Accepted Manuscript

Title: A Shortcut Procedure for Calculation of Process Side

Heat Duty of Refinery Fired Heaters

Author: Constantinos Plellis-Tsaltakis

PII: S0263-8762(17)30321-0
DOI: http://dx.doi.org/doi:10.1016/j.cherd.2017.06.002
Reference: CHERD 2704

To appear in:

Received date: 23-12-2016

Revised date: 2-6-2017
Accepted date: 5-6-2017

Please cite this article as: Plellis-Tsaltakis, Constantinos, A Shortcut Procedure for
Calculation of Process Side Heat Duty of Refinery Fired Heaters.Chemical Engineering
Research and Design http://dx.doi.org/10.1016/j.cherd.2017.06.002

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
A Shortcut Procedure for Calculation of Process Side Heat Duty of Refinery Fired

Heaters

Constantinos Plellis – Tsaltakis *

* Corresponding author, Chemical Engineer Ph.D., M.B.A., Green Industrial Technologies,

18 Glafkonos str., Athens 11632, Greece. E-mail address: cpt@gitech.gr, Tel. +30

6977272843, Fax: +30 211 268 6912

Graphical abstract

Highlights
 Fired heater absorbed duty calculation requires rigorous vapour – liquid equilibrium
calculations.
 Usually costly process simulation software is used.
 Shortcut procedure is proposed, suitable for implementation in a worksheet.

1
  The accuracy of the proposed method is comparable to that of process simulators or to
the indirect method.
 Method may be used on regular basis to reliably monitor fired heater efficiency.

Abstract

Absorbed heat duty of process side of a fired heater often requires rigorous vapour – liquid

equilibrium calculations and is usually performed with costly process simulator software. An

alternative, shortcut method is proposed, which may be easily and efficiently implemented in

a spreadsheet. The results of the proposed procedure were checked against those of a process

simulator and against the results of the indirect efficiency calculation method, for three differ-

ent existing refinery furnaces. The accuracy of the proposed method was found comparable to

both the others.

Keywords

Fired Heater, Furnace, Refinery, Process side, Heat duty, Absorbed duty

1. Introduction

2
Energy efficiency is of paramount importance for modern refineries, mainly due to the high

cost of fuel. Efficient processes also reduce greenhouse gas emissions. Energy audits, metrics

and benchmarking are invaluable tools for achieving maximum efficiency. Fired heaters are

major energy consumers of a refinery, (Devakottai, 2015). Even small reductions in their effi-

ciency, may lead to dramatic increase in fuel consumption. Two methods are available for the

calculation of furnace efficiency, the direct and the indirect method. The indirect method aims

to calculate the heat loss of a fired heater, based on air excess measurements in the flue gas

and the stack temperature, (Abbi and Shashank, 2009), (ASME, 1974). The direct method

compares the released heat duty of fuel’s combustion to the absorbed duty of the heated fluid,

(Abbi and Shashank, 2009). Both methods have advantages and disadvantages. Usually, the

indirect method is used, because it is less demanding in both data and calculations. However,

small errors in the air excess due to instrument deviation or air ingress, may lead to severe

efficiency underestimation and wrong operating decisions (Lieberman and Lieberman, 2014).

On the other hand, direct efficiency calculation may call for vapor – liquid equilibrium calcu-

lations. The rigorous calculations may be performed with a process simulator. However, such

software is not always available; it is costly and demands some training from the user.

This paper aims to provide an absorbed duty calculation procedure for use in direct fired heat-

er efficiency estimation. The procedure calls for a series of calculations which can be per-

formed easily in a spreadsheet, without the use of special software.

The following paragraph describes briefly the direct and indirect calculation methods.

2. Theory

3
2.1 Furnace Efficiency Calculation Methods

The indirect calculation method considers that the heat losses of a fired heater equal the radia-

tion losses from the walls of the heater to the environment plus the losses to the stack, (Abbi

and Shashank, 2009), (ASME, 1974), (Patel, 2005). The radiation losses depend on the out-

side temperature of the furnace walls and their surface area. Typically they are considered to

equal 1.5-3% of the heater’s fired duty, (Arora, 1985).

