A free running axle is a fixed, non-rotating shaft on which each wheel spins independently on its own bearings, used widely on mine cars, ore cars, and trams. Typical underground hutch wagons run 350-450 mm wheels at 80-200 RPM with each wheel free to differentiate around tight 6-10 m radius curves. The design eliminates the scrub and flange wear you get with a fixed wheelset, where both wheels must turn at the same speed. That's why every modern granby car, kibble bogie, and battery-locomotive trailer in operations from Sudbury to the Bushveld runs free wheels on a dead axle.
Free Running Axle Interactive Calculator
Vary curve radius, wheel spacing, and centerline RPM to see the independent inner and outer wheel speeds on a mine-car free running axle.
Equation Used
The calculator compares the arc length demanded by each wheel on a curve. The outer wheel runs on radius R + G/2, while the inner wheel runs on radius R - G/2. A free running axle lets each wheel take its required RPM; a fixed wheelset would have to absorb the RPM mismatch as tread or flange scrub.
- Curve radius R is measured to the axle centerline between the two wheels.
- Wheel rolling diameters are equal and there is no tread slip.
- The axle is fixed to the frame and each wheel rotates freely on its own bearings.
- Wheel spacing G is treated as the distance between the left and right wheel rolling paths.
How the Free Running Axle Actually Works
The principle is simple — the axle does not rotate. It bolts rigidly to the car frame, and each wheel runs on its own pair of bearings (usually tapered roller, sometimes deep-groove ball for light-duty trams) pressed into the wheel hub. When the car rolls into a curve, the outer wheel travels a longer arc than the inner wheel. On a live axle, both wheels are keyed to a common shaft and must rotate at the same RPM, so one wheel has to slip — that's where you get the screech, the flange climb, and the rapid tread wear you see on poorly maintained mine track. With a free running axle (a dead axle in the older British literature), each wheel just turns at whatever speed the rail demands. No scrub, no fight.
The geometry that matters is the bearing fit and the wheel-hub squareness. A typical 350 mm cast-steel mine car wheel runs on a 50-60 mm journal with a bearing radial clearance of 0.05-0.08 mm. Open that up to 0.15 mm through wear and the wheel starts to wobble on the axle, the flange contacts the rail at an angle, and you'll hear it before you see it. The axle itself is usually a forged or hot-rolled medium carbon shaft (1045 or EN8 grade), ground at the journals to about Ra 0.8 µm or better. Rougher than that and bearing inner-race fretting starts within a few hundred operating hours.
Failure modes are predictable. Bearing seizure from water and fines ingress is the number one killer underground — once the seal goes, the bearing fills with rock paste and welds itself solid. The wheel then drags, the tread flat-spots, and you've got a derailment risk. Less dramatic but more common is uneven flange wear because one bearing has more drag than the other, so the car tracks crooked. If you notice one wheel always running hot to the touch after a shift, pull it and check the seal.
Key Components
- Dead Axle (Fixed Shaft): A non-rotating shaft, typically 50-80 mm diameter forged from 1045 carbon steel, bolted or pinned rigidly to the car frame. It carries the vertical load from the car through to the wheels. Journals are precision-ground to Ra 0.8 µm and held to ±0.02 mm on diameter so the bearing inner race seats true.
- Wheel Hub Bearings: Usually a pair of tapered roller bearings per wheel (e.g. Timken 32208 or equivalent) set in opposed configuration to handle thrust loads as the flange contacts the rail. Radial clearance starts at 0.05-0.08 mm and is the wear parameter that decides when the wheel needs reseating.
- Wheel and Flange: Cast or forged steel wheel, 300-500 mm diameter for typical mine cars, with a tread profile matched to the rail head and a flange height of 25-35 mm. The flange does the steering work — wider flange clearance is fine on a free running axle because there's no scrub force fighting it.
- Bearing Seals: Labyrinth or contact lip seals rated for IP65 minimum. Underground mine duty is brutal — water, fines, blast residue — and a failed seal will kill the bearing in under 200 hours. We size seals with a 30% pressure margin over rated wash-down.
- Retainer Nut and Locking Tab: Castle nut with split-pin or tab-washer locking on each axle end. Holds bearing preload at the spec value (typically 0.02-0.05 mm endplay for tapered rollers). Lose the locking tab and the wheel walks off the axle within a shift.
Who Uses the Free Running Axle
Free running axles are the default on almost every wheeled mine vehicle that doesn't drive its wheels directly. Anywhere you have rail-bound haulage with curves tighter than 20 m radius, the differential demand makes a fixed wheelset uneconomic — the wheel and rail wear costs alone justify the bearing complexity. You'll find them on everything from 1-tonne hand-trammed hutches in artisanal workings to 25-tonne granby cars in deep hard-rock mines.
