LightShip Turbo Edition
F-150 PowerBoost · LightShip AE.1
V2V AC Energy Transfer — PowerBoost to LightShip
Across 3,719 miles and 13 legs (July 2026), the V2V chain delivered 7.70 kWh AC per gallon at the ProPower outlet — 22.8% of gasoline LHV (33.7 kWh/gal) arriving as AC at the truck's bed panel. What reaches the LightShip cells is one modelled step further: through the trailer's ~92% onboard charger and the cells' own losses, about 6.9 kWh stored per gallon, ≈ 20.6%. ProPower delivered 74.74 kWh AC over the campaign, attributable to 9.71 gallons — 3.8% of total fuel. The figure proved stable leg to leg across a wide range of conditions. The 557.6-mile June round trip that originally grounded this page is superseded by Campaign 1; see the Test Data reports for per-leg detail.
The fuel cost of charging was measured directly for the first time: a commanded on/off step on flat I-10 at 62 mph, Thermal → Blythe, same road and same truck traffic in both charge states. Switching the trailer charger with the Atlas charge-limit slider — the truck's own Pro Power switch does not interrupt delivery in motion — produced a clean 4.9 kW step at the outlet (5.85 kW bulk against 0.93 kW load-support). Regressed on 65 one-minute windows with grade fitted rather than assumed, R² 0.95: 5.8 kW costs 0.75 gal/hr (band 0.65–0.83), i.e. 7.0–9.0 kWh AC/gal, 7.5 on matched-grade blocks. Campaign 1's 7.70 sits inside the band, near its centre. This number comes from the fuel channel alone and never touches the truck's torque model, which makes it the best-supported figure in the program. On the same leg the charging share of fuel was 5.9% (9.69 kWh AC over 278 miles, charging only Thermal → Blythe) and the trailer went from 74% to 84% state of charge by its own display.
The Architecture — What Is Actually Happening
The ProPower Onboard (PPoB) system is not a standalone generator bolted to the frame. It is a software-defined power export layer that draws from the same high-voltage (HV) bus that the motor-generator unit (MGU) uses for hybrid operation. Understanding this distinction is critical to evaluating the system's efficiency.
What the truck sends down the umbilical is AC power. What the trailer does with it is the trailer's decision, and it is not always charging. Campaign 1 found the converter operating in three distinct states: off entirely, delivering roughly 0.94–0.95 kW, or delivering 5.7–6.1 kW. Only the third puts bulk charge into the pack. The middle state is attributed to load support — energy arriving at about the magnitude of the trailer's own house draw, offsetting the fridge, 12 V systems and controls so the pack discharges more slowly than it otherwise would. It does not raise state of charge, and it is not waste. On the first leg alone it accounted for 9.26 kWh. The three draw magnitudes are measured; the house-load attribution of the middle one is operator judgement and remains under test — 0.94 kW is a narrow band for something that should follow a varying house load, and it may turn out to be a fixed setpoint. Calling the whole capability “charging” makes that energy look like overhead, which is how this project mislabeled it once already.
The MGU is the pivotal component. Because it is physically situated between the ICE crankshaft and the 10-speed transmission input shaft, it rotates at engine speed at all times the truck is in motion. This topology creates the efficiency argument at the heart of the in-motion generation claim.
The In-Motion Efficiency Argument — Theory
The LightShip AE.1 is an all-electric trailer: its house loads, climate control, and TrekDrive axle motor all draw from the onboard HV battery. On a multi-day trip that battery has to be refilled, and the assumption that it can be topped up at destinations or en route does not hold in practice. Public charging built for cars is largely unusable for a 50-foot rig — most sites can't physically accommodate the length, pull-through access is rare, and DC fast charging carries real interoperability problems. Even ordinary AC charging in the field is a lottery: campground pedestals vary in amperage, condition, and availability, and a leg can end somewhere with no usable charge at all. In-motion V2V AC charging removes that dependency. The tow vehicle's inverter is always present, always compatible, and delivers energy across the entire drive rather than only at stops — so the trailer arrives with a useful state of charge regardless of what charging did or didn't exist along the route. The Charging Infrastructure page covers this problem in full.
