Table of Contents
Executive Summary
Industry: Consumer Electronics - Smartphone Manufacturing Component: Eccentric weight for linear resonant actuator (LRA) haptic motor Challenge: Produce 0.5g precision weight with ±0.015 mm runout at <$0.08 per piece (2M units/month) Solution: High-speed powder metallurgy with tungsten heavy alloy Results:
- ✅ 50% cost reduction ($0.15 → $0.075 per weight)
- ✅ Runout: 0.012 mm average (20% better than specification)
- ✅ 3× production speed (8 sec → 2.6 sec cycle time)
- ✅ Zero secondary machining (near-net-shape with ±0.010 mm tolerances)
- ✅ Scalability: 2.5M units/month on single production line
Background & Application
The Smartphone Haptic Feedback Revolution
Modern smartphones use linear resonant actuators (LRA) to deliver precise haptic feedback—replacing simple vibration with nuanced tactile sensations that enhance user experience:
- Typing feedback - Mimics mechanical keyboard feel
- Gaming immersion - Directional impact feedback
- UI interaction - Confirms button presses, swipes, selections
- Accessibility - Tactile alerts for hearing-impaired users
LRA motors require a precisely balanced eccentric weight (0.3-0.8g) that oscillates at 150-250 Hz to generate haptic effects. Quality demands are extreme:
Performance Requirements
| Parameter | Specification | Impact if Out-of-Spec |
|---|---|---|
| Mass | 0.50 ± 0.02 g | Wrong vibration intensity |
| Runout (Eccentricity) | 0.45 ± 0.015 mm | Frequency drift, noise |
| Dimensional Tolerance | ±0.010 mm (critical features) | Assembly failure |
| Surface Finish | Ra <0.8 µm | Bearing wear, noise |
| Density | >17.0 g/cm³ | Insufficient inertia |
| Cost Target | <$0.08 per piece | Economics don't work |
Traditional Approach: CNC Micro-Machining
Conventional Manufacturing:
- Material: Tungsten heavy alloy (W-Ni-Fe, 17.5 g/cm³) rod stock
- Process: Swiss-type CNC lathe with live tooling
- Cycle time: 8 seconds per part
- Secondary operations: Deburr, clean, balance check
Pain Points:
- High Cost: CNC machining = $0.15 per weight (material + machine time + tooling wear)
- Material Waste: 45% of tungsten rod becomes chips (expensive to recycle)
- Tooling Wear: Tungsten's hardness (500+ HV) causes rapid carbide tool wear
- Capacity Constraint: Single CNC cell produces max 450K/month (client needs 2M+/month)
- Supply Chain Risk: Single-source CNC supplier, no backup capacity
Client Goal: Flagship smartphone model launching in 6 months requires 24M weights annually (2M/month peak). Cost target: <$0.08 per weight. Existing CNC supplier can only deliver 40% of volume.
Powder Metallurgy Solution
Material Selection: Tungsten Heavy Alloy PM
We proposed press-and-sinter tungsten heavy alloy to replace CNC machining:
Material Composition:
- 93% Tungsten (W)
- 4.5% Nickel (Ni)
- 2.5% Iron (Fe)
Why This Alloy:
- ✅ Density: 17.2 g/cm³ (98% of theoretical 17.5 g/cm³)
- ✅ Ductile matrix: Ni-Fe phase holds brittle tungsten particles together
- ✅ Machinability: If needed, easier than pure tungsten (though PM goal is near-net-shape)
- ✅ Non-magnetic: Fe content low enough to avoid interference with motor magnets
- ✅ Cost-effective: Powder form uses 98% of material vs. 55% for machining
Manufacturing Process: High-Speed Compaction
Production Flow:
1. Powder Preparation
- Tungsten powder: 3-8 micron (fine for high density)
- Ni-Fe powder: Pre-alloyed, 5-12 micron
- Mix for 4 hours in ball mill with 0.5% wax lubricant
- Granulate to improve flow characteristics
2. High-Speed Compaction
- Equipment: 40-ton servo-electric press (faster than hydraulic)
- Die: 4-cavity hardened steel die (produce 4 weights per stroke)
- Compaction pressure: 600 MPa
- Green density: 11.5 g/cm³ (67% of final)
