Encyclopaedia Index

NAME

See also PHENC entry reserved names

------ PIL character*4 array; group 7 -----

NAME.... Command to give a name to a solved or stored variable, thus: NAME(22)=VORT which might be used if variable 22 were being used for the storage of the vorticity.

Names cannot be more than four characters long.

The default values of the variable names are:-
Name <Index No.> : P1<1>, P2<2>, U1<3>, U2<4>, V1<5>, V2<6>, W1<7>, W2<8>, R1<9>, R2<10>, RS<11>, KE<12>, EP<13>, H1<14>, H2<15>, C1<16>, C2<17>, C3<18>, C4<19>, C5<20> ...... C10<25>......C35<50>.

NAMFI

----- PIL character; default='CHAM '; -

NAMFI is the name of file from which initial fields are read at the beginning of re-start runs, when the FIINIT value of any stored or solved-for variable = READFI.

If NAMFI is not reset by the user, the file used will be PHI or PHIDA, according to the setting PHIDA=t or PHIDA=F in the P file (ie EARCON). When set by the user, the name is limited to four characters.

See also the PHENC entry RESTRT.

NAMGRD

---- PIL character; default='NONE' -

NAMGRD....is used to select by name one of the following 'special GROUND' subroutines, which are called from the subroutine grex3:

 name subroutine   location
 FURN FURNGR \phoenics\d_earth\d_spe\d_furnac\furngr.htm
 ESTR ESTRGR \phoenics\d_earth\d_spe\d_ester\estrgr.htm
 HTBX HTBXGR \phoenics\d_earth\d_spe\d_hotbox\htbxgr.htm
 TACT TACTGR \phoenics\d_earth\d_spe\d_tact\tactgr.htm
 CONV GXCONV \phoenics\d_earth\d_core\gxconvec.htm
 CVD  GXCVD \phoenics\d_earth\d_spe\d_cvd\gxcvd.htm
 MICA  MICA \phoenics\d_earth\d_spe\d_mica\micagr.htm
 F1  F1GRD \phoenics\d_earth\d_spe\d_f1\f1grd.htm
 NONE  

Naming variables

(see GROUP 7 and NAME)

NAMPAT

------- PIL character Array -------------

NAMPAT(IR) is the name of the PATCH number IR.

NAMSAT

---- PIL character*4; default='NONE'; -

NAMSAT....is used to select by name a user's own SATLIT subroutine.

NAMXYZ

---- PIL character; default='CHAM'; group -

NAMXYZ...name of file to which the BFC grid information is written. If NAMXYZ is not reset by the user, the name of the file is set (usually to XYZ(DA)) in the Earth and Satellite CONFIG files (EARCON and SATCON).
When set by the user, the name is limited to four characters.

In the case of multi-runs, the BFC grid information is usually appended to the grid file of the previous run, unless NAMXYZ for the current run is different from the previous run, in which case a new file will be opened, and further data will be written to the new file. Earth will expect to find files with the same names, containing the same number of data sets, available when it runs.

NCG1

------- PIL integer; Default=10 -----------

Satellite only, the dimension of CG array.

NCOLCO

---- PIL integer; default=45; group 23 -- -

NCOLCO....sets the number of columns ( ie. horizontal width ) used in CONTUR plots. See NROWCO.

NCOLPF

---- PIL integer; default=67; group 23 -- -

NCOLPF....sets the number of columns ( ie. the horizontal width ) used in PROFILE plots when IPROF=0.

NCRT

------ PIL integer; default=-1; group ----

NCRT....dictates the sweep frequency of the calculation of the cartesian resolutes of the velocities in body-fitted-coordinate calculations. The default of -1 flags EARTH to set NCRT equal to the print-sweep frequency NPRINT. These resolutes are always calculated on the last sweep, so that PHOTON has the the most up-to-date fields to plot. If the k-epsilon turbulence model is used, NCRT is reset to 1, as the cartesian components are used to calculate the generation term.

NDBCMN

---- PIL character*4; default='DBCM'; gro -

NDBCMN....is the name given to the common block in a debug SEARCH.

NDBF0

----- PIL character*4; default='DBF0'; gro -

NDBF0....is the name given to the common block element in a debug SEARCH.

NDIREC

-------------------------------------

Ground integer.

