BASE DATA
If you are burning heavy oil ,it is strongly recommended that before the actual evaluation a sample of the fuel should be obtained and sent to a suitable Laboratory for detailed analysis. If you don't do this, you can usually get by with specs and and data from your files for the time being. Alternately, your fuel supplier should be able to provide you with detailed analysis of the fuel purchased.
It will also be vital that suitable gas sampling holes exist in the flue gas system. It is strongly recommended that holes large enough to accept 1/4" copper tubing be drilled and plugged if they do not already exist.
Using the survey sheet, obtain all of the basic characteristics of the system such as the boiler rating and the cost of the fuel.
BASELOAD
Have the boiler in question placed on manual control and adjust the load to an appropriate level that can be maintained for 30 to 40 minutes. It will usually take from 5 to 15 minutes for the boiler to stabilize at steady thermal equilibrium.
DATA COLLECTION
As soon as the boiler appears to be holding a steady load, arrange to take all of the survey readings within a 5-minute period and accurately record the values. If working as a team, arrange that one member takes the oxygen and stack temperatures at a specific time while the other team member takes the boiler load and control panel readings at exactly the same time. In order to use the steam meter and fuel integrator readings it will be necessary to take the readings at a precise time.
These readings should be entered into the computer program as soon as taken for a rough or preliminary calculation.
COMPUTATION
Enter all of the data exactly as prompted on the screen being careful to use the appropriate units. The program is designed to trap major inputting errors such as reporting steam rates in lb/h rather than in thousands of lb/h but you must still take responsibility for the data entry.
If the first set of readings entered into the program fails to close exactly, the program will indicate that the input/output efficiency does not agree with the loss efficiency. At this point it is desirable to try and establish the possible source of any measurement errors so choose error analysis at the appropriate prompt.
ERROR ANALYSIS After displaying both the input/output and the loss efficiency various messages will appear based on the magnitude of the error.
Next you will be given an opportunity to choose between the fuel, the feedwater or the steam as being the most probable or probably correct value. If the boiler is equipped with a feedwater meter, this would be an appropriate time to check this. In the majority of cases, the steam meter will be more accurate than the fuel meter but, most plants will have a fairly good estimate of the error inherent in the fuel meter based on drawdown from the oil tanks, monthly purchases, etc.
If the boiler is exposed to outside conditions, then you should answer YES to this question and try to determine the ambient conditions the boiler is exposed to. It's possible to estimate the overall boiler skin temperature by taking some spot readings and an estimate will have to be made of the ambient air velocities in the neighborhood of the boiler. Although the program can compensate for additional radiation and convection loss as a function of temperature differences and air velocities, it does assume that the boiler casing is dry and, therefore, a heavy rain could introduce an error that cannot be compensated for.
You should recheck the blowdown percentage using chlorides, conductivity and/or silica depending on the geographical location.
If the boiler is equipped with an air preheater, you should take both temperature, oxygen and CO2 readings both ahead and after the air preheater.
In the majority of cases, on-line oxygen monitors are installed ahead of the air preheater and therefore may be reading oxygen that are significantly different than the oxygen measured at the stack. Since we must use the stack temperature in the efficiency calculations, it is vital that the true oxygen level at the stack be taken.
If two fuels are being burned simultaneously, for example, hog and oil, it is generally better to use the more reliable fuel measurement as the base for calculation at this time. The secondary fuel can then be estimated by the computer program.
The program does not use the CO2 readings directly but it will calculate a theoretical CO2 to enable you to cross-check your oxygen meter calibration.
It is not unusual to find the stack thermo couples to be significantly off calibration. Using the portable thermo couple and the digital Thermometer, you can get a good approximation of the flue gas temperature.and also on the fuel flow meters. In th the steam flow will be more reliable or more consistent than the instantaneous chart readings. These integrator readings should be taken with a stop watch over a 10 to 15 minute period.
FORCED CLOSURE The program will automatically calculate the fuel rate, or the steam rate or the fuel BTU value required to close the heat balance. It will then display the most probable efficiency and the pounds of steam generated per pound of fuel consumed and the net cost of generating 1000 LB of steam.
There will now be several optional directions that you can take within the program. You will probably want to display the estimated heat balance of the system or perhaps rerun the error analysis portion. Normally you will wish to enter new survey data at this point.
NEW SURVEY DATA
We recommend you take 3 sets of data at each constant condition. You should then change the conditions significantly. Change more than one variable at a time!! That is; change the load and the excess air or change the load and the oil/gas ratio at the same time. The computer needs the widest variation you can provide. Do not try to keep everything the same only changing one variable at a time!
You should repeat the process described above for a minimum of 5 different sets of readings, preferably at different fixed loads and/or different excess oxygen conditions differing by 10% each time. You should expect the stack temperature to change upwards with increasing loads and normally the stack temperature will increase with increasing excess oxygen.
Once you have a minimum of 5 reliable sets of data in the program, you may call up the regression analysis portion of the program. The program will have developed three different mathematical models relating loss efficiency, input/output efficiency and stack temperature against varying boiler loads and excess oxygen conditions.
You can use the mathematical models to play, "WHAT IF" by entering hypothetical sets of conditions. However, when asked for a boiler load, you should input a load not too far below the lowest reading you have taken. The second load should be a value not more than about 20% higher than the highest load that you actually tested at.
The program will then automatically print out a graph relating efficiency to load and excess oxygen.
(not available in the LITE version)
If the same graph is used for both the loss efficiency and the input/output efficiency, the difference in the ide an indication of possible instrumentation discrepancies that are a function of changing load or changing oxygen conditions.
These graphed results will be utilized as a comparison with the next time this same program is run from the same boiler. It will, therefore, be possible to detect relatively small differences at some specific standardized load and excess oxygen condition without having to exactly duplicate these conditions during the test run.
The results from the regression printout and the resulting graphs are therefore the most important result of this survey and the computer analysis.
THE PAYOUT
The most useful information that will be developed for you will be the Best Efficiency and Lowest Cost steam loads (often different). You may think of these conditions as the sweet spots for operation. By tracking these values and the Standard Condition values over time you will develop a clearer picture of possible fouling or changing conditions such as air distribution or leakage.