Component 25: Air Preheater

In This Topic

Component 25: Air preheater

Specifications

Line connections
1	Primary inlet (cold flow, inside tubes)
2	Primary outlet (cold flow, inside tubes)
3	Secondary inlet (hot flow, outside tubes)
4	Secondary outlet (hot flow, outside tubes)
5	Control inlet for KAN - Design Heat Transfer Capability (as H)

General User Input Values Characteristic Lines Physics Used Displays Example

General

This component primarily helps in modeling the heat exchangers between the gaseous mediums. However, it can also be used for liquids (water, oil) and solids (coal). When using this component for water, it must be noted that in this case no phase transition is observed.

In this component, only the heat exchange is considered, and not the material exchange. To model the slippage normal in the air preheater, a mass flow branching e.g. of 5% is to be integrated before the air preheater.

Radiation losses can be modelled with the help of a referred loss factor.

In the design case, either the upper or the lower terminal temperature difference (see "Heat Exchanger General" )

can be specified. From this, EBSILON®Professional calculates the nominal value for k*A, KAN. However, this is possible only in the counter flow operation. In case of direct-current operation, KAN must be specified and the component must be used in the off-design calculation mode.

Logic inlet (Connection point 5) for controlling component properties

(see also : Editing components --> Ports)

To make component properties like efficiencies or heat transfer coefficients (variation quantity) accessible from the outside (for control or reconciliation) it is possible to place the respective value on an auxiliary line as an indexed measured value (specification value FIND). In the component, the same index must then be entered as specification value IPS.

It is also possible to place this value on a logic line that is directly connected to the component (please see FVALKA=2,
Variation variable: KAN, Dimension: Enthalpy).

The advantage is that the allocation is graphically visible, and errors (e.g. when copying) are thus avoided.

MODEL:

For more information on general notes applicable to most common heat exchangers, (see "Heat Exchanger General" )

The (k*A)-value of the off-design calculation results from the (k*A)-value of the design calculation multiplied by a correction factor that is determined by one or more characteristic lines. The corrections of the characteristic lines indicate the dependency of the k-number of the primary and secondary mass flow.
The characteristics be corrected or replaced by an adaption polynomial or a Kernel expression. The control is effected via the Flag FADAPT.

The easiest method of deactivating a heat exchanger without removing it from the connection is to set FFU = off. Pressure losses are, however, accounted for.

If the terminal temperature differences are specified, it is important that depending upon the heat value (M*cp), either the lower or the upper terminal temperature difference leads to correct results.

When using complex cycle arrangements for heat exchangers, it is often not easy to determine the settings of all terminal temperature differences in a useful manner. Error messages occur frequently because an inoperable heat exchanger may also cause some errors in other heat exchangers. To avoid this (k*A) can be entered instead of terminal temperature differences. This can be done in the design mode by using the mode "local off-design". An input of (k*A) always leads to physically possible results.

There are two identification modes for this component: T2-default (FIDENT = 2) and T4-default (FIDENT= 4). Based on this default, k*A is calculated in all the load cases. In the design case, the terminal temperature difference is not used, in the off-design mode, KAN and the characteristic lines are not used.

For this heat exchanger it is possible, to specify the pressure externally (P2, P4).
It is also possible to use the flag FVOL to determine whether the off-design calculation of the pressure drop shall consider only the mass-flow (approximation for incompressible fluids) or mass
and volume flow.

Pinch point violations in the case of heat exchangers: (see "Heat Exchanger General" )

Up to release 10.0, a pinchpoint violation was only determined subsequently in partial load, i.e. KA was calculated for the respective load case and from this the transferred heat quantity and then it was checked whether this heat quantity can be transferred at all at the correct temperature level. Since in the case of evaporation or condensation the temperature remains constant despite heat input or heat removal, there are cases where heat transfer is not physically possible despite the overall balance being correct. In this case, an error message was issued in Ebsilon.

