Why my 85% efficient cast iron boiler is really only 78% efficient

jesmed1

As a homeowner/mechanical engineer who's always trying to understand how things work, it's bugged me over the years that even some heating pros disagree about whether an 85% boiler is really 85% efficient. What does it really mean when the boiler tech sticks his combustion analyzer probe in the flue, and the analyzer spits out the number 85% efficient?

I think I finally understand one source of confusion/disagreement. Heating pros, please feel free to correct me if I go wrong (and I know you will!)

For starters, I'm talking only about "combustion" efficiency. That means, how much energy gets lost up the flue vs. how much gets transmitted into the boiler mass itself and the water that passes through it.

The other efficiency measure, "thermal" efficiency, is a different subject for another time.

Back to combustion efficiency. In doing some online reading, I found this is a useful guide/reference.
https://tsi.com/getmedia/02417ee5-cccc-4dc7-80bc-f7f10924d20a/CA-basic-2980175?ext=.pdf

It seems that a major source of confusion/debate stems from the fact that when oil or natural gas burn, they produce two types of combustion products: "dry gas" (O2, CO, CO2, etc) and water vapor.

Typical combustion analyzers measure the amounts of dry gas (O2, CO, CO2) and the flue gas temperature, from which they can calculate (or use pre-programmed lookup tables to calculate) how much heat energy remains in the exhaust gas stream, as a percentage of the total available energy in the fuel. That's energy that gets lost up the chimney.

For example, in my 85% efficient oil boiler, the tech sticks his probe into the flue, and the analyzer measures the amount of O2, CO, CO2, and the flue temperature, and deduces that the exhaust gas stream still has 15% of the heat energy that was in the oil before it burned.

However, the combustion analyzer does NOT measure how much water vapor is also in the flue gas, and how much energy that vapor is carrying away. This is also called "latent heat." And the percentage of latent heat lost in water vapor varies by fuel. In heating oil, it's about 6-7%. It's higher in natural gas, because gas combustion produces more water.

So the REAL combustion efficiency of my oil boiler is 100-15-7=78%. 15% is lost to "dry gas" in the flue, and 7% is lost to "latent heat," or water vapor.

So if I want to calculate how many useful BTU's my boiler is outputting per gallon, I can multiply the BTU value of oil (138,490 BTU/gal) times 0.78 = 108,022 BTU useful heat per gallon. From my 85% (but really only 78%) efficient boiler.

Now to complicate things further, the heating industry is well aware of the fact that boilers like mine cannot condense and recover that water vapor that's costing me an extra 7% efficiency loss, because condensation would ruin my boiler. So in recognition of that fact, the heating industry gives me two DIFFERENT BTU values for heating oil: a "high heating value" (HHV) which I used in the above example, and a "low heating value" (LHV) which already has the latent heat of the water vapor subtracted from it. I found slightly different values of HHV and LHV for heating oil, but the two that I'm using here are:

HHV: 138,490 BTU/gal
LHV: 129,488 BTU/gal

Again, the HHV includes the latent heat of water vapor, and the LHV does not.

So, another way to calculate the useful heat output of my 85% (?) efficient boiler is to use the low heating value of 129,488 BTU/gal, knowing that the BTU's lost as latent heat of water vapor have already been subtracted.

So now I don't have to subtract that 7% for the water vapor. So my 85% boiler really is 85% efficient, if I use the low heating value for oil that already has the latent heat of water vapor subtracted from it.

So 0.85 x 129,488 = 110,065 BTU/gal useful heat output. (Note that this is about 2,000 BTU different from the previous calculation because I used a round number of 7% for the latent heat earlier, where the difference here between HHV and LHV is 6.5%)

So the truth is that my 85% efficient boiler is really 85% efficient ONLY if I start with the low heating value (LHV) of oil, which is about 130,000 BTU/gal.

But if I start with the high heating value (HHV) of oil, which is about 139,000 BTU/gal, then my 85% efficient boiler is really only 85-7=78% efficient.

That's my current understanding as a non-pro homeowner/engineer, and I welcome correction if I have gone astray. But this example helps me understand one of the sources of confusion about combustion efficiency and what it does (and does not) measure.

