Case Study: Railcar Wash Boom Retrofit From Spray Bars to Rotary Jets
A railcar wash bay swapped static spray bars for rotary jet heads. Water per car fell from 1.05 to 0.70 cubic meters and rewashes from 14 percent to 5.
Exterior railcar washing is a coverage business run at industrial scale. Tank cars and hopper cars move food-grade syrups, petroleum coke, and cement dust. The shell has to come clean between loads.
A North American tank car cleaning operator runs a two-bay facility at 60 to 90 cars a day. Each bay washed cars behind static fixed spray bars. The bars flooded the visible shell and left every shadow zone dry.
That is where rewashes are born. The shadow behind a hand-brake wheel looks clean from the gantry and tests dirty on the floor. This case covers one retrofit at that class of facility. Rotating jet heads went onto the wash booms, wands stayed for the corners, and the recirculation loop was reworked. Every figure is a working figure for this class of rail car washing equipment, checked against published bands rather than a plant record.
Case Snapshot
The figures below are the generic shape of this class of railcar wash installation, not a plant record.
| Item | Value |
|---|---|
| Industry | Rail freight services, tank car cleaning |
| Asset washed | Tank car and hopper car exteriors, 14.5 m class |
| Bay layout | Two drive-through bays, 6 m wash tunnel per bay |
| Throughput | 60 to 90 cars per day across both bays |
| Before | Static fixed spray bars, 224 orifices per bay at 4.2 bar |
| After | Eight rotary jet heads plus 40 static orifices per bay at 5 bar |
| Duty cycle | 16 h per day, two shifts, winter freeze protection in service |
| Water basis | Fresh mains metered per car, caustic pre-soak recirculated |
| Before rewash rate | 14 percent of cars given a second pass |
| After rewash rate | 5 percent working target |
A 14.5 m car at 0.15 m/s through a 6 m tunnel spends about 110 seconds inside the spray zone. Every sizing number in this case hangs off that dwell window, from head count to orifice wear.
Challenge
The line ran a caustic pre-soak followed by a high-pressure rinse, which is the normal shape of a heavy-duty railcar wash. Fixed spray bars wrapped the car on both sides, over the top, and at each end. The bars delivered plenty of water. They delivered it at low impact, and they delivered it in straight lines.
Straight lines are the problem. A tank car exterior is a forest of hardware. The list includes the hand-brake wheel and its chain guard, ladder brackets, and the brake cylinder with its clevis. Add the safety appliances on the top deck and the rocker assembly under the running board.
Each item casts a dry shadow a few hundred millimetres wide on the shell behind it. Pre-soak softens the film there, the bar keeps it wet, and nothing strips it.
The operator tracked consumption per car the way most serious shops do. The metered figures for this class sat near 1.05 m3 of fresh water per car. Caustic ran about 6.3 L per car on a once-through pre-soak. The pain number, though, was the rewash rate. Roughly one car in seven went round a second time, and about one in twenty still needed a manual wand pass before release.
A rewash is not a rounding error. Each second pass adds a full increment of water, chemical, bay time, and labour. It also puts the car behind schedule for its next load. On 75 cars a day, a 14 percent rewash rate is more than ten cars of hidden work. Rail car washing equipment sized on the first pass quietly loses the whole argument on the second one.
The audit pressure made it worse. Cars carrying food-grade cargoes need documented cleaning, and a bay whose outcome depends on which operator holds the wand cannot document much. The plant needed coverage it could prove, at a water number it could hold through a permit review.
Soil and the Shadow Map
The soil decides the energy the nozzle has to deliver. On this duty the shell carries three families of residue, and each one behaves differently under a spray.
| Soil family | Where it bonds | What strips it |
|---|---|---|
| Dust and fugitive powder (petcoke, cement) | Upper shell, running boards | Low energy, high coverage: keep it out of the soak |
| Dried film (molasses, corn syrup, vegetable oil) | Shadow zones, weld seams | Caustic soak plus patterned impact |
| Grease and brake dust | Truck-side lower shell, brake beam | Hot soak plus the highest impact on the boom |
The dust family is a coverage problem. You win by intercepting it before the soak, which is why the pre-rinse ahead of the caustic matters more than its volume. The film and grease families are impact problems. A sheet of wet caustic will sit on dried syrup overnight and do very little. Stripping it takes a jet that hits the same spot repeatedly with real velocity behind it.
So the first engineering document was not a hydraulic drawing. It was a shadow map: every fixture on a representative car listed with its stand-off distance from the shell. That inventory set the target zone list for the boom layout.
| Typical fixture | Stand-off from shell | Treatment |
|---|---|---|
| Hand-brake wheel and chain guard | 150 to 250 mm | Rotary head pattern, wand finish |
| Ladder bracket and grab iron | 100 to 200 mm | Rotary head pattern edge |
| Brake cylinder and clevis | 200 to 300 mm | Low side boom head, 4 to 5 bar |
| Top deck valves and hatch cover | Complex | Retained wand, not automated |
Solution
The retrofit replaced the fixed spray bars on the two side booms and the two end booms. Each bay received eight gear-driven rotary jet heads: four per side at mid-height and two per end. A reduced field of flat-fan car wash nozzles stayed in the top boom for the pre-soak curtain.
Each rotary head runs at about 45 L/min at 5 bar at the head. The class envelope published for these devices runs roughly 25 to 60 L/min at 2 to 4 bar. Heavy-duty wash sets see comparable flow at 4 to 7 bar, and the chosen point sits inside that band.
