The previous commit called BE12599 a second failure mode and claimed the
device "never enters S3 mode at all" and that no inbound work could reach it.
That was an overclaim built on a single slow_drip attempt, and Brian was right
to push back.
It is the same disease. Method B's step 1 worked fine on BE12599 — clearing
the Destination did stop the dial-outs. It was step 2 that did not land, on
one attempt, run ~90 s after a modem reboot with a dead session visible in the
log in that same window; BE9558H needed hours of attempts before one landed.
And the AT-init loop the ALEOS log revealed is almost certainly what BE9558H
was doing too — we just never turned on serial debug in May to look. The
device speaks S3 fine; it handshook cleanly the moment it had a session.
What is genuinely new is the cure, and it deserves to be the default rather
than a footnote. Racing a Stop into the gaps between dial-outs is a coin
flip. Intercepting is deterministic: the unit dials every ~75 s, so give it
somewhere to dial and answer it. It will not answer us because it is on the
phone — so be the one it calls.
Restructures accordingly: a "two cures" table up top, the intercept promoted
to Method A with its own procedure (listener before modem, stop at step 1.5,
drain before disabling ACH, restore the Destination and confirm it), and the
original inbound procedure kept intact as Method B for when there is no
listener the modem can reach.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qcu9ByJfuKBQxmrWb8rSrN
The wedged_unit_recovery runbook covered exactly one failure mode. BE12599
turned out to be a second one wearing the same symptoms, and the existing
procedure did not work on it.
Adds a "TWO failure modes" table up front so the next incident branches
correctly, and a full second-incident section covering what the ALEOS serial
debug log revealed: the device repeating a 29-byte AT modem-init string
(ATQ1/ATE0/ATS0=2, no ATD) every 75 s, never getting an OK because the modem
is in TCP data mode, and therefore never entering S3 mode at all. Inbound
cannot win against that, no matter how well framed.
Also records the two red herrings, since together they cost ~90 minutes:
the RV50 trusted-IP whitelist drops non-listed sources silently (presents as
a connect timeout, and Brian's dynamic dev IP had rotated off the list), and
sfm/server.py returns 502 for BOTH "Protocol error:" and "Connection error:",
so a 502 was misread as "TCP connected, device mute" and a theory built on it.
And the gotchas worth never re-deriving: slow_drip's send_error=null plus a
full duration is not success (only bytes_received > 0 is); stopping monitoring
removes the call-in trigger, so it costs you the channel; --events-only skips
the device-info step, so the serial is never read and ach_state keys on
peer:ephemeral_port, silently breaking dedup and re-downloading the same event
every session.
The plan doc captures the tool Brian wants built out of this — a rescue
listener with a real lifecycle and, critically, a confirmation gate before
shutdown, because leaving the modem's Destination pointed at a dead listener
is worse than never having started. Open questions are listed rather than
guessed at.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qcu9ByJfuKBQxmrWb8rSrN
sfm/compliance.py renders the velocity-vs-frequency blasting compliance chart
Blastware draws on its Event Report:
- limit_at()/limit_curve() — the RI8507 Fig B-1 / 30 CFR 816.67 curve as data
(Drywall 0.75 + plaster 0.50 lines): 0.030in low-freq bound, plateau, 0.008in
rising diagonal to a 2.0 in/s cap at ~40 Hz, drawn continuous.
- channel_compliance_points() — the per-cycle (freq, peak-velocity) scatter by
the zero-crossing method (matches Blastware; cloud ceiling = channel PPV).
- draw_compliance_chart() — matplotlib rendering (both lines + scatter, BW tick
scales + channel markers).
Verified against 7 BE12844 Blastware reports. docs/ri8507_compliance_curve.md
captures the curve construction, the SHM basis, and the scatter method.
Not yet wired into report_pdf.py — that placeholder is the next step.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YDXjZCr4RqT2U3QvMDhgzf
data_block_len() rejected any `40 NN` block with NN > 0x08. That guard had
no evidence behind it: every corpus available when it was written used only
NN in {1,2,3,4,8}, so it was never exercised. Loud UM12947 events use NN of
12, 16, 20 ... up to 196.
Because walk_body/run stop at the first unrecognised tag rather than
raising, rejecting those blocks surfaced as silently short channels -- e.g.
Tran 1812 / Vert 2132 / Long 2324 on a file whose export carries 2324 for
all three. The real bound is the buffer; the caller additionally clamps to
the record end.
Verified against Thor's own CSV exports for UM12947 (2025-07-14 .. 09-25,
167 waveforms, supplied as CSV.zip):
length mismatches 22 -> 0
per-sample exact 1,476,242 / 1,476,249
These are NOT truncated recordings, which was the competing hypothesis --
the exports carry the full sample count.
tests/test_waveform_codec.py asserted the cap as intended behaviour. That
assertion encoded an assumption, not a verified fact, and is replaced with
one pinning the opposite plus the evidence.
Across all three ground-truth corpora: 459 waveform files,
3,807,158 / 3,807,165 samples exact. Production IDFW is now 575/575 with
zero truncations and zero decode failures (median PPV error -0.0007% across
8 units). Series-3 re-verified unchanged at 14,338/14,338.
The 7 residual samples each differ by one 4th-decimal tick and are Thor's
own rounding: intersecting the per-sample rounding constraints over that
corpus is infeasible (binding pair contradict by 2.3e-11, 7e-5 relative),
so no single linear LSB reproduces every printed value. _GEO_LSB_IPS is
already pinned to ~1e-11; do not retune it to chase these.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ru8Lg9HkkYvX9VWWo65SmL
Verified against a second Thor corpus (9-10-26-csv-req: UM11402, UM12947,
UM20147) with per-sample CSV exports: 139/139 waveforms exact
(1,273,380/1,273,380 samples) and 877/877 histograms within 2% of Thor's
reported PPV -- up from 66.9% and 56.6%.
Some units run with the microphone disabled, which changes two structural
things that were both hardcoded to the 4-channel shape:
- Waveform body head sat below the scan floor. A 3-channel unit has a
shorter fixed header and puts its record chain head at 0x0dba, under the
old _BODY_SCAN_FLOOR of 0x0E00. The scan could not see it and fell
through to the Vert segment-0 record, decoding a body shifted one
position around the channel rotation -- Vert came up exactly 512 samples
short. Floor lowered to 0x0C00. The body-offset scoring also had to stop
requiring four channels, or `equal` is permanently False for these events
and the pick falls back to raw sample count.
- Histogram interval record is 56 bytes, not 72. It is
16 * n_channels + 8, and is not inferable from the segment length alone.
The interval count now comes from the segment's cumulative counter
(n = counter - prev_counter) and the stride is derived from it. Assuming
72 read 7 intervals out of every 10-interval segment, then walked off
alignment into garbage that decoded as ~10 in/s peaks -- inflating some
files' PPV by up to 191,000%. Also recovers 4 files that previously
decoded no intervals at all.
Combined across both corpora: 292/292 waveform files,
2,330,916/2,330,916 samples exact. Production IDFW truncations 41 -> 22.
Series-3 unaffected (no shared-codec change in this commit; last full run
14,338/14,338).