The stack losses consist of the enthalpy that remains in the flue gases before they are sent to

the stack. The hotter the flue gases to the stack, the higher the losses. The temperature of the

flue gases have a lower limit, below of which acid gases may condense and corrode the heat

transfer equipment. This limit strongly depends on the amount of sulfur in the fuel and usually

is between 130 – 170oC, (Devakottai, 2015). The amount of heat lost to the stack increases as

the flow of flue gas increases. The mass flow of flue gas equals the amount of fuel burnt plus

the combustion air. Therefore, the higher the air excess, the higher the flue gas flow and the

heat losses to the stack. Air excess is usually maintained at a practical optimum of 3% vol. O2

in the flue gas, (Devakottai, 2015), which provides enough excess air to achieve complete

combustion and in the same time does not lead to excessive heat losses to the stack. The indi-

rect method uses excess O2 measurements in the flue gas and stack temperature to estimate

the heat losses.

The direct method on the other hand, calculates the absorbed duty, as the enthalpy change of

the process side fluid and compares it to the heating value of the fuel burnt. Usually, this

method is more difficult to implement than the indirect one, because it needs accurate meas-

urements of fuel and process side flow rates. Moreover, the calculation of the enthalpy change

4
of the process side may be difficult without the use of specific, costly software. However,

most modern refineries measure accurately enough both the process and the fuel flows in their

fired heaters. They also perform enough lab analysis to allow an adequate characterization of

the process side fluid and therefore an accurate estimation of its enthalpy change along the

furnace. The present work aims to provide a relatively easy and reliable method for the calcu-

lation of the absorbed duty, without resorting to special software solutions.

The absorbed duty is analyzed to its main components, namely heating duty, vaporization du-

ty, superheating duty for steam if superheating coils are present in the convection zone, su-

perheating duty for velocity steam if applicable, superheating duty for hydrogen if applicable

(hydrotreater fired heaters mainly). Each component’s contribution is calculated separately.

The proposed procedure has been developed with the aim to combine as much as possible

ease with accuracy. Therefore, liquid specific heats are not treated as constants, but their vari-

ation with temperature and feed composition are taken into account. Gas (steam and hydro-

gen) isobaric specific heat variation with temperature haw been considered. Hydrocarbon en-

thalpy of vaporization is estimated by the Riedel equation, (Riedel, 1954), taking into account

the composition of the vaporizing stream. Percentage of vaporization is not calculated by

complex, trial and error liquid – vapor equilibria, but is instead estimated using the generally

available distillation curves of the fired heater’s feed, after adjustment for the pressure under

which it takes place.

The following paragraph describes the mathematical equations needed for the absorbed duty

calculation of refinery fired heaters.

2.2 Process Side Heat Duty Calculation

5
The presented method calculates the process side heat duty of a refinery’s fired heater as the

sum of the following duties:

 Heating duty of the feed.

 Vaporization duty for partial evaporation of feed.

 Heating duty for injection steam (velocity steam) superheating.

 Heating duty for steam superheating, (in case one or more steam superheating coils

exist in convection section).

 Heating duty for hydrogen heating, (hydrotreaters)

2.2.1 Heating duty of the feed

The calculation of the heating duty of the feed assumes that the feed remains at the liquid

state from furnace inlet to furnace outlet temperature. The duty is calculated as:

𝐶𝑂𝑇
𝑄1 = 𝑚 ∫ 𝐶𝑑𝑇 (1)
𝐶𝐼𝑇

Where:

Q1: The heating duty of the feed, (kW)

m: The mass flow rate of the feed, (kg/s)

CIT: The furnace inlet temperature, (K)

COT: The furnace outlet temperature, (K)

C: The specific heat of the feed at liquid state, (J/gK)

6
Various correlations for C have been proposed, (Watson and Nelson, 1933), (Lee and Kessler,

1976). For the purposes of this work we have implemented the approach of Watson and Nel-

son, (1933). The correlation is summarized in equation (2).

𝐶 = 4.1868𝐴1 (𝐴2 + 𝐴3 𝑇) (2)

Where:

C: The specific heat of the feed at liquid state, (J/gK)

A1,A2,A3: Parameters related to feed’s composition, defined by equations (3),(4) and (5).

