setcal.f

103 lines · 1 definitions · on GitHub (7be4133)

Python version: setcal.py, 78 lines, on the right (switch to Python above the code).

Variables on this page (18)

calsolage, cmixla, endage, iresca, lendag, lfirst, lsenv0a, lsetdt, nmodls, numrun, rescal, rsclzc, rsclzm1, rsclzm2, senv0a, setdt, xenv0a, zenv0a

Purpose

Prepares the list of runs ("kind cards") for the automatic calibration of a solar model: it fills 48 runs, 16 repetitions of a three-run pattern, from the first three runs given by the user. The calibration itself (adjusting $X$ and the mixing length between repetitions; with $Z$ or $Z/X$ held fixed, setting $X$ sets the initial helium $Y = 1 - X - Z$, which is why the documentation speaks of calibrating the helium) is done elsewhere, by CHKCAL.

How it fits

Who calls it: the main program (main.f), when LCALS is true, before the loop over runs. It passes FACAGE, an age factor, now always 1 (main.f sets FACAGE = 1.0D0, main.f, line 369 ↗; it was drawn per model in the Monte Carlo option, removed since rp_1), and LSOL, a local of the main program that this routine sets to false (the flag CHKCAL sets when the calibration has converged).

Question 1 (line 2 ↗; lines 77–82 ↗; main.f, line 369 ↗). FACAGE still an argument. With the Monte Carlo option removed, main.f sets FACAGE = 1.0D0 and nothing changes it, so the factor on the end ages and SETDT of runs 2 and 3 is always 1. Can the argument go, or is it kept for a planned use?
Inputs and outputs, through modules (besides LSOL): NMLPAR_MOD, the namelist's per-run arrays (NUMRUN, NMODLS, LFIRST, ENDAGE, SETDT, XENV0A, ZENV0A, CMIXLA, LSENV0A, SENV0A, RSCLZC, RSCLZM1, RSCLZM2, and CALSOLAGE, the solar age), and RUNLOOP_MOD (RESCAL, IRESCA, LENDAG, LSETDT). These are shared with PARMIN (which reads them), CHKCAL, RSCALE, STARIN and the main program.

The method

NUMRUN = 48; the composition and mixing length of run 1 are copied to runs 2–48.

The comment (lines 52–55 ↗) is stale: Since "mhp 5/96" (line 56 ↗) the code sets NUMRUN = 48 (line 57 ↗), not "THE MAXIMUM (50)," and copies the first three runs, not two, to runs 4–48, 15 further series of three (the loops at line 65 ↗, line 83 ↗, and line 92 ↗, in steps of 3); its "24" fits the two-run pattern that SETSCAL still uses (setscal.f, lines 112–131 ↗).

Question 2 (lines 52–55 ↗). A stale comment. Can the comment be brought up to date, to say that the first three runs are copied to the next 15 series of three (runs 4–48)?
Runs 4, 7, …, 46 repeat run 1 (a new starting model with its rescaling; LFIRST true).
Run 2 ends at $10^8$ yr × FACAGE, and run 3 at CALSOLAGE × FACAGE; both time steps SETDT are scaled by FACAGE too. Runs 5, 8, …, 47 repeat run 2, and runs 6, 9, …, 48 repeat run 3, each continuing from the previous model (IRESCA = 1, LFIRST false).

Question 3 (lines 76–77 ↗). The hard-wired $10^8$ yr. Run 2's end age is fixed at $10^8$ yr regardless of the input ENDAGE(2) (the older line, line 78 ↗, is commented out). What is this run for (probably reaching the zero-age main sequence)? Would a namelist value be clearer?

The stop criteria END_DCEN, END_XCEN, and END_YCEN are not copied (suspected bug 1).

From YREC's documentation

Control Options (Calibration): LCALS calibrates the solar helium abundance and mixing length automatically, to match the Sun's radius and luminosity at its age CALSOLAGE (default 4.57D9 yr; the documentation calls it a "tolerance"), within TOLL and TOLR (1.0D-5, 1.0D-4); with LCALSOLZX, $Z/X$ is held at CALSOLZX (0.02292, GS98), otherwise $Z$ is held fixed.