Known open, diagnosed but NOT verified: the remaining 22 unequal + 1 failing
production IDFW files (all UM12947, 2025-07-14..09-23) stop the block walker
on tag 40 0c. data_block_len() caps the 40 NN int16 block at NN > 0x08 while
those files use NN up to 196. Both verified corpora only ever use
NN in {1,2,3,4,8}, so the cap is untested there and lifting it leaves both at
100.000% -- which is not evidence it decodes these correctly. Deliberately
not shipped; needs Thor CSV exports for UM12947 in that date range.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ru8Lg9HkkYvX9VWWo65SmL
Verified against Thor's own CSV exports, which carry a per-sample
four-column block beside every binary (CSV/<name>.IDFW.csv). Those 1,012
paired files were in the corpus all along; the decoder had been pinned to
a superseded walker on the stated grounds that "Thor has no ASCII ground
truth in the corpus and its geo scaling is separately suspect". Both
premises were false.
IDFW per-sample exact 39.1% -> 100.000% (1,057,536/1,057,536)
IDFW files fully exact 0/153 -> 153/153
IDFW PPV median error -3.32% -> -0.002%
IDFH within 2% of Thor PPV 51.1% -> 100.0% (858/858)
prod IDFW, 8 units -3.3% -> -0.001%
Four independent root causes:
- Geo LSB was 0.0003, the 4-dp *display rounding* of the real
0.000310308 mistaken for the LSB, so every series-4 geophone sample
read 3.3% low. Pinned to +-6e-11 by intersecting 991,415 rounding
constraints; corroborated by the +-full-scale seed (+-32226) left in
unwritten IDFH slots. IDFH had a separate, also wrong, 10.0/32768.
- IDFH histograms were capped at 250 intervals: the segment validator
required the interval counter's high byte to be zero, but the counter
is a uint16 cumulative index, so every segment past interval 255 was
rejected. Runs over ~4 hours lost their tail, often the peak.
540/858 corpus files affected.
- Record mode 00 00 (raw int16, 10-byte header) was unhandled and fell
through the dispatch, silently dropping each channel's first 512
samples -- the long-standing "loud events truncate" symptom.
MODE_ABSOLUTE is now also accepted as a segment-0 preamble.
- The body-offset search matched 00 02 00 *inside* record headers,
selecting a candidate part-way down the chain and decoding a
rotation-shifted body. It now anchors on record headers and takes the
chain head (6 ms/file).
Also fixes the separately tracked "UM-series decodes ~1000x low" bug.
Series-3 re-verified unchanged at 14,338/14,338 exact after the shared
waveform_codec change.
Known open: 41/575 prod IDFW files (7%, mostly UM12947/UM20147) decode
with unequal channel lengths and also fail metadata extraction -- a
different header variant with no Thor export in the store.
NOTE: this is a codec change; the Thor store owes a regeneration via
scripts/backfill_thor_events.py.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ru8Lg9HkkYvX9VWWo65SmL
Brian noticed BE12599's 2026-08-09 event reports no ZC frequency because the
trace never crosses zero. That is the best detector in this investigation.
A geophone has no DC response, so its output must integrate to ~zero over a
record. |mean|/peak is therefore ~0 for real motion and ~1 for anything
electrical. Across 12,068 channel-events with peak >= 0.05 in/s the statistic
is bimodal with a 1.09% dead zone, and at mp >= 0.8 it returns exactly the five
confirmed units -- from physics rather than a tuned threshold. Two detectors on
different principles agreeing is the strongest corroboration the list has had.
It also settles BE11007 as NOT an offset: mp 0.75-0.89 but frac_neg 0.99 at
peaks of 7.4-9.4 in/s, i.e. a one-sided near-full-scale blast.
Journal 8e diagnoses BE12599 specifically. Its August waveforms are unipolar
impulses with an RC tail (26 ms -> 118 ms -> never recovers over 14 days), and
the fault MOVES between Long and Tran while the sensor self-check passes on
every event. A failing element cannot hop channels; a connector can -- which
also explains why the swing test never fails and why an autozero rarely helps.
Corrects 8c's claim that the spread gate is blind to onsets: of 87 BE18438|Vert
events it rejected one, the transitional record. Narrower than stated.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Records the mechanism investigation in journal 8c. The headline: still unknown,
but the shape is now constrained and a long list of dead ends is closed.
Onset is a ramp of minutes-to-hours, not a step — BE18438 Vert resolved to
one-minute cadence via the histogram corpus, 50% of the excursion in 7 minutes,
>=25 intermediates, validated 75/75 against Blastware's own ASCII. That kills
both poles of the original dichotomy: not a latched digital step, not slow
component wear. What survives is a reversible two-time-constant settling
process, which is a shape constraint and not a mechanism.
Thermal, ground-motion shock, handling/redeployment, accumulated duty, age,
firmware and a mechanical element fault are each refuted or explicitly bounded,
with the power behind every negative stated.
Retracts two claims this journal carried: polarity consistency was a tautology
of offset_scan3's spread gate, and the fleet is 8-9 units rather than 5 once
that gate is dropped. Also notes the gate is blind to onsets by construction --
it rejects a moving floor, and it rejected the one record where the ramp shows.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
BA9229, BA10060, BA10895 and BA15957 were reported throughout as BE — the
scanners synthesised the family prefix, which the BW filename does not carry.
Corrected across the journal with a note recording why, so the mistake is
legible rather than silently patched.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
offset_scan3.py covers only waveforms (6,577 unique binaries). The archive
also holds 63,535 unique histograms, which the pre-trigger method cannot
touch: a histogram carries no samples, only a per-interval per-channel peak.
scratch/offset_hist_scan.py scans them — 63,505/63,535 decoded (99.95%),
43 units, 77.9M intervals. It emits every candidate floor statistic per
(file, channel) rather than deciding anything, so thresholds get calibrated
against the waveform ground truth instead of guessed.
Journal §8b records the outcome. What survives is a site-quiet-gated
cross-channel differential that independently confirms BE18438|Vert and
BE9558|Tran+Long with a clean 2.5x separation gap and 0.037% day-level false
alarm, threshold-insensitive across a 2.3x span — the first operating point
in this investigation to pass that test cleanly.
What it does not do, recorded just as plainly: it finds 2 of the 5 confirmed
units, not 5. DC leakage into the interval peak is bimodal (0.9 on BE18438,
0.02 on BE12599), so a negative histogram result is not evidence of health.
Per-channel attribution is not established (channel-scramble p = 0.769) and
timing resolves to ~a month, not a day.
Two dead ends buried for good: the absolute floor is retired (66% of its
discrimination is a day/site confound), and zero-fraction is structurally
impossible — the device clamps every interval peak at >= 1 A/D count.
Two findings independent of the histograms:
- offset_scan3's spread<=0.02 gate discards 18.8% of rows with |pre|>=0.025,
concentrated on 41 unit-channels currently labelled clean; 4 would be
sustained positives without it. The fleet label is three-state, not two.
- The waveform corpus observes ~7% of the days a unit was deployed.