𝐴1 = 0.055𝐾𝑊 + 0.35 (3)

Where:

KW: The Watson characterization factor, defined by equation (6) , (unitless)

𝐴2 = 0.6811 − 0.308𝑆𝐺60 (4)

Where:

SG60: The feed’s specific gravity at 15.6oC (60oF)

𝐴3 = 0.000815 − 0.000306𝑆𝐺60 (5)

Where:

SG60: The feed’s specific gravity at 15.6oC (60oF)

𝐾𝑊 = 3√1.8𝑇50 /𝑆𝐺60 (6)

Where:

KW: The Watson characterization factor, (unitless)

T50: The normal mean boiling point of the feed, (K)

7
SG60: The feed’s specific gravity at 15.6oC (60oF)

Integrating equation (1), using equation (2), leads to equation (7).

𝐴1 𝐴3 (1.8𝐶𝑂𝑇 − 459.688)2
𝑄1 = 2.3263[𝐴1 𝐴2 (1.8𝐶𝑂𝑇 − 459.688) + − 𝐴1 𝐴2 (1.8𝐶𝐼𝑇
2

𝐴1 𝐴3 (1.8𝐶𝐼𝑇 − 459.688)2
− 459.688) − ] (7)
2

Where:

Q1: The heating duty of the feed at liquid state, (kJ/kg)

A1,A2,A3: Parameters related to feed’s composition, defined by equations (3),(4) and (5).

CIT: The furnace inlet temperature, (K)

COT: The furnace outlet temperature, (K)

2.2.2 Vaporization duty for partial evaporation of the feed

The calculation of the vaporization duty for partial evaporation of the feed assumes that part

of it vaporizes isothermally, at T = COT. The amount of feed vaporization and the molar va-

porization duty of the feed are needed to calculate the heat duty.

The calculation of the amount of feed vaporized is usually the result of complex vapor – liq-

uid equilibrium calculations. For the needs of this paper, these calculations are substituted by

the feed’s True Boiling Point (TBP) distillation curve. TBP distillation curves are not very

common for materials other than crude oil. Other distillation curves are easier to obtain, such

as atmospheric distillations as per ASTM D-86 for naphtha, kerosene and gasoil, vacuum dis-

tillation as per ASTM D-1160 for vacuum distillates and atmospheric residue and simulated

8
distillation methods as per ASTM D-2887 for heavier materials. There are correlations for

conversion of one standard distillation curve to another, (Daubert and Danner, 1997), (Edmis-

ter and Pollock, 1948), (Riazi, 2005). This paper assumes that either original TBP distillation

data or curves derived from other standard distillation methods are available. TBP distillation

curves are expressed at an absolute pressure of 100kPa(a). Therefore, in order to calculate the

amount of material vaporized at the COT the curve must be calculated at the actual partial

pressure of hydrocarbons at operation. The actual operating hydrocarbons partial pressure is

calculated using equation (8).

𝑀𝑜𝑙𝐻/𝐶
𝑃𝐻/𝐶 = 𝑃 (8)
𝑀𝑜𝑙𝐻/𝐶 + 𝑀𝑜𝑙𝐼𝑛𝑒𝑟𝑡𝑠

Where:

PH/C: The partial pressure of hydrocarbons at furnace’s outlet, at operation, (kPa(a))

P: The absolute pressure of hydrocarbons at furnace’s outlet, at operation, (kPa(a))

MolH/C: The molar flow rate of hydrocarbons, (kmol/h)

MolInerts: The molar flow rate of inerts, (steam, H2, H2S etc.), (kmol/h)

The TBP distillation curve points at the PH/C can be calculated using equations (9)-(11),

(Maxwell and Bonnell, 1957).

