Marine Propeller Hub: C95800 Bronze 5-Axis CNC Case Study

A propeller hub for a workboat/trawler application. The requirements are specific enough to narrow the choices considerably: continuous seawater immersion, cavitation exposure at the blade roots, and classification society approval. This case covers the machining approach for a C95800 nickel aluminum bronze hub, from material selection through final balancing.

Project at a Glance

Key Parameters

Item	Spec
Application	Workboat / trawler propeller hub
Primary Material	C95800 nickel aluminum bronze (NiAlBr)
Standard	ASTM B148
Hub Bore Tolerance	H7 (+0.025 / 0 mm)
Blade Slot Positional Accuracy	±0.05 mm
Surface Roughness (mating)	Ra ≤ 1.6 μm
Compliance	DNV/GL, Lloyd's Register, ISO 9001
Annual Volume	10 – 200 pcs

Lead Time

Phase	Duration
Prototype (1 – 5 pcs)	15 – 20 days
Production batch (10+ pcs)	6 – 8 weeks
Testing & certification	Included above
Material form	Cast blank (sand or investment cast)
Machining center	5-axis CNC + CNC boring mill
Balancing	Static & dynamic (ISO 1940 G6.3)
Corrosion testing	Salt spray per ASTM B117

1. Material Selection

Propeller hubs are submerged in seawater for their entire service life. The material must resist general corrosion, pitting, and cavitation erosion while providing sufficient strength to transmit the torque from the shaft to the blades. Classification society approval further constrains the options.

Material	Seawater Corrosion Resistance	Strength	Cavitation Erosion Resistance	Machinability	Cost Index
C95800 (Ni-Al bronze)	Excellent — forms protective aluminum oxide film	≥ 586 MPa tensile	Good — superior to stainless in high-velocity flow	Moderate — abrasive on tools	1.0x
C63000 (Al-bronze)	Very good	≥ 620 MPa tensile	Moderate	Moderate	0.85x
316L stainless	Good — susceptible to pitting in stagnant seawater	≥ 485 MPa tensile	Fair — poor in high-velocity cavitation zones	Good	0.7x
17-4 PH stainless	Moderate — requires coating for long-term immersion	≥ 1,000 MPa tensile	Fair	Moderate	0.9x

Why C95800: The combination of excellent seawater corrosion resistance, good cavitation erosion performance, and established DNV/GL approval makes C95800 the standard choice for marine propeller hubs. While it costs more than stainless steel and is harder on cutting tools, the service life in seawater is substantially longer, which matters more for a part that is difficult and expensive to replace in the field.

2. Why C95800 for This Application

C95800 (UNS C95800), also known as nickel aluminum bronze (NiAlBr), is a copper-based alloy with additions of nickel, aluminum, iron, and manganese. The alloy is defined by ASTM B148 and is widely used for marine propellers, pump impellers, and valve components in seawater service.

Property	Value	Design Implication
Tensile Strength	≥ 586 MPa	Adequate for workboat propeller shaft torque transmission
Yield Strength (0.2%)	≥ 241 MPa	Provides margin against plastic deformation under peak loads
Elongation	≥ 15%	Reasonable ductility for a cast bronze — absorbs impact from debris
Hardness	HB 170–210	Wear-resistant at blade root interfaces
Density	7.64 g/cm³	Comparable to steel — no unusual handling requirements
Thermal Conductivity	26.6 W/m·K	Moderate — heat dissipates during machining
Corrosion Rate (seawater)	< 0.05 mm/year	Long service life in continuous immersion

The nickel addition (typically 4.5–5.5%) improves general corrosion resistance and strengthens the alloy matrix. Aluminum (8.5–9.5%) provides de-aluminization protection by forming a thin, adherent aluminum oxide film on the surface. This film is self-healing in oxygenated seawater and is the primary reason C95800 outperforms stainless steels in continuous immersion service.

Cavitation erosion resistance is where C95800 has a clear advantage over austenitic stainless steels. In high-velocity seawater flow — particularly at the blade root slots where pressure fluctuations are most severe — C95800 retains its surface integrity considerably longer than 316L or even 17-4 PH. This is partly due to the aluminum oxide film and partly due to the alloy's ability to work-harden locally under cavitation impact.

Casting quality matters. C95800 is typically supplied as a sand casting or investment casting. Internal defects — shrinkage porosity, gas porosity, and microshrinkage — are common in bronze castings. Ultrasonic testing (UT) is required to verify casting soundness before committing machining time. A casting with internal porosity that passes visual inspection can still fail during UT or, worse, in service. Source castings from foundries with marine certification experience.

