Reconstruct the 2017 EDR Super Clawbot into a speed-focused Override V2 robot: 55W drivetrain (2.75″ or 4″ wheels), 4-DOF articulating arm reaching from back-loader pickup to tall-goal placement, aluminum chassis with weight-saving measures. Math + decisions + build plan.
What we're keeping from the Super Clawbot, what we're cutting, and why the architecture works for Override.
What's Worth Keeping from the Super Clawbot
The 2017 Super Clawbot was designed as a teaching robot, but its architecture has three things going for it as an Override starting point:
Articulating arm with a real wrist. Most modern V5 lifts (4-bar, DR4B, cascade) lock the gripper orientation. The Super Clawbot has independent wrist control — useful in Override where cup orientation matters (per SC3: opaque vs transparent face affects scoring).
Multi-segment reach. Shoulder + elbow + wrist gives a workspace closer to a hemisphere than a single-axis lift. This is the architecture you need to grab from a back loader and place at a front goal in one continuous motion.
The chassis is salvageable. The drivetrain mounting is straightforward; existing C-channels can be replaced piece-by-piece with aluminum.
What Has To Go
The 393 motors. Cortex-era hardware is non-V5RC-legal. Replace with V5 11W smart motors and 5.5W half-motors throughout.
The 4WD steel chassis. Replaced with a wider aluminum 4WD or 6WD configuration sized for the 18″ envelope.
The basic claw. The original gripper was sized for foam balls. Override needs a claw sized for the 3.15″ cup AND the tapered 1.4″-3.16″ pin.
✨
Good news: the original 4-DOF arm (shoulder + elbow + wrist + claw) stays. The Override game manual caps wattage (88W total, 55W drivetrain), not motor count. With 5.5W half-motors mixed into the arm, you can run a 5-motor 55W speed-focused drivetrain AND retain all four arm degrees of freedom. See Section 1 for the full power budget breakdown.
The Match Cycle (What This Robot Does)
Drive backward toward loader. Speed-focused drivetrain at 5.5-6.5 ft/s makes the back-and-forth fast.
Loader pickup. Articulating arm reaches over the back of the robot, claw at H≈3-4″ (loader output height). Grab the cup-on-pin combo.
Drive forward toward goal. Arm holds combo close to chassis to keep CoG low.
Goal placement. Articulating arm extends forward + up, claw at H≈12″ (over 8.77″ tall goal). Drop combo on goal.
Reset. Arm folds back. Drive back to loader. Repeat.
🎯
Why this works for Override: match loaders deliver pre-stacked cup/pin combos at predictable height. Articulating arm grabs and places in one motion (no manipulator handoff). Speed-focused drivetrain means high cycle volume. Single mechanism handles all three goal heights via different arm angles.
Where This Architecture Sits in the Lift Selection Space
Compared to other Override lift options (see /lift-selection):
Architecture
Cycle Speed
Tall Goal Reach
Build Complexity
Override Verdict
Super Clawbot V2 (4-DOF arm)
High (no handoff)
Yes (via geometry)
Medium-high
★ Strong for cup-on-pin combos
4-bar + V5 claw (V1.5)
Medium
Marginal
Low
Good baseline
4-bar + chain bar + pneumatic
Medium
Yes
High
Mid-V2
DR4B
Low (vertical only)
Easy
Very high
Wrong tool
SECTION 1 / 5
Power Budget — Wattage, Not Motor Count
The Override game manual caps wattage (88W total, 55W drivetrain) — there is NO motor count limit. With 5.5W half-motors mixed in, you can run 9-11+ motors total.
The Two Wattage Caps (No Motor Count Cap)
R10a: Total robot motor power ≤ 88W. This is the only total-robot limit. No 8-motor cap — you can run as many motors as you want as long as their wattage sums to ≤ 88W.
R11a: Drivetrain (Subsystem 1) motor power ≤ 55W.
R11b: No PTO from drivetrain to other mechanisms.
📋
What this means in practice: 11W full motors and 5.5W half-motors are now both tools in your toolkit. A 5.5W half-motor counts as 5.5W toward the cap (not 11W), so two half-motors equal one full motor in budget terms. This unlocks high-DOF arm configurations that would have been impossible under an 8-motor cap.
