Page 129 - Boiler_Operators_Handbook,_Second_Edition

114 Boiler Operator’s Handbook

some efficiencies listed at other firing rates in your losses from 100 to get percent efficiency. Surprisingly
boiler documentation you’ll notice that those others are it is easier and far more accurate to determine some of
labeled “predicted performance” and only the full load the heat losses as a percent of the input so determining
is guaranteed. The problem with that wisdom is the efficiency using the heat loss method is the most widely
boiler seldom, if ever, operates at full load. Whenever accepted method.
you have input, suggest that any new boiler you pur- The Power Test Code (PTC-4.1) provides a struc-
chase be guaranteed for performance at a load you will tured basis for calculating boiler efficiency by two meth-
have, say 50% or 75%. That doesn’t violate the ABMA’s ods, input-output and heat loss. All the larger boilers
rule. Today some chiller manufacturers, and possibly by we installed while I worked for Power and Combustion
the time this book is printed some boiler manufactur- were tested using both methods in a modified form of
ers may, guarantee the part load operating efficiency of the Power Test Code. The cost of performing those tests
their equipment. in strict accordance with the Code could not be justified
Occasionally you will see a boiler efficiency guar- for even the larger boilers (up to 200,000 pounds per
anteed at something around 50% to 75% load. That is hour of steam) that we installed. The primary modifica-
probably a sales tactic because the maximum operating tions we made to the Test Code included shorter test
efficiency of a boiler is typically in that range. As the runs (three hours instead of the required eight to twelve)
load and firing rate decreases the volume of flue gas and less frequent measurements (every twenty minutes
decreases. The heating surface, on the other hand, stays instead of every ten) so we could get two test runs in
the same. Therefore the flue gas spends more time in within one day and with only one man collecting data.
contact with a proportionally larger heating surface so Of course in those days we used an actual Orsat ana-
more heat is transferred. lyzer which took some time to operate, not one of those
You should notice that when you create your own nice electronic analyzers we have today.
performance documentation because the stack tempera- An examination of the results of the hundreds of
ture will drop as you reduce firing rate from full load. test runs we made revealed a typical deviation in the
Somewhere lower the efficiency will start to drop off input-output efficiency of as much as five percent while
because the flue gas is channeling so only a small por- the heat loss results were normally within one percent.
tion of it is contacting the heating surface. As the firing That’s why I can say, with a reasonable degree of confi-
rate decreases it becomes more difficult for the fuel and dence, that the heat loss method is very acceptable.
air to mix completely so excess air must be increased to I always get a kick out of some organizations indi-
prevent CO and efficiency suffers further. The radiation cating that they conducted hundreds of boiler efficiency
losses also become more significant as the load decreas- tests. During my twenty years at PCI we only ran about
es. All these factors influence the operating efficiency of two hundred boiler efficiency tests using that modi-
the boiler to different extents at different loads. fied approach to the Test Code. Each test did consist of
Heat loss efficiency is determined by backing into several test runs so I can say we made hundreds of test
the value. An efficiency is considered to be the output runs. Those were formal tests that included a printed
(what you get out of it) divided by the input (what you report with all the calculations, records of collected data,
put into it) with the result of the division multiplied by and fuel analysis. They were not boiler tests conducted
100. in strict accordance with the Test Code but they were a
Efficiency = Output ÷ Input × 100 lot closer than what some people call a boiler efficiency
test.
The loss, and in the case of a boiler it’s a loss of I don’t consider a strip of narrow paper with a list
heat, is the difference between the input and output. of analysis values, temperatures, and a calculated boiler
Therefore, the output is equal to the input less the heat efficiency representative of a boiler test. Some firms that
losses. By substituting input less losses for the output in claim they’ve done hundreds of tests haven’t included
the formula we get a formula that doesn’t include out- one fuel analysis. Unless you have the fuel analysis the
put at all. test is simply flawed because the hydrogen to carbon
ratio of fuels varies considerably. The modern flue gas
Efficiency = (Input – losses) ÷ Input × 100 analyzer contains programmed calculations based on an
assumed fuel analysis and the odds that your fuel and
If we can calculate the losses as a percent of the the values used by that program are identical are slim to
input then all we have to do is subtract the percent none. The results are only representative and based on