Combustion efficiency and analyzers

Gordan

When the tech was doing combustion analysis with his Testo (I failed to note the model) on my Prestige Solo 60, it was reporting 91.6% efficiency. I know that this is different from thermal efficiency, which is in turn different from AFUE, but I'd expect to see better numbers. Shouldn't combustion efficiency be in the high 90s, with thermal efficiency dependent on return temps and burner modulation?

Mark Eatherton

What are the numbers used to arrive at that efficiency factor?

I NEVER pay any attention to that number because there are no less than 1000 ways to get there. People who don't understand how to use these tools are focused on that number, and really impressed by it. What is important is what the appliance does to the oxygen content, stack temp (net) and CO production. Did he give you a read out?



It is also not uncommon to see low combustion efficiency numbers on the read out during high fire.



Mod cons do best working against a load. going up hill, with a small fire...



ME

Mike Kusiak_2

Combustion efficiency and condensing

Exactly as Mark said, combustion efficiency measurements are subject to a number of factors and variables. One of the most significant is the amount of condensing actually occurring, which can only be determined by measuring the amount of condensate produced, along with the chemical composition and temperature of the flue gas. Here is a quote from the Testo combustion guide, regarding the variables in calculating combustion efficiency:



"It should be noted that there is not a national industry standard for calculating measured

efficiency with a combustion analyzer. Manufacturers of analyzers use differing

calculations to derive efficiency values.  Oftentimes this discrepancy is due to values that

have been extrapolated into the condensing range.

Heat removed from the flue gasses on a condensing furnace is latent or hidden heat.  A

combustion analyzer that measures only temperature and not volume of condensate

cannot measure the quantity of heat removed from the flue gas during the condensing

process. Although terms of thermal and combustion efficiency are often used

interchangeably on non-condensing units, they cannot be used in the same manner on

condensing appliances.

The thermal efficiency of a condensing appliance and combustion efficiency will be

different.  The only way to calculate the actual thermal efficiency of an appliance is to

measure the exact airflow across the heat exchanger and the change in air temperature

across the heat exchanger and input the measured values into the sensible heat

formula to calculate the heat energy input into the conditioned air. There will be some

minimal loss to the furnace cabinet by radiation and conduction. Depending on how

much of the heat energy is extrapolated from the water in the flue gas, an average of 970

BTU per pound, the efficiency readings can differ by as much as 10%.  This assumes

that either all latent heat energy was extracted from the flue gasses after they reached

the dew point or none of the latent heat energy was extracted.

This extrapolation of values is distorted, and has led manufacturers of appliances to

inadvertently post higher than actual thermal efficiency numbers. Due to the readings

achieved on their analyzer. (NOTE: This calculation does not affect the AFUE numbers,

which are derived by a different means.) By not taking this discrepancy into account,

some in the industry have suggested that fuels are being delivered with low BTU levels.

This leads them to suggest that fuel pressures be raised to provide the net heat output

that the manufacturer has published. For this reason, Testo is recommending that the

fuel pressure be set per the manufacturer's instructions. The combustion efficiency will

then be a function of the actual dry flue gas and not of the thermal efficiency of a

condensing appliance. This avoids use of a calculated rather than a measured parame-

ter. Testo has chosen to use a combustion calculation that does not extrapolate the ther-

mal efficiency values of flue gasses below the dew point, as those values are not

representative of the heat that is removed from the flue gasses during the condensing

process. Although this may result in the appearance of a lower thermal efficiency of the

appliance, the science used for measuring combustion efficiency is not artificially high.

Once differences in combustion and appliance thermal efficiency are understood, the

methodology of scientific measurement verses extrapolation of measured values can be

appreciated and applied, allowing manufacturers to publish combustion and thermal

efficiencies that are representative of the actual efficiency of their appliance, thereby

creating a standard that is based upon actual measurement rather than an extrapolation."



Most efficiency measurements use the "higher heating value" of the fuel used, which includes the latent heat of condensing the water vapor, so if no condensing occurs, you are limited to about 90% efficiency even if the combustion is ideal. The Europeans do not typically use the higher heating value,  and therefore come up with efficiency numbers like 109% when in full condensing mode.



Certainly makes for confusion?

Tim McElwain

Gordan glad you got your system running

Just a little to add to Mark and Mikes excellent explanations.



The real demon if you will to good efficiency is to be able to control excess air. On atmospheric equipment were air is taken from the room the appliance is operating in can be all over the place.



On Modulating/Condensing equipment the ability to control the gas/air mix is much better and excess air is not usually factor as we can control operation by adjusting the throttle screw on most of the gas valve/combustion blower adaptations.



As excess air is increased two things will be affected:



1. Stack temperature will go up as we have a larger package we are trying to vent.



2.CO2 will go down and therefore we see a reduction in actual efficiency.



One other part of the difficulty with these systems is that they modulate and we can't adjust them at every step. We try to adjust on high fire and then low fire but it is still not an exact science.



The Lamda Pro system from Viessmann is the answer to the future as it will adjust itself on each step in the overall ratio of firing rates AUTOMATICALLY. No need to have to do any adjustments at start-up.



One last thing, I have tested equipment with six different testers being used to do a combustion analysis. Interesting they are very often very close in their analysis. The variable is usually CO air free due to some tester calculating the NOX out of the analysis.