How This Changes the V2 Architecture
Earlier drafts of this guide assumed the 4-DOF Super Clawbot arm had to drop a degree of freedom to fit a 5-motor 55W drivetrain. That was wrong. With 5.5W half-motors, the wrist comes back:
Speed-focused 4-DOF V2 (corrected):
Drivetrain: 5 × 11W = 55W ✓ R11a cap
Arm: shoulder (11W) + elbow (11W) + wrist (5.5W) + claw (5.5W) = 33W
Total: 88W ✓ R10a cap exactly
Motor count: 9 motors (5 drive + 2 full arm + 2 half arm)
You keep all 4 DOF AND have a 5-motor speed drivetrain.
Recommendation: Option A (Speed + Full 4-DOF). 5-motor speed drivetrain at 55W gives you 6.28 ft/s (with 4″ wheels at 5:3 reduction). Full 4-DOF arm preserves cup orientation control via the wrist. The shoulder gets the full 11W cartridge (it's the highest-load joint); wrist and claw get 5.5W half-motors (low-load joints, fixed at 200 RPM). 88W exactly at R10a cap.
Half-Motor Tradeoffs (5.5W vs 11W)
Why use half-motors at the wrist and claw, but not the shoulder?
High-load or speed-tunable tasks (drivetrain, lift, shoulder)
Recommended Final Allocation (Option A)
Subsystem
Motors
Cartridge
Power
Notes
Drivetrain (5 motors)
5 × 11W
Blue 600 RPM
55W
Ports 1-5; speed-tuned via 5:3 reduction (see Section 2)
Shoulder
1 × 11W
Red 100 RPM
11W
Port 6; full power for highest-load joint; rubber band assist mandatory
Elbow
1 × 11W
Red 100 RPM
11W
Port 7
Wrist
1 × 5.5W half-motor
n/a (fixed 200 RPM)
5.5W
Port 8; just enough torque to rotate the claw; speed adequate
Claw
1 × 5.5W half-motor
n/a (fixed 200 RPM)
5.5W
Port 9; rack-and-pinion grip; 200 RPM = ~0.5 sec full open/close
Total
9 motors
—
88W
R10a cap exactly; R11a cap exactly; 4 DOF preserved
⚠️
Half-motor caveat at the wrist: 5.5W half-motors are fixed at 200 RPM with stall torque ~10.5 in-lb. If your wrist needs to rotate a heavy end-effector (claw + cup + pin = ~360 g) at 8″ from the wrist axis, the static torque load is ~6.3 in-lb. That's 60% of stall — workable but high. Reduce wrist load by:
Mounting the wrist motor close to the rotating axis (not at the end of a lever)
Using a 2:1 or 3:1 gear reduction at the wrist (gives more torque, slower rotation)
Swapping the wrist to an 11W motor and dropping somewhere else (move shoulder to 5.5W mirrored pair = 11W, freeing 5.5W for wrist upgrade)
SECTION 2 / 5
Speed-Focused Drivetrain — 2.75″ vs 4″
For a speed build with 5×Blue motors, which wheel and gear ratio? Math compares both options at speed-tuned ratios.
Setting the Speed Target
For Override, "built for speed" usually means 5.5-6.5 ft/s at the wheel. Faster than that and the driver loses precision; slower than that and you're indistinguishable from a push-heavy build. The 5.5-6.5 range is where speed-focused robots live.
Configuration A: 2.75″ Wheels + Speed Gearing
Reduction
Gear Pair
Wheel RPM
Linear Speed
Force/wheel
Verdict
1:1 (none)
direct drive
600
7.20 ft/s
10.2 lb
Top end
4:3
36T → 48T
450
5.40 ft/s
13.6 lb
Sweet spot
5:4
32T → 40T
480
5.76 ft/s
12.7 lb
★ Speed-focused
1.2:1
30T → 36T
500
6.00 ft/s
12.2 lb
Speed-focused
2.75″ speed-tuned (5:4 reduction = 480 RPM):
Linear speed = (480 × π × 2.75) / 720 = 5.76 ft/s
Force per wheel = (14 in-lb × 1.25) / 1.375″ = 12.7 lb
5-wheel total at stall = ~63 lb push (theoretical max)
Top speed (1:1 direct drive): 7.20 ft/s — possible but very fast. Use only with experienced drivers and PID control.
