ADR-022: Linear Bearing Selection

Status

Accepted

Context

ADR-021 establishes dual 8mm smooth rods as the reference motion system. This ADR addresses what rides on those rods - the linear bearings that enable smooth, low-friction motion.

Bearing Options Available

  1. Ball bearings (LM8UU, LM8LUU): Recirculating steel balls in a cage
  2. Polymer bushings (IGUS Drylin): Self-lubricating polymer sleeves
  3. Printed bearings: DIY polymer sleeves with adjustable clamping
  4. Bronze bushings: Oil-impregnated sintered metal

Scavenger Context

Two donor Prusa-style printers typically provide: - 24× LM8UU bearings (12 per printer) - Amalgam Dual-8 requires 16 bearings - Surplus of 8 bearings available for spares or selection

Rod Material Requirements

Critical insight: Ball bearings and polymer bushings have different rod material requirements.

Ball Bearings (LM8UU/LM8LUU) Require Hardened Steel

Linear ball bearings use the rod as their inner race. The steel balls “plow” into soft materials, causing rapid wear.

Rod Material Typical Hardness Ball Bearing Life Factor Notes
Hardened steel (induction) HRC 60-65 1.0 (100%) Required for ball bearings
Hardened stainless (440C treated) HRC 54-60 0.7-1.0 Acceptable
Stainless 304/316 (annealed) HRC 20-25 Very poor Galling risk - do not use
Aluminum HRC ~15 Incompatible Will destroy bearings

Key finding: Using HRC 40 shaft with ball bearings gives only 20% of rated life compared to HRC 60. (Linear Motion Tips)

Warning: Standard 304/316 stainless steel is prone to galling (cold welding) with steel ball bearings. The balls create grooves and the rod surface tears. (Nickel Institute)

Polymer Bushings (IGUS Drylin) Work on Multiple Materials

IGUS Drylin bushings use self-lubricating polymer (iglidur J4) that glides rather than rolls. This enables compatibility with softer materials:

Rod Material IGUS Compatibility Notes
Hardened steel ✅ Excellent Works perfectly
Stainless steel (any grade) ✅ Excellent No galling risk
Aluminum ✅ Good Lightweight option
Carbon fiber ✅ Good Premium/experimental
Chrome-plated steel ✅ Excellent Common scavenged rods

Source: IGUS Official - Drylin RJ4JP

Can Scratched Rods Be Salvaged?

A common scavenger question: can damaged hardened rods be reused?

Damage Type Ball Bearings IGUS Bushings
Light scratches May click/jam on grooves Can bridge over (larger contact area)
Deep gouges Unusable - will damage balls May work, test carefully
Bent rod Unusable Unusable
Surface rust Clean and test Clean and test

The IGUS advantage: Because polymer bushings have a large contact area (vs point contact of balls), they can “bridge” minor surface imperfections that would jam ball bearings.

The Long Bearing Question

LM8UU vs LM8LUU comparison:

Spec LM8UU LM8LUU
Length 15mm 24mm
Inner diameter 8mm 8mm
Outer diameter 15mm 15mm
Ball rows 4 6
Load capacity Lower ~50% higher
Moment resistance Lower Higher

Long bearings (LM8LUU) provide better moment resistance - they resist tilting and racking forces better than standard length bearings.

Decision

We recommend LM8LUU for X-axis and LM8UU for Y and Z axes as the reference specification.

Bearing Assignment

Axis Bearing Quantity Rationale
X (Plough carriage) LM8LUU 4 Highest dynamic loads, pitch/yaw resistance critical
Y (gantry, per side) LM8UU 4 (8 total) Dual rods provide stability, standard bearings sufficient
Z (bed support) LM8UU 4 Slow movement, minimal dynamic loads
Total 16

Why LM8LUU for X Only?

The X-axis carriage (Plough) experiences the most severe dynamic forces:

  1. Y-acceleration creates pitch moment (nose-dive)
  2. X-acceleration creates lateral shift forces
  3. Extruder retraction creates small Z-moments
  4. Belt tension creates constant preload

Long bearings on X provide: - Better pitch resistance during Y-acceleration - More stable carriage during direction changes - Reduced “racking” tendency under load

Why LM8UU is Sufficient for Y and Z?