The calculation has now been changed in such a way that the transferred heat quantity is reduced as far as it is still physically possible, with the minimum pinch point
can be set in a default value PINPMIN. This results in a correspondingly reduced KA.
The user is informed of this by a warning message ("KA reduced to avoid pinchpoint violation") and can then adjust the part-load characteristic curve or the part-load exponent for KA accordingly so that the warning no longer occurs. The advantage, however, is that one gets a physically possible result in any case.

Furthermore, at the end of the calculation there is a check if there is a pinch point violation due to curved course of Q(T) (caused by significant changes of cp depending on the temperature). This can be verified by dividing the heat exchanger into individual sections.
This case can occur, for example, when on the hot side the cp at the inlet is significantly smaller than at the outlet (for example, steam that has a cp of about 2 kJ/kgK at high superheat, but more than 5 just above the boiling line). This means that this steam provides more heat at a lower temperature level than at a high one. At appropriately low degrees, this can be a limitation on the amount of heat transfer that is possible.
The QT diagrams take into account the non-linearity (curvature of the curves) in areas without phase change.

The flag FSPEC (deprecated) has been divided into two flags:

FSPECD indicates in which way the component is to be designed
FIDENT serves to activate a component identification mode, i.e. the respective condition of the component is determined on the basis of temperature specifications

Note:

When loading a model that was created with Release 11 (or older), the corresponding values for FSPECD, and FIDENT are determined from the value of the flag FSPEC, and
FSPEC is set to “void” (-999). The model then calculates the same result values. If required, however, the flag FSPEC can still be used as well. This is necessary so that the existing EbsScripts
in which a switchover of FSPEC into an identification mode is carried out will continue working. If FSPEC is not “empty” (-999) but has a value of -4 or -5 (the old values for the
identification modes), the new flag FIDENT will be ignored, and the component will behave according to the setting of FSPEC (this is indicated in a comment).

To remove ambiguity, the terms “primary side” and “secondary side” respectively have been replaced by “cold side” and “warm side” in the input screens. The cold side (previously “primary”) is the flow from Pin 1 to Pin 2 that is heated. The warm side (previously “secondary”) is the flow from Pin 3 to Pin 4 that gives off the heat.

Design in the Case of Concurrent Flow, (see "Heat Exchanger General" )

In the heat exchanger (Components 25) it is possible to carry out a design via the upper and lower terminal temperature difference also in the case of concurrent flow (FFLOW=1).
If both inlet temperatures are specified, the upper terminal temperature difference can only be determined iteratively. Usually this is no problem. If convergence problems occur in more
complex models, another design mode will have to be used.

Effectiveness Method

See Heat Exchangers General Information Effectiveness Method

Flag FDQLR

In Release 13, you can use the FDQLR flag to define how DQLR (factor for modeling heat losses) should be interpreted.

Note about the result values:

Specific heat capacity : CP12 / CP34

The mean specific heat capacity is displayed as result value on the cold side (CP12) and on the hot side (CP34).

The mean specific heat capacity results from the quotient of the enthalpy difference and the temperature difference.

If no temperature difference is present (e.g. in the two-phase range or when the heat exchanger is shut off), however, it is not possible to calculate this quotient. In this case, the specific heat capacity at the respective temperature is used, provided that it is defined. Otherwise the result value will remain blank.

Performance factor RPFHX

The quotient from the current value for k*A (result value KA) and the k*A expected in the respective load point due to the component physics and characteristic lines respectively (result value KACL) serves to assess the condition of a heat exchanger.

The quotient KA / KACL is displayed as result value RPFHX.