- Underground Hard Rock Mining: Granby-style side-dump ore cars used at Vale's Creighton mine in Sudbury — 8 to 12 tonne capacity, 400 mm cast steel wheels on dead axles with Timken tapered roller bearings.
- Coal Mining: Mine cars and supply cars on locomotive haulage in Appalachian room-and-pillar operations, typically 3-tonne to 6-tonne capacity running 24-inch gauge.
- Tunnel Construction: Muck cars behind roadheaders and TBMs, such as the Schöma narrow-gauge locomotive trains used in Gotthard Base Tunnel construction — every trailing wagon ran free-wheeling axles.
- Shaft Sinking: Kibble bogies and cactus-grab transfer cars used in shaft-bottom loading stations at operations like the Cigar Lake uranium mine.
- Heritage and Tourist Mining: Restored ore cars at the Britannia Mine Museum in BC and Park's Canada's heritage workings — dead axles are simpler to rebuild than live wheelsets and parts are still readily forged.
- Aggregate and Quarry: Side-tipping skip cars on inclined-haulage quarry tramways, common in older slate and limestone operations across North Wales.
The Formula Behind the Free Running Axle
The most useful calculation for a free running axle is the differential RPM between inner and outer wheels on a curve — this tells you whether the design is even justified versus a fixed wheelset, and how hard the bearings work in a typical curve. At the gentle end of the typical mine-track range (say a 30 m radius main-haulage curve), the differential is small and a worn wheelset might just about cope. At the tight end (6 m radius around a sub-level loading pocket) the differential is large enough that a fixed axle would scrub the rail head visibly within a week. The sweet spot for free running axle bearing life sits at moderate speed and moderate curve radius where bearing RPM stays under 200 and the differential demand is well within the bearing's capacity.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| ΔN | Differential RPM between outer and inner wheel on the curve | RPM | RPM |
| v | Car forward speed along the curve centreline | m/s | ft/s |
| Dw | Wheel tread diameter | m | ft |
| G | Track gauge (rail centre to rail centre) | m | ft |
| R | Curve radius at track centreline | m | ft |
Worked Example: Free Running Axle in a copper mine ore car fleet in Zambia
An operator on the Zambian Copperbelt is specifying replacement wheelsets for a fleet of 6-tonne side-dump ore cars running 600 mm gauge track in a sub-level haulage drift. Wheel diameter is 380 mm, average tramming speed is 1.8 m/s, and the tightest curve in the circuit is a 12 m radius bend at a loading pocket. They want to know the differential RPM the bearings see on that curve, and how it compares against gentler 30 m main-line curves and the tightest 6 m radius spur into a maintenance bay.
Given
- v = 1.8 m/s
- Dw = 0.380 m
- G = 0.600 m
- Rnom = 12 m
Solution
Step 1 — calculate the nominal wheel RPM at tramming speed. This sets the baseline bearing load.
Step 2 — at the nominal 12 m radius curve, the differential between outer and inner wheel is the gauge over the radius times the wheel RPM:
That's a comfortable working point — about 5% differential, well inside the slip tolerance of the rail-tread interface. The bearings on each wheel are running at slightly different speeds but neither is highly loaded.
Step 3 — at the easy end of the operating range, a 30 m main-line curve:
Barely any differential. Honestly, a worn fixed wheelset would survive this curve too — you'd see scrub but not much. This is why old surface tramways with gentle curves got away with live axles for decades.
Step 4 — at the tight end, a 6 m radius spur into the maintenance bay:
That's a 10% speed difference between the two wheels. On a fixed wheelset one wheel would have to skid every revolution, peeling steel off the tread and the rail head both. The free running axle handles it without complaint — the outer bearing just spins a bit faster than the inner.
Result
Nominal differential RPM is 4. 5 at the 12 m curve, with each wheel turning at around 90 RPM. In practice you'd hear nothing unusual at this operating point — just the normal rumble of cast wheel on rail. Across the full range, the differential goes from 1.8 RPM on gentle main-line curves up to 9.1 RPM on the tightest maintenance spur, which tells you the bearings are well within their duty envelope across the entire circuit. If you measure flange or tread wear progressing faster than predicted, suspect three things in this order: (1) bearing radial clearance opened past 0.15 mm letting the wheel cant on the axle, (2) journal Ra above 1.6 µm from corrosion or fretting, generating drag that prevents true differentiation, or (3) one bearing dragging from a partially failed seal so the car tracks crooked and the flange runs in continuous contact instead of intermittent contact.