NDIREC....is a variable set by EARTH for use in Sections 8 through 15 of GROUND group 8. The notes provided in the listing of GREX3 should be consulted for further details.

NDIREC=1 signifies the y-direction, NDIREC=3 signifies the x-direction,NDIREC=5 signifies the high z-direction, and NDIREC=6 signifies the low z-direction.

NDIRECA

NDIRECA....is a variable set by earth for use in section 8 of ground group 8.NDIRECA is used to identify the direction as follows:

NDIRECA=1,3 and 5 signify the x-,y-, and z-directions respectively.

NDST

------- PIL integer; Default=0 ------------

Used when running test battery. Points to the name of the current DSTL file in EARCON.

NEIGHBOUR patches

Activation of the subroutine GXNEPA enables sources to be provided to a cell which are proportional to the excess of a neighbouring variable over the values of the variable in the cell in question. The proportionality constant is the 'CO' of the associated 'COVAL'.

These neighbours may be:

Neighbour patches are activated by PATCH and COVAL commands within the Q1 file, as is explained by notes at the top of, and comments within, the said subroutine.

The use of the neighbour-cell technique is illustrated in Q1 files which:

  1. use it to represent the flow of a second fluid along the tubes of a heat exchanger, as in cases: 118, 790, 791, 792, 793, 794, 795, 796, 797, 798 and z110.

  2. test the conservative low-dispersion algorithm , and specifically in library cases n101, n122 and n301.

Since the introduction of In-Form it has become possible to achieve the same effects as are provided by neighbour patches by use of the In-Form source statement in the Q1 file.

Core library case 710 illustrates this.

NEWENL

---- PIL logical; default=T; group 8 ----

NEWENL....this does for ENUL what NEWENT does for ENUT.

NEWENT

---- PIL logical; default=T; group 8 ----

NEWENT....signals, when T, that the GROUND-set kinematic turbulent viscosity is to be re-calculated at each slab directly after the solution of the variables other than velocity. In consequence, the latest viscosities are used in the solution of the velocities at the current slab.

If the GROUND-set turbulent viscosity is not a strong function of the solved variables, set NEWENT to F for economy.

NEWRH1

---- PIL logical; default=T; group 8 ----

NEWRH1....signals, when T, that the first-phase GROUND-set density is to be re-calculated at each z slab prior to the construction of the continuity errors (which form the residuals of the pressure-correction equation), at a point when all other dependent variables at the slab have been solved. In consequence, the densities used in the construction of the continuity errors will be in conformity with the latest values of variables of which they are functions.

If the density is not a strong function of the field variables, or if it depends only on the pressure, set NEWRH1 to F for economy.

NEWRH2

---- PIL logical; default=F; group 8 ----

NEWRH2....this does for RHO2 what NEWRH1 does for RHO1.

NewText

------------------------------------ Photon Help ----

[NewText] is for writing screen text. The input character string can be typed into the input space near the bottom of the window. Once the text has been typed in, PHOTON asks for the screen position to put in the text. Use the cursor to locate the bottom-left of the text.

Next

(Grid Edit Menu)--------------------------------------- Photon Help ----

[Next] loads the attributes of the next GRID element in the stack. After the last element, it will return to the top and load the first again.

See also: [Grid No.]

Next

(Vector Edit Menu)--------------------------------------- Photon Help ----

[Next] loads the attributes of the next VECTOR element in the stack. After the last element, it will return to the top and load the first again.

See also: [Vector No.]

Next

(Text Edit Menu) --------------------------------------- Photon Help ----

[Next] loads the attributes of the next TEXT element in the stack. After the last element, it will return to the top and load the first again.

See also: [Text No.]

Next

(Contour Edit Menu)--------------------------------------- Photon Help ----

[Next] loads the attributes of the next CONTOUR element in the stack. After the last element, it will return to the top and load the first again.

See also: [CONTOUR.]

Next

(Streamline Edit Menu)--------------------------------------- Photon Help ----

[Next] loads the attributes of the next STREAMLINES element in the stack. After the last element, it will return to the top and load the first again.

See also: [Stream No.]

Next

(Surfaces Edit Menu)--------------------------------------- Photon Help ----

[Next] loads the attributes of the next SURFACE element in the stack. After the last element, it will return to the top and load the first again.

See also: [Surface No.]