Similar components:

For more information on how this heat exchangers compares to other heat exchangers, see Heat Exchanger General Components

User Input Values

FSPECD	Design specification method Like in Parent Profile (Sub Profile option only) Expression = 0: Effectiveness given as EFF = 1: Lower terminal temperature difference given asDTN = 2: Upper terminal temperature difference given asDTN = 4: Both hot stream temperatures and one cold stream temperature given = 5: Both cold stream temperatures and one hot stream temperature given
FIDENT	Component identification (only in off-design Like in Parent Profile (Sub Profile option only) Expression = 0: No identification = 2: Outlet temperature of cold stream (T2) given externally in off-design, KA calculated = 4: Outlet temperature of hot stream (T4) given externally in off-design, KA calculated In the design case, variant FIDENT=2 is identical with FSPECD=5 and FIDENT=4 with FSPECD=4. In order to prevent contradictory specifications, the flag FIDENT is therefore only used in off-design for these components. Please note: The peculiarity for these components is that in the design case identification (namely the calculation of KAN from temperature specifications) takes place by default. This is not the case with other components.
DTN	Terminal temperature difference (nominal, see FSPECD) Depending upon the value of FSPEC either the upper (=T3-T2) or the lower (=T4-T1) terminal temperature difference is to be entered. DTN in concurrent flow (FLOW) see under "General"
EFF	Effectiveness
FDP12N	Pressure drop handling stream 12 Like in Parent Profile (Sub Profile option only) Expression = 1: Calculated by DP12N = -1: P2 given from outside
DP12N	Cold side pressure drop (nominal)
FDP34N	Pressure drop handling stream 34 Like in Parent Profile (Sub Profile option only) Expression = 1: Calculated by DP34N = -1: P4 given from outside
DP34N	Hot side pressure drop (nominal)
FVOL	Flag for cartload pressure drop Like in Parent Profile (Sub Profile option only) Expression =0: Only depending on mass flow =1: Depending on mass- and volume flow
FDQLR	Heat loss handling Like in Parent Profile (Sub Profile option only) Expression =0: Constant (DQLRQN in all load cases) DQLR refers to the design value QN (which equals the heat quantity given off by the hot flow in the design case) in all load cases, i.e. it has a constant value in all load cases. If, however, this value exceeds 10 percent of the heat quantity given off by the hot flow, the heat loss will be limited to this value, and a warning will be output. =1: Relative to actual heat input (DQLRQ354) DQLR refers to the heat quantity given off by the hot flow. If the corresponding warning is ignored, losses of more than 10 percent can be modelled here too.
DQLR	Heat loss through radiation, relative (with reference to QN)
TOL	Maximum permissible tolerance in energy balance for the internal iteration
FMODE	Flag for calculation mode Design/Off-design Like in Parent Profile (Sub Profile option only) Expression =0: GLOBAL =1: local off-design (i.e. always off-design mode, even when a design calculation has been done globally) =2: special local off-design (special case for compatibility with the earlier Ebsilon-versions, should not be used in newer models, because the results of the real off-design calculations are not consistent) = -1: local design
FFLOW	Direction of flow (see "Heat Exchanger General" ) Like in Parent Profile (Sub Profile option only) Expression =0: Counter Current flow =1: Concurrent flow
FADAPT	Flag for adaptation polynomial ADAPT/ adaptation function EADAPT Like in Parent Profile (Sub Profile option only) Expression =0: Off =1: Correction for kA [KA = KAN Char Line factor * polynomial] =2: Calculation of kA [KA = KAN polynomial] =1000: Not used but ADAPT evaluated as RADAPT (Reduction of the computing time) = -1: Correction for kA [KA = KAN Char Line factor * adaptation function] = -2: Calculation of kA [KA = KAN adaptation function] = -1000: Not used but EADAPT evaluated as RADAPT (Reduction of the computing time)
EADAPT	Adaptation function for KA
FFU	Flag for activating the component Like in Parent Profile (Sub Profile option only) Expression =0: heat exchanger deactivated =1: heat exchanger activated
FVALKA	Validation of k*A (only in off-design) Like in Parent Profile (Sub Profile option only) Expression =0: KAN used without validation =1: Pseudo measurement point identified by IPS used (can be validated) =2: KAN given by enthalpy on control inlet 5
IPS	Index for pseudo measurement point
PINPMIN	Minimum value for the pinch point (KA is reduced automatically if the pinch point would fall below this value)
FSPEC (deprecated)	Deprecated specification combi switch Like in Parent Profile (Sub Profile option only) Expression = -999 : Unused (FSPECD und FIDENTused instead) Old values: =1: In the design case, DTN is interpreted as the lower terminal temperature difference (DT41N), in off-design mode, calculation is done with characteristic lines =2: In the design case, DTN is interpreted as upper terminal temperature difference (DT32N), in off-design mode, calculation is done with characteristic lines =4: In the design case T3 and T4 as well as one of the two temperatures T1 or T2 are specified from outside, in off-design mode, calculation is done with characteristic lines =5: In the design case T1 and T2 as well as one of the two temperatures T3 or T4 are specified from outside, in off-design mode, calculation is done with characteristic lines =-5: T2 is specified from outside (in all load cases), DTN or the characteristic lines are not used =-4: T4 is specified from outside (in all load cases), DTN or the characteristic lines are not used
KAN	Heat transfer coefficient * area (nominal) - Design Heat Transfer Capability
M1N	Cold side mass flow (nominal)
M3N	Hot side mass flow (nominal)
QN	Heat exchanger capacity (nominal) (Q34N)
V1N	Specific vol. at cold side inlet (nominal) or Design Heat Transfer Capability resp.
V3N	Specific vol. at hot side inlet (nominal)