The central premise is this: when the truck is moving, the mechanical energy required to spin the MGU rotor is already embedded in the drivetrain's friction budget. The bearing drag, windage, and rotor inertia of the MGU are paid for by the wheels turning, not by dedicated fuel combustion. Therefore, the incremental fuel cost to begin extracting electrical power from the MGU is limited to the thermodynamic conversion cost of combustion producing additional crankshaft torque — not the cost of spinning up a cold generator from a standing start.
This is subtly but importantly different from a conventional claim. A standalone generator at idle must consume fuel to: (a) overcome its own mechanical friction, (b) run through incomplete combustion at low load, and (c) manage thermal losses at a non-optimal RPM. The PowerBoost MGU in motion does none of these things — it is already thermally stabilized, mechanically spun to road speed, and operating in a regime where the ICE is in or near its BSFC-efficient cruise band.
To extract 5.8 kW from the MGU while moving, the ICE must produce approximately 6.6 kW of additional shaft power (5.8 kW ÷ an ≈ 88% modelled chain: MGU ≈ 93%, bus and cabling ≈ 97%, ProPower inverter ≈ 98% — the inverter stage now measured, see § 04). At 30% BTE that is about 22 kW of additional heat release, or ~0.65 gal/hr of incremental fuel; with the rotor drag already paid by the drivetrain (an estimated 0.7 kW) the theory lands at roughly 0.58–0.65 gal/hr. The measured figure is 0.75 gal/hr (band 0.65–0.83), from the Campaign 2 step test. The theory is about 15% optimistic, and the fuel channel cannot say whether the shortfall is a cruise BTE nearer 26% than 30% or a chain losing more than modelled; the truck's torque model puts crank-to-outlet at ≈ 61–69%, but that channel under-responds to load and the figure is indicative only.
Compare this to static generation (parked, ICE cycling at ~1,000 RPM): community-reported data and the physics of low-load ICE operation suggest 0.8–1.0 gal/hr for ~6 kW output, and the one instrumented third-party parked test on record gives 6.55–7.20 kWh AC/gal (see the ProPower System page). The implied brake thermal efficiency at that operating point is only 18–22%, versus 26–30% at cruise load. This is the efficiency advantage of in-motion generation — the ICE is running at a load point where it converts heat to shaft work more effectively. Measured against those parked figures the advantage is ≈ 10–15% (7.70 against 6.55–7.20 kWh/gal), not the 30% an idealised model suggests; a parked figure on this truck with this project's own instrumentation remains an open bench measurement.
Full Conversion Loss Chain — AC Outlet to LightShip Cells
The 5.8 kW leaving the ProPower 240V outlet undergoes four conversion stages before resting as stored energy in the LightShip battery. Each stage introduces a measurable loss.
| Stage | Loss mechanism | Loss (W) | Loss (%) | Cumulative efficiency |
|---|---|---|---|---|
| NACS cable (~15 ft, 10 AWG equiv.) | I²R resistive heating at 24.2A | 23 W | 0.40% | 99.6% |
| NACS connector contacts | Contact resistance ~4 mΩ | 2 W | 0.04% | 99.5% |
| LightShip onboard charger (OBC) | AC-DC rectification, switching, magnetics | 464 W | 8.0% | 91.6% |
| LightShip battery cells | I²R cell resistance, coulombic loss | 107 W | 1.8% | 90.2% |
The LightShip onboard charger (AC→DC rectifier) consumes ~464 W converting the 240V AC from the ProPower outlet to the LightShip HV DC bus voltage. This is 8% of the input, or roughly 61% of all conversion losses in the chain. Cable and connector losses are essentially negligible at this current level. Battery coulombic efficiency at this very low C-rate (~0.075C into a 77 kWh pack) is excellent. Net delivery: 5.23 kW into the cells from 5.8 kW AC input — a 90.2% chain efficiency from outlet to stored energy. This whole section is modelled: the project meters the truck side of the umbilical, not the trailer side, and the 92% charger figure is LightShip's.
Full Chain: Gasoline to LightShip Battery Cells
Combining the in-motion generation efficiency with the AC-to-battery conversion chain gives the end-to-end efficiency of the V2V charging system. The chain is now anchored at the outlet by measurement; the two stages either side of it remain modelled.