NFMAX

----- PIL integer; default=999999; group 2 -

NFMAX....is a device used to trip out-of-core working for testing purposes irrespective of the actual dimension of the F-array.

NFUSER

The user is allowed to reserve for himself a part of the F-array. In the satellite, he must set NFUSER to the number of locations which he needs. Then, in GROUND coding, he can refer to the I'th of his reserved locations as: F(L0USER + I). See also entries for F-array and for GXMAKE.

NIG1

------- PIL integer; Default=20 ----------- Read only.

Satellite only, the dimension of IG array.

NLG1

------- PIL integer; Default=20 ----------- Read only.

Satellite only, the dimension of LG array.

NMAT

NMAT is the number of materials for which properties are to be read from a file.

Note that NMAT is not a PIL variable. It should appear only in a line starting: MATFLG=T.

See PHENC entries: MATFLG and PROPS.

NNORSL

A linear-equation-solver enhancement for NONORT=T.

NNORSL is a PIL logical variable, defaulted = F, which should be set = T when significant computer time is spent in the linear equations solver for pressure and the grid is highly non- orthogonal.

The reason is that the linear-equation solver is iterative in character, and re-calculates residuals between iterations. Built into the solver is a residual-calculation formula which, though quick to execute, is exact only for orthogonal grids. The result is that, even for linear problems for which one pass through the solver, and only one sweep, should suffice , many sweeps re needed.

When NNORSL = T, a longer-to-execute but exact calculation of the residuals is employed, with the reult that one sweep will indeed suffice for linear problems. Large total-computer-time savings may ensue.

No.

---------------------------------------- Photon Help ----

The maximum number of steps for each of the streamlines to be tracked.

No.of lines:

---------------------------- No Photon Help ----

Number of streamlines in the current STREAMLINES element.

NOADAP

----------- PIL logical Switch ----------

NOADAP... when set true, disables the setting of the Ap coefficient to FIXVAL in blocked cells.

NOCLEAR

------------------------------------ Photon Help ----

NOC[lear].... is a REPLAY command which causes the display not to clear the screen before drawing a frame. NOCLEAR is used when it is necessary to construct composite pictures by over-plotting frames. The CLEAR command can be used to re-establish screen clearing between frames.

RESET will cause a CLEAR command to be issued.

See also : REPLAY, NOCLEAR, DRAW, REDRAW, RESET

NOCHKFLO

This is the name of a file, existence of which in the working directory will ensure that the satellite will not, at pressure-defining patches for which the user has created no COVALs for velocity, gratuitously introduce some.

Why should the satellite ever do such a thing? The practice arose as a precaution in the early days of PHOENICS; and it has been retained because some users may have come to rely upon it.

The objection to the practice is that the COVAL is of the form:
  COVAL(patch name, velocity name, ONLYMS, 0.0)

If the velocity in the cell should be finite, this introduction of zero-velocity mass may seriously distort the flow-field.

NOCOMM

---------- PIL logical, default F -------

NOCOMM=T, stops comments from being written to the COPYQ1 file. This is primarily of use for menu writers.

NOCOPY

---------- PIL logical, default F -------

NOCOPY=T stops any commands from being written to the COPYQ1 file.

Nodal points

In PHOENICS, the nodal points,i.e.the nodes at the centres of the control cells, are labelled IX, IY and IZ. These indices run from 1 to NX, 1 to NY and 1 to NZ respectively. The surfaces of the control cells are labelled I, J and K, which indices run from 1 to NX+1, 1 to NY+1 and 1 to NZ+1

Node At

------------------------------------ Photon Help ----

The position of the grid node in a cell. It can be either VERTEX or CENTRE.

NODE ID

A unique identifier (usually a CPU number or a hard-disc serial number) for a machine on which PHOENICS is to be installed. The NODE ID has to be supplied to CHAM for the issuing of an "unlocking string". The NODE ID is displayed on the screen when any PHOENICS module is run while PHOENICS is locked.

NODEF

----------- PIL logical, default F -------

When set to T, NODEF, will prevent the defaults from being re-set when the STACK is re-loaded. This allows a menu system to be exited and re-entered without losing previous data settings. Error messages relating to attempts to declare variables which already exist are suppressed while NODEF=T.