BE10895 is reclassified from transient to a genuine Vert fault of a different
subtype: 49.4% single-axis-dominant events, the highest in the fleet, all on
Vert. The other six marginal units are clean.
Not done: the 11 thin-coverage units were not screened, and no completeness
audit was run.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Brian's method, and better than v2's whole-record median: the pre-trigger
window is definitionally quiet (the buffer captured before the trigger fired),
whereas a median is merely robust to the event. Requiring the floor to hold
across pre-trigger / middle / end rejects transients that a median cannot.
per channel: pre/mid/end medians, spread = max - min
offset when |pre| >= floor AND spread <= 0.02 in/s
real fault >= 3 consecutive flagged events on that channel
The empirical noise floor justifies the threshold and validates the decoder:
across 19,244 non-flagged channel-events the pre-trigger floor is 62.7% exactly
0.000, 94.5% within +/-1 quantisation unit, median +0.0000, mean -0.0008. There
is no systematic zero-point bias — an independent confirmation of the
32000-count geo scale.
The result is threshold-insensitive across a 2x range (0.020 to 0.040 in/s),
which is what separates a real signal from a tuned one:
FINAL: 5 of 45 units (11%) — BE9558, BE11529, BE12599, BE13117, BE18438
BE11007 and BE10895 drop out; the spread test identifies them as transients
rather than pedestals. v1's 11% headline was right by luck — it included
BE11007 and named the wrong channel on most units.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Brian challenged the v1 finding that offsets "come and go", against field
experience that a unit which develops one stays broken until the geophone is
replaced. He was right; v1 had two flaws, both of which manufactured false
recoveries:
1. It scored only the axis with the largest peak, so a real event on one axis
hid a persistent pedestal on another. BE12599 on 2026-08-21 read "clean"
because Long had a 1.065 in/s event, while Tran sat at +0.4732 in/s and was
never examined.
2. It used the mean, which a real transient perturbs. The median is the resting
baseline and a blast does not move it. Same event, Long channel:
mean +0.0783 vs median -0.0050.
offset_scan2.py flags a CHANNEL when |median| >= 0.025 in/s (5 A/D counts,
Instantel's own criterion) and treats >=3 consecutive flagged events as the
real signal. No m/p ratio guard is needed — that existed only to compensate for
the mean.
Corrected results:
units with any flagged event 6 -> 19 of 45
units with a sustained pedestal 8 of 45 (18%)
runs >=3 consecutive 29; 1-2 event runs (noise) 69
Also corrected: the affected channel is most often Vert, not Tran (v1 named
whichever axis had the largest peak, so it was frequently wrong). BE10895 and
BE18003 were invisible to v1. BE12599's fault began 2026-08-14, not 08-17.
The decode itself was never in question and is confirmed against Blastware's
own ASCII export: on K558LJN3.BK0W, BW shows Tran parked at +0.265..+0.375 in/s
for the entire record while Vert and Long sit at ~0.005 — Instantel's "parallel
lines above or below the zero line".
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Adds docs/offset_investigation.md — a dated journal of the "offset" hardware
fault, in the style of the codec status docs: findings with provenance, dead
ends kept with the reason they died, and per-unit case files.
Contents:
- base rate 5-6 of 45 units (11-13%) over the DL2 archive, 2018-2026,
confirming rather than overturning the earlier 2-of-21 estimate
- the detector, with the rationale for each term and its known blind spot
(event traces carry real motion, so only trace-dominating offsets show)
- the bimodality result: relaxing the amplitude floor 11x adds no new units
- Instantel's own procedure and thresholds from their FAQs 13-0-21 / 12-0-10:
A/D-mode ">5 counts", the autozero key sequence, and the 2027-2069 X1/X8
acceptance window that explains the ~10% field success rate of a re-zero
- four ruled-out hypotheses, each with the evidence that killed it:
condensation, clipping, the sensor check as a predictor (102 offset events,
zero failures — a grossly offset unit passes its own self-check), and the
calibration-timing correlation (confounded, one unit per time bucket)
- open questions, chiefly whether SUB 0x0E carries the autozero numbers
Cross-referenced from CLAUDE.md and Appendix E of the protocol reference
(whose CRLF line endings are preserved).
Separately: CLAUDE.md had 793 NUL bytes appended after its last line. They
predate this work (present at least as far back as e42956a / v0.21.0) and made
grep treat the file as binary, silently skipping it. Stripped.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Sub-minute histogram intervals are packed several to a block so that
every block still covers exactly one minute of data:
interval intervals/block stride
1 minute 1 32 <- the standard big-endian block
15 s 4 92
2 s 30 612
stride = 12 + n * 20
Block = [00][segment][ctr uint16 LE][0a][00], then n x 20-byte records of
8 x uint16 LITTLE-endian values (T_peak, T_halfp, V_peak, V_halfp,
L_peak, L_halfp, M_peak, M_halfp) plus a 2-word tail whose first word is
0000 on every real interval, then a 6-byte block trailer.
The standard 32-byte block is BIG-endian; this variant is LITTLE-endian.
The tail-word check matters: a session ending mid-block leaves buffer
garbage in the remaining interval slots, which decoded as peaks
thousands of times the real value. Stride detection also requires at
least 2 records, since a 1-record block would have stride 32 and
collide with the standard block.
Recovers 415 files that decoded to nothing: 216 on BE18193 (2 s
intervals) and 199 on BE9440 (15 s). Before decoding to nothing they
were being accepted by the WAVEFORM codec, which returned garbage
peaking up to 400x the device-reported PPV.
Ground truth BE9440/K440L3AQ.T70H (5,710 intervals) matches its
Blastware ASCII export exactly: 17,130/17,130 geo peaks, 22,840/22,840
frequencies, 5,710/5,710 mic dB(L). Across all 455 affected files,
1,354/1,365 channel peaks (99.2%) match the device-reported PPV; the 11
that don't are under-reads on BE9440 where the walk stops early.
Fixture (binary + ASCII) saved under tests/fixtures/, which is
gitignored per repo practice — the ground-truth test skips when absent.
Tests: 258 passed, failure list unchanged from baseline.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
backfill_sidecars.py skipped the .h5 write when a file produced no
samples, with the stated intent of not replacing it with an empty
placeholder. That silently preserved output from a superseded decoder.
After the record-chain fix, 415 histogram files stopped decoding (216 on
BE18193, 199 on BE9440) but kept .h5 files whose peaks ran up to 400x
the device-reported PPV. Those were feeding charts and the
false-trigger detector with nothing marking them. The .h5 is now
removed in that case and the run reports stale_h5_removed.
Store-wide effect, series-3, decoded peak vs device-reported PPV:
waveform 1307/1307 (100%), mean abs ratio error 0.00000
histogram 4434/4435 (100%)
Both were 99% with a tail of 18 and 25 wrong files respectively.
The 415 files are a genuine unmapped format variant, not a regression:
their bodies open `00 00 00 01 0a 00` (valid block header, marker 0a at
[4], block_ctr 256) but block[28:32] matches neither known tail, and no
stride from 8 to 64 bytes places a marker at [4] consistently. Bodies
are very large (one is 360,573 bytes). They were previously being
decoded by the WAVEFORM codec, which accepted them and returned garbage
- so the gap pre-dates today's work; the fix only exposed it. Logged as
an open question in the protocol reference.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Supersedes the segment-header model entirely, including the fixes made
earlier today. Found via multi-agent structural analysis of the 25 files
that stalled the walker, then verified independently.