Idealised, the chain would run 30% × 88% × 90.2% ≈ 24%, or about 8 kWh stored per gallon. What is measured is 7.70 kWh AC per gallon at the outlet — 22.8% of the gasoline's 33.7 kWh — and one modelled step behind it, about 6.9 kWh stored in the LightShip battery per gallon, ≈ 20.6%. The ProPower inverter's own DC-to-AC ratio was measured at ≈ 0.98 in Campaign 2 from the inverter's HV-side input current; the earlier 93% assumption for that stage was too pessimistic, which moves the modelled shortfall upstream to the engine and the motor-generator.
At 7.70 kWh/gal delivered to the umbilical and gasoline at $3.50/gal, energy costs $0.45/kWh at the outlet and ≈ $0.50/kWh stored in the LightShip battery — roughly 2.7–3× the U.S. average residential rate of ~$0.17/kWh (modelled; scales with the pump price). However, this is mobility-delivered energy with no charging infrastructure required, delivered at 62 mph. The correct comparison is not residential electricity but rather: (a) a campground hookup where available, or (b) the opportunity cost of a Supercharger stop adding significant time to the journey.
Contrast with Measured Telemetry
Two campaigns provide the empirical ground truth against which to test the model — Campaign 1's 3,719 miles across 13 legs, from 80°F to 104°F ambient and from Colorado mountain grades to Nebraska plains, and Campaign 2's Prescott ⇄ Thermal legs at 46 PIDs and 1 Hz, which added the bus-load channels and the commanded step test.
| Metric | Theoretical (in-motion, modelled) | Measured | Basis |
|---|---|---|---|
| Incremental fuel for 5.8 kW | ~0.65 gal/hr | 0.75 gal/hr (0.65–0.83) | Campaign 2 step test, I-10 at 62 mph, 65 windows, R² 0.95 |
| kWh AC per gallon (fuel→outlet) | ~8.9 | 7.70 (C1) · 7.0–9.0, matched-grade 7.5 (C2) | C1 seven legs; C2 from the row above. Theory ≈ 15% optimistic |
| Fuel→outlet efficiency | ~26% | 22.8% | 7.70 ÷ 33.7 kWh/gal LHV |
| Fuel→cells efficiency | ~24% | ≈ 20.6% (modelled step) | 22.8% × 0.902 onboard-charger chain; not measured on the trailer side |
| Delivery rate, bulk charge | 5.8 kW | 5.69–6.06 kW (C1) · 5.85 kW mean (C2) | No thermal derate at any ambient through 104°F |
| ProPower inverter DC→AC ratio | 0.93 assumed | ≈ 0.98 | C2, from the inverter's HV-side input current |
| ProPower delivered | — | 74.74 kWh AC (C1) · 9.69 kWh AC (C2 return) | Bulk-charge basis, ≥ 2.0 kW |
| Fuel attributable to ProPower | — | 9.71 gal · 3.8% (C1) · 1.26 gal · 5.9% (C2 return) | C2 return charged Thermal → Blythe only |
| Trailer state of charge gained | — | 74% → 84% | C2 return, Atlas display at stops; coarse endpoint instrument |
| Overall campaign economy | — | 14.38 mpg (C1) · 13.0 mpg (C2 return) | C1 258.59 gal; C2 return 278 mi with a 5,150 ft net climb, charging half the leg |
| Fuel reconciliation vs Trip Fuel PID | — | mean 0.21% (C1) · 0.07% (C2 return) | C1 worst leg 0.87% |
The 7.70 kWh AC/gal figure held stable across Campaign 1 despite dramatically varied operating conditions — mountain grades and plains, 80–104°F ambient, seven distinct ProPower legs — and Campaign 2's direct step test put it inside a measured band of 7.0–9.0. The structural model survives: generation efficiency is set by chain losses and ICE efficiency at cruise, both stable across the operating range, so grade and headwind change the base towing fuel without materially altering the incremental generation cost. What the model gets wrong is the level, by about 15% — the measured cost is 0.75 gal/hr against a theoretical 0.65. Earlier revisions of this page closed that gap with a "theoretical 8.54 kWh/gal" that turned out to be the measurement divided by the onboard-charger efficiency, not an independent prediction; it is withdrawn. What the June round trip suggested with two data points, Campaign 1 established across 3,719 miles and Campaign 2 measured directly.
The per-leg loss-chain breakdown, bus-load accounting and incremental regressions are documented in the individual nightly reports rather than reproduced here. The earlier June figures for those sub-tables (0.697–0.753 gal/hr allocated rate, PMU throttling events at specific mileposts) were single-trip observations superseded by the campaign datasets; the throttling observation in particular was not reproduced in either campaign. See the Test Data reports for current per-leg telemetry, and the Specifications page for the reconciled figures.