Free Running Axle vs Alternatives
Free running axles are the default for rail-bound mining haulage, but they aren't free — the bearing complexity adds cost and a maintenance footprint that a fixed wheelset doesn't have. Here's how the choice stacks up against the two real alternatives you see underground.
| Property | Free Running Axle | Live Axle (Fixed Wheelset) | Independent Suspension Hub |
|---|---|---|---|
| Differential capability on curves | Full — each wheel rotates independently | None — both wheels locked to common shaft | Full — each hub fully independent |
| Tightest practical curve radius | 6 m or less | 30 m before scrub becomes severe | 5 m or less |
| Bearing maintenance interval | 1500-3000 hours typical | 3000-5000 hours (one wheelset bearing pair) | 1000-2000 hours (more bearings to seal) |
| Capital cost per axle assembly | Moderate — 4 bearings, 2 seal sets | Low — 2 outboard bearings only | High — full hub carrier per wheel |
| Typical wheel RPM range | 80-300 RPM | 80-300 RPM | 100-500 RPM |
| Failure mode if neglected | Bearing seizure, wheel drag | Flange and rail wear, derailment | Hub bearing seizure, suspension lockout |
| Fit for underground mine haulage | Excellent — industry standard | Acceptable for straight gentle track only | Overkill — used on rubber-tyre LHDs not rail |
Frequently Asked Questions About Free Running Axle
Squeal on a free running axle almost always traces to bearing drag, not the axle design itself. If both bearings have equal drag the wheel still rotates, just unhappily — but if one bearing has more drag than the other (typically from over-preload during reassembly, or a partially collapsed seal), the wheel resists differentiating and you get the same scrub you'd get from a live axle.
Check endplay first. Tapered rollers want 0.02-0.05 mm of free play after the castle nut is torqued and backed off. Zero endplay or any preload at all and the bearing fights every revolution. Spin each wheel by hand off the rail — it should turn 3-4 revolutions from a moderate push. If it stops in one, that bearing is too tight.
Run the differential RPM number for your tightest curve at your normal tramming speed. If the differential exceeds about 3% of wheel RPM, you'll see measurable tread wear within a few hundred operating hours on a fixed wheelset — that's the threshold where free running axles pay back their bearing cost.
As a rule of thumb, anything below 25 m radius at 600 mm gauge wants free wheels. Heritage tramways with 50 m+ curves can run live axles indefinitely with little wear. Anywhere you have switchback curves into loading pockets or maintenance spurs, free running is the only sensible answer.
Temperature asymmetry on a free running axle is a seal warning, full stop. The bearing that runs hot has either lost its lubricant (seal failed outward, grease purged) or contaminated it (seal failed inward, water and fines got in). Both scenarios end the same way — bearing seizure within 50-200 operating hours.
Pull the hot wheel before next shift. If the grease comes out grey or gritty, you've got ingress. If it comes out dry or scorched-smelling, you've got loss. Either way the bearing is on borrowed time and reusing it is a derailment risk.
For underground mine duty, grease is almost always the right call — and it's what we specify on FIRGELLI mine car builds. Oil-bath setups need a perfect seal to retain oil, and underground seals don't stay perfect. Grease tolerates a partially compromised seal far better because it doesn't run out the moment the seal lip lifts.
Use an EP2 lithium-complex grease with a dropping point above 200°C. Repack at half the bearing's calculated L10 hours, not at failure — by the time you can hear or feel a bearing problem, the rollers have already brinelled the race.
Because it's not really non-rotating in service — it's oscillating. Every time the car hits a rail joint or flange-contacts a curve, a small torque pulse goes through the axle into the frame mount. Over thousands of cycles this works any clamp or bolted joint loose unless it's positively locked.
Two fixes: pin the axle to the frame with a transverse dowel pin (don't rely on bolt clamping alone), or use a serrated-flange mount that bites into the frame casting. We've seen plain bolted mounts walk loose in under 600 hours on rough underground track — the dowel pin solution lasts the life of the car.
Almost never worth it underground. Ceramic hybrids shine at high RPM where centrifugal loading on steel rollers becomes a problem — not relevant at the 80-300 RPM a mine car wheel turns. The real failure mode underground is contamination, and ceramics don't help with that.
Spend the same money on better seals (double-lip with a flinger ring) and a tighter repack schedule. You'll get more service life per dollar than chasing exotic bearing materials.
References & Further Reading
- Wikipedia contributors. Axle. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