Records are self-delimiting: off+2 is a uint16 BE length, next_record =
off + 2 + len, and the chain ends on a record whose chan_id is 0x06.
off+8 carries a 3-valued mode enum:
02 00 14-byte header, 2 anchors, then CUMULATIVE delta blocks
01 00 10-byte header, no anchors, blocks are ABSOLUTE values
00 03 10-byte header, NO TAGS AT ALL - raw 12-bit packed absolute
`40 NN` is an ordinary int16 BE data block (2*NN + 2), never a header.
Reading it as a 2*NN + 16 header is what made walks drift — the
"variable-prefix segment descriptors" reported earlier today were not a
format feature, just walker drift of exactly
4 - (old_stop - true_record_start), on all 25 affected files.
Measured on the production snapshot:
all four channels equal length 156/1388 -> 1388/1388
ASCII sample-count exact 72/75 -> 75/75
ASCII fully exact 70/75 -> 73/75
device PPV waveform (live) 1288/1306 -> 1306/1306 (mean err 0.00000)
device PPV histogram (live) 4434/4459 -> 4458/4459
Also eliminates the walker-over-read class: 24 of those 35 files were
histograms that read_blastware_file fed to the waveform codec first; the
old walker accepted them and returned garbage (one yielded 98,923
"intervals"), while the record-chain decoder returns None so they fall
through to histogram_codec.
00 03 records are DECODED, not skipped. Skipping them silently shifts
the time base of everything after them on that channel — BE9558/
K558LOF2.820W had MicL displaced by exactly 512 samples with nothing
marking the gap.
Footer detection now prefers the 0e 08 candidate whose body yields a
chain terminating on 0x06; the signature can occur inside a sample
stream. Blast radius 1 file of 1388.
The superseded model survives as decode_waveform_legacy, pinned by
micromate/idf_file.py: its Thor IDFW body-offset search trial-decodes
candidates and keeps whichever yields the most samples, so the new
decoder returning None where the old returned garbage changes that
heuristic's winner. Deferred until that search uses the record chain.
Tests: 253 passed (+11), failure list unchanged from baseline. The 9
tests pinning the superseded model are retargeted at
decode_waveform_legacy, which still implements it.
NOTE: stored .h5 files need regenerating — nearly all get longer.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Two errors in the series-3 histogram block model, both found by diffing
against the per-interval data table in the preserved Blastware ASCII
exports (1211 files in the prod snapshot — far stronger ground truth
than the header PPV used previously).
1. The block is uniformly BIG-ENDIAN. Peaks and half-periods are uint16
BE (T_peak [5:7], T_halfperiod [7:9], V_peak [9:11], V_halfperiod
[11:13], L_peak [13:15], L_halfperiod [15:17], M_peak [17:19],
M_halfperiod [19:21]); only block_ctr [2:4] is little-endian.
The old uint8-peak model silently CLIPPED any peak above 1.275 in/s:
the final interval of BE18193/T193LQ9K.OE0H reads 8.270 in/s in BW's
export (1654 counts = 0x0676) and decoded as 0x76 = 118 = 0.590.
The byte documented as a per-channel "annotation" was never an
annotation — it is the half-period's high byte, which is exactly why
it was non-zero on the sub-Hz intervals BW renders as "<1.0".
The marker is block[4] alone. Testing [4:6] as a uint16 LE marker
forced block[5] == 0, which is what capped the peak at one byte.
2. The final block of each stream carries tail 9c 06 00 42 instead of
1e 0a 00 00, and holds arbitrary bytes at [21:23]. Rejecting it
dropped the last interval of nearly every histogram — frequently the
interval holding the event peak, so the file's PPV read low.
Verified end to end through the production path: 1211/1211 histograms
decode exactly (interval count + every per-interval peak), plus 842,442
per-interval frequency comparisons with zero mismatches. Previously
1 of 1196 files was fully correct.
decode_histogram_body_full records expose `is_terminal` in place of the
removed `annotations` tuple. +6 tests. No regressions: full-suite
failure list unchanged from baseline.
NOTE: stored histogram .h5 files need regenerating to pick this up.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Re-measured properly. The first pass compared h5 max against the ASCII
header PPV and reported "26% of channels miss the peak". The histogram
ASCII actually carries a full per-interval data table (Tran/Vert/Long
peak + freq + PVS per interval), which is real ground truth, so the
comparison should have been per-interval from the start.
Per-interval result, n=1196 series-3 histograms:
- decoded VALUES are right: 1031/1196 (86%) match within 1 LSB across
the overlapping prefix
- the interval COUNT is short in 1195 of 1196 files: median 1 missing,
1088 short by 1-2, 65 by 3-10, 39 by 11-100, 3 by >100 (max 205)
- decoded max falls below the device PPV in 169/1196 files (14%), not
26% — that happens when a dropped interval held the peak
So it is a termination bug in histogram_codec.decode_histogram_body,
the same family as the waveform-walker truncation fixed earlier today,
rather than mis-decoded interval values. Series-3 only; there is no
preserved series-4 ASCII in the snapshot to compare against.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Operator report: attaching a different geophone to an affected unit
makes the offset go away. That rules out the unit's analog front-end
and any stored per-channel zero constant (a constant lives in the unit
and would survive a sensor swap).
The stored data agrees — MicL, a separate transducer on its own cable,
shows no offset during either episode (|mean|/peak 0.17 and 0.02) while
the geo channels on the same unit at the same moment are pinned.
Two distinct sensor-side patterns recorded:
BE18438 Vert 0.97, Tran 0.16, Long 0.18 -> one conductor pair
BE9558 Long 0.99, Tran 0.90, Vert 0.81 -> shared return / ground
Candidate mechanisms narrowed to three, since a geophone coil is passive
and cannot generate sustained DC: galvanic corrosion at a connector or
splice (matches the ~46 mV referred to the ADC input), a leakage path to
shield, or changed coil DC resistance interacting with the amplifier's
input bias current.
Also records the confound: swapping a sensor requires a monitoring
restart, and these units run Sensor Check "Before monitoring", so the
restart re-zeros too. The swap does not cleanly separate "new sensor"
from "the restart re-zeroed it". Controls and the single best
measurement (open-circuit DC across the suspect connector) documented.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Brings the protocol reference, CLAUDE.md and the codec RE status doc up
to date with everything confirmed in this pass.
instantel_protocol_reference.md
- Changelog row for the five findings.
- S7.6.1: scope table showing the 32000 scale correction applies to
series-3 waveform, series-3 histogram and series-4 Thor alike, with
the measured before/after ratios for each.
- S15: closed "Full channel ID mapping in SUB 5A stream" — resolved by
the segment-header channel id ([channel][00][00][segment], 0x46=Tran
0x47=Vert 0x48=Long 0x49=MicL, 1697/1697 verified). Four new open
questions: variable-prefix segment descriptors, the histogram codec
missing peak intervals (26% of channels), UM-series IDF decoding
~1000x low, and the Thor per-count LSB residual.