5.994296 − 0.972546 log10 (0.1333𝑃𝐻/𝐶 )

𝑋= (9)
2663.129 − 95.76 log10 (0.1333𝑃𝐻/𝐶 )

Where:

X: Parameter for equation (10)

PH/C: The partial pressure of hydrocarbons at furnace’s outlet, at operation, (kPa(a))

9
 748.1𝑋𝑇
𝑇𝑃𝐻/𝐶 = (10)
1 + 𝑇(0.3861𝑋 − 0.00051606)

Where:

X: Parameter, defined by equation (9)

TPH/C: The TBP distillation curve temperature point at the partial pressure of hydrocarbons at

furnace’s outlet, at operation, (K)

T: The TBP distillation curve temperature point at 100 kPa(a), (K)

𝑓 = 0, 𝑖𝑓 𝑇𝑃𝐻/𝐶 < 366𝐾

𝑇𝑃𝐻/𝐶 − 366
𝑓= , 𝑖𝑓 𝑇𝑃𝐻/𝐶 < 366𝐾 < 477𝐾
111

𝑓 = 1, 𝑖𝑓 𝑇𝑃𝐻/𝐶 < 366𝐾 (10)

Where:

f: Correction parameter, for equation (11)

TPH/C: The TBP distillation curve temperature point at the partial pressure of hydrocarbons at

furnace’s outlet, at operation, (K)

𝑇𝐶𝑃𝐻/𝐶 = 𝑇𝑃𝐻/𝐶 + 1.389𝑓(𝐾𝑊 − 12)(log10 (0.1333𝑃𝐻/𝐶 ) − 2.8808) (11)

Where:

TCPH/C: The TBP distillation curve temperature point at the partial pressure of hydrocarbons

at furnace’s outlet, at operation, corrected for the aromaticity of the petroleum fraction, (K)

f: Correction parameter, defined by equation (10)

KW: The Watson characterization factor, as defined in equation (6) (unitless)

PH/C: The partial pressure of hydrocarbons at furnace’s outlet, at operation, (kPa)

10
The amount of feed vaporization may be estimated by linear interpolation within the TBP dis-

tillation curve temperature points, at the partial pressure of hydrocarbons at furnace’s outlet,

at operation, corrected for the aromaticity of the petroleum fraction, as calculated by equation

(11).

Riedel equation is used to estimate the molar vaporization duty of hydrocarbons, (Riedel,

1954).

ln(𝑃𝐶 /100) − 1.013

𝛥𝛨𝑉𝐵 = 1.093𝑅𝑇𝐵 (12)
𝑇𝐵
0.93 − 𝑇
𝐶

Where:

ΔΗVB: The molar vaporization duty of the hydrocarbon, at its normal boiling point,

(Joule/mol)

R: The universal gas constant, (8.314 Joule*mol-1*K-1)

TB: The normal boiling point of the hydrocarbon, defined as the temperature of 50% evapora-

tion of the TBP curve, at 100 kPa(a) pressure, (K)

PC: The critical pressure of the hydrocarbon, as estimated by equation (13), (kPa(a))

TC: The critical temperature of the hydrocarbon, as estimated by equation (14), (K)

The critical properties of a hydrocarbon fraction may be estimated by the Kesler - Lee equa-

tions, (13) and (14), (Kesler and Lee, 1976) from its average normal boiling point and specific

gravity. The equations have been modified for use of SI instead of the original imperial units.

11
 0.0566 4.12164 0.213426
ln 𝑃𝐶 = 17.2019 − − (0.43632 + + 2 ) 10−3 𝑇𝐵
𝑆𝐺60 𝑆𝐺60 𝑆𝐺60

11.81952 1.5301548
+ (4.75794 + + 2 ) 10−7 𝑇𝐵2
𝑆𝐺60 𝑆𝐺60

9.9 −10 3
− (2.4505 + 2 ) 10 𝑇𝐵 (13)
𝑆𝐺60

Where:

lnPC: The natural logarithm of critical pressure of the hydrocarbon, (Pa)

TB: The normal boiling point of the hydrocarbon, defined as the temperature of 50% evapora-

tion of the TBP curve, at 100kPa(a) pressure, (K)

SG60: The average specific gravity of the hydrocarbon at 15oC, as estimated by equation (6)

when solved for SG60.

𝑇𝐶 = 189.83 + 450.56𝑆𝐺60 + (0.4244 + 0.1174𝑆𝐺60 )𝑇𝐵

(0.1441 − 1.0069𝑆𝐺60 )105

+ (14)
𝑇𝐵

Where:

TC: The critical temperature of the hydrocarbon, (K)

TB: The normal boiling point of the hydrocarbon, defined as the temperature of 50% vol.

evaporation of the TBP curve, at 100 kPa(a) pressure, (K)

SG60: The average specific gravity of the hydrocarbon at 15oC, as estimated by equation (6)

when solved for SG60.

