Big Baseline Boilers - COMBUSTION efficiencies!?

4 posts / 0 new
Last post

Hi everyone!

90.1-2007 Table 6.8.1F for boilers prescribes a minimum combustion efficiency (EC) of 82% for gas boilers larger than 2.5MBtu/h. This is the burner's efficiency before flue losses.

The same boiler's thermal efficiency, inclusive of flue losses, would be a lesser figure.

When modeling such boilers for an Appendix G baseline, what is actually appropriate to enter for the boiler's HIR input? As I understand it, this HIR input in eQuest is the inverse of thermal efficiency (net input vs. net output), and would be inclusive of flue losses.

This is for a LEED project, so I'd like to cite something solid to demonstrate what thermal efficiency is appropriate to assume for a baseline boiler of this size. Has anyone gone down this path before?


[cid:489575314 at 22072009-0ABB]


Nick-Caton's picture
Joined: 2011-09-30
Reputation: 805


It has always been my understanding that combustion efficiency includes flue losses as it is calculated based on CO2% or O2% in the stack gas and net stack temperature (stack temp.-boiler room ambient temp entering the burner). The only other thing I can think of that would affect the thermal efficiency or useful Btu's output as steam or hot water/fuel Btu's input would be radiant and convective jacket losses from the boiler shell at 100% firing rate. Although manufacturers are reluctant to publish data on jacket losses, they could typically range from about 1%-4% based on the design and mfg. This will be a fixed quantity of Btu's no matter what the firing rate is, so this loss as a % of boiler load increases as % load decreases.

As an example, look at an Ajax or Rite Boiler. They look pretty similar and are both manufactured within a fairly close proximity of each other. The front and back ends have uninsulated steel head plates about 1" thick to gain access to the tube bundle. Below the head plates are uninsulated thin sheet metal panels to gain access to the burners. With the boilers firing, I've measured temperatures of the upper plates in the 200-300 deg. range and the lower sheet metal panels in the area of 450 deg. This is kind of an extreme scenario, but the jacket losses are probably a significant number even though the rated or measured combustion efficiency might be 82%.

The other piece of the puzzle when it comes to seasonal boiler efficiency I don't know how DOE-2 handles would be cycling losses. If you take that Rite or Ajax boiler equipped with an atmospheric burner, the seasonal efficiency could easily approach 50%. During every off cycle, boiler room air will flow up through the boiler cooling the firebrick and mass of the boiler. The next firing cycling, those draft cooling losses need to be replaced by additional Btu's. A cycling boiler with a power burner could also have a low seasonal efficiency as the result of cooling losses from pre-purge and post-purge operation to get rid of unburned fuel.

I don't know if that helps or confuses the matter. After that long dissertation, I would suggest using the rated combustion efficiency.

My best,

Mike Busman

Michael R. Busman, CEM

Busman, Michael R's picture
Joined: 2012-07-12
Reputation: 0

Mike - thanks a bunch for your response!

I can appreciate and relate to your experience with the boiler industry marketing folks... it's very easy to get the run-around, and is often difficult/impossible to get information needed for modeling from product literature. I've come to the same conclusion as you advised, and here's my accounting/reasoning for those interested:

The footnote in 90.1 Table 6.8.1-F defines combustion efficiency as "100% less flue losses," so I that was the basis of this query. There's additionally reference to this procedure: 10 CFR part 431

That procedure defines combustion and thermal efficiency as follows:
Combustion efficiency for a commercial packaged boiler is determined using test procedures prescribed under ? 431.86 and is equal to 100 percent minus percent flue loss (percent flue loss is based on input fuel energy)

Thermal efficiency for a commercial packaged boiler is determined using test procedures prescribed under ? 431.86 and is the ratio of the heat absorbed by the water or the water and steam to the higher heating value in the fuel burned.

Clearly they're not quite the same thing, where combustion efficiency does not account for jacket/standby/other losses affecting net output.

I haven't seen numbers to verify this, but perhaps ultimately jacket losses are such a small fraction (at these largest capacities) that it's accepted to be irrelevant. Put another way, the actual values for Et and Ec for boilers larger than 2.5MBtuh/h just aren't very different.

Running with that assumption, and unless anyone can answer further to this, I'll treat baseline thermal and combustion efficiencies identically moving forward for large boilers: HIR = 1/Ec = 1/Et. If nothing else, it's a "safe/conservative" approach for LEED, in the sense that if there is an error of procedure here it is only artificially helping the baseline and would in turn prove a beneficial correction for the performance rating down the road.

Thanks again!


[cid:489575314 at 22072009-0ABB]


Nick-Caton's picture
Joined: 2011-09-30
Reputation: 805

I smile at this quandary (sorry
), having asked it myself, but not in the
context of eQuest
I?ve rationalized that the combustion and thermal
efficiencies are identical when the boiler is in a conditioned space,
because the jacket losses heat the space. I typically use a 1-2% downrate
of the combustion efficiency to get to thermal, if the boiler is in an
unconditioned space.

Jim Fowler, PE, CBCP, CEA

Jim Fowler, PE, CEM
Jim Fowler's picture
Joined: 2012-12-14
Reputation: 1