Y-axis (gantry): - Dual rods per side already provide excellent moment resistance - Load is distributed across 4 bearings per side - Lower acceleration than X-axis - Long bearings would add cost without proportional benefit

Z-axis: - Very slow movement (typically 5-10mm/s) - Primarily static load (bed weight) - No significant dynamic moments - Standard bearings have decades of proven service in this role

Scavenged Bearing Protocol

Donor printer bearings should be:

  1. Inspected: Roll on a clean rod section - should be smooth, no grinding
  2. Cleaned: Flush with isopropyl alcohol, remove old grease and debris
  3. Dried: Allow to fully dry before regreasing
  4. Regreased: Light machine oil or lithium grease
  5. Tested: Should roll smoothly with minimal resistance

Reject bearings that: - Have visible rust or pitting - Make grinding sounds when rolled - Have excessive play (wobble on rod) - Have flat spots or notchy feel

With 24 bearings from 2 donors, you can afford to reject questionable units.

Consequences

Benefits

  1. Optimized performance: Long bearings where they matter most (X)
  2. Cost-effective: Standard bearings where sufficient (Y, Z)
  3. Fully scavengeable: All bearings from 2 donor printers
  4. Proven technology: Ball bearings have decades of RepRap validation
  5. Low maintenance: Occasional regrease, no adjustment needed

Trade-offs

  1. LM8LUU may need purchase: Some donors have only LM8UU
  2. Slightly longer X-carriage: LM8LUU housings are 9mm longer
  3. Not “pure” scavenger: May need to buy 4× LM8LUU (~$5-8 AUD)

Scavenger Decision Tree

Use this flowchart to select your bearing/rod combination:

What does your donor printer have?
│
├─► Hardened steel rods + LM8UU bearings (Prusa clone, Anet A8)
│   │
│   └─► Are the rods in good condition?
│       │
│       ├─► YES: Use scavenged rods + bearings (clean & regrease)
│       │       Cost: $0 | Result: Standard "printer hum"
│       │
│       └─► NO (scratched/grooved):
│           │
│           ├─► Option A: Try IGUS bushings on scratched rods
│           │   Cost: ~$30-40 | Bridges minor scratches
│           │
│           └─► Option B: Buy stainless rods + IGUS
│               Cost: ~$70-90 | Silent, maintenance-free
│
├─► V-slot rollers, no rods (Ender 3, CR-10)
│   │
│   └─► Must purchase rods. Choose:
│       │
│       ├─► Hardened steel rods + LM8UU
│       │   Cost: ~$50-70 | Standard ball bearing setup
│       │
│       └─► Stainless steel rods + IGUS (Recommended for V-slot donors)
│           Cost: ~$70-90 | Silent, corrosion-proof
│
└─► Mixed donors (e.g., Ender + Prusa clone)
    │
    └─► Use rods from rod-bearing donor
        Use best bearings available
        Consider IGUS if rods are damaged

Quick Reference: Two Valid Paths

Path Rods Bearings Cost Characteristics
Scavenger (reference) Hardened steel (donor) LM8UU/LM8LUU (donor) $0-8 Standard printer sound, proven reliability
Silent/Premium Stainless steel (buy) IGUS Drylin (buy) ~$70-90 Silent operation, maintenance-free, corrosion-proof

Both paths produce quality prints. Choose based on your donor situation and noise preferences.

Alternative Paths (Documented for Reference)

Alternative A: All LM8UU (Pure Scavenger)

X: 4× LM8UU
Y: 8× LM8UU
Z: 4× LM8UU
Total: 16× LM8UU

Pros: - Guaranteed scavengeable from 2 donors - Zero bearing purchase cost - Simpler parts list

Cons: - Reduced X-carriage stability - May see more racking on X during fast Y moves - Acceptable for quality-focused (not speed-focused) printing

Verdict: Acceptable fallback if LM8LUU unavailable.