NOGRID

---- PIL logical; default=F; group 6 ----

NOGRID...when =T, has two effects namely:

  1. When BFC=T, it instructs the satellite NOT to write a BFC grid file for the run in question. This may save some time on a mature BFC project, when the grid is well-established and is not changing frequently.
    The user must then ensure that a grid file of the correct name is available during the EARTH run.
    See the entry on NAMXYZ for details of naming the grid file.

  2. When BFC=F, it instructs the Satellite to accept precisely the grid setttings implied by the GRDPWR, XFRAC etc statements in the Q1 file and NOT either to adjust the grid to fit VR-objects or to make any would-be-improving modifications whatsoever.
    Warning The NOGRID=T statement must however appear in the Q1 file below the GRDPWR and/or XFRAC etc statements. Otherwise it will nullify them.

Non-linear data transforms in AUTOPLOT

The commands which perform non-linear transforms are:

POWER X, POWER Y, RECIPROCAL X, RECIPROCAL Y

POWER X and POWER Y raise the x- or y-axis values of an element, or range of elements, to a specified power. Similarily the RECIPROCAL commands invert the x or y values. The operation is performed on the data stored in memory, and a SCALE command is required to display the transformed data.

Non-orthogonal BFC grids, setting of

NONORT

---- PIL logical; default=F; group 6 ----

NONORT....set NONORT to T when a body-fitted grid is significantly non-orthogonal. This action causes EARTH to calculate the cell-face mass fluxes taking into account the contributions from not-in-face velocities.

For highly skewed grids, these contributions can exceed the contribution of the in-cell-face velocity to the cell-face mass flux.

NONORT has no efffect when GCV = T; for the GCV solver always takes account of the not-in-cell velocities.

Norm

(View Menu)--------------------------------------- Photon Help ----

Turn the view and up direction to the nearest axis of the current coordinate system and fit the image in the current graphic window.

NORMAL /NORMAL X /NORMAL Y

---- Autoplot Help ----

NO[RMAL] [X or Y] {i j} Data elements i-j will have the x and/or y axes normalized between 0.0 - 1.0. SCALE will give a correctly scaled plot.

Normalisation of data in AUTOPLOT

It is possible to normalise data so that all values lie between 0.0 and 1.0 by means of the NORMAL X or Y commands. These commands operate on individual elements, or ranges of adjacent elements. SCALE will be required to produce the normalised plot.

NORTH

----- PIL real flag; value= 4.0; group 13 -

NORTH....is a PATCH type used for setting sources per unit north ( ie. larger-y ) area by way of COVAL in group 13.

North face porosity

(see NPOR)

NOWIPE

---- PIL logical; default=F; group 2 ----

NOWIPE....Set NOWIPE= T in run I in order to prevent the re-setting of data defaults at the start of run I+1.

This device ensures that run I+1 starts with the run-I data, which the input file of run I+1 can then modify.

However, its use is definitely not recommended, because it is easy to forget to set NOWIPE=F so as to switch off its effect when it is no longer needed.

Much safer is to precede the number of the to-be-loaded file by a dollar sign; for this applies only to that file.

NOx Production Model

The term NOx refers to nitrogen oxide (NO) and nitrogen dioxide (NO2), which are two major combustion-generated air pollutants. In combustion, NO is the dominant of the two components, NO2 mainly being derived from NO.

The 3 principal sources of NO emissions in combustion are:
1) thermal NO - the oxidation of molecular nitrogen;
2) prompt NO - the reaction of fuel-derived radicals with
nitrogen, ultimately leading to NO;
and  
3) fuel NO     - the oxidation of fuel-bound nitrogen.

The relative importance of these 3 sources to total NO emissions depends on operating conditions and on fuel composition.

Large thermal NOx production are associated with high tempurature and high atomic oxygen concentrations. Prompt NOx is significant where temperatures are low and residence times are short. Fuel NOx is a major source of emissions. NOx in fossil fuel combustion.

There is no switch-on NOx production model in PHOENICS, but rather there are library cases which exemplify the use of a very simple thermal NOx model coded in the GX file GXCHEM.FOR ( this file resides in the directory phoenics/d_earth/d_opt/d_chem ). Library cases C112 and C113 consider gaseous combustion, whereas cases C109 to C111 inclusive consider coal combustion.