- Cycle time: 2.5 seconds per stroke = 0.625 sec per weight
3. Sintering
- Atmosphere: Hydrogen (H₂) gas at 1,450°C
- Time: 2 hours in batch furnace (500 parts per batch)
- Mechanism: Ni-Fe melts (liquid phase sintering), tungsten particles rearrange
- Shrinkage: 20% linear (predictable, compensated in die design)
- Final density: 17.2 g/cm³ (98% theoretical)
4. Quality Inspection
- 100% automated runout measurement (laser + rotary stage, 1.2 sec per part)
- Weight check (every 50th part, ±0.005g tolerance)
- Visual inspection (automated camera, detects cracks/chips)
Total Cycle Time: 2.6 seconds per weight (including inspection)
Design Optimization
Eccentric Weight Geometry
Specifications:
| Dimension | Value | Tolerance | PM Achievability |
|---|---|---|---|
| Outer Diameter | 4.80 mm | ±0.010 mm | ✅ ±0.008 mm |
| Inner Bore | 1.50 mm | ±0.008 mm | ✅ ±0.006 mm |
| Length | 8.00 mm | ±0.015 mm | ✅ ±0.012 mm |
| Eccentric Offset | 0.45 mm | ±0.015 mm | ✅ ±0.012 mm |
| Mass | 0.50 g | ±0.02 g | ✅ ±0.015 g |
Design Modifications for PM:
Original CNC Design:
- Sharp internal corners (0.1 mm radius)
- Surface finish Ra 0.4 µm (from turning)
- Bore tolerance ±0.005 mm (tight for assembly)
Optimized PM Design:
- Internal corner radii increased to 0.2 mm (better powder flow, less stress concentration)
- Surface finish Ra 0.6-0.8 µm (acceptable for bearing interface)
- Bore tolerance relaxed to ±0.008 mm (assembly confirmed compatible)
- Added 0.05 mm chamfer on bore edges (prevents chipping during sintering)
Net Result: 100% form-fit-function compatible with motor assembly. No tooling or process changes required at motor manufacturer.
Performance Validation
Dimensional Accuracy Results
Inspection Data (10,000-part production sample):
| Feature | Specification | Mean | Std Dev (σ) | Cpk | Pass Rate |
|---|---|---|---|---|---|
| Outer Diameter | 4.80 ± 0.010 mm | 4.798 mm | 0.003 mm | 2.22 | 100% |
| Inner Bore | 1.50 ± 0.008 mm | 1.502 mm | 0.002 mm | 3.00 | 100% |
| Length | 8.00 ± 0.015 mm | 7.998 mm | 0.004 mm | 2.50 | 100% |
| Eccentric Offset | 0.45 ± 0.015 mm | 0.448 mm | 0.005 mm | 2.00 | 99.8% |
| Mass | 0.50 ± 0.02 g | 0.498 g | 0.006 g | 2.78 | 100% |
Quality Achievement: All features meet Cpk >1.67 (automotive/electronics standard). Eccentric offset Cpk 2.0 indicates very capable process.
Runout (Dynamic Balance) Performance
Runout Measurement (100% inspection, 50K sample):
| Metric | Specification | CNC Machined | PM Sintered | Improvement |
|---|---|---|---|---|
| Mean Runout | <0.015 mm | 0.013 mm | 0.012 mm | ✅ 8% better |
| Std Deviation (σ) | — | 0.004 mm | 0.003 mm | ✅ 25% tighter |
| Max Runout (worst case) | <0.020 mm | 0.018 mm | 0.017 mm | ✅ Improved |
| % Within Spec | 100% | 99.2% | 99.6% | ✅ Better yield |
Why PM Achieves Better Runout:
- Symmetric sintering shrinkage (material compresses uniformly from all directions)
- No machining-induced stresses (CNC cutting can warp thin walls)
- Isotropic material properties (no grain direction from metal flow)
Motor Performance Testing
LRA Motor Assembled with PM Weights (vs. CNC Baseline):
| Test Parameter | CNC Weight | PM Weight | Result |
|---|---|---|---|
| Resonant Frequency | 175 Hz | 174 Hz | ✅ Equivalent |
| Vibration Amplitude | 2.2 G | 2.25 G | ✅ +2.3% (within tolerance) |
| Frequency Stability | ±3 Hz | ±2.5 Hz | ✅ 17% more stable |
| Noise Level | 38 dB(A) | 37 dB(A) | ✅ Quieter |
| Power Consumption | 85 mW | 84 mW | ✅ Equivalent |
| Life Test (5M cycles) | 0 failures | 0 failures | ✅ Pass |
User Perception Testing: Blind A/B test with 50 users showed no detectable difference in haptic feedback quality between CNC and PM weights. PM approved for production.