- NEW Appendix E — Known Device Faults. Documents the field-observed
"offset" fault: symptom, why it floods the ACH queue (pedestal
exceeds the unit's own geo trigger level), the episode table, the
detection rule that works, what the data rules out (not the
geophone, not the battery, not environmental, not condensation),
and the two remaining candidate mechanisms with the test that
separates them. Explicitly flags that it is NOT a decode artifact,
since that mistake has already been made once.
CLAUDE.md
- Body-codec section: the four framing cases and the channel-id
finding, with the corpus result.
- "What's NOT solved": replaced the stale walker-edge-cases bullet
with the four genuinely open items.
waveform_codec_re_status.md
- Scale scope table matching the protocol reference.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
Two independent bugs, both found by diffing 75 production events against
their preserved Blastware ASCII exports (<store>/<serial>/<file>_ASCII.TXT).
1. Geo full scale was wrong — every geophone reading was 2.34% low.
The codec emits geo samples in 16-count units with a documented LSB of
exactly 0.005 in/s, and decoded_to_adc_counts multiplies by 16, so one
ADC count is 0.005/16 in/s and 10.000 in/s is 10.0/(0.005/16) = 32000
counts. sfm/event_hdf5.py and minimateplus/event_file_io.py both
divided by 32768 (2^15), scaling every sample and derived peak down by
1 - 32000/32768. The error scales with amplitude, so it was invisible
on quiet events and worst on the loud ones that matter for compliance.
Mic is unaffected (it back-solves its scale from the device peak).
216 per-channel comparisons: 32768 -> 151/216 exact, worst error 0.238
in/s on a 10 in/s event; 32000 -> 216/216 exact, worst 0.005 = 1 LSB.
2. walk_body silently truncated channels on four unhandled framing cases.
An unrecognised tag ends the walk and decode_waveform_v2 returns
whatever it got, so this surfaced as short channels, never an error:
- wide-NN RLE `0X NN` (runs longer than 252 samples)
- `30 NN` with NN > 0x10 (the old cap was arbitrary)
- variable-width `40 NN` headers: NN counts previous-channel
continuation deltas, so the header is 2*NN + 16 bytes; `40 01`
and `40 03` occur alongside `40 02`
- tagless segment headers: no `40 NN` tag at all, just the 14-byte
tail [field2:2][len:2][channel_id:4][marker:2][anchors:4]
Also: the header field documented as a "monotonic uint32 LE counter" is
really [channel_id][00][00][segment_index], with 0x46=Tran 0x47=Vert
0x48=Long 0x49=MicL — verified on 1697/1697 segment headers, zero
disagreements. decode_waveform_v2 now takes the channel from that field
instead of rotation position, which was fragile: one missed header
desynced every channel after it.
parse_segment_header now returns n_prev_deltas/prev_deltas/marker/
anchors/channel/segment_index; the old fixed_pattern (02 00 00 01)
conflated the 2-byte marker with the first anchor.
Ground-truth corpus, end to end through the production path:
exact 37 -> 72, truncated 23 -> 3, full-length value errors 15 -> 0.
Store-wide, 729 of 1388 series-3 waveform events decode differently and
728 gain samples; the scale fix changes float values on all of them, so
stored .h5 files need regenerating.
Still open: 3 events truncate at a header variant with a variable-width
prefix (2/4/6 bytes) before the channel id and an `01 00` marker.
Documented in docs/instantel_protocol_reference.md with byte offsets.
+20 tests. No regressions: the byte-exact fixture suite still passes and
the full-suite failure list is unchanged from baseline (16 pre-existing
failures from gitignored fixtures).
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HgTe8CamXAHcAmaQ6QNcog
7-task TDD plan: shape DSP module (crest factor + points-near-peak), shape_*
columns + auto-migrate, insert_events persistence, ingest population in the
save paths, backfill script, /db/events exposure + v0.24.0 bump. Phase B
(terra-view scoring/UI/review) gets its own plan once this feed is live.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YDXjZCr4RqT2U3QvMDhgzf
the BE9558 / BE18003 extension-byte case
The bytes at [7]/[11]/[15]/[19] are an annotation field (purpose still
unclear — empirically non-zero on intervals with sub-Hz or unmeasurable
freq), NOT the high byte of the peak count. The N844 fixture corpus
the original RE was done against had zero values in those bytes for
every block, so uint8 and uint16 LE were equivalent there — but on
real BE9558 Tran-drift events and BE18003 Histogram+Continuous events
the uint16 LE interpretation produced peaks up to 268 in/s and 35×
inflated PVS sums.
Cross-correlated against BW's per-interval ASCII export on:
- K558LKZU/LL1P/LL3K → 100% T/V/L/M peak match (1435 blocks each)
- T003LKZR/LL0O/LL1M → 100% T/V/L, 99.3% M (0.05 dB rounding only)
- N599LKZS/LL0L → 100% all channels
- N844 fixture corpus → 100% all channels (unchanged)
Annotations preserved on every record for future RE; the defensive
_MAX_PEAK_COUNT bound is no longer needed (uint8 maxes at 1.275 in/s,
well below any physical limit).
Synthetic regression test added using the verbatim K558LKZU.RE0H
interval-12 block.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The histogram-mode event body is now byte-exact decodable.
Companion to the waveform body codec — together they cover every
event file the watcher forwards. Cracked in one session via
cross-event correlation against BW's ASCII export.
The §7.6.2 spec in instantel_protocol_reference.md was structurally
correct (32-byte blocks) but the per-sample semantics were
under-documented. Cross-checking block 130 of N844L6Z8.ZR0H
against its TXT row revealed the layout perfectly:
slot[0] = 10 (constant marker)
slot[1] = T_peak_count (× 0.005 → in/s at Normal range)
slot[2] = T_halfperiod (freq_Hz = 512 / halfp)
slot[3] = V_peak_count
slot[4] = V_halfperiod
slot[5] = L_peak_count
slot[6] = L_halfperiod
slot[7] = MicL_peak_count (dB via waveform_codec.mic_count_to_db)
slot[8] = MicL_halfperiod
The `>100 Hz` sentinel is halfperiod ≤ 5 (since 512/5 = 100 Hz).
Mic dB uses the SAME formula as the waveform codec (sign × (81.94
+ 20·log10(|count|))) — they share the mic ADC calibration constant.
Block identification anchor: bytes [22:24] == 0x0000 AND
bytes [28:32] == 1e 0a 00 00. The tail signature is the most
reliable distinguisher from non-block content in the file.