Configuration B: 4″ Wheels + Speed Gearing
Reduction
Gear Pair
Wheel RPM
Linear Speed
Force/wheel
Verdict
5:3
36T → 60T
360
6.28 ft/s
11.7 lb
★ Speed-focused
3:2
24T → 36T
400
6.98 ft/s
10.5 lb
Top end
4:3
36T → 48T
450
7.85 ft/s
9.3 lb
Too fast
2:1
36T → 72T
300
5.24 ft/s
14.0 lb
Sweet spot, not speed
4″ speed-tuned (5:3 reduction = 360 RPM):
Linear speed = (360 × π × 4) / 720 = 6.28 ft/s
Force per wheel = (14 in-lb × 1.667) / 2.0″ = 11.7 lb
5-wheel total at stall = ~58 lb push
Head-to-Head at Speed-Tuned Ratios
Factor
2.75″ @ 5:4
4″ @ 5:3
Winner
Linear speed
5.76 ft/s
6.28 ft/s
4″ (~9% faster)
Pushing force per wheel
12.7 lb
11.7 lb
2.75″ marginal
Ground clearance
1.4″
2.0″
4″ clearly
Pinion size on motor
32T (manageable)
36T (standard)
4″ cleaner
Drivetrain weight (5 wheels)
~250 g
~425 g
2.75″ saves 175g
Acceleration from stop
Marginally faster
Marginally slower
2.75″ marginal
Available chassis-internal space
More room (smaller wheel boxes)
Less room
2.75″ better
✓
Recommendation: 4″ wheels at 5:3 reduction (36T → 60T) → 6.28 ft/s. Wins on the two factors that matter most for an articulating-arm robot:
Higher actual top speed (6.28 vs 5.76 ft/s) — 9% faster cycles, meaningful at scale
Higher ground clearance — Override scatters cups + pins to midfield, and 1.4″ clearance will catch on debris when driving fast
The 2.75″ option only wins on weight (175g), which is well within noise for a 6kg robot.
5-Motor Drivetrain Layout
5 motors on a tank drivetrain means asymmetric placement. Three valid layouts:
Layout
Description
Pros
Cons
3+2 split
3 motors on one side, 2 on the other (with chain or gear connecting)
4 motors driving 4 wheels directly, 5th motor connects via chain to add power to an existing wheel pair
Symmetric drive feel
Chain alignment + tension management
6WD center-drop (recommended)
6 wheels in a row (3 on each side), 5 motors split 3+2 or 2+2+1 with the unmotored wheel passive
Wider footprint, better stability with arm extended
Larger wheelbase, less internal chassis space
📐
Recommended layout: 6WD center-drop with 5 motors driving 5 wheels (1 wheel passive). The center wheel on the drive-side with 3 motors can be drop-center for turning agility, with the unmotored wheel as a free spinner on the 2-motor side. Total of 6 wheels for stability under articulating arm dynamic load. Documented in /drivetrain-architectures.
SECTION 3 / 5
3-DOF Articulating Arm Geometry
Shoulder + elbow + claw. Reach math from back-loader pickup to tall-goal placement, with kinematic SVG.
Reach Targets
Action
Claw position (Y, H)
Notes
Back-loader pickup
(0″, 4″)
Y=0 = back edge of robot; H=4″ = loader output height (estimated)
Transit position
(9″, 8″)
Tucked above chassis center, low CoG
Alliance goal placement
(20″, 6.5″)
1″ past front edge, 3.25″ goal + clearance + cup half
Short neutral placement
(20″, 9″)
Front edge area, 5.77″ goal + clearance
Tall center placement (target)
(20″, 12″)
The hardest reach — defines arm length
Sizing the Arm
Shoulder pivot at chassis: position Y=6″ from back, height H=4″ above ground (mounted on a small tower above the chassis top).
Distance from shoulder to tall-goal target:
Distance from shoulder (6, 4) to target (20, 12):
d = √((20−6)² + (12−4)²) = √(196 + 64) = √260 = 16.1″
Distance from shoulder (6, 4) to back-loader (0, 4):
d = √((0−6)² + (4−4)²) = 6.0″
Required arm reach: max(16.1″, 6.0″) = 16.1″ with some margin.