The parameters marked in blue are reference quantities for the off-design mode. The actual off-design values refer to these quantities in the equations used.

Generally, all inputs that are visible are required. But, often default values are provided.

For more information on colour of the input fields and their descriptions see Edit Component\Specification values

For more information on design vs. off-design and nominal values see General\Accept Nominal values

Characteristic Lines

1^st Characteristic line CKAM1 FK1 = f (M1/M1N)

2^nd Characteristic line CKAM3 FK2 = f (M3/M3N)

(k*A) / (k*A)N = FK1 * FK2

Characteristic line 1: (k*A)-Characteristic line CKAM1: (k*A)1/(k*A)N = f (M1/M1N)

     XX-axis        1          M1/M1N                     1^st point
                        2          M1/M1N                     2^nd point
                        .
                      N         M1/M1N                     last point

     Y-axis          1          (k*A)1/(k*A)N             1^st point
                        2          (k*A)1/(k*A)N             2^nd point
                        .
                        N         (k*A)1/(k*A)N             last point

Characteristic line 2: (k*A)-Characteristic line CKAM3 : (k*A)2/(k*A)N = f (M3/M3N)

     X-axis       1          M3/M3N                    1^st point
                        2          M3/M3N                    2^nd point
                        .
                        N         M3/M3N                    last point

     Y-axis          1          (k*A)2/(k*A)N             1^st point
                        2          (k*A)2/(k*A)N            2^nd point
                        .
                        N         (k*A)2/(k*A)N             last point

Physics Used

Equations

Design case (Simulation flag: GLOBAL = design case and FMODE = design case)

if the lower terminal temperature difference is given with FSPEC, then {

P4 = P3 - DP34N

T4 = T1 + DTN

H4 = f(P4,T4)

M4 = M3

Q4 = M4 * H4

DQ = (Q3 - Q4)*(1-DQLR)

P2 = P1 - DP12N

Q2 = Q1 + DQ

M2 = M1

H2 = Q2/M2

T2 = f(P2,H2)

DTLO = T4 - T1 (for FFLOW = counter current)

DTUP = T3 - T2 (for FFLOW = counter current)

LMTD = (DTUP - DTLO)/(ln(DTUP) - ln(DTLO))

KAN = DQ/LMTD

KAN*LMTD = M2*H2 - M1*H1

KAN*LMTD = (M3*H3 - M4*H4)*(1 - DQLR)