3. Machining Strategy

3.1 5-Axis CNC Milling — Hub Geometry

The propeller hub has a complex geometry: a tapered central bore, multiple blade root slots arranged radially, oil distribution channels, and external flange faces. 5-axis CNC milling handles the curved surfaces and angled features that would require multiple setups on a 3-axis machine.

Rough mill: Remove bulk material from the casting. Leave 1.0–1.5 mm stock on all machined surfaces. Focus on establishing datum surfaces first.
Semi-finish mill: Machine blade root slots, oil channels, and external profile. Leave 0.3–0.5 mm on mating surfaces and bore.
Finish mill: Final pass on external surfaces, flange faces, and blade root slot profiles. Ball-nose end mill 5-axis contouring for blade slots.
Surface finishing: Manual polishing of all mating surfaces to Ra ≤ 1.6 μm. Blade root slot surfaces require particular attention for proper blade fit.

3.2 CNC Boring — Hub Bore

The hub bore (typically tapered for a taper-lock fit to the propeller shaft) is the most dimensionally critical feature. The H7 tolerance (+0.025/0 mm) requires precision boring followed by honing.

Tooling: Carbide inserts with PVD coating (TiAlN or AlTiN). The aluminum oxide in C95800 is highly abrasive — uncoated carbide tools wear rapidly.
Boring strategy: Single-point precision boring in multiple passes. Leave 0.02–0.03 mm for honing.
Honing: Single-pass mandrel hone to final bore dimension and surface finish.
Coolant: Water-soluble coolant with good lubricity. Maintain minimum 12–15 L/min flow. Insufficient coolant contributes to work hardening and accelerated tool wear.

3.3 Tool Wear — The Abrasive Bronze Problem

C95800 contains hard aluminum oxide particles distributed through the copper-nickel matrix. These particles act as an abrasive during machining, causing tool wear that is significantly faster than with carbon steel or even stainless steel at equivalent hardness.

Use carbide inserts with PVD coating (TiAlN preferred). Tool life improvement over uncoated carbide is 2–3x.
Reduce cutting speeds by 20–30% compared to free-machining bronze. Higher speeds do not improve productivity because tool changes become the bottleneck.
Maintain steady coolant flow to flush chips and prevent chip re-cutting. Recutting chips accelerates insert edge wear.
Monitor tool wear with in-process gauging where possible. Bore diameter drift is the first indicator of boring tool wear.

Surface finishing tip: After CNC machining, the mating surfaces (bore, flange face, blade root slots) require manual polishing to achieve Ra ≤ 1.6 μm. Start with 120-grit abrasive, progress through 240, 400, and finish with 600-grit. The final polish also helps close any residual porosity on the machined surface, improving corrosion resistance in the field.

4. Quality Testing

Test	Method	Criteria	Frequency
CMM dimensional inspection	Coordinate measuring machine	Hub bore (H7), blade slot position (±0.05 mm), flange face flatness, all critical dimensions per drawing	100% of units
Ultrasonic testing (UT)	Contact UT per ASTM E2375	No indications exceeding reference level. Verifies casting soundness — no shrinkage porosity or gas defects.	100% of castings (pre-machining)
Hardness testing	Brinell HB, per ASTM E10	HB 170–210 (per ASTM B148)	Per piece or per lot
Salt spray corrosion test	ASTM B117, 1,000 hours	No red rust or significant corrosion product. Surface condition documented photographically.	Per lot (sample)
Static & dynamic balancing	ISO 1940 grade G6.3	Residual unbalance within G6.3 limits for the operating speed range	100% of units
Visual inspection	Surface examination, 10x magnification	No visible porosity on machined surfaces, no surface cracks, no embedded debris	100% of units
Blade slot angular accuracy	CMM or dedicated fixture with dial indicator	±0.5° from nominal blade angle	100% of units
Material certification	Foundry mill cert + PMI verification	Chemical composition per ASTM B148, traceable to heat/lot number	Per casting lot

UT testing before machining. Ultrasonic testing should be performed on the raw casting before any machining. If a casting has internal defects, discovering this after machining is a waste of time and material. Some foundries provide UT reports with the casting; if not, factor this into the timeline.