Total arm length L_total ≥ 16.5″ (with 0.4″ safety margin)
Two-Segment Arm Sizing
Splitting 17″ total reach across upper arm (shoulder→elbow) and forearm (elbow→claw):
Configuration
Upper arm
Forearm
Pros
Cons
Equal split
8.5″
8.5″
Symmetric load distribution
Folded arm is asymmetric on chassis
Slightly forearm-heavy
9″
8″
Easier elbow torque math; better folded position
Slightly larger upper-arm stress
Long forearm
7″
10″
Flexible reach near shoulder
Forearm cantilever is long; claw at end has high inertia
📏
Recommended: 9″ upper arm + 8″ forearm = 17″ total. Easy gear-reduction math (both segments use 36T → 60T = 5:1 reduction), good folded position behind the chassis, well-distributed bending stress.
Side-View Kinematic Diagram
4-DOF Arm — Pickup, Transit, Place Positions
↑ Three arm positions overlaid. Pickup (green): arm folded back to grab cup/pin combo from back loader at (Y=0″, H=4″). Transit (yellow): arm tucked over chassis center for stable driving with combo at (Y=9″, H=8″). Placement (orange): arm extended forward + up to drop combo on tall center goal at (Y=20″, H=12″). Shoulder pivot fixed at (Y=6″, H=4″).
Joint Torque Math
Worst case for shoulder torque: arm fully extended forward at placement position. The shoulder must support the entire arm + cup+pin combo at 16″ horizontal distance.
Shoulder torque calculation (arm fully extended, placement position):
Upper arm aluminum: ~150 g, center of mass at 4.5″ from shoulder
Elbow joint + motor: ~280 g, at 9″ from shoulder
Forearm aluminum: ~140 g, center of mass at 9 + 4 = 13″ from shoulder
Claw + motor: ~280 g, at 9 + 8 = 17″ from shoulder
Cup + pin combo: ~80 g, at 17.5″ from shoulder
Actually proper math (g·in to in-lb): 1 g·in = 0.00220 in-lb (1 lb = 453.6 g)
Total torque g·in = 11,175 g·in
Total torque in-lb = 11,175 × 0.00220 = 24.6 in-lb at shoulder
Shoulder motor capability (1× Red 100 RPM at 5:1 reduction):
Stall torque at cartridge output: ~21 in-lb
Through 5:1 reduction: 21 × 5 = 105 in-lb stall at shoulder pivot
Load utilization: 24.6 / 105 = 23% of stall ✓ Comfortable
With rubber band assist: bands provide ~5-8 in-lb of pre-load toward up direction
Effective load: 24.6 − 7 = 17.6 in-lb → 17% of stall ✓ Excellent
Elbow torque (forearm + claw + cup at full extension):
Forearm CoM × distance: 140 g × 4″ = 560 g·in
Claw + motor: 280 g × 8″ = 2,240 g·in
Cup+pin: 80 g × 8.5″ = 680 g·in
Total at elbow: 3,480 g·in × 0.00220 = 7.7 in-lb
Elbow motor (1× Red 100 RPM at 5:1): 105 in-lb stall
Load utilization: 7.7 / 105 = 7% of stall ✓ Excellent No rubber band assist needed at elbow.
Wrist torque (1× 5.5W half-motor, no reduction):
The wrist rotates the claw around its mounting axis. Load is the offset claw + cup mass.
Claw + cup CoM offset from wrist axis: ~1.5″ (claw center, cup hangs below)
Combined mass: claw 280 g + cup+pin 80 g = 360 g
Static torque at wrist (worst case, claw horizontal): 360 g × 1.5″ × 0.00220 = 1.2 in-lb
5.5W half-motor stall torque: ~10.5 in-lb
Load utilization: 1.2 / 10.5 = 11% of stall ✓ Comfortable
Wrist rotation speed: 200 RPM = 3.33 RPS = ~1.2 sec for 90° rotation. Adequate for cup orientation correction during transit.
Claw torque (1× 5.5W half-motor, rack-and-pinion grip):
Required grip force on cup: ~0.1 lb (cup mass × accel = 0.07 lb gravity)
Required grip force on pin: ~0.5 lb (pin's tapered shape needs more pinch)
Standard claw rack-and-pinion gives ~5-8 lb of grip force at the tips Claw motor at 11% of stall during grip cycle. Easy. 200 RPM = 0.5 sec full open-close cycle.
