search

sonofeft / RocketCEA

RocketCEA Wraps The NASA Fortran CEA Code
GNU General Public License v3.0
94 stars 34 forks source link

Identical return values on get_Temperatures despite varying fuel combination mass ratios #21

Closed RasmusAPedersen closed 2 years ago

RasmusAPedersen commented 2 years ago

When I run this code:

from rocketcea.cea_obj_w_units import CEA_Obj

combustor = CEA_Obj(oxName='GOX', fuelName='C3H8', 
                   pressure_units='Pa',
                   cstar_units='m/s',
                   temperature_units='K',
                   sonic_velocity_units='m/s',
                   enthalpy_units='kJ/kg', # Need to be aware that this is kJ
                   density_units='kg/m^3',
                   specific_heat_units='J/kg-K',
                   thermal_cond_units='W/cm-degC') # Need to be aware of cm here

P_comb = 5.31e5 # 5.31 bar
er = 1

print(combustor.get_Temperatures(Pc=P_comb, MR=8.9,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.0,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.1,eps=er))

print('---')

from rocketcea.cea_obj import CEA_Obj
combustor = CEA_Obj(oxName='GOX', fuelName='C3H8')
print(combustor.get_Temperatures(Pc=P_comb, MR=8.9,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.0,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.1,eps=er))

I get

[2994.1379217115486, 2852.486415593293, 2852.4864153875865]
[2915.777219727355, 2767.0212830485807, 2767.02128304865]
[2915.777219727355, 2767.0212830485807, 2767.02128304865]
---
[6431.538299794149, 5929.8059041449515, 5929.805904145204]
[6065.983851953892, 5566.474112026709, 5566.474112026699]
[6065.983851953892, 5566.474112026709, 5566.474112026699]

The two returns of MR=9.0 and MR=9.1 are identical. I assume this might have something to do with caching, but I haven't been able to find the error in the source code.

sonofeft commented 2 years ago

In the first run, you are using units of Pa for pressure and degK for output temperature. In the second, you are using units of psia for pressure and degR for output temperature.

The following code will correct the values.


from rocketcea.cea_obj_w_units import CEA_Obj

combustor = CEA_Obj(oxName='GOX', fuelName='C3H8', 
                   pressure_units='Pa',
                   cstar_units='m/s',
                   temperature_units='K',
                   sonic_velocity_units='m/s',
                   enthalpy_units='kJ/kg', # Need to be aware that this is kJ
                   density_units='kg/m^3',
                   specific_heat_units='J/kg-K',
                   thermal_cond_units='W/cm-degC') # Need to be aware of cm here

P_comb = 5.31e5 # 5.31 bar
P_comb_psia = 0.000145038*P_comb # == 77.015 
print( 'P_comb=%g,  P_comb_psia=%g'%(P_comb, P_comb_psia) )
er = 1

print(combustor.get_Temperatures(Pc=P_comb, MR=8.9,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.0,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.1,eps=er))

print('---')