Gordan

Thanks for the explanation, guys!

If I'm getting this right, the combustion analyzer doesn't account for latent heat recovery through condensation - correct? That would explain why efficiency didn't change from low to high fire (much) nor from low supply temps (=condensing) to high supply temps.



I had thought that the number it was reporting was the completeness of combustion, which was a base-level misunderstanding that had me concerned. But to do that, it would have to detect unburnt fuel components (other than CO) which, looking at the specs, it clearly doesn't.

Combustion_guy

More info on condensing eff

The issue gets even more complicated.   Combustion efficiency must be divided into 2 sections Dry (sensible) and wet (latent) losses.  Combustion analyzers only read the dry losses.  For a natural gas burner the latent losses represent 9.7% of the total heat available.  To calculate how much of this energy you are recovering you need to first measure how many pounds of water per hour you are producing.  Next you need to know how many pounds of water per hour you are actually producing.  Simply put, a good rule of thumb is, for every 24000 btu of gas burnt, you make 2.2 pounds of water.  This means, for an acurate calculation, you need to actually clock the gas firing rate (cfh), know the barometric pressure (gas laws) and the rated btu per cubic feet of the gas in your area.  Natural gas is actually a blend of different gases and can have a standard BTU/cubic foot value between 970 and 1150 when measured at sea level.  When you know how much you produced, you can calculate the ratio of recovery and (rule of thumb) calculate your % efficiency recovery rate.

 I also want to point out that the key to understanding a condensing appliance is realizing that the flue gases have a dew point.  Whenever the heat exchanger surface temperature is below the dew point, water will condense.  Basically dew point is the ratio of the dry flue mass and the water vapor mass.  If you add extra combustion air (excess combustion air) the dry mass increases and the dew point temperature drops.  This makes it harder to condense any water.

Another interesting fact is the amount of condensing taking place is very dependant on the temperature differance between the dew point and the entering air or water around the heat exchanger.  The standard AFLUE and other tests do not take this factor into account.  During testing we have actually plotted a 0.3% efficiency increase for every 10 F drop in inlet temperature. 

Additional factors effecting efficiency include the actual discharge temperature.  Typically in the field, the first thing that is done is to drive the unit to high fire.  This raises the heat exchanger temperature such that the temperature is now above the dew point and condensation stops.  The only way to get a true efficiency number is to be there on a design day and measure all the parameters.  At that point the numbers need to be entered into a combustion program that takes into account all the variables.  We can measure all the variables under lab conditions, but for field conditions it is much more difficult.  At best the numbers can be 2 or 3% out.   

Tim McElwain

Gordan we must by

now be blowing your mind. Al the postings from all of are correct. However for a simple adjustment of your unit a basic combustion test is all you are going to get out of most technicians. If you find one that rolls a bunch of laboratory equipment into your basement let me know I want to hire him.



Thanks for all the good input everyone.

Gordan

Tim

This is not blowing my mind, but it sure is feeding it! What blows my mind is just how much excellent knowledge and advice is freely dispensed at this site!

Mike Kusiak_2

Question, Tim

I am a little confused by something you wrote above:



"As excess air is increased two things will be affected:







1. Stack temperature will go up as we have a larger package we are trying to vent.







2.CO2 will go down and therefore we see a reduction in actual efficiency."





If you increase the excess air, I would expect it to result in lower combustion temperatures because you are heating additional air which cannot actually chemically react and contribute to combustion. I would think it would result in a larger volume of exhaust gas, but at a lower stack temperature.



What am I missing?

Tim McElwain

Yes Mike it would seem that way but it is not

especially on atmospheric equipment. The Mod/Con stuff is pretty locked in with air mix and there is not much room for error.



Here is the deal excessive excess air decreases efficiency. The stack temperature to a point will increase especially in the range of 50 to 150% excess air. As you continue out on the O2 and CO2 curves it starts to cool down. In the lower area of excess air however you are attempting to put a larger package flue gases plus extra air. This will somewhat increase the flow velocity of the flue gases pulling heat from the inside of the heat exchanger that otherwise would have contributed to heat transfer into the exchanger.



The combination of increase in O2 and the obvious decrease in CO2 along with the change in velocity can really drop efficiency way down. Things like leaving doors off, boilers on cement blocks, use of draft hoods instead of on gas double swing barometric all contribute to this. By the way when all of this is happening Carbon Monoxide goes up as we are actually cooling the flame therefore increasing the amount of unburned carbon.



I do not have the exact figures handy on the flow velocity increase versus heat loss but it is down at my training center if I can find it.

Mike Kusiak_2

Thanks

What I was not taking into account was the decrease in heat transfer to the exchanger caused by the greater volume and velocity of flue gas with excess air. So even though the combustion temperature decreases, the reduced heat transfer more than makes up for it, and the stack temp increases.



I actually noticed the effect when checking out a badly adjusted oil burner recently. It was running about 150% excess air ( with terrible CO) and a stack temp of 640F. I partially covered the burner air intake, and the excess air and CO dropped immediately along with the stack temp to about 610F. It puzzled me at the time because it seemed counter-intuitive, but with your explanation it now makes sense.



Turns out the oil service company recently changed the nozzle to a smaller size, but never adjusted the air to compensate.