In-Motion vs Static V2V — When to Use Each
| Mode | ICE operating point | BTE | Fuel rate (6 kW) | kWh/gal delivered | Best use case |
|---|---|---|---|---|---|
| In-motion (measured) | Cruise load, 1,400–1,800 RPM | 26–30% (modelled) | 0.75 gal/hr (0.65–0.83) | 7.0–9.0 · reference 7.70 | Highway towing, any distance |
| Static, engine warm (third-party) | ~1,000–1,200 RPM, cycling | 20–24% | 0.80–0.95 gal/hr | 6.55–7.20 (instrumented press test) · 6.3–7.5 (community) | Campsite top-up, short sessions |
| Static, cold start (modelled) | <1,000 RPM, enrichment, cycling | 15–18% | 0.90–1.10 gal/hr | 5.4–6.7 | Avoid — worst efficiency |
The in-motion advantage is real and consistent, but smaller than an idealised model suggests: ≈ 10–15% more energy per gallon than the third-party parked figures (7.70 against 6.55–7.20 kWh/gal), more against a cold start. This is the payoff of the hybrid architecture: the ICE never operates at the low-efficiency idle cycling mode that a conventional generator must use when stationary, and the discount is largest in motion and merely smaller at rest. The parked figures are published third-party data, not this project's measurement; the ProPower System page carries them, their sources, and the withdrawn claim that a purpose-built generator would beat a parked PowerBoost.
Community-reported experience (F150gen14 forum) confirms 0.8–1.0 gal/hr for approximately 6 kW static output, with the ICE cycling on/off to maintain HV battery SOC. One reported observation: the ICE runs almost continuously at ~1,000 RPM when ProPower is at full load — the 1.5 kWh HV battery (with only ~600 Wh usable headroom) depletes within seconds at a 6+ kW extraction rate, so the engine cannot pulse-and-coast as it does under lighter loads. This is why static efficiency is limited: the engine is locked into a low-RPM, partial-load operating point with no ability to burst into a more efficient range.
Charging Timeline and Practical Range Implications
All four tiles are modelled from measured pieces: the 5.8 kW outlet rate and the 0.75 gal/hr fuel cost are measured, the 92% onboard-charger step and the 77 kWh gross pack are LightShip's figures (usable capacity ≈ 69.2 kWh). The real-world implication: an overnight drive of 700+ miles would theoretically deliver a full charge to a depleted LightShip battery. For practical trip segments of 250–300 miles at full charging duty, the umbilical delivers roughly 23–28 kWh AC, about 21–25 kWh into the cells — enough to materially offset TrekDrive discharge during that leg or provide a day or two of campsite electrical autonomy. Both campaigns confirmed the rate at scale: seven Campaign 1 legs delivered a campaign total of 74.74 kWh AC to the trailer, and the Campaign 2 return delivered 9.69 kWh AC over the charging half of a 278-mile leg while the trailer's own display moved from 74% to 84% — 9.69 kWh through a 92% charger is ≈ 8.9 kWh, against ≈ 7.7 kWh for ten points of a 77 kWh pack, agreement within the display's integer resolution and house draw. Both are consistent with the per-leg delivery rates the model predicts. Charging only runs while the trailer wants it: a trailer that starts full pays none of this fuel.
The cumulative energy-delivery curve shown here previously plotted the single June round trip. Campaign 1's per-leg delivery is documented across the individual nightly reports; the campaign total of 74.74 kWh AC over seven ProPower legs supersedes the earlier 42.7 kWh / 557.6-mile figure. See the Test Data reports for per-leg delivery detail.
The 17 MPG Sea-Level Observation — Now Measured
Forum posts from the LightShip owner noted that early testing with the modified PowerBoost at sea level with no wind at 62 mph showed 17 mpg with the ProPower umbilical connected. Earlier revisions of this page argued the figure from a road-load estimate. Campaign 2 measured it.