12
Kesler and Lee (1976), presented also a useful correlation for the mean molecular weight of a

hydrocarbon fraction, which will be used in this paper together with equation (8), for the cal-

culation of the partial pressure of the hydrocarbons at the furnace outlet:

𝑀𝑊 = −12272.6 + 9486.4𝑆𝐺60 + (8.37414 − 5.99166𝑆𝐺60 )𝑇𝐵

222.466 7
(0.7465 −
2 ) 𝑇𝐵 ) 10
+ (1 − 0.77084𝑆𝐺60 − 0.02058𝑆𝐺60
𝑇𝐵

+ (1 − 0.80882𝑆𝐺60

17.3354 12
(0.32284 −
2 ) 𝑇𝐵 ) 10
− 0.02226𝑆𝐺60 (15)
𝑇𝐵3

Where:

MW: The mean molecular weight of the hydrocarbon, (g/mol).

SG60: The average specific gravity of the hydrocarbon at 15oC, as estimated by equation (6)

when solved for SG60.

TB: The normal boiling point of the hydrocarbon, defined as the temperature of 50% vol.

evaporation of the TBP curve, at 100kPa(a) pressure, (K).

Equations (12-14) may be used to calculate the vaporization duty of the hydrocarbon at its

normal boiling point. However, since the evaporation of hydrocarbons does not take place at

100kPa(a), the enthalpy of vaporization must be calculated at the operating conditions. For

this purpose, Watson correlation (Watson, 1933, 1943) may be used:

𝑇1 0.38
1−𝑇
𝐶
𝛥𝛨𝑉 = 𝛥𝛨𝑉𝐵 ( ) (16)
𝑇𝐵
1−𝑇
𝐶

Where:

13
ΔΗV: The vaporization duty of the hydrocarbon, at the operating temperature, (Joule/mol)

ΔΗVB: The vaporization duty of the hydrocarbon, at its normal boiling point, (Joule/mol)

T1: The operating temperature, (K)

TB: The normal boiling point of the hydrocarbon, defined as the temperature of 50% vol.

evaporation of the TBP curve, at 100kPa(a) pressure, (K)

2.2.3 Heat duty for steam superheating / gas heating

Many fired heaters have one or more steam superheating coils in their convection sections.

Moreover, live velocity steam may be injected in the coils of some furnaces. Hydrotreater /

hydrocracker furnaces also usually heat hydrogen (or the recycle gas of the unit). Such duties

may be calculated by equation (1) if the isobaric specific heat (CP) of each gas is known. For

the needs of this paper, Shomate equation, (NIST, 2015), is used for CP.

𝐸
𝐶𝑃 = 𝐴 + 𝐵𝑡 + 𝐶𝑡 2 + 𝐷𝑡 3 + (17)
𝑡2

Where:

Cp: The isobaric specific heat of the gas at temperature T, (J/molK)

t: A function of absolute temperature T, t=T/1000, (K)

A~E: Constants specific to each gas.

Combining equations (1) and (17), the enthalpy change of gas superheating may be calculat-

ed:

𝐵 2 𝐶 𝐷 1 1
𝛥𝛨𝐺𝑆𝐻 = 𝐴(𝑡2 − 𝑡1 ) + (𝑡2 − 𝑡12 ) + (𝑡23 − 𝑡13 ) + (𝑡24 − 𝑡14 ) − 𝐸( − ) (18)
2 3 4 𝑡2 𝑡1

Where:

14
ΔΗGSH: The enthalpy change of the gas, heated from T1 to T2, (J/mol)

t: A function of absolute temperature T, t=T/1000, (K)

A~E: Constants specific to each gas.

Values for the constants A~E are available in the bibliography for a wide range of materials.

Tables 1 and 2, (NIST, 2015), (Chase, 1998), may be used for steam and hydrogen.

The next paragraph presents a procedure based on equations (1) - (18) for the calculation of

the process side duty of a fired heater.