(Please also note that this does NOT get into combustion efficiency measurements for condensing boilers, which try to recover most or all of that latent heat by condensing the water vapor. Since I do not have a condensing boiler, that's a different discussion. But the principles are the same, ie that some heat is lost as "dry gas," and some (but hopefully much less, and ideally none) is lost as the latent heat of the water vapor.)

STEVEusaPA

You've put a lot of thought into discovering what Jim Davis @ NCI has been preaching and teaching for almost his entire life.

jesmed1

STEVEusaPA said:

You've put a lot of thought into discovering what Jim Davis @ NCI has been preaching and teaching for almost his entire life.

Wait, is Jim Davis the @captainco here? A few hours ago I just found his 7-year old thread about why combustion efficiency numbers were always wrong...with other pros disagreeing...which got me wondering who was right...which got me to reading up for myself...

Now I just have to find that discussion again...
https://forum.heatinghelp.com/discussion/162380/is-it-possible-to-test-a-boiler-efficiency/p1

Jamie Hall

You have also hit -- I suspect accidentally -- on why condensing furnaces running on gas can advertise high efficiencies -- they are capturing the latent heat in the water vapour, so if you are playing with the higher heating value all the time, the numbers work. This is also, however, one reason (there are others) why you don't see condensing furnaces on fuel oil: the amount of water vapour produced is much less, relatively speaking, and so the higher heating value isn't as different from the lower heating value, and there isn't much point in trying to capture it.

pedmec

To me you are incorrect in your assessment because when you are measuring your flue gas you are only measuring sensible heat, not latent heat. The process by which there is water vapor would mean that the heat energy required to create a water vapor is already absorbed, the latent heat of vaporization. so that heat has already lowered the flue gas temperature. That would be why you are not deducting the 7%. the products of flue gas that you read in the flue have already completed the process of combustion and latent heat of vaporization. You gain that 7% when you condense the flue gases in the secondary heat exchanger of a condensing boiler (7% is a number you are using so i will use that) .

So i totally disagree with your statement. That's is why anybody that has combustion analysts of boilers and furnace for 25 years like myself will tell you that i never see an actual 95-97% efficiency number because the combustion analyser cannot measure the latent heat, only sensible heat. And i have done well over 1000 test.

jesmed1

pedmec said:

To me you are incorrect in your assessment because when you are measuring your flue gas you are only measuring sensible heat, not latent heat.

I agree that the combustion analyzer only measures the sensible heat of the dry gas, and not the latent heat of the water vapor, but I'm not following the rest of your post. So let's look at how the combustion analyzer works.

If you read the link I posted above with the technical details of combustion analysis, you'll find a formula on page 21 for the heat loss due to dry gas. That formula takes the concentrations of CO, CO2, O2, and N2 as measured in the flue gas flow. And it measures the flue gas temperature. It then calculates the thermal energy in that gas flow based on the known heat capacities per degree F of those gases, relative to room temperature. Another way to put it is that the analyzer finds out how much energy we could get back from those hot gases if we cooled them back down to room temperature.

The analyzer doesn't know or care about the water vapor. It only knows/cares about these "dry gases," and it's programmed only to calculate how much energy we could get back from those hot gases if they were cooled back down to room temperature, in BTU's per pound. Again, that number of BTU's we could recover from these "dry gases" is totally independent of the water vapor's latent heat.

Now the analyzer takes that BTU-per-pound number that is "locked up" in the dry gases only, and divides it by the number of BTU's in a pound of heat ing oil. When the analyzer does that math on my flue gases, it finds the "dry gas" has about 15% of the total BTU's that can be gotten from burning a pound of oil.

I actually just ran that "dry gas" heat loss calculation by hand based on the analyzed composition of my flue gases and stack temperature, and guess what...it came out to 15%, exactly what my boiler tech's combustion analyzer calculated. No water vapor/latent heat involved.

So the latent heat of the water vapor is a totally independent calculation. That's on page 22 of the link. That calculation just takes the percent of hydrogen in the fuel, figures out how much water weight it produces when burned, and then how much energy we'd get back from it if we could condense the water, again as a percentage relative to the total BTU value of oil per pound.