Alternative B: IGUS Drylin + Stainless Steel (Silent Build Path)

This alternative is recommended for V-slot donors (Ender 3, CR-10) that provide no scavengeable rods or bearings.

Configuration:

Rods: 8× stainless steel 8mm (420mm length)
Bearings: 16-22× IGUS Drylin RJ4JP-01-08
Cost: ~$70-90 AUD total

Pros: - Silent operation: No “printer hum” from ball bearings - Maintenance-free: Self-lubricating polymer, no grease needed - Corrosion-proof: Stainless rods + polymer = zero rust risk - Tolerant of imperfect rods: Can bridge minor scratches - Works with any stainless grade: No galling risk (unlike ball bearings) - Vibration damping: Polymer absorbs some mechanical noise

Cons: - Higher friction: 2-3× static friction (“stiction”) vs ball bearings - Higher cost: ~$70-90 vs $0 for scavenged - Not scavengeable: Must be purchased new - Requires housing compliance: Needs slight clamping pressure to seat - Lower speed ceiling: Not ideal for 300mm/s+ printing

When to choose this path: 1. V-slot donor (Ender 3) - no rods to scavenge anyway 2. Silent operation is priority (bedroom/office printer) 3. Scavenged rods are scratched/damaged beyond ball bearing use 4. Humid environment where rust is a concern

Cost breakdown (AUD): | Part | Quantity | Unit Price | Total | |——|———-|————|——-| | 8mm stainless rod 420mm | 8 | ~$5-6 | ~$40-48 | | IGUS RJ4JP-01-08 | 20 | ~$1.50-2 | ~$30-40 | | Total | | | ~$70-88 |

Installation notes: 1. Chamfer rod ends: File 45° bevel to avoid slicing IGUS internal ribs 2. Do NOT lubricate: IGUS bushings are self-lubricating; grease reduces life 3. Housing design: Include compliance gap for slight clamping pressure 4. Cutting stainless: Use angle grinder with thin “Inox” cutoff wheel

Verdict: Excellent alternative for V-slot donors or silent builds. Not reference spec due to higher friction and cost, but a legitimate “premium” path.

References: - Tom’s 3D IGUS analysis - IGUS Official Drylin RJ4JP

Alternative C: Printed Adjustable Bearings

DIY polymer sleeve bearings with split-clamp adjustment:

    ┌─────────────┐
    │  ╭───────╮  │  ← Printed PETG sleeve
    │  │  rod  │  │
    │  ╰───────╯  │
    └──────┬──────┘
         screw    ← Adjustable preload

Pros: - Zero cost (print them) - Adjustable preload compensates for wear - Works with slightly imperfect rods - Silent operation - True self-replication (RepRap philosophy)

Cons: - 2-3× higher friction than ball bearings - Requires careful tuning (too tight = binding, too loose = play) - Will wear over time - Needs periodic relubrication (PTFE grease, not oil) - Unproven at Amalgam’s 360mm spans

Best candidate axis: Z (slowest, lowest risk)

Material recommendation: PETG with PTFE/silicone grease

Verdict: Experimental option for Z-axis. Not recommended for X or Y in reference spec.

Reference: RepRap Printed Bearings Wiki

Alternative D: Bronze Bushings

Oil-impregnated sintered bronze sleeves.

Pros: - Very long service life - Self-lubricating - High load capacity - Industrial proven

Cons: - Not scavengeable from printers - Require precise rod diameter match - Higher friction than ball bearings - Heavier than alternatives

Verdict: Not aligned with scavenger philosophy.