5. Cost Drivers

Cost Driver	% of Unit Cost	Notes
Raw material / casting	25–35%	C95800 castings are expensive. Marine-certified foundries charge a premium. Material cost is the single largest variable.
5-axis CNC machining	30–40%	Complex geometry requires multiple setups and long cycle times. Tool wear on abrasive bronze adds to cost. Low volume means no fixture amortization.
CNC boring & honing	5–10%	Precision bore work with H7 tolerance. Boring tools and honing mandrels are setup costs amortized over the batch.
Surface finishing & polishing	5–10%	Labor-intensive manual polishing to Ra ≤ 1.6 μm on all mating surfaces. Cannot be fully automated for complex hub geometry.
Testing & certification	10–15%	UT, CMM, hardness, salt spray, material certification. DNV/Lloyd's Register surveyor fees if third-party witnessing is required.
Balancing	5–8%	Static and dynamic balancing per ISO 1940 G6.3. Balance correction (drilling or milling) adds time if initial unbalance is significant.

Low volume is the primary cost driver for this type of part. At 10–200 pieces per year, there is limited opportunity to amortize fixture costs, optimize tool paths for cycle time reduction, or negotiate volume discounts on castings. The testing and certification burden (10–15% of cost) is proportionally higher at low volume because fixed setup fees for UT, CMM programming, and salt spray testing do not decrease with quantity.

6. Common Mistakes

Mistake 1: Using general-purpose tooling without PVD coating. C95800 contains hard aluminum oxide particles that act as an abrasive during machining. Uncoated carbide inserts wear rapidly — tool life can be less than half that of PVD-coated inserts. The cost difference per insert is small relative to the machine downtime for tool changes. Use TiAlN or AlTiN coated carbide throughout.

Mistake 2: Insufficient coolant flow causing work hardening. C95800 is not as prone to work hardening as 17-4 PH, but it does occur when coolant supply is interrupted or flow is inadequate. The surface generates heat, the local area hardens, and subsequent passes cut into hardened material, accelerating tool wear and degrading surface finish. Maintain a steady 12–15 L/min flow to the cutting zone at all times.

Mistake 3: Skipping UT testing on the casting. Internal casting defects — shrinkage porosity and gas porosity — are common in bronze castings and are not visible on the surface. A hub with internal porosity can pass dimensional inspection and visual inspection, then fail structurally at sea. UT is the only practical way to verify casting soundness before machining. This is not a step to skip.

Mistake 4: Incorrect blade slot geometry. The blade root slots must match the propeller blade pitch angle and root profile precisely. Angular errors exceeding ±0.5° result in poor blade fit, reduced propulsive efficiency, and uneven load distribution that can lead to blade fatigue. Verify blade slot geometry on CMM against the blade manufacturer's specifications before committing to production.

Mistake 5: Not compensating for material springback in thin-wall sections. C95800 has moderate elasticity for a bronze alloy. In thin-wall sections of the hub (typically the web between bore and blade slots), material springback after machining can cause bore distortion or blade slot width variation. Leave finishing stock on both sides of thin walls and take light final passes to minimize residual stress.

7. Production Timeline

Phase	Duration	Deliverable
DFM review & quotation	3–5 days	Updated drawing with DFM notes, formal quote with testing and certification breakdown
Casting procurement	10–15 days	C95800 cast blank with foundry cert, UT report, chemical composition verification
UT verification (incoming)	2–3 days	UT report confirming casting soundness before machining
Fixture design & manufacture	5–7 days	5-axis fixtures, boring tools, honing mandrels, CMM program
Prototype machining (1–3 pcs)	5–8 days	Machined hubs with dimensional report, ready for testing
Testing & balancing (prototype)	3–5 days	CMM report, UT, hardness, salt spray (sample), balancing cert
Customer approval / FAI sign-off	3–7 days	Approved first article with full documentation package
Production machining (batch)	3–4 weeks	Batch of finished hubs per order quantity
Production testing & balancing	1–2 weeks	100% CMM, UT, balancing certs per unit; salt spray per lot
Total (prototype: 1– 3 pcs)	4–6 weeks	Finished hubs with full documentation
Total (production: 10+ pcs)	8–12 weeks	Batch delivery with lot documentation and certification

About this case study This technical analysis is based on marine propeller hub programs produced at Sinbo Precision. Specific customer details, vessel configurations, and proprietary design features have been modified or omitted. All process parameters, material data, and tolerance values are representative of typical workboat and trawler propeller hub requirements.

Have a Similar Project?

Send us your drawing — we'll return a DFM review and quote within 3 business days.

Get a Quote →