The version 210 of MFIRE is upgraded from version v2.01/2.0. It is applied 
for the simulation of dry and wet conditions. It was compiled with Lahey 
Fortran (V4.1L) for the Intel 80386 based computers. Users who wish to modify 
the source code can apply other fortran compilers too.

    The modification in MFIRE v 2.10

       1.  Add all the modified parts in version 1.30 to version 2.10 except
the third modification in version 1.30 because the error in handling fan char- 
 cteristic input data in the user specified time table has not existed in version 2.01/2.0.

      2.   The error in handling the input data of the used specified time table

       In the version 2.10/2.0, the error mensioned above is eliminated
 because it is not necessary to make the parameter transfer for every related
 arrays. It induces, however, a new error. When a fan stops,  in the subroutine 
CDCH of version v2.01/2.0, only the type of airway NWTYP, in which the 
stopped fan is installed, is send to zero. The other parameters of that fan
in arrays NOF, RGRAD, NFREG, QF, PF, and variable NFNUM, the 
number of operating fans, are unchanged. The simplification  causes a new
 conflict between the unchanged variable NFNUM, array NOF and the 
changed array NWTYP. The conflict will cause problem. The airways arranged
by following their types form array INU in subroutine ARR. The wrong type of 
airway will give a wrong array INU and variable NBU and then cause the 
problems at later calculation and final result output. The modified part 
offers the combination of unchanged array NOF and changed array NWTYP
 to avoid above mistake.

      3.  The error in input data check 

 a. In input data set, if an airway is a decline airway and the input length of 
the airway is smaller than the elevation difference of two end junctions of the
 airway. The correction of input data in subroutine CCDATA will wrongly 
assign a negative  elevation difference to the length of that airway and cause
a computing problem. In the new version 2.10, the absolute value of elevation
difference of two end junctions of the airway is applied for the correction of
 airway length.

  b. In the subroutine CDJN of former version of MFIRE, the cross sectional 
areas of the surface airways are sent to the values larger than 1000 ft2 so that the evaluations of thermal state variation of those airways are ignored, In 
some mines, however, it is possible the cross sectional areas of some under
ground airways are larger than 1000 ft2, such as in the example data set cote93.mfi. These airways are wrongly treated as surface airways and ignored 
the calculation of thermal state variation. In the new version of v2.10, the 
criterion of cross selection area of a surface airway is increased to 105 ft.
 
  c. If the reference temperature, TR, is not given a value in input 
data file,  the input data correction in subroutine INPUT  of former version 
v2.01/2.0 will fails to assign a default value to the TR. The reference
 temperature TR with zero value will cause large error.  In the new version,
 the correct  statement for the TR with zero value is added. For the same 
reason, the input data correction of junction temperature T(K), and relative
 humidity HJN(K) are added the zero value correction statements.

      4.  The error in fan curve fitting 

     In the former version 1.29, the number 100000 is in the statement of fan curve derivative of subroutine ITR:

                  RQSUN=RQSUM-(RGRAD(L)*100000)

In version 2.01/2.0, the number 100000 is missed in above statement so that it is added in new version v2.10.

       5.   The maximum outer iteration number ITN 

       In version 2.01/2.0, the variable ITN, maximum outer iteration number for
 natural ventilation calculation, which is usually used in the main program,
 is wrongly applied as the maximum inner iteration number for airflow dis- 
tribution calculation, which is usually used in subroutine ITR. The small error 
causes a big problem. The iteration for airflow distribution calculation, the
 inner iteration,  is forced stopped when the iteration time counter ITCT is
 larger than ITN, the maximum outer iteration time and normally given value 
about 10 to 15 in input data set. The ITN  is often too small  to achieve
 convergence of airflow distribution calculation, inner iteration, in the most
 data sets.   It causes the iteration divergence and the unbalancing airflow 
distribution. Finally, it causes a total wrong state simulation results. The 
maximum inner iteration number is replaced by 500 in subroutine ITR in
 the new version 2.10. The problem does not exist in former version 1.29 
because the number 500, not ITN, is used as the maximum inner iteration
number.