Cost Analysis
Detailed Cost Breakdown (Per Weight, 2M Units/Month)
| Cost Element | CNC Machining | PM Sintering | Savings |
|---|---|---|---|
| Raw Material | $0.055 (rod stock, 45% waste) | $0.032 (powder, 2% waste) | -$0.023 |
| Processing | $0.065 (machine time + labor) | $0.028 (press + sinter batch) | -$0.037 |
| Tooling Amortization | $0.012 (carbide inserts) | $0.008 (die wear) | -$0.004 |
| Quality Inspection | $0.008 (manual + automated) | $0.005 (100% automated) | -$0.003 |
| Scrap/Rework | $0.010 (0.8% reject rate) | $0.002 (0.4% reject rate) | -$0.008 |
| Total Cost per Weight | $0.150 | $0.075 | -$0.075 (50%) |
Annual Savings at 24M Units: $1,800,000
Tooling Investment: $85,000 (PM dies + sintering fixtures) vs. $25,000 (CNC tooling) Break-Even Volume: ~800K units (achieved in 4 months at 2M/month production rate)
Production Scaling Success
Volume Ramp Results
Phase 1: Pilot Production (Month 1-2)
- Volume: 50K units/month
- Yield: 96.5%
- Issues: Runout variation ±0.018 mm (refinement needed)
Phase 2: Production Ramp (Month 3-4)
- Volume: 500K units/month
- Yield: 98.8%
- Implemented automated powder feed system
- Die maintenance every 80K cycles (vs. initial 50K)
Phase 3: Full Production (Month 5+)
- Volume: 2.5M units/month (exceeds target)
- Yield: 99.4%
- Runout: 0.012 mm average (consistent)
- Added 2nd production line for redundancy
Capacity & Flexibility Benefits
Production Line Configuration:
- 2× servo-electric presses (4-cavity dies) = 8 weights/stroke
- Cycle time: 2.5 seconds → 11,520 weights/hour per press
- Operating schedule: 22 hours/day, 6 days/week
- Capacity: 3.0M weights/month (20% buffer above peak demand)
Flexibility Advantage:
- Quick die change: 45 minutes to switch between weight variants (different offsets for different phone models)
- CNC required 4-6 hours to reprogram + new fixtures for each variant
- PM enables cost-effective multi-SKU production
Challenges & Solutions
Challenge 1: Sintering Distortion Control
Problem: First production batches showed ±0.025 mm runout (67% out of spec).
Root Cause: Non-uniform sintering temperature across batch furnace (±15°C variation) caused differential shrinkage.
Solution:
- Installed convection fan system in furnace (improved temperature uniformity to ±5°C)
- Redesigned part tray with uniform spacing (eliminates shadowing effects)
- Added thermocouple monitoring at 9 furnace locations
- Result: Runout improved to 0.012 mm ± 0.003 mm (99.6% yield)
Challenge 2: Bore Surface Finish
Problem: As-sintered bore surface Ra 1.2-1.5 µm exceeded Ra 0.8 µm specification (motor shaft bearing surface).
Root Cause: Sintering atmosphere contained trace oxygen (oxidized tungsten particles at bore surface).
Solution:
- Upgraded hydrogen purity from 99.5% to 99.95% (lower O₂ contamination)
- Added palladium catalyst purifier in gas line
- Implemented core rod pre-lubrication (smoother bore surface after ejection)
- Result: Bore surface improved to Ra 0.6-0.8 µm (within spec, no secondary polishing needed)
Challenge 3: High-Speed Die Wear
Problem: Initial die life only 40K cycles (target 100K) due to abrasive tungsten powder.
Root Cause: Tungsten's extreme hardness (500 HV) wore die surfaces rapidly.
Solution:
- Upgraded die material from D2 tool steel to tungsten carbide inserts (core rods)
- Applied DLC (diamond-like carbon) coating to punch faces
- Implemented automated die lubrication every 20 cycles
- Result: Die life extended to 120K cycles (exceeds target, reduces tooling cost/part)
Environmental & Sustainability Impact
Material Efficiency Comparison
| Metric | CNC Machining | PM Sintering | Improvement |
|---|---|---|---|
| Material Utilization | 55% (45% chips) | 98% (2% recyclable) | +43 percentage points |
| Tungsten Waste | 0.41g per weight | 0.01g per weight | -97.6% |
| Annual Tungsten Saved | — | 9,840 kg | $590,400 value |
| Energy per Part | Baseline | 0.6× (less machine time) | -40% energy |
| Coolant/Oil Use | 15 ml/part | 0 ml/part | Zero coolant waste |
Sustainability Win: PM's near-net-shape approach saves 9.8 tons of tungsten annually—a critical metal with supply chain constraints. Client received recognition for "Green Manufacturing Innovation."
Customer Testimonial
"We were skeptical that powder metallurgy could achieve the precision needed for smartphone haptics, but the results exceeded expectations. Not only did we cut costs in half, but quality and consistency actually improved. The rapid scaling capability allowed us to meet our flagship launch deadline—something our CNC supplier couldn't guarantee. PM is now our standard for all LRA motor weights across our product line."