For coal combustion, fuel NO is main contributor to the total NO. The NOx model coded in PHOENICS allows in some sense for fuel NO, but it underestimates the amount of fuel NO because no account is taken of the formation of HCN.

The NOx model embodied in PHOENICS is that described by Pericleous et al (1991). A two-stage approach is used to model the formation of NO. The first stage involves the main exothermic hydrocarbon reaction in air, and the second step involves the Zeldovich reaction scheme for the production of NO, i.e:

N2 + O = NO + N                                 (1)
O2 + N = NO + O                                 (2)
OH + N = NO + H                                 (3)

These reactions are decoupled from the main fuel-burning reaction of stage 1, and as the mass fractions of the pollutants are present in trace amounts, they do not affect the density and specific heat of the gas mixture. In addition, the species appearing in these reactions are assumed to form part of the product stream resulting from the combustion model of stage one.

The product species O2, O, OH, H, H2, CO & CO2 are calculated in such proportions as are appropriate to equilibrium stoichometric combustion at the prevailing pressure and enthalpy. This calculation is performed from a minimization of the Gibbs free energy via the CREK program of Pratt and Wormeck (1976). This program is embodied in the GX file GXCHEM.FOR.

The calculation then computes the mass fractions of NO & N via the Zeldovich reaction mechanism by solving species conservation equations for NO and N which have as convection and diffusion fluxes the ones determined from the first-stage combustion calculation. It is assumed that the mass fractions of the other species remain invariant, with the justification that they are produced by reactions which are considerably faster than those involving nitrogen. The temperature and species mass fractions used in the Zeldovich reaction source terms do not account of the influence of turbulent fluctuations.

  1. K.A.Pericleous, I.W.Clark and N.C.Markatos, 'Mathematical modelling of Thermal NOx emissions in Combustion Chambers', In 'Heat Transfer in Radiating and Combusting Systems', Ed. M.G.Carvalho, F.Lockwood, J.Taine, Proc. EUROTHERM Seminar No.17, p393, Cascais, Portugal, Springer Verlag, (1991).
  2. 'Fossil Fuel Combustion: A Source Book', Ed. W.Bartok and A.F. Sarofim, John Wiley, p228, (1991). 
  3. J.J.Wormeck and D.T.Pratt, 'CREK A computer program for the calculation of combustion reaction equilibrium and kinetics in laminar or turbulent flow', In 'Studies in Convection', Vol. II, Ed. B.E.Launder, Academic Press, (1975).
  4. Y.B.Zeldovich and Y.P.Raizer, In 'Physics of Shock Waves and High Temperature Phenomenon', Vol. I, Ed. W.D.Hayes and R.F. Probstein, Academic Press, (1966).

NPHI

-----------------------------------------Integer;default=150.

NPHI....is the number of whole-field variables (dependent or auxiliary). It is defaulted to 150, which corresponds to the default dimension of arrays in SATELLITE and EARTH

It may be set to smaller values with advantage (but never to less than 13) if only a few variables require 3D storage; for then "search" loops in EARTH of the kind DO MPHI=1,NPHI are more quickly executed.

NPLT

------ PIL integer; default=-1; group 23 -- -

NPLT....is the frequency, in number of sweeps ( or slab-iterations for PARAB), of storage and tabulation and/or plots of spot- values and/or residuals. The default of -1 ensures that if the user has not set NPLT, EARTH will set it as follows: NPLT = MAX0( 1, (IPLTL-IPLTF+1)/20 ). This ensures that 20 values are printed.

NPNAM1

----- PIL integer; Default=200 ----------

Read only.

Satellite only, the dimension of NAMPAT array.

NPOR

------ PIL integer name; group 7 ----------

NPOR.... indicates which whole-field store will be used for the porosity of the north face in response to the command STORE(NPOR). Once stored, this field should be initialized in group 11. The field values are set to 1.0 by default.

NPRINT

---- PIL integer; default=10000; group 23 -

NPRINT....governs the sweep interval of the field print-out. This print-out occurs only for sweep numbers that are whole-number multiples of NPRINT. Print-out is provided on the last sweep irrespective of NPRINT.

In parabolic calculations, NPRINT governs the slab-wise iteration ( see LITHYD) interval of field print-out. This print-out occurs only for iteration numbers that are whole-number multiples of NPRINT.