On the Campaign 2 return, the flat interstate stretch at 62 mph with truck traffic, below sea level to Blythe, ran ≈ 17 mpg grade- and charge-corrected — the umbilical delivering bulk charge for most of it. The figure includes drafting behind interstate truck traffic and is not a clean-air number: the program's own clean-air speed curve, measured the same week on the Salton Sea shore with no charging, gives 14.8 mpg at 62 mph at sea-level density, and the Campaign 1 Phase A pure-tow baseline of 15.43 and 15.27 mpg (Days 5 and 6, 1,080 miles, ProPower off; both legs ran light tailwinds and are flattered, unequally) sits between them. The owner's early observation was right for the conditions it was made in, and the conditions — interstate traffic — matter as much as the speed.
What charging costs in economy follows from the step test without any road-load estimate. At 62 mph, 0.75 gal/hr is 0.012 gal per mile: on the 14.8 mpg clean-air curve that is 14.8 → 12.6 mpg with the charger on, and on a 17 mpg drafting stretch about 17 → 14 (modelled from measured pieces). The earlier road-load arithmetic on this page (40–50 hp at 62 mph) is withdrawn; the truck-side force balance is the subject of an open torque-channel question described on the Specifications page, and no drag figure is quoted here.
Why the Measured Cost Is 0.75 gal/hr, Not the 0.5 gal/hr Document Claim
The original concept document put the ProPower generation load at 0.50 gal/hr. The measured figure is 0.75 gal/hr (band 0.65–0.83). Earlier revisions of this page offered four explanations for the gap; two of them did not survive the campaigns and are recorded here as withdrawn rather than deleted, because the reversal is itself a finding.
| # | Factor | Status after two campaigns |
|---|---|---|
| 1 | Allocation method vs. incremental method | Resolved. The 0.5 gal/hr is not reproduced by any method. Allocation of total fuel during charging gave 0.70 gal/hr (June, single trip). Whole-leg regressions with weak state contrast return lower numbers — 0.505 gal/hr on the full Campaign 2 return at R² 0.58, where half the leg had the umbilical dark — and those fits are underdetermined, not more accurate. The controlled step test, with the state switched on matched ground at held speed, gives 0.75 (0.65–0.83) at R² 0.95 and is the measurement. |
| 2 | PMU power throttling during climbs | Withdrawn. A single June observation of 5.36 kW average output was not reproduced: delivery held 5.69–6.06 kW on every Campaign 1 leg and 5.85 kW mean on Campaign 2, at every ambient through 104°F and on every grade. When the system runs, it runs at rate; duty is set by the trailer's battery-management setpoints and by operator command, not by heat or load. |
| 3 | High ambient temperature → open-loop enrichment | Withdrawn — the finding is reversed. The screen (sustained rich λ and load above 60% and charge-air temperature above 140°F, transients excluded) fired once in Campaign 1, on a mountain grade, and never on a 95°F plains leg; it fired zero times in Campaign 2 at ambients to 93°F, because charge-air temperature never got there. Ten of thirteen Campaign 1 legs ran above 86°F and the mechanism did not appear. The hot-weather fuel penalty that does exist is electrical — the air-conditioning compressor and 12 V system on the same HV bus, 1.28 kW and 0.62 kW mean on the Campaign 2 return — not enrichment. The earlier link to an enrichment analysis on the TrekDrive page is removed; there is none. |
| 4 | Engine BTE at partial load | Restated. The "measured implied BTE of 27.0%" quoted earlier came from the scan tool's engine-power channel, which is computed from fuel rate at a fixed 30.3% and so cannot measure efficiency at all. Campaign 2's torque-derived channel reads 0.31–0.33 on steady cruise windows but under-responds to load changes (about 0.8 of truth on known grades) and is not trusted for this purpose. What the fuel side alone says: 0.75 gal/hr against a 0.65 theory is fully explained by a cruise BTE near 26% instead of 30%, or by a generation chain nearer 77% than 88% — or a share of each. Which it is remains open; the measured cost does not depend on the answer. |
Every crank-side figure on this page (crank-to-outlet ≈ 61–69%, BTE at cruise) rests on the PCM's engine-torque model, which the Campaign 2 closure gates showed responds to a load change at roughly 80% of truth in the cruise regime. Figures built on it carry ±10% and are labelled indicative. The charging cost of 0.75 gal/hr and the 7.70 kWh AC/gal reference are fuel-side measurements and do not inherit this caveat — they are the numbers to quote.
Summary — Key Numbers
If you have questions or comments, please visit the LightShip RV Owners Club on RVForums.com and post them to this topic. I monitor that site regularly and would enjoy your feedback.