Files:
minimateplus/histogram_codec.py (new) — decoder + public API
matching the waveform codec's shape:
walk_body(body) -> records
decode_histogram_body(body) -> {Tran, Vert, Long, MicL}
decode_histogram_body_full(body) -> [per-interval dicts]
half_period_to_hz, geo_count_to_ins helpers
minimateplus/event_file_io.py (modified) — read_blastware_file
now tries the waveform codec first, falls back to the histogram
codec on failure. Same output shape, same downstream pipeline.
tests/test_histogram_codec.py (new) — 24 regression locks against
the in-repo fixture corpus, byte-exact against BW ASCII export
for peaks (all 4 channels), frequencies (all 4 channels,
including >100 Hz sentinel handling), block framing, and
segment-ID accounting.
scripts/backfill_sidecars.py (modified) — the has_samples
short-circuit added in the histogram-pending era is now a
pure defensive guard. Histograms in prod will regen .h5 files
correctly on the next backfill run.
docs/histogram_codec_re_status.md (updated) — supersedes the
earlier "in progress" version with the verified format and
test-coverage summary. Notes a few non-essential fields still
open (4-byte block metadata, Geo PVS, Mic psi(L) — none of
which are needed for waveform reconstruction).
Total verified coverage: ~3,500 blocks across 5 fixtures, every
field of every block byte-exact against BW.
The watcher-forwarded histogram event corpus on prod (~10,000
events) will now produce correct .h5 sidecars on the next backfill
run. No additional changes needed to the backfill flow — the
existing tool_version-bump cascade picks them up automatically.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures everything learned in the 2026-05-20 session before scope
forced a pause:
- Block framing is solved: 32-byte blocks, one per histogram
interval, signature byte pattern `[22:24]=0x0000` +
`[28:32]=0x1e 0x0a 0x00 0x00` reliably identifies data blocks.
- Block count = interval count (791 blocks in N844L20G.630H for
a TXT-reported 792 intervals).
- Sample[0] = Tran peak in 0.0005 in/s/count units (verified on
one event — needs cross-event confirmation).
- Samples 1-8 → channel/metric mapping is still open. None of
the obvious layouts (peak-then-freq alternating, all-peaks-
then-all-freqs, per-channel 3-tuples) match the TXT values
across multiple blocks. Likely needs a higher-activity
fixture (current N844 corpus is all noise-floor data) to
disambiguate.
- `>100 Hz` sentinel encoding in the binary is unknown.
- 4-byte variable metadata field at block[24:28] needs
correlation work against TXT columns.
Doc mirrors the structure of docs/waveform_codec_re_status.md so
a future RE session has a familiar entry point. Includes the
suggested attack plan + the code seam where the eventual decoder
will land (minimateplus/histogram_codec.py).
The §7.6.2 spec in instantel_protocol_reference.md is structurally
correct but doesn't pin down per-sample semantics — this doc
supersedes it where they conflict on confidence level.
No code shipped on this branch. When the codec is cracked, the
plan is to land minimateplus/histogram_codec.py + wire into
event_file_io.read_blastware_file() + remove the has_samples
short-circuit from scripts/backfill_sidecars.py.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When NN exceeds 0xFC, the codec extends to 12-bit NN by using the
low nibble of the TYPE byte as the high nibble of NN:
1X NN → nibble-delta block, NN = (X << 8) | NN_byte
2X NN → int8-delta block, same NN encoding
Walker and decode_waveform_v2 now handle both narrow (X=0) and wide
(X != 0) forms uniformly.
Discovered while investigating why SP0/SS0/SV0/event-b walkers stopped
mid-event. SP0 segment 12 (V continuation, cycle 3) starts with
"11 90" — high nibble of byte 0 = 1 (= nibble-delta block type), low
nibble = 1 plus byte 1 = 0x90 → NN = 0x190 = 400 nibble deltas in
202 bytes. Walker was rejecting "11" as a non-tag.
Sample count went from 47,364 to 72,972 verified byte-exact:
event-a: 9984 (full) was 9984 (full)
event-b: 6912 (full) was 738
event-c: 3840 (full) was 3840 (full)
event-d: 3840 (full) was 3840 (full)
JQ0: 9984 (full) was 9984 (full)
V70: 9984 (full) was 9984 (full)
SP0: 9984 (full) was 5122
SS0: 9222 (-7 tail) was 1758
SV0: 9222 (-7 tail) was 2114
7 of 9 fixtures now decode end-to-end across all 3 geo channels.
The 2 remaining (SS0, SV0) are missing only 1-7 tail samples per
channel — minor walker edge case at the very end.
74 tests pass (was 71).
Replaces the broken legacy int16 LE decoder in client.py with the
verified multi-channel codec. Three changes:
1. blastware_file.extract_body_bytes(a5_frames) — new helper that
factors out the body-reconstruction logic from write_blastware_file
so both writers (BW binary) and decoders (sample arrays) can use
the same canonical bytes.
2. waveform_codec.decode_a5_frames(a5_frames) — production entry point.
Returns the raw_samples dict consumers expect (Tran/Vert/Long as
int16 ADC counts; MicL as native ADC counts). Internally:
A5 frames → extract_body_bytes → decode_waveform_v2
→ decoded_to_adc_counts (geos ×16; mic pass-through)
3. waveform_codec.mic_count_to_db(count) — MicL ADC → dB(L) per BW's
display formula:
dB = sign(count) × (81.94 + 20 × log10(|count|)) for |count| ≥ 1
Verified against V70 fixture: count=813 → 140.14 dB (BW PSPL 140.1).
client.py:_decode_a5_waveform is reduced to a thin wrapper that calls
decode_a5_frames and populates event.raw_samples. Original implementation
preserved as _decode_a5_waveform_LEGACY (dead code; reference only).
Also fixed a tail-end bug in decode_waveform_v2 where trailer-section
"40 02" markers (containing ASCII serial bytes, NOT real segment headers)
were being mis-interpreted, producing 2 spurious samples per channel at
the end of each event. Added bytes [12:14] == "02 00" validation to
reject non-header markers.
7 new pytest tests cover the new helpers and dB conversion. Total:
71 passing (up from 64).
Known limitation (carried over from before): the walker still stops
mid-event on the loudest fixtures (SP0/SS0/SV0/event-b) at some
mid-segment edge cases not yet characterized. Every sample reached
is decoded correctly; the walker just doesn't reach all of them.
Loud events still yield 5,000–15,000 byte-exact samples each.
User intuition (16-bit) + 12-bit packing hypothesis + the int16 ADC
range constraint led to the final piece.
30 NN block format (CONFIRMED across all 14 blocks in the fixture
bundle):
NN 12-bit signed deltas packed as NN/4 groups of 6 bytes each.
Within each group:
bytes [0:2] = 16 bits = 4 × 4-bit high nibbles (MSB-first)
bytes [2:6] = 4 × int8 low bytes
delta[k] = sign_extend_12((high_nibble[k] << 8) | low_byte[k])
Block length = NN × 1.5 + 2 bytes (tag included). Earlier walker
used NN × 4 which is only correct in the TRAILER section.
Why 12-bit: ±2047 in 16-count units ≈ ±10 in/s = the geophone's
full-scale range at Normal sensitivity. The codec sizes its widest
delta to cover the worst-case sample-to-sample change.
Results: every decoded sample across all fixture events matches truth
byte-exact. ZERO divergences.
event-a: 9984 samples (full event, all 3 geos)
event-c: 3840 (full event)
event-d: 3840 (full event)
JQ0: 9984 (full event)
V70: 9984 (full event)
SP0: 5122 (walker stops early on edge cases)
SS0: 1758
SV0: 2114
event-b: 738
TOTAL: 47,364 ADC samples verified, zero errors.