from rocketcea.cea_obj import CEA_Obj
combustor = CEA_Obj(oxName='GOX', fuelName='C3H8')
print( [t/1.8 for t in combustor.get_Temperatures(Pc=P_comb_psia, MR=8.9,eps=er)] )
print( [t/1.8 for t in combustor.get_Temperatures(Pc=P_comb_psia, MR=9.0,eps=er)] )
print( [t/1.8 for t in combustor.get_Temperatures(Pc=P_comb_psia, MR=9.1,eps=er)] )

```output is:
P_comb=531000,  P_comb_psia=77.0152
[2923.661514085015, 2775.686615971293, 2775.6866159713377]
[2915.793048468526, 2767.0340827573127, 2767.0340827572713]
[2907.9173515455013, 2758.358534637467, 2758.3585346375326]
---
[2923.661662450697, 2775.6867365248213, 2775.686736524833]
[2915.793194752911, 2767.034201047843, 2767.034201047821]
[2907.9174957518553, 2758.358650669505, 2758.3586506694633]
RasmusAPedersen commented 2 years ago

I am aware of the units. I remade the CEA object in imperial units to show that the issue was similar for both unit sets.

You will notice that the outputs of my function calls are identical for MR=9.0 and 9.1

That is the issue

sonofeft commented 2 years ago

The problem is the use of an integer for area ratio, for future versions I will consider casting the input to float. When I run the following code:

from rocketcea.cea_obj_w_units import CEA_Obj

combustor = CEA_Obj(oxName='GOX', fuelName='C3H8', 
                   pressure_units='Pa',
                   cstar_units='m/s',
                   temperature_units='K',
                   sonic_velocity_units='m/s',
                   enthalpy_units='kJ/kg', # Need to be aware that this is kJ
                   density_units='kg/m^3',
                   specific_heat_units='J/kg-K',
                   thermal_cond_units='W/cm-degC') # Need to be aware of cm here

P_comb = 5.31e5 # 5.31 bar
er = 1.0

print(combustor.get_Temperatures(Pc=P_comb, MR=8.9,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.0,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.1,eps=er))

"""
I get 
[2923.661514085015, 2775.686615971293, 2775.6866159713377]
[2915.793048468526, 2767.0340827573127, 2767.0340827572713]
[2907.9173515455013, 2758.358534637467, 2758.3585346375326]
"""
RasmusAPedersen commented 2 years ago

Correcting from

er = 1 to er = 1.0

Does not seem to fix the issue for me, unfortunately. In fact it seems sort of random when it uses the same value or not.

RasmusAPedersen commented 2 years ago

Tinkered a little more with it. Running this code:

from rocketcea.cea_obj import set_rocketcea_data_dir
set_rocketcea_data_dir( r'D:\temp\RocketCEA' )
from rocketcea.cea_obj_w_units import CEA_Obj

combustor = CEA_Obj(oxName='GOX', fuelName='C3H8', 
                   pressure_units='Pa',
                   cstar_units='m/s',
                   temperature_units='K',
                   sonic_velocity_units='m/s',
                   enthalpy_units='kJ/kg', # Need to be aware that this is kJ
                   density_units='kg/m^3',
                   specific_heat_units='J/kg-K',
                   thermal_cond_units='W/cm-degC') # Need to be aware of cm here

P_comb = 5.31e5 # 5.31 bar
er = 1.0

print(combustor.get_Temperatures(Pc=float(P_comb), MR=float(8.9),eps=float(er)))
print(combustor.get_Temperatures(Pc=float(P_comb), MR=float(9.0),eps=float(er)))
print(combustor.get_Temperatures(Pc=float(P_comb), MR=float(9.1),eps=float(er)))

Gives me this:

Setting ROCKETCEA_DATA_DIR = D:\temp\RocketCEA
[2994.1379217115486, 2852.486415593293, 2852.4864153875865]
[2915.777219727355, 2767.0212830485807, 2767.02128304865]
[2915.777219727355, 2767.0212830485807, 2767.02128304865]

Which is still the same issue. If I change MR to 10, it gives me a different return, and it only seems to update its output on whole integers of MR.

sonofeft commented 2 years ago

Running CEA with no nozzle at all (i.e. an area ratio of 1.0) is a bit unusual. Try putting a small nozzle on. For example a 2:1 nozzle. When I run er = 2.0, I get the same chamber and throat temperatures as er = 1.0.