3. Algorithm

Equations (1) – (18) can be used to calculate the process side (absorbed) duty of a fired heat-

er. The required data are:

 Fired heater mass feed rate.

 Fired heater velocity steam feed rate (if any).

 Fired heater hydrogen feed rate (if any).

 Superheater coil steam feed rate (if any).

 Feed TBP or other distillation curve.

 Feed Specific Gravity at 15.6oC (60oF).

 Coil Outlet Temperature and Pressure. In the case of distillation column feed preheat

furnaces, the flash zone conditions may be used instead.

 Coil inlet temperature.

15
  Superheater coil inlet and outlet temperatures, (if any).

 Velocity steam inlet temperature (if any).

 Hydrogen coil inlet temperature (if any).

The proposed algorithm consists of 18 steps and may be easily used to develop a computer

program. The following section describes the steps. The algorithm’s flow chart is depicted in

Figure 1. The number on the left of each step in Figure 1 corresponds to the step number in

the algorithm description below.

1. If a TBP distillation curve of the feed is available, proceed to step 4.

2. Convert all feed distillation data to TBP distillation curve. The conversion procedures

need as input the Watson characterization factor, KW. In case KW is not available, sup-

pose KW = 12.

3. Use the temperature of 50% vol. evaporation of the TBP curve calculated in step #2 as

input (T50) for equation (6), to calculate a new value for KW. Return to step #2 and use

the new value of KW as input for the calculation of the TBP curve. Repeat steps #2 and

3 until KW is calculated.

4. Calculate the mean molecular weight of the feed, using equation (15).

5. Using the result of step #4, the hydrocarbons feed rate, the coil outlet pressure, the ve-

locity steam feed rate if any and the hydrogen feed rate if any, calculate the partial

pressure of the hydrocarbons at the furnace outlet,(PH/C) utilizing equation (8).

6. Use the PH/C calculated in step #5, the KW of either step #1 or step #3 and the TBP

curve of the feed at 100 kPa(a), (either from step#1 or #3), to calculate the TBP distil-

lation curve points at the partial pressure of the hydrocarbons, implementing equations

(9)-(11).

16
7. Use the TBP curve calculated in step #6 and the coil outlet temperature of the furnace,

to calculate the %volume of evaporation (%EV), of the liquid hydrocarbon feed at the

coil outlet conditions, by linear interpolation within the points of the step #6 TBP

curve.

8. The volume average normal boiling point of the fraction of the feed that was evapo-

rated inside the furnace may be calculated from the original TBP curve, (either step #1

or step #2) as the temperature that corresponds to (%EV)/2 volumetric evaporation.

Use (%EV) from step #8 to calculate the volume average normal boiling point of the

evaporated fraction of the feed, (TEV50).

9. Use KW of either step #1 or step #3 and the TEV50 calculated at step #8 as input for

equation (6), in order to calculate the mean specific gravity of the fraction of the feed

that was evaporated inside the furnace. Equation (6) must be rearranged for the calcu-

lation of SG60.

10. Use the TEV50 of step #8 and SG60 of step #9 with equations (13) and (14) to calculate

the critical properties of the fraction of the feed that was evaporated inside the furnace.

11. Use the critical properties calculated in step #10 and the TEV50 of step #8 as input for

equation (12), to calculate the vaporization duty of the fraction of the feed that was

evaporated, at its normal boiling point.

12. Use equation (16), to calculate the vaporization duty of the fraction of the feed that

was evaporated, at the operating temperature. Other input for this step are TEV50 of

step #8, (as the normal boiling point of the hydrocarbon), the critical temperature of

step #10 and the operating temperature at the coil outlet.

13. Use the specific gravity of the whole feed and the KW of either step #1 or step #3, with

equations (3),(4) and (5) to calculate respectively the parameters A1, A2 and A3.

17
 14. Use the coil inlet temperature, the coil outlet temperature and the parameters A1, A2

and A3 from step #13 to calculate the heating duty of hydrocarbons along the coil,

with equation (7).

15. Use equation (18), the steam inlet temperature and the coil outlet temperature to calcu-

late the heating duty of the velocity steam, if any.