NPRMNT

---- PIL integer; default=10000; group 22 -

NPRMNT....time-step interval for print-out of correction- equation spot values and slab-wise residuals at IZMON, the monitor slab. Arguments 3 and 4 of OUTPUT determine which variables are to have these quantities printed.

NPRMON

---- PIL integer; default=10000; group 22 -

NPRMON.... sweep interval for print-out of correction- equation spot values and slab-wise residuals at the slab identified by IZMON. Arguments 3 and 4 of OUTPUT determine which variables are to have these quantities printed.

In parabolic calculations, NPRMON controls the iteration interval of this print-out at the slab IZMON.

NREGT

------ PIL integer; default 1; GROUP 2----

NREGT....is the number of regions in the time direction. By default, every region contains 1 time step and has an extent of 1 sec.

NREGX

------ PIL integer; default 1; group3 ----

NREGX....is the number of regions in the X direction. By default, every region contains 1 cell and has an extent of 1 metre (or 1 radian if CARTES=F).

NREGY

------ PIL integer; default 1; group 4 --- -

NREGY....is the number of regions in the Y direction. By default, every region contains 1 cell and has an extent of 1 metre.

NREGZ

------ PIL integer; default 1; group 5 --- -

NREGZ....is the number of regions in the Z direction. By default, every region contains 1 cell and has an extent of 1 metre.

NRG1

------- PIL integer; Default=100 ----------

Read only.

Satellite only, the dimension of RG array.

NROWCO

---- PIL integer; default=20; group 23 -- -

NROWCO....sets the number of rows ( ie. vertical width ) used in CONTUR plots. See NCOLCO.

NSAVE

----- PIL character; default= 'CHAM'; -

NSAVE....is the four-character name of saved-for-restarts file. This variable is most often used on restart runs in which SAVE is T, for then it can be used to differentiate the file from which the restart is to be made (named by NAMFI= 'hers', say), from the file SAVEd. When this is not done, the file SAVEd will overwrite the re-start file.

If NSAVE is not reset by the user, the name of the file is set (usually to PHI(DA)) in the EARTH portion of CONFIG file (ie EARCON)).

When set by the user, the name is limited to four characters.

The following statements, when insert in a Q1 file:

RESTRT(ALL)
NAMFI=hers
SAVE=T
NSAVE=mine

will restart from the file named 'hers' and save the resulting fields in a file named 'mine' and therefore not 'phi', which is the default.

NTPRIN

---- PIL integer; default=10000; group 23 -

NTPRIN....governs the time-step interval of field print-out. It occurs only for time steps for which ISTEP is a whole-number multiple of NTPRIN. Print-out is provided on the last step irrespective of NTPRIN.

NTZPRF

---- PIL integer; default=1; group 23 --- -

NTZPRF....Time-wise (or z-wise if PARAB) plotting frequency of PROFIL plots.

When STEADY=F, the commands ( for example ), PATCH(name,PROFIL,1,1,1,1,1,1,1,LSTEP) and PLOT(name,P1,0.0,0.0) produce a profile plot of P1, the ordinates of which are at time steps that are an integral multiple of NTZPRF.

When PARAB=T, the commands ( for example ), PATCH(name,PROFIL,NX,NX,NY,NY,1,NZ,1,1) and PLOT(name,W1,0.0,0.0) produces a profile plot of P1, the ordinates of which are at IZ steps that are an integral multiple of NTZPRF.

NULL, a PHOENICS-VR object type

A NULL object which is used to create regions in the grid. It has no other attributes. See the description in the PHOENICS_VR Reference Guide, TR326

NULLPR

---- PIL logical; default=F; group 20 --- -

NULLPR....is a PIL variable which, when set TRUE, inhibits the output that EARTH otherwise writes to the RESULT file.

See DISTIL for further information.

Number

(Contour Menu)------------------------------------- Photon Help ----

Cells are filled with a specified number of polygons (the default and also the minimium is 4).

Number of iterations at a slab

(see LITC and LITHYD)

NUMCLS

---- PIL integer; default=5; group 23 --- -

NUMCLS....specifies the number of columns printed across the page or on the VDU screen in the tabulation of field print-out.

As each column occupies 12 characters, the default setting of 5 arranges the occupation of the complete width of an alpha-numeric VDU screen display. This setting is also suitable for hardcopy print-out for use in A4 reports. For a 132-character-field printer, NUMCLS=10 occupies the complete width of paper.