from rocketcea.cea_obj_w_units import CEA_Obj

combustor = CEA_Obj(oxName='GOX', fuelName='C3H8', 
                   pressure_units='Pa',
                   cstar_units='m/s',
                   temperature_units='K',
                   sonic_velocity_units='m/s',
                   enthalpy_units='kJ/kg', # Need to be aware that this is kJ
                   density_units='kg/m^3',
                   specific_heat_units='J/kg-K',
                   thermal_cond_units='W/cm-degC') # Need to be aware of cm here

P_comb = 5.31e5 # 5.31 bar
er = 2.0

print(combustor.get_Temperatures(Pc=P_comb, MR=8.9,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.0,eps=er))
print(combustor.get_Temperatures(Pc=P_comb, MR=9.1,eps=er))

"""
with 1.0 I get 
[2923.661514085015, 2775.686615971293, 2775.6866159713377]
[2915.793048468526, 2767.0340827573127, 2767.0340827572713]
[2907.9173515455013, 2758.358534637467, 2758.3585346375326]

with 2.0 I get 
[2923.661514085015, 2775.686615971293, 2376.9687762950966]
[2915.793048468526, 2767.0340827573127, 2364.698294153291]
[2907.9173515455013, 2758.358534637467, 2352.3556307779886]

"""
RasmusAPedersen commented 2 years ago

The issue remains still.

Is there some kind of printout of my system that I can give you?

sonofeft commented 2 years ago

OK... try the following code to generate standard CEA output. Since you are running such low pressure and extremely oxidizer rich, it is possible that the FORTRAN in CEA, when driven by RocketCEA, is leaving some internal iteration termination criteria as satisfied when the mixture ratio changes so slightly (that's a total guess on my part)

Each of the runs should give a detailed output with the correct results.

Also, since I can not reliably reproduce your error, tell me what version of python you are using and a little about your Windows environment.

from rocketcea.cea_obj import CEA_Obj
combustor = CEA_Obj(oxName='GOX', fuelName='C3H8')

P_comb = 5.31e5 # 5.31 bar
P_comb_psia = 0.000145038*P_comb # == 77.015 

er = 2.0

print(combustor.get_full_cea_output(Pc=P_comb_psia, MR=8.9,eps=er, short_output=1))
print(combustor.get_full_cea_output(Pc=P_comb_psia, MR=9.0,eps=er, short_output=1))
print(combustor.get_full_cea_output(Pc=P_comb_psia, MR=9.1,eps=er, short_output=1))
RasmusAPedersen commented 2 years ago

I ran the code as you suggested.

from rocketcea.cea_obj import CEA_Obj
combustor = CEA_Obj(oxName='GOX', fuelName='C3H8')

P_comb = 5.31e5 # 5.31 bar
P_comb_psia = 0.000145038*P_comb # == 77.015 

er = 2.0

print(combustor.get_full_cea_output(Pc=P_comb_psia, MR=8.9,eps=er, short_output=1))
print(combustor.get_full_cea_output(Pc=P_comb_psia, MR=9.0,eps=er, short_output=1))
print(combustor.get_full_cea_output(Pc=P_comb_psia, MR=9.1,eps=er, short_output=1))

And the output is below.

It's rather curious that CEA reports the same OF being calculated twice despite it seemingly having understood the input?

 *******************************************************************************

         NASA-GLENN CHEMICAL EQUILIBRIUM PROGRAM CEA, OCTOBER 18, 2002
                   BY  BONNIE MCBRIDE AND SANFORD GORDON
      REFS: NASA RP-1311, PART I, 1994 AND NASA RP-1311, PART II, 1996

 *******************************************************************************

 reac
  fuel C3H8(L) C 3 H 8     wt%=100.
  h,cal=-30372.     t(k)=231.08   rho=0.5808
  oxid O2(G)  O 2
  h,cal=0.      t(k)=298.15       wt%=100.

 prob case=RocketCEA,
  rocket equilibrium   p,psia=77.015178,  supar=2.000000,
  o/f=8.900000,

 output calories  short  transport
 end

              THEORETICAL ROCKET PERFORMANCE ASSUMING EQUILIBRIUM

           COMPOSITION DURING EXPANSION FROM INFINITE AREA COMBUSTOR

 Pinj =    77.0 PSIA
 CASE = RocketCEA,     

             REACTANT                    WT FRACTION      ENERGY      TEMP
                                          (SEE NOTE)      CAL/MOL       K  