— Dr. Zhang Wei, Senior Hardware Engineer, [Major Smartphone OEM]
Key Takeaways for Consumer Electronics PM
When to Choose PM for Miniature Components
✅ Good Fit:
- High-volume production (>500K units/month)
- Small, precision parts (0.1-5g, tolerances ±0.010-0.025 mm)
- Expensive materials (tungsten, cobalt, stainless) where waste reduction critical
- Geometric complexity (eccentric shapes, multi-level features)
- Cost-sensitive consumer products
⚠️ Challenging:
- Ultra-tight tolerances (<±0.005 mm) may require post-sinter grinding
- Very small features (<0.3 mm) difficult with conventional PM (consider MIM)
- Low volumes (<100K units) where CNC or MIM more economical
- Surface finish <Ra 0.4 µm (requires secondary polishing)
Design Guidelines for Miniature PM Parts
- Tolerance Allocation: ±0.010-0.015 mm achievable as-sintered; ±0.005 mm requires sizing/grinding
- Wall Thickness: Minimum 0.4-0.5 mm for tungsten alloys (thinner prone to cracking)
- Corner Radii: Use 0.15-0.25 mm internal radii (aids powder flow, reduces cracking)
- Draft Angles: Not required for PM (unlike MIM), but 1-2° can improve ejection
- Bore Finish: Specify Ra 0.6-1.0 µm as-sintered; Ra <0.5 µm needs polishing
- Mass Tolerance: ±0.015-0.025g achievable with density control
- Shrinkage Compensation: 18-22% linear shrinkage (varies by material, precisely predictable)
Next Steps: Explore PM for Your Consumer Electronics Application
Powder metallurgy is enabling next-generation consumer electronics through precision miniature components at unprecedented cost efficiency.
Our Consumer Electronics PM Capabilities: ✅ Miniature parts (0.3-5g, ±0.010 mm tolerances) ✅ Tungsten, stainless, aluminum, copper alloys ✅ High-speed production (1M+ units/month capacity) ✅ 100% automated inspection (vision + dimensional) ✅ Rapid prototyping to mass production (4-6 week ramp)
Request Consumer Electronics PM Feasibility Study →
Engineering Support: Free design review for PM suitability Certifications: ISO 9001:2015, IATF 16949 for electronics manufacturing
Internal Links
- Consumer Electronics PM Components - Overview of PM in consumer devices
- Tungsten Heavy Alloy Materials - Material properties and applications
- High-Volume Miniature PM - Production capabilities
- Powder Metallurgy vs CNC Machining - Detailed process comparison
- Medical Device Miniature Components - Another precision PM application
Frequently Asked Questions
Can PM achieve the same precision as CNC micro-machining?
For features ±0.010-0.015 mm, yes—PM matches CNC capability as-sintered. For tighter tolerances (±0.005 mm), PM can use sizing or light grinding, still at 30-40% lower cost than full CNC. For ultra-precision (<±0.003 mm), CNC or PM+grinding hybrid is better.
What's the smallest PM part that can be produced economically?
Conventional PM handles parts down to ~0.3g (2-3 mm diameter). Below this, metal injection molding (MIM) becomes more suitable. However, tungsten's high density enables 0.5-1.0g parts in very compact envelopes (5-8 mm dimensions).
How does PM surface finish compare to CNC?
PM as-sintered: Ra 0.6-1.2 µm typical. CNC turning: Ra 0.2-0.6 µm. For non-bearing surfaces, PM finish is acceptable. For bearing/sealing surfaces, PM can achieve Ra 0.4-0.8 µm with optimized process, or add light polishing if <Ra 0.4 µm needed.
What production volumes justify PM tooling investment for small parts?
Break-even typically 200K-500K units for miniature tungsten components. At 1M+ volume, PM delivers 40-60% cost savings vs. CNC. For prototyping (<50K), CNC is more economical. Consider PM when scaling from prototype to mass production.
Can PM handle other consumer electronics components besides haptic weights?
Yes! PM works well for: camera module structural parts (aluminum PM), phone frame components (stainless PM), connector housings (bronze/brass PM), hinge mechanisms (stainless PM), speaker magnets (ferrite PM). Any high-volume, precision metal part is a candidate.
Related Resources
Use these internal links to keep moving through the most relevant guides, service pages, and technical references for this topic.
Consumer Electronics PM Parts
See where miniature PM components fit haptics, wearables, hinge hardware, and precision consumer assemblies.
FC-0000 Soft Iron Guide
Review a soft magnetic PM material route often considered in compact motor and electronic component programs.
Case Studies
Browse more PM project examples covering medical, robotics, aerospace, and industrial applications.
Request a Quote
Send your miniature PM component geometry, tolerance, and annual demand for feasibility review and pricing support.