Suspected bugs, checked by running the code

Each bug is marked by what running the code showed: (changes results, large) or (small), the fix moving a run clearly beyond a null change; (stops the run); (wrong output), the fix changing only what is written; (harmless in the runs), within the noise; (never occurred), the code ran but the fault did not; (not reached); (cannot be reached), a guard prevents it; (withdrawn), the suspicion was wrong.

1. (never occurred) The new stop criteria are not copied to the repeated runs.

(lines 83–100 ↗; nmlpar_mod.f, lines 261–263 ↗; main.f, lines 1089–1115 ↗) The loops that repeat runs 2 and 3 copy ENDAGE, LENDAG, SETDT, and LSETDT, but not END_DCEN, END_XCEN, and END_YCEN ("MHP 10/24 ADDED STOP CRITERIA," line 32 ↗), per-run arrays in NMLPAR_MOD (nmlpar_mod.f, lines 261–263 ↗) that the main program reads for each run (main.f, lines 430–459 ↗; main.f, lines 1089–1115 ↗). What follows: In a calibration with such a stop set for run 2 or 3, only run 2 (or 3) itself stops at it; its repetitions, runs 5 to 47 (or 6 to 48), run to their end ages. SETSCAL has the same omission (setscal.f, lines 131–139 ↗). Only with LCALS and a stop criterion set for run 2 or 3.

A possible fix:

setcal.f

@@ lines 88–93 @@
          LENDAG(I) = LENDAG(2)
          SETDT(I) = SETDT(2)
          LSETDT(I) = LSETDT(2)
+C 11/10/26 KC via Claude
+C code changed to copy the stop criteria of run 2 too.
+         END_DCEN(I) = END_DCEN(2)
+         END_XCEN(I) = END_XCEN(2)
+         END_YCEN(I) = END_YCEN(2)
       END DO
       DO I = 6,48,3
          IRESCA(I) = 1
@@ lines 97–102 @@
          LENDAG(I) = LENDAG(3)
          SETDT(I) = SETDT(3)
          LSETDT(I) = LSETDT(3)
+C 11/10/26 KC via Claude
+C code changed to copy the stop criteria of run 3 too.
+         END_DCEN(I) = END_DCEN(3)
+         END_XCEN(I) = END_XCEN(3)
+         END_YCEN(I) = END_YCEN(3)
       END DO
       LSOL = .FALSE.
       RETURN

Checked by running YREC (Nov 10, 2026; a line added at each suspected path counted how often it was taken, the outputs otherwise unchanged): SETCAL ran once, in YREC's one calibration case, examples/run_standard_solar_model/Test_solarcal_noGS_norot (NUMRUN = 48 after it), and no END_DCEN, END_XCEN, or END_YCEN was set for run 2 or 3: A line counting any nonzero one never fired, so copying them would change nothing. The fix was not run. What would reach it: a calibration (LCALS) with one of these stops set for run 2 or 3, which no case of YREC's sets. Intent: none found.

Definitions

subroutineSETCAL

lines 2–103 · GitHub

Uses modules NMLPAR_MOD, RUNLOOP_MOD

Called by (file level)

Change log

The changes since rp_1

Changed since rp_1. LSOL is an argument (formerly in /CALSUN/, whose other members this routine never used); declarations only otherwise: COMMON blocks to modules, IMPLICIT NONE.

