Bug in make_psr_make for pulsars using FB0 for orbital frequency

[scottransom]

@SafferA and I just found that several of the NANOGrav black-widows are not having their t-scrunch times set appropriately because of the line below. That line requires a PB to be present in the stored ephemeris, but that does not exist for those pulsars using FB* params. That means that those pulsars then default to the standard scrunch time.

nanopipe/scripts/make_psr_make

Line 256 in 78996d7

pb_sec = float(arch.get_ephemeris().get_value('PB').replace('D','e'))*86400.0

I don't know if there is an easy workaround in nanopipe since vap -E for those pulsars doesn't print FB0 either! So this might also be a PSRCHIVE bug.

(pulsar) [sr00075@link J0023+0923]$ vap -E vegas_60280_83223_J0023+0923_0002_0001.fits

PSRJ              J0023+0923
ELONG             9.0703848506822169D+00
ELAT              6.3091082323120879D+00
PEPOCH            57405.0000000000000000
F                 3.2784702050177981D+02
F1               -1.2278318405605864D-15
PMELONG          -1.3881858402574196D+01
PMELAT           -4.0989108213472647D-01
POSEPOCH          57405.0000000000000000
PX                9.7313554249001555D-01
DM                1.4326208874275428D+01
START             59007.5038422969810199
FINISH            59936.2168303359721904
CLK               TT(BIPM)
EPHEM             DE436
TZRMJD            56853.3680633734190053
TZRFRQ            4.3156200000000001D+02
TZRSITE           AO
BINARY            ELL1
A1                3.4841416700845666D-02
EPS1              6.5756014817169405D-06
EPS2             -2.6220510747491727D-06
TASC              57404.9565064508715295
NTOA              1950
UNITS             TDB
TIMEEPH           FB90
T2CMETHOD         IAU2000B
DMEPOCH           5.7405000000000000D+04