Note that these two calculations are entirely independent of each other. The combustion analyzer measures only the composition and temperature of the "dry gases" and figures out how much energy they carry, and again, this is totally independent of how much water vapor is in the flue gas.

Another way to think of this is that the combustion analyzer tells you only how much energy you would recover if you could cool ONLY the "dry gases" back down to room temperature. In my case, it's 15%. Then if I had a condensing boiler that could also condense the water vapor back down to room temperature, I'd get back another 7%.

I think what's confusing is that, in practice, it's impossible to separate out the "dry gases" and only cool THEM down to room temperature, without also condensing the water vapor. So in practice if we tried this, we'd end up cooling BOTH the mixed gas and vapor back down to room temp, and we'd end up recovering 15% "dry gas" heat + 7% water vapor heat = 22% of our BTU's.

Which is why the combustion analyzer confuses people, because it's based on an impractical question, which is "how many BTU's could we get back if we cooled ONLY the dry gases back down to room temperature, WITHOUT cooling the water vapor at all?" Of course, in practice, this is impossible. And the combustion analyzer has no way of measuring how much water vapor, and resulting latent heat, is in the flue gas. So we end up with a combustion analysis that tells us ONLY how much heat the "dry gases" are carrying away, without telling us anything about how much heat the water vapor is carrying away.

Daveinscranton

I saw my first condensing boiler 45 years ago.  I don’t think that it modulated.  It might have.  It was a guy who grew up in the trades.  I was there to buy a Selkirk stainless flue for a wood stove.

I stopped when I saw an odd constellation of parts in his shop.  It looked like a Reznor garage heater on top of a smallish boiler for hydronic baseboard heat.  He was burning outside air.  The garage heater was a condenser.  Natural gas came into his boiler, burned outside air, and warm water was headed towards the drain.  As I recall, he had to blow the exhaust gases out of the unit because they were pretty cold.  He laughed and told me about the day that the gas company showed up with the police as the gas company was convinced that his building used too little gas.  They tried to buy one of his contraptions as they left but he wasn’t selling as he built it as a bit of a hobby.

My guess is that a lot of mod cons are not maximizing efficiency.  The condensation has to take place in the boiler, or very near piping.  You will see a steady stream of condensate whether it comes from the boiler or the piping through the cold attic.

Does anyone make a retrofit add on to recover latent and otherwise lost heat from older CI boilers?

jesmed1

Daveinscranton said:

I saw my first condensing boiler 45 years ago.  I don’t think that it modulated.  It might have.  It was a guy who grew up in the trades.  I was there to buy a Selkirk stainless flue for a wood stove.

I stopped when I saw an odd constellation of parts in his shop.  It looked like a Reznor garage heater on top of a smallish boiler for hydronic baseboard heat.  He was burning outside air.  The garage heater was a condenser.  Natural gas came into his boiler, burned outside air, and warm water was headed towards the drain.  As I recall, he had to blow the exhaust gases out of the unit because they were pretty cold.  He laughed and told me about the day that the gas company showed up with the police as the gas company was convinced that his building used too little gas.  They tried to buy one of his contraptions as they left but he wasn’t selling as he built it as a bit of a hobby.

My guess is that a lot of mod cons are not maximizing efficiency.  The condensation has to take place in the boiler, or very near piping.  You will see a steady stream of condensate whether it comes from the boiler or the piping through the cold attic.

Does anyone make a retrofit add on to recover latent and otherwise lost heat from older CI boilers?

Dunno, but it may be more trouble than it's worth. You'd get maybe a 10-15% efficiency gain, and then you have to re-engineer the entire draft system for the boiler, which depends on a chimney stack full of hot air to keep the combustion process going correctly.

hot_rod

Now calculate actual run time and cycle efficiency.

Jamie Hall

and unless the chimney is lined to take it -- stainless steel -- it will live a short and unhappy life. Replacing it will eat any savings you might make.