 FUEL        C3H8(L)                      1.0000000    -30372.000    231.000
 OXIDANT     O2(G)                        1.0000000         0.000    298.000

 O/F=    8.00000  %FUEL= 11.111111  R,EQ.RATIO= 0.453543  PHI,EQ.RATIO= 0.453543

                 CHAMBER   THROAT     EXIT
 Pinf/P            1.0000   1.7253   7.5280
 P, ATM            5.2395   3.0369  0.69600
 T, K             2994.14  2852.49  2483.59
 RHO, G/CC       6.1074-4 3.7603-4 1.0172-4
 H, CAL/G         -76.531  -186.49  -452.19
 U, CAL/G         -284.29  -382.07  -617.89
 G, CAL/G        -7235.75 -7007.01 -6390.65
 S, CAL/(G)(K)     2.3911   2.3911   2.3911

 M, (1/n)          28.639   28.983   29.785
 (dLV/dLP)t      -1.02493 -1.02032 -1.00922
 (dLV/dLT)p        1.5385   1.4628   1.2449
 Cp, CAL/(G)(K)    1.2077   1.1202   0.8221
 GAMMAs            1.1249   1.1244   1.1319
 SON VEL,M/SEC      988.9    959.2    885.9
 MACH NUMBER        0.000    1.000    2.001

 TRANSPORT PROPERTIES (GASES ONLY)
   CONDUCTIVITY IN UNITS OF MILLICALORIES/(CM)(K)(SEC)

 VISC,MILLIPOISE   1.0165  0.98285  0.89347

  WITH EQUILIBRIUM REACTIONS

 Cp, CAL/(G)(K)    1.2077   1.1202   0.8221
 CONDUCTIVITY      2.3679   2.1005   1.3179
 PRANDTL NUMBER    0.5185   0.5242   0.5573

  WITH FROZEN REACTIONS

 Cp, CAL/(G)(K)    0.3921   0.3900   0.3832
 CONDUCTIVITY      0.5610   0.5373   0.4745
 PRANDTL NUMBER    0.7106   0.7134   0.7216

 PERFORMANCE PARAMETERS

 Ae/At                      1.0000   2.0000
 CSTAR, FT/SEC              4828.9   4828.9
 CF                         0.6517   1.2046
 Ivac,LB-SEC/LB              184.8    220.7
 Isp, LB-SEC/LB               97.8    180.8

 MOLE FRACTIONS

 *CO              0.03840  0.03072  0.01276
 *CO2             0.17809  0.18837  0.21240
 *H               0.00569  0.00429  0.00146
 HO2              0.00017  0.00012  0.00004
 *H2              0.00711  0.00585  0.00274
 H2O              0.24372  0.25426  0.27999
 H2O2             0.00001  0.00000  0.00000
 *O               0.03140  0.02490  0.01062
 *OH              0.06978  0.05961  0.03346
 *O2              0.42564  0.43187  0.44654

  * THERMODYNAMIC PROPERTIES FITTED TO 20000.K

 NOTE. WEIGHT FRACTION OF FUEL IN TOTAL FUELS AND OF OXIDANT IN TOTAL OXIDANTS

 *******************************************************************************

         NASA-GLENN CHEMICAL EQUILIBRIUM PROGRAM CEA, OCTOBER 18, 2002
                   BY  BONNIE MCBRIDE AND SANFORD GORDON
      REFS: NASA RP-1311, PART I, 1994 AND NASA RP-1311, PART II, 1996

 *******************************************************************************

 reac
  fuel C3H8(L) C 3 H 8     wt%=100.
  h,cal=-30372.     t(k)=231.08   rho=0.5808
  oxid O2(G)  O 2
  h,cal=0.      t(k)=298.15       wt%=100.

 prob case=RocketCEA,
  rocket equilibrium   p,psia=77.015178,  supar=2.000000,
  o/f=9.000000,

 output calories  short  transport
 end

              THEORETICAL ROCKET PERFORMANCE ASSUMING EQUILIBRIUM

           COMPOSITION DURING EXPANSION FROM INFINITE AREA COMBUSTOR

 Pinj =    77.0 PSIA
 CASE = RocketCEA,     

             REACTANT                    WT FRACTION      ENERGY      TEMP
                                          (SEE NOTE)      CAL/MOL       K  
 fuel        C3H8(L)                    100.0000000    -30372.000    231.000
 oxid        O2(G)                      100.0000000         0.000    298.000

 O/F=    9.00000  %FUEL= 10.000000  R,EQ.RATIO= 0.403149  PHI,EQ.RATIO= 0.403149

                 CHAMBER   THROAT     EXIT
 Pinf/P            1.0000   1.7296   7.6469
 P, ATM            5.2395   3.0294  0.68518
 T, K             2915.78  2767.02  2364.69
 RHO, G/CC       6.3999-4 3.9416-4 1.0674-4
 H, CAL/G         -68.878  -174.15  -427.75
 U, CAL/G         -267.14  -360.27  -583.20
 G, CAL/G        -6875.00 -6633.04 -5947.51
 S, CAL/(G)(K)     2.3342   2.3342   2.3342

 M, (1/n)          29.225   29.543   30.228
 (dLV/dLP)t      -1.01876 -1.01443 -1.00498
 (dLV/dLT)p        1.4184   1.3411   1.1408