16. Use equation (18), the steam inlet temperature and the steam outlet temperature to cal-

culate the heating duty of the steam superheating convection coils, if any.

17. Use equation (18), the hydrogen inlet temperature and the coil outlet temperature to

calculate the heating duty for the heating of hydrogen, if applicable.

18. Calculate the total absorbed duty of the furnace from the results of steps #12, 14, 15,

16 and 17 and the respective flow rates.

4. Comparison with other methods

The proposed algorithm has been implemented in a computer program. The program has been

used to calculate the absorbed duty of existing refinery fired heaters. The absorbed duty of

those furnaces has been also calculated by the indirect method and by a commercial rigorous

simulation package, using the Soave – Redlich – Kwong equation of state for flash calcula-

tion. Table 3 compares the results of the calculations.

The results show that the proposed algorithm has results comparable to a commercial, rigor-

ous simulator. Moreover, both results are very close to those of the usual indirect method. Ta-

18
ble 4 provides the details of the absorbed duty calculation for all 3 cases used in Table 3, per

algorithm step. Each line represents the corresponding step’s result.

5. Conclusions and Recommendations.

An algorithm for the calculation of the absorbed, (process side) duty of fired heaters was pre-

sented. The required data are usual operating and laboratory data (specific gravity and distilla-

tion curve of the feed). The proposed method is easy to implement in a usual spreadsheet and

in the same time has accuracy comparable to process simulators and to the indirect method.

The results of the proposed method are almost identical to those of absorbed duty calculation

by process simulator, for the same set of analytical and operational data.

The correlations used in this work, may be substituted with other, more specific if it is

deemed necessary. API Databook, (Daubert and Danner, 1997) correlations for critical prop-

erties may be used instead of Kesler – Lee, (Kesler and Lee, 1976) equations, or NASA multi

parameter isobaric specific heat correlations instead of Shomate polynomials.

As is the general case of direct method, the absorbed duty calculation is sensitive to errors in

flow rate measurement. However, modern refineries use adequate instrumentation and usually

are able to provide accurate enough measurements. On the other hand, daily laboratory feed

analysis needed for process side characterization, are usually more accurate than field flue gas

O2 measurements. Moreover, simultaneous use of direct and indirect method may indicate

excessive tramp air ingress and thus aid to equipment efficiency improvement.

19
20
References

Arora V.K. Check Fired Heater Performance. Hydrocarbon Processing May 1985; 85-7.

Abbi Y.P., Shashank J. Handbook on Energy Audit and Environment Management. New
Delhi, India: The Energy and Resources Institute, TERI; 2009.

American Society of Mechanical Engineers (ASME) Performance Test Code for Steam Gen-
erating Units PTC 4.1”, 1974.

Chase Jr. M.W., NIST-JANAF Thermochemical Tables. 4th Edition J. Phys. Chem. Ref. Data,
Monograph 9, 1998; 1-1951

Daubert T.E., Danner R.P. API Technical Databook – Petroleum Refining. 6th ed. Washington
D.C.: American Petroleum Institute (API); 1997.

Devakottai B.S. Energy efficiency in furnaces and boilers. In: Rossiter A.P., Jones B.P. Ener-
gy Management and Efficiency for the Process Industries. Hoboken: Wiley; 2015. p. 107-27.

Edmister W.C., Pollock D.H. Phase Relations for Petroleum Fractions. Chem. Eng. Progr.
1948; vol. 44: 905-26.

Ganapathy V. Applied Heat Transfer. 1st ed. Tulsa, Oklahoma: Pennwell Publishing Co.;
1982.

Kesler M.G., Lee B. I. Improve prediction of enthalpy of fractions. Hydrocarbon Processing,

March 1976; 153 – 8.

Lieberman N.P., Lieberman E.T. A Working Guide to Process Equipment. 4th ed. New York:
McGraw-Hill; 2014.

Linstrom P.J., Mallard W.G., Eds. NIST Chemistry WebBook NIST Standard Reference Da-
tabase Number 69. National Institute of Standards and Technology, Gaithersburg MD, 20899,
http://webbook.nist.gov, (retrieved July 15, 2015).