EBEBRATT-Ed

This is why in a standard boiler (non condensing) that gas can never be as efficient as oil. Gas contains more water vapor than oil and that goes along for the ride. This is why on a cold day a chimney for a gas boiler you can see the water vapor exiting from the chimney as it starts to cool. You sometimes do on oil if it is really cold but oil has much less water vapor. Learned this in school 50 years ago

jesmed1

hot_rod said:

Now calculate actual run time and cycle efficiency.

@hot_rod Uh oh, more homework.

I honestly have no idea how to calculate cycle efficiency according to your example charts, so I’m going to take a different approach to estimating overall system efficiency based on an earlier analysis I did to find out where all the BTU’s from a 45-minute boiler run go.

Note that because we have oversized cold start boilers, every boiler run is roughly the same length of time based on the thermostat swing setting, and there's a lot of time between runs, giving the thermal masses enough time to cool back down to room temperature. So every run starts at 68 degree water temperature and ends at 140 degree water temperature in the boiler, with the average water temp throughout the system somewhat lower due to delta T. So the following numbers apply to every boiler cycle for the heating season, and as a result can safely be taken as a season average.

After adding up all our thermal masses, pipe lengths and diameters, etc, and applying basic delta T math and specific heat capacities for different materials (water, cast iron), I get the following results.

For Weil-McLain WGO-5 running 45 minutes:

Oil input gph: 1.18
Oil BTU/gal: 138,500
BTU input rate: 163,430 BTU/hr
BTU input during 45 minute run: 122,573

Stack Losses
15% dry gas loss: 18,386 BTU
7% latent heat loss: 8,580 BTU

Useful heat output after stack losses: 95,607 BTU (78% of input)
(Let’s round up to 96,000 BTU to simplify)

So where did the useful heat go during the 45 minute burn? Most of it got absorbed into the water and radiator thermal masses as stored energy that remained after the boiler shut down. Some of it was radiated into the house by the radiators and supply pipes. Some of it got absorbed into the boiler cast iron and remained as stored energy. Here’s the rough breakdown:

Stored in water: 52,400 BTU
Stored in rads: 22,100 BTU
Stored in boiler cast iron: 8,700 BTU
Emitted by rads: 7,200 BTU
Emitted by pipes: 5,600 BTU

Total: 96,000 BTU

So most of our useful heat ended up stored in the thermal mass. Now where does it go? Most of it gets emitted into the house as the thermal mass cools. The remainder is lost up the flue as air continues flowing through the boiler, cooling the heat exchanger and the water content.

We can estimate how much of the residual heat could be lost up the flue by making a worst-case assumption that all of the BTU’s stored in the boiler cast iron and internal water volume get lost up the flue. By coincidence, the BTU’s stored in the water volume inside the boiler equals the BTU’s stored in the boiler cast iron, 8,700 BTU each. So that’s 17,400 BTU total stored inside the boiler thermal mass. If our worst-case assumption is true, all of that goes up the flue after boiler shutdown.

That leaves us with 96,000-17,400=78,600 BTU’s emitted into the house. Then dividing by the total BTU input during the 45-minute run,

78,600/122,573=.64, or 64% overall efficiency.

But our worst-case assumption that all the BTU’s left stored in the boiler thermal mass went up the flue probably isn’t correct. Water continued to circulate by gravity through the boiler to carry some of that residual heat back into the house. What if half of it got carried back into the house, and half went up the flue? Then

96,000-8,700=87,300 BTU
87,300/122,573=.71, or 71% overall efficiency.

So, worst case, our overall efficiency was as low as 64%. Best case would be if no residual heat went up the flue, giving us an upper limit of 78%. The reality is probably somewhere in the middle, maybe around 71%.

MikeAmann

This is a good discussion. Keep it going.

jesmed1

Thanks, Mike. I expect @hot_rod will be back soon to grade my homework!

ChrisJ

I've wished for years that there was an easy way to measure the condensate coming back from my system accurately. If I could do that I could tell how much actual steam the system is producing vs the gas it's using.

hot_rod

jesmed1 said:

Thanks, Mike. I expect @hot_rod will be back soon to grade my homework!

Tell us more  about hour system. The type of heat emitters, is it multiple zones?
Nearest weather station

ethicalpaul

ChrisJ said:

I've wished for years that there was an easy way to measure the condensate coming back from my system accurately. If I could do that I could tell how much actual steam the system is producing vs the gas it's using.