Three full 3-sec events decode end-to-end across all three geo
channels. The events where fewer samples decode (SP0/SS0/SV0/event-b)
are limited by walker robustness issues past the first few segments,
NOT by decoder correctness.
64 tests pass (up from 55). Files: minimateplus/waveform_codec.py
(new 30 NN decode + corrected walker length), tests/test_waveform_codec.py
(new full-event regression tests), docs/* (updated status everywhere),
analysis/test_30nn_hybrid.py (new — the analysis script that confirmed
the format).
User asked the right question: do events without 30 NN blocks decode
fully? Answer: YES.
event-a: Tran 3328 ✓ Vert 3328 ✓ Long 3328 ✓ (28 segments, 0 '30 NN')
event-c: Tran 1280 ✓ Vert 1280 ✓ Long 1280 ✓ (12 segments, 0 '30 NN')
event-d: Tran 1280 ✓ Vert 1280 ✓ Long 1280 ✓ (12 segments, 0 '30 NN')
17,664 ADC samples decoded byte-exact against BW's ASCII export.
Zero divergences across event-a, event-c, event-d.
This means the codec is FULLY SOLVED for any event without 30 NN
blocks. The remaining gap is the 30 NN block format only — used for
high-amplitude regions where deltas exceed int8 range. For quiet
events (or quiet stretches of loud events), the decoder is complete.
9 new regression tests bring the total to 55, all passing.
Files: tests/test_waveform_codec.py + docs/waveform_codec_re_status.md
+ new analysis/verify_quiet_bundle.py.
The segment-channel scoring analyzer (from scratch/next_experiment_skeleton.py)
ran and immediately confirmed the rotation hypothesis:
SP0 seg 0: best fit Vert 508/508 ✓
SP0 seg 1: best fit Long 508/508 ✓
SP0 seg 3: best fit Tran 508/508 ✓ (Tran continuation)
SP0 seg 5: best fit Long 508/508 ✓
SP0 seg 9: best fit Long 508/508 ✓
V70 seg 0: best fit Vert 508/508 ✓
V70 seg 1: best fit Long 508/508 ✓
Channels rotate Tran → Vert → Long → MicL per 40 02 segment header.
Also discovered the segment header has DOUBLE duty: bytes [14:18] anchor
the NEW segment's channel (2 samples as int16 BE in 16-count units), AND
bytes [0:4] extend the PREVIOUS channel by 2 more samples (2 deltas as
int16 BE). This is the same "2 anchors + delta stream" structure as the
body preamble for Tran.
decode_waveform_v2 now returns full per-channel sample dicts.
Byte-exact verified ranges:
V70: Tran 512, Vert 512, Long 512 (all first segments)
JQ0: Tran 512, Vert 258
SP0: Long 1536 (all 3 L segments)
Still open: the 30 NN block format (high-amplitude packed deltas) —
appears mid-segment when single-byte deltas can't carry the magnitude.
6 new tests bring the count to 46. All passing.
Three "truth layers" had drifted apart between commits. Fixed:
1. waveform_codec.py docstring rewritten from the 2026-05-08
"structural framing only" state to the 2026-05-11 "Tran segment 0
solved + segment-header partially decoded" state. Killed stale
"~80 sample-sets per segment" language (real segments are
flash-page-byte-sized, not sample-count-sized; observed first-segment
sizes are 42-510 samples depending on signal). Killed stale
"preamble is 7 or 9 bytes" language (always 7).
2. docs/instantel_protocol_reference.md §7.6.1: added a clear
"CURRENT STATUS" box at the top with a status table. Replaced the
stale "~80 sample-sets" line with the verified per-event segment
sizes. Merged two redundant segment-header field-table sections.
3. docs/waveform_codec_re_status.md (NEW): clean working-status doc.
Solved / not solved / hypothesis / next experiment / fixtures /
tests. The protocol reference remains the historical Rosetta
Stone; this new file is the current-truth working note that
shouldn't accumulate fossil layers.
4. CLAUDE.md §"Waveform body codec": prominent warning box at top —
"DO NOT TRUST decoded sample arrays yet." BW binary passthrough
is the only sample-bearing output to trust until the decoder
lands. Added a "Next experiment" subsection pointing the next
pass at the segment-channel scoring analyzer.
40 tests still pass.
User uploaded a Vert-heavy event (JQ0) and a Mic-heavy event (V70).
Those two were exactly what was needed to crack the next piece:
- 00 NN block = run-length-encoded zero deltas in the current channel.
Append NN copies of the current cumulative value (no change).
- find_data_start now recognizes 00 NN as a valid first tag (some events
begin with a leading 00 NN RLE block).
- decode_tran_initial now decodes the FULL segment 0 (not just the first
data block).
Results across 5 fixture events:
- M529LL1A.SP0 (loud-all-channels) : 510 / 510 ✓
- M529LL1L.JQ0 (Vert-heavy) : 510 / 510 ✓
- M529LL1L.V70 (Mic-heavy) : 510 / 510 ✓
- M529LL1A.SV0 (loud-from-start) : 58 / 58 ✓
- M529LL1A.SS0 (loud-from-start) : 42 / 502 (stops at first 30 04)
The 30 04 block (only seen in loud-from-start events) hasn't been
decoded yet — likely a channel-switch marker for the high-amplitude
regime.
Also discovered: segment header (40 02) payload bytes [0:2] = T_delta
at first sample of new segment, [6:8] = byte length to next segment.
Multi-segment Tran decoding still diverges after sample 512 because
the per-segment channel ordering after the header is unknown.
Tests: 40 pass (up from 36).
Files:
- minimateplus/waveform_codec.py: find_data_start fix, RLE handling,
full segment-0 decode in decode_tran_initial
- tests/test_waveform_codec.py: synthetic RLE test, full segment 0
tests for JQ0 and V70
- tests/fixtures/5-11-26/: M529LL1L.JQ0, M529LL1L.V70 + TXT exports
- docs/instantel_protocol_reference.md §7.6.1: RLE + segment-header docs
User uploaded 3 high-amplitude events (PPV 6-7 in/s — shook the geophone
hard) to decode-re/5-11-26/. These cracked the Tran codec:
- Preamble bytes [3:5] and [5:7] = Tran[0] and Tran[1] as int16 BE
in 16-count units (LSB = 0.005 in/s). Confirmed across all 7
fixtures.
- First data block carries Tran deltas from sample 2 onward:
* 10 NN block: NN/2 bytes of payload, each byte = two 4-bit signed
nibble deltas (high nibble first)
* 20 NN block: NN int8 signed deltas
Verified 22+42+46 = 110 Tran samples across SP0/SS0/SV0 with 0 errors
against BW's ASCII export.
Why the earlier 96-combination brute force failed: the quiet 5-8
events all had T[0] = T[1] ≈ 0 so the preamble's per-channel encoding
was undetectable. Loud events made the encoding obvious.
What's solved:
- minimateplus.waveform_codec.decode_tran_initial: returns first
N Tran samples in 16-count units for any body.
- Walker length formula for in-data 30 NN blocks (NN*2 instead of NN*4).
