Great project — and a smart instinct to check how the thruster behaves *once the hull is moving*, because that's where most back-of-envelope propulsion plans go wrong. Here's the short version of what I found, followed by a complete HTML page you can drop into your site.
**Headline results (best estimates, anchored to the Shinmaywa rating):**
| MPH | Thrust | kW |
|-----|--------|-----|
| 0 | 3,200 N (720 lbf) | 3.20 |
| 0.5 | ~2,720 N (610 lbf) | ~2.89 |
| 1.0 | ~2,250 N (505 lbf) | ~2.58 |
| 1.5 | ~1,770 N (400 lbf) | ~2.27 |
Key physics: thrust falls roughly **linearly** with speed (blades see reduced angle of attack), hitting zero around **~3.4 mph**. Power draw falls much more slowly (blade profile drag keeps absorbing torque), so **don't size your solar array below ~3.2 kW per unit** if you want full thrust on demand. Against my estimate of your hull's drag (~21 m² effective drag area from the four angled columns + cables), one unit gives a calm-water top speed of **~1.0 mph**, two units **~1.3 mph**. All figures ±25–30% until you calibrate with a dockside bollard test.
Here's the full HTML page:
```html
Seastead Propulsion — Mixer Thrust & Power vs Speed
Seastead Propulsion Analysis
Shinmaywa-type submersible mixer (2.5 m propeller, 3,200 N / 720 lbf bollard thrust,
2–3.2 kW) repurposed as a low-speed thruster — estimated thrust and electrical power draw vs vessel speed.
Bottom line: Thrust falls roughly linearly with boat speed; power draw falls much more slowly.
Expect about 55–60% of rated thrust at 1.5 mph while still drawing ≈2.3 kW.
Estimated calm-water top speed: ≈1.0 mph with one unit, ≈1.3 mph with two (see hull-drag section).
These are engineering estimates (±25–30%) — calibrate with a dockside bollard test and GPS speed trials.
Estimated Performance vs Vessel Speed
Vessel speed (mph)
(m/s)
Thrust (N)
Thrust (lbf)
Power draw (kW)
Propulsion efficiency*
0 (stationary)
0
3,200
720
3.20
— (bollard)
0.5
0.22
2,720
610
2.89
21%
1.0
0.45
2,250
505
2.58
39%
1.5
0.67
1,770
400
2.27
52%
2.0 (extrapolated)
0.89
1,290
290
1.96
59%
*Efficiency = thrust × speed ÷ electrical power. It rises with speed (the rotor unloads),
but absolute thrust margin shrinks. Thrust reaches zero at ≈3.4 mph; power draw never reaches zero
(≈1.1 kW remaining at that point, absorbed by blade profile drag). Running two units simply doubles
both columns.
Optimistic: upper bound from ideal momentum theory assuming the drive holds input power
constant as speed rises. Real fixed-RPM hardware will not beat this.
Best estimate: fixed shaft speed, fixed pitch (how these AC-driven mixers actually behave).
Conservative: shorter effective blade pitch and/or nozzle drag at speed.
Note the power band is narrow in all scenarios: budget ≈3.2 kW per unit for solar sizing.
How the Estimates Were Made
Step 1 — Decode the bollard rating
Momentum theory gives two exact identities for a rotor in stationary water:
power P = T × vᵢ (where vᵢ is the induced velocity at the rotor plane),
and the far slipstream moves at w = 2·vᵢ. With the rated 3,200 W and 3,200 N:
Induced velocity at rotor: vᵢ = 3200/3200 = 1.0 m/s
Bollard slipstream speed: w ≈ 2.0 m/s
Equivalent ideal actuator disk: Aₑ = T³/(2ρP²) ≈ 1.56 m² (D ≈ 1.41 m)
The physical propeller is 2.5 m (disk area 4.91 m²), so the unit performs like an ideal
actuator about 1.4 m across — a figure of merit of ≈0.68 versus a perfect 2.5 m disk
(which would give ≈4,700 N at this power). That is entirely typical for real thruster hardware,
so the published rating looks credible and is used here at face value.
Step 2 — Zero-thrust speed
At fixed shaft speed and fixed pitch, thrust vanishes when the advance speed reaches the blades'
zero-lift pitch speed. For heavily loaded, low-pitch mixers this occurs at roughly
0.6–0.8× the bollard slipstream speed, i.e. Vₔₜ ≈ 1.2–1.6 m/s.
Adopted central value: 1.5 m/s ≈ 3.4 mph.