Easy, I think, especially at the pressure you run. Make your wet return go into a graduated bucket. Run the pipe or hose low into the bucket and start with the bucket like 1/2 full.

now i have another video to make

jesmed1

hot_rod said:

Tell us more  about hour system. The type of heat emitters, is it multiple zones?
Nearest weather station

@hot_rod

Weather station KOWD (Norwood MA airport)
Average HDD is around 5600
1200 gal/yr consumption, for heating only (no DHW)
Cast iron radiators, converted gravity system
Total EDR=960 sq ft
Total occupied area=4800 st ft
Design heating load @ 0 degrees = 96,000 BTU/hr, or 20/BTU/hr/sq ft

Heat comes from two Weil McLain WGO-5 boilers, each running 1.18 gph input, for 2.36 gph max total
One zone per boiler (each boiler heats one side of the house, and each side has one zone)

Boilers are 85% dry gas efficiency, minus 7% latent heat loss for oil, so 78% combustion efficient
Cold start with typical run time 45 minutes, followed by 3-6 hours off
Starting water temp=65F, ending temp=140
On a design day, boilers will run 45 minutes on, then 90 minutes off, for a 33% duty cycle
During -10F polar vortex with 40 mph winds (Feb 2023), boilers ran just 44% duty cycle
Massively oversized!

hot_rod

Since you seem to like crunching numbers..
I think this BIN data can be useful for analyzing heating systems, performance, efficiency, etc.

Some bucket time needed to look into these efficiency numbers.

I suspect few boilers operate at steady state condition for much of their life. If so the math is simple.

Steady State= heat output in Btu/hr ÷ energy input of boiler

128,000 ÷ (1.18) ( 140,000 Btu/ gal #2)= 77% efficient


Cycle Efficiency= total heat output over period of time ÷ energy content of fuel used over that time. Obviously a lower % number

Run Fraction= burner on÷total elapsed time

5 minutes on ÷ 5 min + 20 min = 20%

Look at a partial load condition with 128K boiler supplying a 40K load
40,000 ÷ 128,000= 31%

Plot that 31% on this Brookhaven developed chart, at 30% efficiency goes south quickly.

If you know the actual heat load of the building you can go a step further

Heat load of 100,000 (70°-40°) ÷ 70°-0°= 20,000 BTU/hr - minus internal gains, call it 20,000

22, 860 ÷128,000 = 17.8%. on the graph, looks like you are slipping into the mid to low 70% range

Use your actual numbers. An unknown is which boiler output number to use IBR net assumes 15% heat loss. To tighten that number a piping heat loss could be used. In a system with large uninsulated piping, boiler in cold spaces that 15% number could be off by a bit?

Formulas from Modern Hydronic Heating & Cooling 4th edition Chapter 3

EBEBRATT-Ed

@ChrisJ A valve in your return line at the boiler with a tee upstream of the valve. with valve shut .Run the tee into a bucket and weigh the buckets' on a scale to get pounds of water. Pound of water is a pound of steam. 970 btus makes a pound of steam from and at 212 deg.

Not included is any steam in the header that returns to the boiler via the equalizer.

And measuring the sensible temp rise in the boiler would be a little more difficult. Measure boiler water temp. Fire the boiler and clock gas meter and stop clocking when the header starts to steam. None of this will be 100% accurate but should be close enough

EdTheHeaterMan

This is all very interesting, however, having the knowledge of the efficiency to the accuracy of within say .001% and accurately measuring the fuel to the .001 cubic centimeter and the actual liquid propane measured to the accuracy of within .001 cubic centimeter then also knowing the exact amount of natural gas used to within .00001 cubic feet, then comparing the actual healing valvue of each fuel and the exact amount of heat loss from each home or building that a given appliance will get you some very interesting information to make a detailed report about your findings.

After all that, you will still pay the same thing amount of money to heat your home at any given outdoor temperature based on the existing equipment and conditions you encounter in real time.

But it is an interesting discussion. I may have to actually draw an illustration on how @ethicalpaul might construct a condensate retrieval video. I'm on the edge of my seat