- Walker now handles bodies that start with 20 NN (in addition to 10 NN).
What's still open:
- Tran past the first data block (multi-block channel switching).
- Vert / Long / MicL channel encodings.
- Walker correctness past offset ~427 in event-b.
Tests: 36 pass. decode_waveform_v2 still returns None — the full
multi-channel decoder is not wired up. decode_tran_initial is the
new verified entry point.
Files: minimateplus/waveform_codec.py, tests/test_waveform_codec.py
(adds 5-11-26 fixtures + decode_tran_initial tests), and
docs/instantel_protocol_reference.md §7.6.1 (Tran codec spec).
Decoded the structural framing of the Blastware waveform body — the bytes
between the 21-byte STRT record and the 26-byte file footer. The body is
a sequence of tagged variable-length blocks, NOT raw int16 LE. Five tag
types (10/20/00/30/40 NN) and their lengths are now confirmed against the
4-event May 2026 fixture bundle. Body splits cleanly into ~16 segments
(for a 1280-sample event) separated by 40 02 segment headers carrying a
monotonically incrementing uint32 LE counter at bytes [8:12].
What's done:
- minimateplus/waveform_codec.py — block walker, segment splitter, segment
header parser. decode_waveform_v2 is a stub returning None until the
byte-to-sample mapping is solved; client.py is unchanged.
- tests/test_waveform_codec.py — 31 tests covering block detection, lengths,
contiguous-walk, segment splitting, segment-header parsing, and counter
monotonicity. All pass.
- tests/fixtures/decode-re-5-8-26/ — bundled fixtures (4 events, BW binary
+ Blastware ASCII export each).
- docs/instantel_protocol_reference.md §7.6.1 — replaced retraction box
with the verified structural decoding plus an explicit list of what's
still open.
What's still open: the per-byte mapping inside 10 NN / 20 NN blocks. 96
channel-permutation × nibble-order × sign-convention combinations were
brute-force tested; none match BW's ASCII export to within ±1 ADC count.
The codec is more elaborate than uniform 4-bit deltas — likely a hybrid
variable-bit-width scheme with segment-anchor resync points. Next
recommended step: capture an event with a known calibration tone to pin
down magnitude scaling.
Walker also bails out partway through event-b (open issue documented in
both the module and the protocol reference).
Three-pass audit of docs/instantel_protocol_reference.md against
CLAUDE.md and the minimateplus/ implementation. Closes long-standing
discrepancies that had accumulated as the protocol understanding
evolved month over month.
Major corrections:
- §2/§3: S3 frames terminate on bare ETX, not DLE+ETX; payload
byte[1] is flags / byte[2] is SUB (was wrongly DLE/ADDR).
- §4.2: probe responses do not carry data length; DATA_LENGTH
is a per-SUB hardcoded constant.
- §5.1: dropped stale duplicate "SUB 1C = TRIGGER CONFIG READ"
row; SUB 0A lengths corrected from 0x30/0x26 to 0x46/0x2C.
- §5.3: added the missing write-frame mechanics (BW_CMD-only
doubling, DLE-aware checksum, offset = data[1]+2, ack format,
SUB 71 chunk parameters).
- §7.6.x: switched compliance-anchor convention from the unstable
10-byte form to the canonical 6-byte `\xbe\x80\x00\x00\x00\x00`;
recording_mode confirmed at anchor−8 in both read and write
(the prior anchor−3/−4 split caused anchor drift on write).
Sample_rate at anchor−6, histogram_interval at anchor−4 (now ✅),
record_time at anchor+6. Geo_range row added at channel_label+33.
- §7.5b/§8: added the 10-byte sub_code=0x03 continuous-mode
timestamp variant; peak vector sum location corrected from
fixed offset 87 to label-relative tran_pos−12.
- §7.7.2: SUB 1E/1F token byte at params[7], not params[6].
- §7.7.3: SUB 0A length disambiguation rewritten.
- §7.8.4/§7.8.7: fi==9 skip marked FIXED; metadata-page TODO
replaced with current decoder state.
- §11: POLL example wire bytes corrected; SUB 5A row added to
checksum table.
- §13/§14: device-under-test updated to BE11529/S338.17; TCP
Idle Timeout consistency fix (0→2 min); Data Forwarding
Timeout units clarified.
- §15 (renumbered from second §14): open-question entries
already resolved in CLAUDE.md closed out.
- Appendix D: extension taxonomy rewritten — extensions encode
a timestamp (AB0T scheme), not recording mode.
Navigation note added to §7 acknowledging the organic-growth
duplicate section numbers (§7.5/§7.5b, §7.6, §7.7, §7.8, §7.9)
and pointing readers to the canonical sections for each topic.
https://claude.ai/code/session_019tWZybD94YUsBaEGhnM5A2
Tighten the Series III / Series IV boundary so UI and storage dispatch
on a clean signal instead of sniffing filenames or applying magnitude
heuristics.
Phase 1 — events.device_family column ("series3" | "series4"):
self-applying migration with filename-based backfill of existing rows
(1,132 backfilled on prod 2026-05-20); plumbed through every import
path (BW endpoint, IDF endpoint, ACH server, BW CLI, sidecar
backfill); UPSERT preserves via COALESCE; UI dispatches on it.
Phase 2 — extract micromate/ package alongside minimateplus/:
native IdfEvent / IdfReport / IdfPeaks / IdfProjectInfo /
IdfSensorCheck (mic in dB(L), not pseudo-psi); moved
idf_ascii_report.py from sfm/ to micromate/; refactored
save_imported_idf to use IdfEvent and bridge to minimateplus.Event at
the SQL-insert boundary; idf_file.py stub for the future binary codec.
Phase 3 prep — docs/idf_protocol_reference.md captures the two
observed Thor binary header signatures (1,012 newer-firmware files vs
2 old files whose layout is byte-for-byte BW-STRT-compatible), file-size
hints suggesting int8 sample encoding, open questions in dependency
order, and a concrete first-session plan for cracking the codec.
Also rolled in the v0.18.1 hotfixes that motivated this work:
- idf_ascii_report parser now handles "<0.005 in/s" (below-threshold)
and "N/A" markers without leaving raw strings in numeric DB columns.
- sfm_webapp.html: defensive _ppvFmt / mic formatter so future
data-shape drift can't kill the whole events table render.
All 1,014 example-data sidecars round-trip through the new package.
See CHANGELOG.md for full notes.
- Created a comprehensive runbook (`wedged_unit_recovery.md`) detailing the recovery process for units stuck in a call-home loop, including symptoms, recovery steps, and explanations of the failure mode.
- Added `blind_stop.sh` script to send stop-monitoring commands in a tight loop for unresponsive devices.
- Introduced `rescue_device.sh` script to disable Auto Call Home and erase events from a busy device.
- Implemented `slow_drip.sh` script to send stop-monitoring frames at a slow rate to prevent UART overrun.
- Developed `spam_stop.sh` script to rapidly send stop-monitoring commands to a device.
- Created `watch_unit.sh` script for passive monitoring of device reachability, logging results over time.
serversdownandClaude Opus 5
claude