 Cp, CAL/(G)(K)    1.0235   0.9279   0.6362
 GAMMAs            1.1298   1.1312   1.1488
 SON VEL,M/SEC      968.1    938.6    864.4
 MACH NUMBER        0.000    1.000    2.005

 TRANSPORT PROPERTIES (GASES ONLY)
   CONDUCTIVITY IN UNITS OF MILLICALORIES/(CM)(K)(SEC)

 VISC,MILLIPOISE  0.99688  0.96114  0.86251

  WITH EQUILIBRIUM REACTIONS

 Cp, CAL/(G)(K)    1.0235   0.9279   0.6362
 CONDUCTIVITY      1.9194   1.6479   0.9228
 PRANDTL NUMBER    0.5316   0.5412   0.5946

  WITH FROZEN REACTIONS

 Cp, CAL/(G)(K)    0.3818   0.3794   0.3714
 CONDUCTIVITY      0.5319   0.5077   0.4407
 PRANDTL NUMBER    0.7155   0.7183   0.7270

 PERFORMANCE PARAMETERS

 Ae/At                      1.0000   2.0000
 CSTAR, FT/SEC              4708.3   4708.3
 CF                         0.6540   1.2075
 Ivac,LB-SEC/LB              180.3    215.0
 Isp, LB-SEC/LB               95.7    176.7

 MOLE FRACTIONS

 *CO              0.02613  0.01954  0.00583
 *CO2             0.17270  0.18145  0.19983
 *H               0.00364  0.00254  0.00058
 HO2              0.00016  0.00011  0.00003
 *H2              0.00480  0.00373  0.00131
 H2O              0.22907  0.23875  0.26128
 H2O2             0.00001  0.00000  0.00000
 *O               0.02526  0.01895  0.00609
 *OH              0.05866  0.04836  0.02262
 *O2              0.47956  0.48657  0.50244

  * THERMODYNAMIC PROPERTIES FITTED TO 20000.K

 NOTE. WEIGHT FRACTION OF FUEL IN TOTAL FUELS AND OF OXIDANT IN TOTAL OXIDANTS

 *******************************************************************************

         NASA-GLENN CHEMICAL EQUILIBRIUM PROGRAM CEA, OCTOBER 18, 2002
                   BY  BONNIE MCBRIDE AND SANFORD GORDON
      REFS: NASA RP-1311, PART I, 1994 AND NASA RP-1311, PART II, 1996

 *******************************************************************************

 reac
  fuel C3H8(L) C 3 H 8     wt%=100.
  h,cal=-30372.     t(k)=231.08   rho=0.5808
  oxid O2(G)  O 2
  h,cal=0.      t(k)=298.15       wt%=100.

 prob case=RocketCEA,
  rocket equilibrium   p,psia=77.015178,  supar=2.000000,
  o/f=9.100000,

 output calories  short  transport
 end

              THEORETICAL ROCKET PERFORMANCE ASSUMING EQUILIBRIUM

           COMPOSITION DURING EXPANSION FROM INFINITE AREA COMBUSTOR

 Pinj =    77.0 PSIA
 CASE = RocketCEA,     

             REACTANT                    WT FRACTION      ENERGY      TEMP
                                          (SEE NOTE)      CAL/MOL       K  
 fuel        C3H8(L)                    100.0000000    -30372.000    231.000
 oxid        O2(G)                      100.0000000         0.000    298.000

 O/F=    9.00000  %FUEL= 10.000000  R,EQ.RATIO= 0.403149  PHI,EQ.RATIO= 0.403149

                 CHAMBER   THROAT     EXIT
 Pinf/P            1.0000   1.7296   7.6469
 P, ATM            5.2395   3.0294  0.68518
 T, K             2915.78  2767.02  2364.69
 RHO, G/CC       6.3999-4 3.9416-4 1.0674-4
 H, CAL/G         -68.878  -174.15  -427.75
 U, CAL/G         -267.14  -360.27  -583.20
 G, CAL/G        -6875.00 -6633.04 -5947.51
 S, CAL/(G)(K)     2.3342   2.3342   2.3342

 M, (1/n)          29.225   29.543   30.228
 (dLV/dLP)t      -1.01876 -1.01443 -1.00498
 (dLV/dLT)p        1.4184   1.3411   1.1408
 Cp, CAL/(G)(K)    1.0235   0.9279   0.6362
 GAMMAs            1.1298   1.1312   1.1488
 SON VEL,M/SEC      968.1    938.6    864.4
 MACH NUMBER        0.000    1.000    2.005

 TRANSPORT PROPERTIES (GASES ONLY)
   CONDUCTIVITY IN UNITS OF MILLICALORIES/(CM)(K)(SEC)

 VISC,MILLIPOISE  0.99688  0.96114  0.86251

  WITH EQUILIBRIUM REACTIONS

 Cp, CAL/(G)(K)    1.0235   0.9279   0.6362
 CONDUCTIVITY      1.9194   1.6479   0.9228
 PRANDTL NUMBER    0.5316   0.5412   0.5946

  WITH FROZEN REACTIONS

 Cp, CAL/(G)(K)    0.3818   0.3794   0.3714
 CONDUCTIVITY      0.5319   0.5077   0.4407
 PRANDTL NUMBER    0.7155   0.7183   0.7270

 PERFORMANCE PARAMETERS

 Ae/At                      1.0000   2.0000
 CSTAR, FT/SEC              4708.3   4708.3
