 your min-fire output is more than 50% of your true design condition load. The high thermal mass of the water volume + iron in your radiation keeps it from short-cycling, but you would have gotten significantly more modulation out of the smaller boiler. Even though the ECO 70 would be the appropriate boiler for 95% of homes in the US, I'd be willing to bet they are selling far more ECO110s. In your case it's neither an efficiency or longevity hit, the way it would be on many fin-tube baseboard heating systems.

Last year I ran into a situation in CT where the homeowner had done the math three ways and was soliciting bids for a ECO70, and could not find a single contractor willing to install anything smaller than the ECO110. The contractor with low-bid proposal for the ECO110 was asked to re-submit with the ECO70, and he pushed back, saying he would neither install or support an ECO70, and the the homeowner would be crazy to install it. The load based on fuel-use calculations was about 32,000 BTU/hr, and based on an I=B=R spreadsheet calc was about 37,000 BTU/hr. That contractor did not get the business.

For the napkin math on right-sizing mod cons, read this bit o' bloggery.

 

[Flat_Twin]

Good reading there! From your blog...

" With more heat going into the water than is coming out, the water temperature rises, and is eventually above boiler’s outdoor reset curve temperature, at which point the burner turns off, even as the water continues to circulate. The burner re-fires only when the water temperature drops below the reset temperature. "

This is a very good description of what I've observed when outside temperatures are warmer than say 45 or higher. The ODR target temperature has a default range of 5 degrees either side so there's a bit of room before the boiler stops firing when it exceeds the target. Sometimes it will ride for several hours 3 or 4 degrees higher than the target. Between this 5 degree allowance and the buffering effect of the cast iron system, the boiler can get away with providing more heat than required.... for a while. The shortest cycle I've seen since dialing in the ODR is 20 minutes on / 45 minutes off with a ODR target of 95 degrees or less. The radiators don't feel warm at all but they are apparently still providing heat.

I regret not doing my own homework ahead of replacing the oil boiler. I trusted the heating contractor to size it. This is the same company that's serviced the oil boiler for years. The house was measured, windows and doors taken into account, questions about insulation and wall construction were talked about. Once it was installed and running, they left the boiler at default settings per the manual for cast iron radiators. 180 max /130 min. He did show me how to get into the ODR programming but no real explanation of how to go about it. The result was short cycling 8 minutes on / 8 minutes off. That's when I started reading all I could on the subject. I'm a recently retired telco technician so I've got time and plenty of curiosity. Initially I started by gradually lowering the max and min temperatures. I'd wait a day or so to see how the lower settings played out, then adjust them further. I also used the "raise the thermostat to 80 and see where ODR takes you" method. The current settings seem to be a good compromise though as you pointed out, a lower minimum firing rate would have been even better. I'm sure the heating contractor adds a little safety factor to his heat loss calculation rather than come up short and if there's similar safety factor built in to whatever method he uses, you end up with a larger than necessary boiler. As careful as I was with the slant fin app, I still came up with 70k so maybe I wouldn't have done any better with boiler selection. Considering the many stories I've read about grossly oversized boilers I realize we got lucky. And, the goal of saving money on heating costs (40% less) is looking pretty good.

Thanks for your input.

 

[Dana]

It's all water under the bridge, but you probably would have hit pretty close to reality doing a fuel-use based load calculation using the oil fired beastie-boiler as the measuring instrument. With the errors introduced by being so oversized and possible well-aged, it might have been off by 25%, but not 70% or 100%.

For a boiler like that to be seriously short-cycling would be sub-2 minute burns with 10-15+ burns per hour, which adds up to a lot of wear & tear and real efficiency hit. There are plenty of those out there!

But with your high-mass radiation delivering 8 minute minimum burn times, and ~4 burns per hour or less, the operating cost of the more appropriately sized smaller boiler wouldn't have been any different that what you've got, so the only real down side was the uptick in up front cost. Be happy!

 

[Flat_Twin]

If I can bend your ear just a bit more.... About figuring fuel use and heating degree day data. I pulled some numbers for my house and my area from last winter. Based on 800 gal fuel oil, 100k btu oil boiler rated 80% eff = 88,768,000 btu

Per data from degreedays.net
based on 65 degrees / heating degree days Oct-May = 5218 divided into total btu above = 17,012 btu or per hour = 709 btu
@ 60f = 4135 = 21,467 btu per hour = 894 btu


Design temperature for this area is 5 degrees so...

Balance point @ 65 - 5 = 60 x 709 btu = 42,540 btu
Balance point @ 60 -5 = 55 x 894 btu = 49,170 btu


If I did this right it explains a lot about why I'm able to heat with low boiler temps

 

[Dana]

Yes- those numbers are really an upper bound, since the boiler is clearly more than 2x oversized for the load. If you limit the fuel use period to mid-winter rather than including the shoulder seasons it's usually more accurate, since the idle losses become a super-large fraction at the very low duty of the Oct-Nov and the April-May seasons, which exaggerates the apparent heat load. If you're curious you can play around with the NORA Oil Heat FSA calculator, which does a more sophisticated fuel-use load calculation based on boiler models created at the Brookhaven National Lab testing.

A typical tightened-up 2x4 framed house will usually come in at about 15 BTU/hr per square foot of conditioned space @ 0F, so one might expect something closer to 30KBTU/hr than 40 - 50K for an 1874' house. Air leakage or high ventilation rates can often make a 10K difference. You can run the same fuel-use calc on your current boiler as a sanity check, which should be delivering something like 95% efficiency at your low water temps.