BOM Implications

Reference Build (2 Donors)

Scavenged: | Part | Quantity | Source | Notes | |——|———-|——–|——-| | LM8UU | 12 | Donors | For Y (8) + Z (4) |

Purchased (if needed): | Part | Quantity | Cost (AUD) | Notes | |——|———-|————|——-| | LM8LUU | 4 | ~$5-8 | For X-axis carriage |

Alternative (pure scavenger): | Part | Quantity | Source | |——|———-|——–| | LM8UU | 16 | Donors (select best 16 of 24) |

Donor Bearing Yield

Donor Type LM8UU LM8LUU Rods Notes
Prusa i3 clone 11-12 0 6× 8mm Standard config
Anet A8 10-12 0 6× 8mm Standard config
Prusa MK2/MK3 9 3 6× 8mm Y-carriage uses LM8LUU
Creality Ender 3 0 0 None Uses V-slot rollers
CR-10 0 0 None Uses V-slot rollers

If one donor is MK2/MK3 style, you may get LM8LUU for free.

V-Slot Donor Build (Ender 3, CR-10)

If your donor uses V-slot rollers, you must purchase the motion system:

Option 1: Traditional (Hardened + Ball Bearings) | Part | Quantity | Cost (AUD) | |——|———-|————| | 8mm hardened steel rods 400mm | 8 | ~$40-50 | | LM8UU bearings | 16 | ~$15-20 | | LM8LUU bearings (X-axis) | 4 | ~$8-10 | | Total | | ~$63-80 |

Option 2: Silent (Stainless + IGUS) - Recommended | Part | Quantity | Cost (AUD) | |——|———-|————| | 8mm stainless steel rods 420mm | 8 | ~$40-48 | | IGUS RJ4JP-01-08 bushings | 20 | ~$30-40 | | Total | | ~$70-88 |

For V-slot donors, the IGUS path costs only ~$10 more but provides silent operation and zero maintenance.

Maintenance Protocol

Initial Setup

  1. Clean all bearings with IPA
  2. Dry completely
  3. Apply light lithium grease or machine oil
  4. Install with bearing seated fully in housing

Ongoing Maintenance

  • Every ~100 print hours: Wipe rods, check for debris
  • Every ~500 print hours: Remove bearings, clean, regrease
  • Signs of wear: Grinding sound, increased resistance, visible play

Lubrication Options

Lubricant Pros Cons
Lithium grease Long-lasting, good protection Can attract dust
PTFE/silicone grease Clean, low friction Needs more frequent application
Light machine oil (3-in-One) Easy to apply, low friction Needs frequent reapplication
Sewing machine oil Very light, clean Minimal protection

Reference spec recommendation: Lithium grease for long service intervals.

Implementation Notes

X-Carriage (Plough) Bearing Layout

     LM8LUU        LM8LUU
    ┌──────┐      ┌──────┐     ← Top rod
    │      │      │      │
    │ 24mm │      │ 24mm │
    │      │      │      │
    └──────┘      └──────┘
       ├── ~60mm ──┤            ← Bearing spacing for stability

    ┌──────┐      ┌──────┐     ← Bottom rod
    │      │      │      │
    └──────┘      └──────┘
     LM8LUU        LM8LUU

4× LM8LUU total, spaced for maximum wheelbase.

Y-Gantry Bearing Layout (Each Side)

    ═══════════════════  ← Top Y rod
         │      │
       LM8UU  LM8UU      ← 2 bearings, spaced apart
         │      │
    ═══════════════════  ← Bottom Y rod
         │      │
       LM8UU  LM8UU

4× LM8UU per side (8 total), standard length is sufficient with dual-rod stability.

Z-Axis Bearing Layout

    ┌─LM8UU─┐        ┌─LM8UU─┐
    │       │        │       │
    │  Z    │        │   Z   │
    │  rod  │        │  rod  │
    │       │        │       │
    └─LM8UU─┘        └─LM8UU─┘

4× LM8UU total (2 per rod), handles bed weight with large margin.

Quality Verification

New Bearing Test

  1. Roll on clean rod - should be silent and smooth
  2. Check for lateral play - minimal acceptable
  3. Verify dimensions with calipers

Scavenged Bearing Test

  1. Visual inspection for rust, pitting, damage
  2. Roll test - reject if grinding or notchy
  3. Play test - reject if excessive wobble
  4. Clean, regrease, retest before installation

References

Technical Sources