 CF                         0.6540   1.2075
 Ivac,LB-SEC/LB              180.3    215.0
 Isp, LB-SEC/LB               95.7    176.7

 MOLE FRACTIONS

 *CO              0.02613  0.01954  0.00583
 *CO2             0.17270  0.18145  0.19983
 *H               0.00364  0.00254  0.00058
 HO2              0.00016  0.00011  0.00003
 *H2              0.00480  0.00373  0.00131
 H2O              0.22907  0.23875  0.26128
 H2O2             0.00001  0.00000  0.00000
 *O               0.02526  0.01895  0.00609
 *OH              0.05866  0.04836  0.02262
 *O2              0.47956  0.48657  0.50244

  * THERMODYNAMIC PROPERTIES FITTED TO 20000.K

 NOTE. WEIGHT FRACTION OF FUEL IN TOTAL FUELS AND OF OXIDANT IN TOTAL OXIDANTS
sonofeft commented 2 years ago

It looks like the FORTRAN CEA code is not reading the input deck correctly. (i.e. the O/F in the input deck is correct, but CEA runs the integer value of the O/F)

What can you tell me about how you compiled the FORTRAN?

sonofeft commented 2 years ago

Perhaps to make sure that is the problem for your calls to get_Temperatures, try the following code. I get the smooth curve shown below, you may get stairsteps.

from pylab import *
from rocketcea.cea_obj import CEA_Obj

combustor = CEA_Obj(oxName='GOX', fuelName='C3H8')
er = 2.0
P_comb_psia = 77.015 

mrL = [imr/10.0 for imr in range(1,100)]
TcL = [combustor.get_Temperatures(Pc=P_comb_psia, MR=mr,eps=er)[0] for mr in mrL]

plot(mrL, TcL)
title("Chamber Temperature")
show()

chk_cea_const plot

RasmusAPedersen commented 2 years ago

I ran the code you pasted, and it outputs this. It's quite clear it only runs a different OF value on whole integers. Something somewhere seems to get casted to an int? image

As for how I compiled Fortran, I don't think I'm qualified enough to give a detailed report on that. I'm on a Win10 machine, and I followed your provided instructions here: https://rocketcea.readthedocs.io/en/latest/installgfortran.html

Previously, I had some issues getting rocketcea to work, as discussed in this issue: #18

sonofeft commented 2 years ago

I'm pretty sure that the FORTRAN compile is the problem. (i.e. RocketCEA creates a correct input deck to the FORTRAN code, but the FORTRAN misinterprets a float as an integer.

The FORTRAN compile is administered by the numpy.f2py code, so it is certainly possible that python version or numpy version could affect it. Anaconda may also have some special handling for FORTRAN.

Because Windows is so tricky to get the compile correct, and I am so unfamiliar with your particular setup, I don't think I can solve the FORTRAN compile from afar.

I have 3 recommendations, none of which are perfect, but they would allow you to run cases on your Windows machine.

1) The most "native" option, would be to install a version of python outside of Anaconda and follow the directions for Windows at: https://rocketcea.readthedocs.io/en/latest/quickstart.html#windows-batch-file

I'm assuming you already have MinGW installed from:

https://rocketcea.readthedocs.io/en/latest/installgfortran.html#windows-10

The most trouble-free version of python is probably 3.7 from:
https://www.python.org/downloads/release/python-379/
I recommend you start with 3.7 and only after that works, try moving to more recent versions of python.

You would need to either set the path in your terminal to find the correct python EXE file, or call the full path each time (e.g. D:\python37_64\python.exe script_name.py). I like batch files for each version (e.g. py37 script_name.py)

2) Run cases on Google Colaboratory as shown in: https://rocketcea.readthedocs.io/en/latest/quickstart.html#google-colaboratory

3) Run cases on Windows 10 with WSL as shown in: https://rocketcea.readthedocs.io/en/latest/quickstart.html#windows-10-with-wsl

Please post a comment of your results (especially if you are able to solve the original FORTRAN problem in Anaconda)

I'm sorry I wasn't more help.