1C $$$$$$ 2 SUBROUTINE SETCAL(FACAGE,LSOL) 3C 10/14/26 MHP via Claude 4C code changed to convert COMMON/NMLPAR/ into NMLPAR_MOD. 5 USE NMLPAR_MOD 6C 10/14/26 MHP via Claude 7C code changed to convert COMMON/RUNLOOP/ into RUNLOOP_MOD. A blanket 8C USE imports exactly the names the COMMON declaration made global here, 9C so it cannot change meaning. See runloop_mod.f for the block history 10C and for why the module declares the types the implicit rules supplied. 11 USE RUNLOOP_MOD 12C 10/20/26 MHP via Claude 13C code changed to replace implicit typing with IMPLICIT NONE and 14C explicit declarations. 15 IMPLICIT NONE 16 REAL*8 FACAGE 17 INTEGER I,J 18 LOGICAL*4 LSOL 19 20C 10/12/26 MHP via Claude 21C code changed to extend COMMON/NMLPAR/ to 234 members, adding the 22C namelist variables that the physics routines read from 27 COMMON 23C blocks which also held derived or working members. Those blocks keep 24C only their non-namelist members; five of them -- NEWMX, VMULT, 25C PMMWIND, CCOUT1 and ENVGEN -- had nothing left and are gone. 26C Derivation-only namelist variables from the same blocks went to 27C COMMON/NMLBYPASS/ instead, not here. 28C 29C Declared identically in all 121 files that hold any of these members. 30C Generated from one member list, so every copy is byte-identical; the 31C linker's 'size of symbol changed' warning is the check that they are. 32C MHP 10/24 ADDED STOP CRITERIA FOR CENTRAL H,D,AND HE4 33C COMMON/SETT/ENDAGE(50),SETDT(50),LENDAG(50),LSETDT(50) 34C 10/13/26 MHP via Claude 35C code changed to replace COMMON/CALSUN/ with an explicit argument. This 36C routine touched only LSOL, which it clears; the block's other eight 37C members were carried here positionally and never used. 38C MHP 8/25 Removed character file names from common block 39C 9/24/26 MHP via Claude 40C code changed to remove dead IOOPAL2 from COMMON/ZRAMP/ 41C 10/9/26 MHP via Claude 42C code changed to remove NK from COMMON/ZRAMP/: it was only ever a 43C redundant copy of the run-index loop variable main.f already passes 44C explicitly to the subroutines that need it (STARIN/RSCALE), and was 45C never read by name in this file. 46C 10/13/26 MHP via Claude 47C code changed to remove COMMON/ZRAMP/: its sole remaining member 48C IOLAOL2 is neither written nor read anywhere. The namelist members 49C that shared this block moved to the consolidated block in stage 2b. 50 SAVE 51 52C SET UP RUN TO CALIBRATE A SOLAR MODEL. 53C THIS CONSISTS OF SETTING THE NUMBER OF RUNS TO THE MAXIMUM (50), 54C AND COPYING THE RELEVANT PARAMETERS FROM THE FIRST TWO RUNS TO 55C THE NEXT SERIES OF 24 CALIBRATING RUNS. 56c mhp 5/96 changed to do solar models in 3 runs rather than 2. 57 NUMRUN = 48 58 DO I = 2,48 59 XENV0A(I) = XENV0A(1) 60 ZENV0A(I) = ZENV0A(1) 61 CMIXLA(I) = CMIXLA(1) 62 LSENV0A(I) = LSENV0A(1) 63 SENV0A(I) = SENV0A(1) 64 END DO 65 DO I = 4,46,3 66 IRESCA(I) = IRESCA(1) 67 LFIRST(I) = .TRUE. 68 NMODLS(I) = NMODLS(1) 69 RSCLZC(I) = RSCLZC(1) 70 RSCLZM1(I) = RSCLZM1(1) 71 RSCLZM2(I) = RSCLZM2(1) 72 DO J = 1,4 73 RESCAL(J,I) = RESCAL(J,1) 74 END DO 75 END DO 76C MHP 06/13 HARDWIRE RUN 2 TO 1D8 YEARS AND RUN3 TO CALSOLAGE YEARS 77 ENDAGE(2) = 1.0D8*FACAGE 78C ENDAGE(2) = ENDAGE(2)*FACAGE 79 SETDT(2) = SETDT(2)*FACAGE 80 ENDAGE(3) = CALSOLAGE*FACAGE 81C ENDAGE(3) = ENDAGE(3)*FACAGE 82 SETDT(3) = SETDT(3)*FACAGE 83 DO I = 5,47,3 84 IRESCA(I) = 1 85 LFIRST(I) = .FALSE. 86 NMODLS(I) = NMODLS(2) 87 ENDAGE(I) = ENDAGE(2) 88 LENDAG(I) = LENDAG(2) 89 SETDT(I) = SETDT(2) 90 LSETDT(I) = LSETDT(2) 91 END DO 92 DO I = 6,48,3 93 IRESCA(I) = 1 94 LFIRST(I) = .FALSE. 95 NMODLS(I) = NMODLS(3) 96 ENDAGE(I) = ENDAGE(3) 97 LENDAG(I) = LENDAG(3) 98 SETDT(I) = SETDT(3) 99 LSETDT(I) = LSETDT(3) 100 END DO 101 LSOL = .FALSE. 102 RETURN 103 END