}

if the upper terminal temperature difference is given with FSPEC; then {

P2 = P1 - DP12N

T2 = T3 - DTN

M2 = M1

H2 = f(P2,T2)

Q2 = M2 * H2

DQ = Q2 - Q1

P4 = P3 - DP34N

Q4 = Q3 - DQ/(1 - DQLR)

M4 = M3

H4 = Q4/M4

T4 = f(H4,P4)

DTLO = T4 - T1 (for FFLOW = counter current)

DTUP = T3 - T2 (for FFLOW = counter current)

LMTD = (DTUP - DTLO)/(ln(DTUP) - ln(DTLO))

KAN = DQ/LMTD

KAN*LMTD = M2*H2 - M1*H1

KAN*LMTD = (M3*H3 - M4*H4)*(1 - DQLR)

}

Off-design case (Simulation flag: GLOBAL = Off-design case or FMODE = off-design case)

F1 = (M1/M1N) ** 2

if GLOBAL = design, then F1=1.0

F3 = (M3/M3N) ** 2

if GLOBAL = design, then F3=1.0

P2 = P1 - DP12N * F1

M2 = M1

FK1 = f(M1/M1N) from characteristic line 1

if GLOBAL = design, then FK1=1.0

FK2 = f(M3/M3N) from characteristic line 2

if GLOBAL = design, then FK2=1.0

KA = KAN * FK1 * FK2

P4 = P3 - DP34N * F3

M4 = M3 + M5

Maximum/Minimum values for the iteration {

H2max = f(P2,T3)

Q12max = M1 * (H2max - H1)

H4min = f(P4,T1)

Q34max = Q3 - M4 * H4min

}

For FFLOW = counter current {

Qmax = min(Q12max,Q34max)

}

For FFLOW = concurrent{

Estimation for iteration 1

QA = min(Q12max,Q34max)

QM = QA*QA/(Q12max+Q34max)

Iteration1{

H2 = H1 + QM*(1-DQLR)/ M2

T2 = f(P2,H2)

T4 = T2

H4 = f(P4,T4)

QN = Q3 -M4 * H4

DQQ_1 = DQQ

DQQ = QM - QN

regula - falsi method {

Gradient = (QM - QM_1)/(DQQ - DQQ_1)

at iteration step 1: gradient of the last global step

QMN = QM - DQQ * Gradient

QM_1 = QM

QM = QMN

}

DQ = | DQQ/((QM+QN)*.5) |

if DQ < TOL, then end the iteration 1

else continue the iteration

}

Qmax = QM

}

Q12 = 0.5*Qmax

Iteration2{

H4 = (Q3 - Q12/(1-DQLR) )/M4

T4 = f(P4,H4)

H2 = H1 + Q12/M2

T2 = f(P2,H2)

DTLO = T4 - T1 (for FFLOW = counter current)

DTUP = T3 - T2 (for FFLOW = counter current)

DTLO = T4 ; T2 (for FFLOW = concurrent)

DTUP = T3 ; T1 (for FFLOW = concurrent)

LMTD = (DTUP - DTLO)/(ln(DTUP) - ln(DTLO))

QQ = KA * LMTD

DQQ_1 = DQQ

DQQ = Q12 - QQ

regula - falsi method {

gradient = (Q12 - Q12_1)/(DQQ - DQQ_1)

at iteration step 1: Gradient of the last global step

Q12X = Q12 - DQQ * Gradient

Q12_1 = Q12

Q12 = Q12X

}

DQ = |DQQ /((Q12+QQ)*.5)|

if DQ < TOL then end iteration step 2

else continue the iteration

}

QN = Q3 -M4 * H4

KA*LMTD = M2*H2 - M1*H1

KA*LMTD = M3*H3 - M4*H4 - QN*DQLR

Component Displays

	Display Option 1		Display Option 2
	Display Option 3		Display Option 4
	Display Option 5		Display Option 6
	Display Option 7

Example

Click here >> Component 25 Demo << to load an example.