fix(filters): use the measured thicknesses and keep calibration foils out of fil_trans

Two ways the filter table disagreed with filter.mac rev 1.19.

The Si thicknesses were nominal where spec used values measured at
18.58 keV: 400/200/3200/100/1600/800 against 345.6/234.6/3303/137.5/
1661.5/833.8. Si100 vs Si137.5 is 37% out, and at 17 keV the pair
Si200+Si1600 gives T=0.060 where the measured Si234.6+Si1661.5 gives
0.052 -- a 16% error, larger than any distinction the search was trying
to make between neighbouring combinations.

The Zr and Cu calibration foils were selectable. filter.mac flagged them
disabled so that selection by transmission could not pick them, and at
17 keV fil_trans(0.05) duly returned 5246 -- a combination built on the
Zr foil, 1 keV below its K edge, where its transmission moves with any
small change in energy.

They are excluded through a new _EXCLUDE_FROM_SEARCH rather than by
clearing their 'enabled' flag, because those two things are not the same
and the flag is load-bearing elsewhere: _fil_trans_report substitutes
T=1.0 for a disabled position, so clearing the flag would have made the
report silently ignore a foil parked in the beam -- precisely the state
an edge scan puts it in. The foils stay enabled, stay addressable via
fil_comb as they were in spec, and are skipped only by the automatic
search.

At 17 keV fil_trans(0.05) now returns 2146 (Si345.6 + Ti200 + Ti20,
T=0.0494) and fil_trans(0.01) returns 2234 (T=0.0099), both matching an
independent calculation over the same CXRO tables.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017kLGTqTyXzTT4swt3CfTkV
This commit is contained in:
menzelandClaude Opus 5 authored and x12sa committed 2026-08-30 13:50:50 +02:00
1 parent f2db469c9b
commit 5c2c0770a9
1 file changed
+39 -14
@@ -74,19 +74,30 @@ class cSAXSFilterTransmission:
# -----------------------------------------
# Exposure-box filter configuration (4 x 6)
#
# Current hardware (per your message):
# Unit 1 (filter_array_1_x): out, Si400, Ge300, Ti800, Zr20
# Unit 2 (filter_array_2_x): out, Si200, Si3200, Ti400, Cu20
# Unit 3 (filter_array_3_x): out, Si100, Si1600, Ti200, Ti3200, Fe5
# Unit 4 (filter_array_4_x): out, Si50, Si800, Ti100, Ti1600, Ti20
# Thicknesses are the values from filter.mac rev 1.19, i.e. effective
# thicknesses measured at 18.58 keV rather than nominal foil thicknesses.
# They are used at every energy, which assumes the tabulated attenuation
# lengths are right and only the thickness was uncertain -- the same
# assumption spec made. The odd Si values are measurements, not typos.
#
# Unit 1 (filter_array_1_x): out, Si345.6, Ge300, Ti800, Zr20
# Unit 2 (filter_array_2_x): out, Si234.6, Si3303, Ti400, Cu20
# Unit 3 (filter_array_3_x): out, Si137.5, Si1661.5, Ti200, Ti3200, Fe5
# Unit 4 (filter_array_4_x): out, Si50, Si833.8, Ti100, Ti1600, Ti20
#
# Positions 1..6 = [out, m1, m2, m3, m4, m5]
# Each entry: ((mat1, th1_um), (mat2, th2_um), enabled_bool)
#
# 'enabled' means the position exists and its transmission can be computed.
# Keeping the Zr/Cu calibration foils enabled is deliberate: they are real
# filters, so _fil_trans_report must account for them when one is parked in
# the beam for an edge scan. Excluding them from automatic selection is a
# separate concern -- see _EXCLUDE_FROM_SEARCH.
# -----------------------------------------
_FILTERS: List[Tuple[Tuple[str, float], Tuple[str, float], bool]] = [
# Unit 1
(("none", 0.0), ("none", 0.0), True), # out
(("si", 400.0), ("none", 0.0), True), # Si400
(("si", 345.6), ("none", 0.0), True), # Si345.6
(("ge", 300.0), ("none", 0.0), True), # Ge300
(("ti", 800.0), ("none", 0.0), True), # Ti800
(("zr", 20.0), ("none", 0.0), True), # Zr20
@@ -94,16 +105,16 @@ class cSAXSFilterTransmission:
# Unit 2
(("none", 0.0), ("none", 0.0), True), # out
(("si", 200.0), ("none", 0.0), True), # Si200
(("si", 3200.0), ("none", 0.0), True), # Si3200
(("si", 234.6), ("none", 0.0), True), # Si234.6
(("si", 3303.0), ("none", 0.0), True), # Si3303
(("ti", 400.0), ("none", 0.0), True), # Ti400
(("cu", 20.0), ("none", 0.0), True), # Cu20
(("none", 0.0), ("none", 0.0), False), # unused
# Unit 3
(("none", 0.0), ("none", 0.0), True), # out
(("si", 100.0), ("none", 0.0), True), # Si100
(("si", 1600.0), ("none", 0.0), True), # Si1600
(("si", 137.5), ("none", 0.0), True), # Si137.5
(("si", 1661.5), ("none", 0.0), True), # Si1661.5
(("ti", 200.0), ("none", 0.0), True), # Ti200
(("ti", 3200.0), ("none", 0.0), True), # Ti3200
(("fe", 5.0), ("none", 0.0), False), # Fe5 (disabled unless data file provided)
@@ -111,12 +122,20 @@ class cSAXSFilterTransmission:
# Unit 4
(("none", 0.0), ("none", 0.0), True), # out
(("si", 50.0), ("none", 0.0), True), # Si50
(("si", 800.0), ("none", 0.0), True), # Si800
(("si", 833.8), ("none", 0.0), True), # Si833.8
(("ti", 100.0), ("none", 0.0), True), # Ti100
(("ti", 1600.0), ("none", 0.0), True), # Ti1600
(("ti", 20.0), ("none", 0.0), True), # Ti20
]
# Calibration foils, not attenuators. filter.mac flagged these "disabled" so
# that selection by transmission would not pick them, while leaving them
# directly addressable via fil_comb -- they are exactly what you park in the
# beam for a K-edge scan. Zr in particular has its edge at 18.0 keV, so near
# that energy its transmission moves with any small energy change.
# (unit, position), both 1-based.
_EXCLUDE_FROM_SEARCH = frozenset({(1, 5), (2, 5)}) # Zr20, Cu20
_UNITS = 4
_PER_UNIT = 6
@@ -520,10 +539,16 @@ class cSAXSFilterTransmission:
for u in range(self._UNITS)
]
# Precompute per-position transmissions
# Precompute per-position transmissions. Positions excluded from the
# search are treated exactly like disabled ones here, i.e. skipped.
per_pos_T: List[List[Optional[float]]] = [
[self._position_transmission(pos_entry, energy_kev) for pos_entry in unit]
for unit in units
[
None
if (u + 1, i + 1) in self._EXCLUDE_FROM_SEARCH
else self._position_transmission(pos_entry, energy_kev)
for i, pos_entry in enumerate(unit)
]
for u, unit in enumerate(units)
]
combos: List[dict] = []