Step 3 — Decay laws (anchored to the rating)
T(V) = 3200 × (1 − V / 1.5) N (V in m/s)
P(V) = 3.2 × (1 − 0.65 × V / 1.5) kW
Intuition: as the seastead moves forward, water already flows through the rotor, reducing each blade's
angle of attack — like a fan on a moving cart. Torque (and therefore electrical power) decays more
slowly than thrust because blade profile drag keeps absorbing torque even after net thrust reaches zero.
This mirrors standard propeller Kₜ/K₌ chart behavior (e.g., Wageningen B-series).
Steps 4–5 — Bounds
Optimistic bound (only reachable if the drive held power constant): the constant-power
actuator-disk equation, solved for T at each speed:
T³ + 2ρAₑ·P·V·T − 2ρAₑ·P² = 0, ρ = 1025 kg/m³
Conservative bound: effective zero-thrust speed lowered to 1.1 m/s (shorter pitch /
shroud drag), power factor 0.545.
Hull Drag & Expected Top Speed (illustrative)
Based on your description — four 4-ft-wide columns, 13 ft long at 45° (vertical extent ≈9.2 ft each
≈ 147 ft² frontal) plus cable/appendage allowance ≈ 175 ft² (16.3 m²),
at Cₖ ≈ 1.3 ⇒ drag area CₖA ≈ 21 m². Drag = ½ρU²·CₖA ≈ 10,760×U² (N):
Speed
Drag (N)
Drag (lbf)
0.5 mph
540
120
1.0 mph
2,150
480
1.5 mph
4,840
1,090
Solving thrust = drag against the best-estimate thrust curve:
≈1.0 mph with one unit (range 0.9–1.2 across scenarios),
≈1.3 mph with two units (range 1.2–1.6).
If your columns are round rather than square, drag drops ~30–40% and speeds rise accordingly —
send me the real cross-section and cable layout and I'll tighten this.
Station-keeping reality check: one unit can hold position against only ≈1.0 knot of
pure current (two units ≈1.5 kt) before wind and wave loads, which will usually dominate.
Plan mooring for weather, thrusters for maneuvering and slow transit.
Practical Notes for the Design
Electrical: 3.2 kW ≈ 67 A at 48 VDC (133 A at 24 V).
For continuous daytime operation, size the array ≥3.2 kW per unit (4 kW+ realistic after
temperature/soiling derating); add battery buffer for night and gust response.
The throttle lever you get for free: propeller power scales ≈ RPM³.
With a VFD at 70% speed you draw only ≈0.9 kW and still make ≈0.7 mph —
an excellent solar-only cruise mode.
Mounting: the 720 lbf bollard thrust reacts into your structure; keep the unit
fully submerged (the motor cools via the water), add debris screening (accepting a few % thrust loss),
and schedule biofouling cleaning — fouled blades lose significant thrust.
Materials: confirm the stainless grade for your waters; 316 is marginal in warm
seawater, duplex is preferred for permanent immersion.
Recommended Calibration (cheap and decisive)
Dockside bollard test: spring scale or load cell astern; log DC volts × amps.
Confirms the 3,200 N / 3.2 kW anchor point.
Speed runs: in calm water, log DC power and GPS speed-over-ground. Your measured
drag curve plus the linear law T = T₀ − kV lets you refit everything with two numbers.
Optional tow test: measures raw thrust-vs-speed independent of hull drag.
Assumptions & limitations: Manufacturer rating taken at face value (±10% typical);
actual shaft speed, pitch, and shroud configuration unknown; fixed-RPM fixed-pitch operation assumed;
seawater density 1025 kg/m³; wind, wave, and current loads excluded; hull drag estimated from
described geometry only. These are engineering estimates for feasibility screening — not a substitute
for prototype testing. Prepared as design support for a seastead propulsion study.
```
**A few things worth flagging:**
1. **The rating is self-consistent.** Feeding 3,200 N / 3.2 kW through momentum theory implies the unit behaves like an ideal actuator disk ~1.4 m across (figure of merit ≈0.68 vs. the physical 2.5 m disk) — normal for real hardware, so I trusted the spec as the anchor point.
2. **Your intuition on power was right, but the drop is modest** — from 3.2 kW to only ~2.3 kW at 1.5 mph. Practically, size solar for the full 3.2 kW.
3. **The biggest lever is RPM, not speed.** Since power scales ~RPM³, a VFD at 70% speed cuts draw to ~⅓ while costing you only ~30% of top speed. For a solar-only seastead, that's probably your default cruise mode.
If you can share the actual column cross-section (square vs. round), cable count/diameter, or the mixer's rated RPM, I can tighten the drag and zero-thrust-speed estimates considerably.