NUMCOL

------ PIL integer; default 0 -----------

NUMCOL....is the number of colours available on the device. After the graphic mode is first entered, NUMCOL will contain the number of colours provided from the the graphic driver. For a monocrome screen, the number is 1; for a colour screen, NUMCOL will normally return a value of 16.

When resetting defaults, NUMCOL is also set to 0.

NWALL

----- PIL real flag; value= 19.0; group 13 -

NWALL....is a PATCH type used in group 13 in conjunction with COVAL for representing the sources resulting from a wall at the north faces of the cells identified by PATCH.

See WALL and WALLS for further information

NX

-------- PIL integer; default=1; group 3 ----

NX....is the number of sub-divisions of the integration domain in the x-direction. In cylindrical-polar coordinates (ie when CARTES=F) the x-direction is the circumferential coordinate, and is measured in radians. The default value of 1 is the smallest NX allowed. There are many practical calculations for which NX=1 is the appropriate setting, for example axisymmetrical flows. For these, CARTES would equal F, and NY and NZ would have such values in excess of 1 as are deemed necessary.

There is a built-in upper limit for the total number of cells and timesteps that can be set in the Q1 file,i.e.the maximum size of the TFRAC, XFRAC, YFRAC and ZFRAC arrays is constrained by the requirement that NX + NY + NZ + LSTEP must be less than or equal to 3000. To exceed this number, increase the parameter MAXFRC in the MAIN programme of the Satellite.

The x-direction is recommended for use when one-dimensional calculations are to be used. Then NX might equal 100, with NY and NZ both equal to 1.

NXPRIN

---- PIL integer; default=1; group 23 --- -

NXPRIN....controls the IX-interval in the tabulations of the field variables that have been selected for print-out by means of the command OUTPUT.

NY

-------- PIL integer; default=1; group 4 ----

NY....number of spatial subdivisions in the y-direction. The default of 1 is appropriate to the one (x)-dimensional class of flows mentioned under the NX entry; or, of course, to a two- dimensional model in which the domain extends in the x- and z-directions only. Two-dimensional flows in turbomachinery cascades may be analysed in this way.

The y-direction can be employed for one-dimensional analyses, it may be mentioned, precisely in the same way as the x-direction. However, the y-direction lacks the cyclic-boundary-condition feature ( see XCYCLE ), which is the reason for recommending the use of x for one-dimensional flows as a general rule.

In cylindrical-polar coordinates (ie when CARTES=F) the y-direction is the radial direction.

NYPRIN

---- PIL integer; default=1; group 23 --- -

NYPRIN....controls the IY-interval in the tabulations of the field variables that have been selected for print-out by means of the command OUTPUT.

NZ

-------- PIL integer; default=1; group 5 ----

NZ....number of spatial subdivisions in the z-direction.

PHOENICS treats the z-direction differently from the x- and y-direction in several respects:

  1. if the flow is parabolic ( PARAB=T ), it is the z-direction that is chosen as the 'predominant direction of flow';
  2. if CARTES=F, z is the axial direction of the cylindrical-polar coordinate system; and
  3. the storage arrangements in PHOENICS are such that, if the core memory of the computer does not suffice for all variables to be accomodated in core, whatever secondary-storage device is available will automatically store the information pertaining to larger and smaller values of z than are being dealt with at any particular stage of the computation.

In computations of parabolic flows, NZ represents the number of z-direction steps to which the integration will proceed. This can be as large as desired, without in any way influencing the storage requirement; for EARTH operates, for parabolic flows, always with a single 'slab' of cells.

For non-parabolic flows, the fact that out-of-core information can be stored and recovered when the integration domain extends in the z-direction is a reason for qualifying the recommendation ( made under the NX entry ) that the x-direction should always be used for one-dimensional calculations. If the number of cells should be very large, the use of the z-direction may be preferable; for then the secondary-storage feature may be employed as needed.

NZPRIN

---- PIL integer; default=1; group 23 --- -

NZPRIN....controls the IZ-interval in the tabulations of the field variables that have been selected for print-out by means of the command OUTPUT.

NZSTEP

-------------------------------------

Ground integer.

NZSTEP....is the number of z-slabs into which the domain is sub-divided.