Salida, CO → Broomfield, CO

Salida to Broomfield, the pitch-stop leg: US-285 north out of the San Luis Valley area, over Kenosha Pass (summit 10,057 ft at 12:34 local), then descending into the Denver metro to LightShip Energy’s Broomfield facility. Ambient temperature ran from the high-70s at the Salida start to a 104°F peak on the descent, consistent with the extreme-heat conditions flagged for this leg. ProPower ran essentially continuously (96.6% of moving time) rather than in discrete on/off test segments — expected behavior under the hot-weather thermal-management plan, not a deliberate Phase A/B design for this leg. TrekDrive was not engaged; no telemetered evidence of State A activity appears in the F-150 data, consistent with the plan to withhold TrekDrive on this hot leg. End-of-day GPS fix (39.989028, −105.009851) lands essentially on top of LightShip Energy’s Broomfield address — the pitch-stop arrival.

Idle rows (gear 70/128/130, 302 of 5,962 rows — all clustered at the Broomfield arrival/park) filtered from duty and fuel accounting below except where noted. Row-integrated fuel (12.806 gal) cross-checks the OBD trip-fuel counter (12.796 gal) to within 0.08%.

This leg does not support an ON/OFF or grade-matched subtraction: ProPower was on for 96.6% of moving time with no genuine matched-speed/matched-grade OFF pair. The table below is therefore bookkeeping against the KB baseline, not a fresh measurement.

Nightly incremental-GPH regression: skipped this leg. The regression needs real variation in both ProPower on/off state and grade. ProPower ran on for 96.6% of moving time here — the only off-window coincides with decelerating into the Broomfield arrival, which is confounded with speed and grade both dropping toward zero simultaneously and is not a usable regression point. Per the standing methodology, the table is omitted rather than forcing a number from an underdetermined fit. Kenosha Pass supplied real grade variation, but without a matched ProPower off-state it can’t isolate the ProPower coefficient.

No State A segment this leg. TrekDrive was deliberately minimized on the Prescott→Broomfield corridor to avoid stacking battery energy debt on top of the extreme-heat thermal load, per the trip plan. This is expected, not a data gap — there is no gross/net table to report because there is no State A activity to characterize. Manual-TrekDrive evidence collection resumes once TrekDrive use resumes (plains legs east of Denver are the primary target going forward).

• IAT2 peaked at 168.8°F — well above the 140°F Tier-3 throttle-closure threshold and above the prior parked-sun peak (143.6°F). Coincides with the 13:16–13:17 enrichment window (highest peak IAT2 of the leg, 154.4°F row-level during sustained rich λ).
• ECT peaked at 232.7°F — a new high for the trip, exceeding the prior non-towing peak of 212.9°F noted in the KB. First ECT reading over 230°F recorded under towing load; worth tracking on subsequent hot climbs.
• LTFT B2–B1 gap: 0.39% (first half of leg) widening to 1.08% (second half) — consistent with the ongoing slowly-widening-gap flag carried from pre-trip data. Leg-average gap (0.74%) sits within the previously observed 1.7–2.1% band's lower edge; continue tracking trend, not yet a fault.
• OAR ranged −0.697 to −0.475 — never reached −1.0 (full timing authority), consistent with prior findings. Knock Control Spark Retard stayed small (IQR 0.0–0.18°, max 5.32°), no sustained knock activity.

• OBDLink frame rate held at fixed 2.0 s across all 42 PIDs for the full leg: median 2.000 s, std 0.0028 s, 100% of intervals in [1.9, 2.1] s.
• TrackLogger confirmed native 1 Hz (11,891 fixes over 11,927 s); merge onto OBDLink’s absolute wall-clock timeline matched at median offset 0.49 s, all 5,962 rows matched within a 2 s tolerance.
• GPS cross-check: OBDLink’s own onboard GPS vs. TrackLogger position at matched timestamps agreed to a median 0.0001° (~36 ft) in both lat and lon — no MST/MDT clock-drift failure mode this leg (both endpoints sit inside Colorado/MDT throughout, unlike the AZ state-line risk called out in the instruction set).
• λ channel behaving as expected: median 0.996 in cruise-load rows (load < 40%), dropping to a leg-minimum of 0.567 momentarily (filtered as transient) and to a sustained floor near 0.80–0.88 during the four flagged high-load climb episodes.
• One all-zero header row (row 0) and one OAT floor-zero dropout row filtered per standard preprocessing; no other dropouts or stale-repeat runs observed.

• New ECT peak (232.7°F) exceeds the prior towing/non-towing benchmarks in the KB — worth a KB update once a couple more hot climbing legs confirm whether this is a one-off (ambient + grade + AC load stacking) or a new sustained ceiling.
• ProPower ran fully continuous for 3+ hours without a single BMS-driven cycle-off, unlike the "up to ~1 hr per cycle, house-load-driven" pattern documented for this hot-weather corridor. Either LightShip house SOC was drawing down continuously without ever hitting the BMS restore-satisfied point, or this leg’s house-load draw ran higher than the modeled 3–4 kWh cycle band. Worth confirming against the LightShip Atlas display if available, and a positive data point for the pitch either way (sustained full-rate delivery, no dropouts, despite a 104°F peak).
• No EVSE thermal-ceiling dropout observed this leg despite OAT reaching 104°F — the 102–105°F EVSE cutoff was never tripped. Consistent with the AeroHub 3–5°F-above-ambient offset keeping EVSE temperature under the ceiling even at peak OAT.
• CAT scale weigh-in documentation (confirmed complete 2026-07-11) still not uploaded to Project Knowledge — mass-dependent calculations continue to use the ~7,053 kg/15,550 lb estimate and flag it as such. Not consequential to today's report (GPH regression was skipped), but will matter once Cd·A speed holds or a regression-capable leg comes up.
• Missing Day 1 first segment (Prescott→Williams-area, pre-dawn 2026-07-11, including the accident delay) remains unresolved — carried forward from the prior nightly report.