1"""Python version of setcal.f (YREC 7be4133): the 48 runs of an automatic 2solar calibration, from the first three.""" 3 4from pyrec import nmlpar_mod 5from pyrec import runloop_mod 6 7 8def setcal(facage: float, lsol: bool) -> bool: 9 """Fill the run list for a solar calibration: 16 repetitions of a 10 three-run pattern (a new starting model, a run to 1e8 yr, a run to 11 the solar age), copied from runs 1-3, with ages and steps scaled by 12 facage. 13 14 Translates: setcal.f#SETCAL (7be4133) 15 16 Arguments: facage, the age factor (always 1 now; question on the page); 17 lsol, the calibration's convergence flag (only set). 18 19 Returns lsol, set to False. 20 21 """ 22 23 n = nmlpar_mod 24 r = runloop_mod 25 26 # Each calibration cycle takes 3 runs (kind cards) (mhp 5/96); run 1's 27 # composition and mixing length copied to all (Fortran lines 57–64). 28 # Question (see the page): the comment says two runs and 24. 29 n.numrun = 48 30 for i in range(2, 49): 31 n.xenv0a[i] = n.xenv0a[1] 32 n.zenv0a[i] = n.zenv0a[1] 33 n.cmixla[i] = n.cmixla[1] 34 n.lsenv0a[i] = n.lsenv0a[1] 35 n.senv0a[i] = n.senv0a[1] 36 37 # Runs 4, 7, ..., 46 repeat run 1 (a new starting model) (Fortran lines 65–75). 38 for i in range(4, 47, 3): 39 r.iresca[i] = r.iresca[1] 40 n.lfirst[i] = True 41 n.nmodls[i] = n.nmodls[1] 42 n.rsclzc[i] = n.rsclzc[1] 43 n.rsclzm1[i] = n.rsclzm1[1] 44 n.rsclzm2[i] = n.rsclzm2[1] 45 for j in range(1, 5): 46 r.rescal[j, i] = r.rescal[j, 1] 47 48 # Run 2 is hard-wired to 1e8 yr, run 3 to CALSOLAGE (MHP 06/13) (Fortran lines 76–82). 49 # Question (see the page): the hard-wired 1e8 yr. 50 n.endage[2] = 1.0e8 * facage 51 n.setdt[2] = n.setdt[2] * facage 52 n.endage[3] = n.calsolage * facage 53 n.setdt[3] = n.setdt[3] * facage 54 55 # Runs 5, 8, ..., 47 repeat run 2, continuing the model (Fortran lines 83–91). 56 # Question (see the page): END_DCEN, END_XCEN, END_YCEN not copied. 57 for i in range(5, 48, 3): 58 r.iresca[i] = 1 59 n.lfirst[i] = False 60 n.nmodls[i] = n.nmodls[2] 61 n.endage[i] = n.endage[2] 62 r.lendag[i] = r.lendag[2] 63 n.setdt[i] = n.setdt[2] 64 r.lsetdt[i] = r.lsetdt[2] 65 66 # Runs 6, 9, ..., 48 repeat run 3 (Fortran lines 92–100). 67 for i in range(6, 49, 3): 68 r.iresca[i] = 1 69 n.lfirst[i] = False 70 n.nmodls[i] = n.nmodls[3] 71 n.endage[i] = n.endage[3] 72 r.lendag[i] = r.lendag[3] 73 n.setdt[i] = n.setdt[3] 74 r.lsetdt[i] = r.lsetdt[3] 75 76 # Fortran line 101. 77 lsol = False 78 return lsol