Reach Verification Table
Action
Target (Y, H)
Distance from shoulder
Reach status
Back-loader pickup
(0, 4)
6.0″
✓ Easy (arm folds)
Transit
(9, 8)
5.0″
✓ Comfortable
Alliance goal placement
(20, 6.5)
14.2″
✓ Reaches (17″ arm)
Short neutral placement
(20, 9)
14.9″
✓ Reaches
Tall center placement
(20, 12)
16.1″
✓ Reaches with 0.9″ margin
SECTION 4 / 5
Aluminum Construction + Weight Saving
Switching from steel to aluminum cuts ~2 kg of robot mass. Where it's safe, where it isn't, and the structural math.
Aluminum vs Steel — VEX Specs
Property
Steel C-channel
Aluminum C-channel
Aluminum advantage
Density
7.85 g/cm³
2.70 g/cm³
~3× lighter per unit volume
Tensile strength
~530 MPa
~310 MPa (6061-T6)
Steel ~70% stronger
Young's modulus (stiffness)
~200 GPa
~70 GPa
Steel ~3× stiffer
Mass per 35-hole 1×2 c-channel (~17.5″)
~250 g
~85 g
~165 g savings per piece
Mass per 25-hole 1×1 angle
~160 g
~55 g
~105 g savings per piece
Cost per piece (typical)
$3-5
$5-8
Aluminum ~50-100% more expensive
Total Weight Savings — Robot-Wide Estimate
A typical V5RC robot uses 12-18 long c-channels + 6-10 short angles + small parts. Conservative count:
Robot structural inventory (typical V5RC build):
14 long c-channels (chassis + arm) × 165 g savings = 2,310 g
8 short angles (brackets + brackets) × 105 g savings = 840 g
Hardware (screws, nuts, spacers) ~ same in both → 0 g savings
Motors, wheels, gears, electronics ~ same in both → 0 g savings
Total mass reduction: ~3,150 g (~3.1 kg)
For a baseline 6 kg steel-c-channel robot:
Aluminum-converted robot mass = 6.0 − 3.15 = 2.85 kg
= ~50% mass reduction
⚠️
The 50% reduction estimate is theoretical maximum. In practice you'll keep some steel for high-stress parts (drivetrain motor mounts, shoulder pivot reinforcement, gear box plates), so realistic savings are 35-45% of total mass. Plan for ~2-2.5 kg actual savings, not 3+ kg.
Where Aluminum Is Safe (Use Throughout)
Chassis side rails (long c-channels): primary load is bending across the long axis; aluminum's lower modulus is acceptable here because spans aren't enormous.
Tower verticals (vertical c-channels above chassis top): primary load is compression; aluminum is fine.
Arm bars (upper arm + forearm): primary load is bending; analyze case-by-case (see below) but generally OK with reinforcement.
Cross-supports between chassis rails: stiffness in shear; aluminum works.
Bumpers and license-plate mounts: non-structural; weight savings free here.
Where Steel Stays (or Aluminum Needs Reinforcement)
Motor mounts (especially drivetrain): high vibration, screw fatigue, shock loading from impact. Use steel here, OR aluminum with internal aluminum spacer reinforcement (see /team-hardware-kit).
Shoulder pivot mount: the highest single point of load on the robot. Use steel c-channel in that one location, OR doubled aluminum with diagonal bracing.
Gear box plates: if gears mesh through aluminum, deflection causes alignment errors. Steel preferred for any plate where two gears engage on opposite sides.
Long unsupported cantilevers (e.g., end of forearm): aluminum's lower stiffness causes more deflection. Either keep span short or add a stiffening bar/triangle.
Bending Math — Forearm Aluminum vs Steel
Will an 8″ aluminum forearm be too floppy? Let's check:
where F = force at end (claw + cup + pin = 360g = 0.79 lb)
L = length = 8″ = 0.203 m
E = Young's modulus
I = moment of inertia of c-channel cross-section
For VEX 1×2 c-channel, I ≈ 4.16 × 10⁻⁹ m⁴ (rough estimate from cross-section)
Steel forearm (E = 200 GPa):
δ = (3.51 N × 0.00837 m³) / (3 × 200 × 10⁹ × 4.16 × 10⁻⁹)
δ = 0.0294 / 2.50 = 0.0118 m = 11.8 mm = 0.46″ deflection at end
Aluminum forearm (E = 70 GPa):
δ = same numerator / (3 × 70 × 10⁹ × 4.16 × 10⁻⁹)
δ = 0.0294 / 0.874 = 0.0337 m = 33.7 mm = 1.33″ deflection at end
Aluminum deflects ~3× more than steel. 1.3″ sag at the claw is significant — affects placement accuracy.