Maxwell J.B., Bonnell L.S. Derivation and Precision of a New Vapor Pressure Correlation for
Petroleum Hydrocarbons. Industrial and Engineering Chemistry 1957; 49: 1187- 96.

Patel S. Simplify Your Thermal Efficiency Calculation. Hydrocarbon Processing, July 2005;
63-9.

Riazi M.R. Characterization and Properties of Petroleum Fractions, American Society for
Testing and Materials, 2005.

Riedel L. Kritischer Koeffizient, Dichte des gesättigten Dampfes und Verdampfungswärme.

Untersuchungen über eine Erweiterung des Theorems der übereinstimmenden Zustände. Teil
III. Chemie Ingenieur Technik 1954; 26: 679–83.

Trambouze P. Petroleum Refining, Materials and Equipment. 1st ed. Paris: Editions Technip;
2000.

21
Watson K.M., Nelson E.F. Improved Methods for Approximating Critical and Thermal Prop-
erties of Petroleum Fractions. Ind. Eng. Chem. 1933; 25 (8): 880–7.

Watson K.M. Thermodynamics of the Liquid States, Generalized Prediction of Properties.

Ind. Eng. Chem. 1943; 35: 398-406.

22
Figure Legends

Figure 1: Algorithm Flow Chart.

23
Table 1, Shomate polynomial constants for steam, (NIST, 2015), (Chase, 1998)

Temperature
(K) 500 - 1700 1700 - 6000
A 30.092 41.96426
B 6.832514 8.622053
C 6.793435 -1.49978
D -2.53448 0.098119
E 0.082139 -11.15764
F -250.881 -272.1797
G 223.3967 219.7809
H -241.8264 -241.8264

Table 2, Shomate polynomial constants for hydrogen, (NIST, 2015), (Chase, 1998)

Temperature
(K) 298 - 1000 1000 - 2500
A 33.066178 18.563083
B -11.363417 12.257357
C 11.432816 -2.859786
D -2.772874 0.268238
E -0.158558 1.97799
F -9.980797 -1.147438
G 172.707974 156.288133
H 0 0

24
Table 3, Absorbed duty as calculated by three different methods

Furnace Type Indirect Direct (pro- Direct (Rig-

(kW) posed algo- orous Simu-

rithm) (kW) lation) (kW)

Case 1, Vacuum distillation furnace, ca- 34,047 34,342 34,265

thedral type, 2 radiant sections, common

convection section, with velocity steam

and 2 convection coils for low and medi-

um pressure steam superheating.

Case 2, Diesel hydrotreater feed & recycle 10,813 10,940 10,633

gas, Vertical cylindrical type.

Case 3, Vacuum distillation furnace, ca- 42,439 42,644 41,061

thedral type, 2 radiant sections, common

convection section, with velocity steam

and 2 convection coils for low and medi-

um pressure steam superheating.

25
Table 4, Details of the absorbed duty calculation for all 3 cases used in Table 3.

Step # units Case 1 Case 2 Case 3

No, ASTM D- No, ASTM D-86 No, ASTM D-
1
1160 available Available 1160 available
2 ok ok ok
3 unitless 11.761 11.792 11.885
4 g/mol 562.62 223.08 644.05
5 kPa(a) 2.753 733.865 2.642
6 ok ok ok
7 %vol 56% vol 37% vol 45% vol
8 K 717.6 490.0 712.0
9 0.9260 0.8133 0.9139
Tc=914.2 K ; Tc=730.5 K ; Tc=910.0 K ;
10 K ; kPa
Pc=1,123.75 kPa Pc=2,171.39 kPa Pc=1,099.11 kPa
11 kJ/kg 185.35 233.26 156.25
12 kW 11321.3 1538.3 5942.8
A1=0.996839; A1=0.998533; A1=1.003666;
13 unitless A2=0.385728; A2=0.421425; A2=0.384465;
A3=0.0005215 A3=0.000557 A3=0.00052
14 kW 28,962.9 8,009.5 27,256.0
15 kW 323.9 Not applicable 198.7
16 kW 2,036.2 Not applicable 944.8
17 kW Not applicable 1,392.0 Not applicable
18 kW 42,644.2 10,939.8 34,342.3

26