⚠️
1.3″ deflection on the forearm is too much. Mitigation options:
Use paired aluminum c-channels back-to-back on the forearm (doubles I, halves deflection to ~0.65″). Adds ~140 g but acceptable.
Add a diagonal stiffener from the elbow to a point along the forearm (creates a triangle, dramatically reduces deflection). Adds ~50 g.
Use steel for forearm only, aluminum elsewhere. Saves more weight robot-wide than aluminum forearm with stiffeners.
Recommendation: steel forearm. Aluminum benefits don't outweigh the precision loss for this single piece.
Weight-Saving Beyond Aluminum
Technique
Savings
Tradeoff
Lightening holes (drilled in c-channels)
10-20 g per piece
Reduces strength ~5-10% per hole; can be done safely on non-critical members
Shorter c-channels (cut to exact length needed)
5-30 g per piece
Locks in geometry, harder to iterate
Polycarbonate plates instead of metal where possible
50-100 g per plate
Lower stiffness; OK for non-load-bearing covers
Aluminum standoffs instead of steel
~10-15 g per standoff × ~20 standoffs = 200-300 g
Lower thread strength; use Loctite
Skip the bumpers (just license plates)
~150 g
Minor scuffing on edges; usually fine
Final Weight Estimate
~6.0 kg
Steel Baseline
Original Super Clawbot architecture rebuilt in steel. Heavy.
~40% weight reduction. Faster acceleration, easier on motors, lower CoG, longer battery life per match.
SECTION 5 / 5
Build Plan + Verdict
Phased rollout, parts list, common failure modes, and an honest assessment of whether this build is right for your team.
Phased Build Plan (V2 Conversion)
Week 1: Disassembly + parts inventory. Strip the original Super Clawbot. Inventory every steel c-channel; mark which ones get replaced with aluminum. Order aluminum stock + V5 hardware (5×11W motors + Blue cartridges for drive, 2×11W + Red cartridges for shoulder/elbow, 2×5.5W half-motors for wrist/claw).
Week 2: New chassis (aluminum). Build the 18″ × 14″ chassis with 4″ omni wheels and 5×Blue motor drivetrain at 5:3 reduction. Battery, brain, motors mounted. Test driving — target 6.0-6.5 ft/s.
Week 3: Shoulder + tower. Mount shoulder tower at Y=6″ from back, 4″ tall (top of pivot at H=4″). Use steel c-channel for the shoulder mount specifically. Single Red 100 RPM motor with 5:1 reduction. Add rubber band assist.
Week 4: Upper arm + elbow. Build 9″ aluminum upper arm with internal spacer reinforcement. Mount elbow joint with single Red 100 RPM motor + 5:1 reduction. Test arm range hand-driven through full shoulder + elbow envelope.
Week 5: Forearm + wrist + claw. Build 8″ steel forearm (steel for stiffness — see Section 4). Mount 5.5W half-motor wrist at end of forearm with 2:1 or 3:1 reduction (gives torque margin). Mount Override-sized claw with 5.5W half-motor for grip. Test wrist rotation independently from claw open/close.
Week 6: Integration + tuning. PID tune all four arm joints (shoulder, elbow, wrist, claw). Test full pickup-transit-place sequence with cup orientation correction during transit (the wrist's job). Measure actual reach to all goals. Verify CoG with calculator at /center-of-gravity.
Parts List Summary
Subsystem
Key Parts
Notes
Drivetrain
5× V5 Smart Motor (276-4840), 5× Blue cartridge, 4×4″ omni (276-1454) + 2×4″ omni for 6WD, gears (5×36T pinion, 5×60T driven)
~$320 in motors + cartridges; ~$60 in wheels; ~$25 in gears
Shoulder + Elbow
2× V5 Smart Motor (276-4840), 2× Red cartridges, gear pairs (12T:60T at shoulder, 12T:60T at elbow)
~$130 motors + cartridges
Wrist + Claw
2× V5 5.5W Half-Motor (276-4842), claw rack-and-pinion assembly, gear reduction 2:1 or 3:1 at wrist axis