The heads were sized from three numbers: flow per head at duty pressure, impact pressure at the target surface, and dwell per car length. Flow per head sets the orifice. Impact at the surface sets the maximum throw and therefore the boom stand-off. Dwell sets how many pattern cycles the shell actually sees.
| Sizing step | Working value | What it locks |
|---|---|---|
| Flow per head | 45 L/min at 5 bar | Equivalent orifice near 5.5 mm |
| Boom stand-off | 1.2 to 1.5 m nominal | Impact stays high at the shell |
| Dwell in tunnel | 110 s per 14.5 m car | Pattern cycles per pass |
| Pattern period | 1 to 2 s per revolution | Dozens of repeat impacts |
Wands stayed on the line on purpose. The top deck of a tank car is crowded with fittings at distances no fixed device can service cleanly. Forcing automation there produces an expensive compromise.
A retained high pressure spray wand at 6 to 7 bar covers the deck and the worst brake shadows. The wand operator now finishes a job the booms have already 90 percent cleaned, instead of washing the whole car by hand. A rotary jet head on the boom and a trained hand on the deck is still the most robust layout in this industry.
The recirculation loop changed with the booms. The once-through pre-soak became a recirculating sump, with chemical makeup dosed on concentration. Filtration ahead of the boom pumps moved to a band around 100 microns. That is the working band for this class of rail car washing equipment. It protects rotary heads and small orifices from grit, without financing a polishing loop.
The 100 micron figure is class data, not a rule. Cars carrying abrasive mineral dust push finer, and cars with soft floc residue do not need it. Head service envelopes, published flow bands, and filtration practice are compared in the tank cleaning applications guide.
Engineering Notes
Six pieces of arithmetic carried the design. All of them are checkable with a tape measure, a pressure gauge, and a flow meter.
Impact pressure against throw. For hollow-cone and fan jets of the fixed type, published data is commonly read as impact falling with the reciprocal of throw. Call it about 1/(throw) while coherence survives. Coherence is exactly what a shadow-prone bay tests.
A jet landing at 2.0 bar impact one metre out gives about 1.0 bar at 2 m, and about 0.67 bar at 3 m. That is why the retrofit moved the booms to a 1.2 to 1.5 m stand-off instead of chasing reach with the pump. Buying impact with pressure costs power at 3/2; recovering it with geometry costs a steel shop a day of boom work.
Coverage and spacing. For a solid stream with total angle theta, pattern width at distance d follows width = 2 * d * tan(theta / 2). Worked at the top-boom rinse nozzles, a 25 degree stream at 3 m gives 6 * 0.2217 = 1.33 m of width. Spacing those orifices at 1.0 m therefore leaves 0.33 m of overlap, about 25 percent of the spray width. Wider pitch opens dry stripes, and a dry stripe on a moving car is a rewash with paperwork attached. The spray angle guide walks through the same geometry for fan and cone families.
Pump duty against pressure. Through a fixed orifice, flow rises with the square root of pressure. The before layout ran a boom field at 4.2 bar and 1,200 L/min.
Holding those orifices and demanding 6.0 bar gives 1,200 * sqrt(6.0 / 4.2) = 1,434 L/min, about 19 percent more water. Hydraulic power is pressure times flow, so it scales as P^1.5. That step is 4.2e5 Pa * 0.02 m3/s = 8.4 kW against 6.0e5 Pa * 0.0239 m3/s = 14.3 kW, about 70 percent more. The retrofit therefore held the pressure target at the head through smaller orifices and shorter throws, not a bigger motor. The nozzle flow calculation guide covers the orifice equation behind these numbers.
Dwell per car length. Dwell = (car length + overtravel) / line speed. At 0.15 m/s, a 14.5 m car with 1 m overtravel each end spends 16.5 / 0.15 = 110 s in the tunnel. At a 1.5 s pattern period, each head revisits one band of shell about 70 times. That is more than enough repeat impact for film soils.
Watch the slow line, not the fast one. Plants running cars at 0.10 m/s to “be careful” gain dwell but flood the sump balance. The shadows stay untouched, because low-impact coverage was never the missing ingredient. This dwell math separates rail car washing equipment sized for the bay from a catalogue of nozzles dropped into one.
Orifice sizing check. A 1 mm equivalent orifice at 1 bar passes about 0.67 L/min. Area 0.785 mm2 times ideal exit velocity 14.1 m/s gives that figure. Flow then scales as Q = 0.67 * d^2 * sqrt(P), with d in mm and P in bar. A head at 45 L/min at 5 bar needs an equivalent orifice of sqrt(45 / (0.67 * 2.24)) = 5.5 mm. That is two 4 mm jets, or one 5.5 mm path, per head.
Ideal exit velocity at 5 bar is sqrt(2 * 5e5 / 1000) = 31.6 m/s. Orifice friction, with discharge coefficient 0.9 to 0.95, puts the real jet slightly below that figure. That is the energy the shadow zones never received from a flooded bar.
Freeze protection. A north-facing bay freezes what it cannot drain. Header slopes on this class of installation run to a minimum of 1 in 100 toward low-point drains. A 15 m boom then sits 150 mm above its drain outlet. Drain valves stay open between cycles, and each boom section empties by gravity in a couple of minutes.
Freeze breaks nozzles and splits headers quietly, at 3 a.m., in January. The winter failure signature is a dry stripe in March. Insulate the chemical loop, trace the wand standpipes, and make the blow-down an interlock, not a habit.
Boom Hydraulics and Sump Balance
The booms are only the visible half of the system. The sump decides whether the loop survives winter and a shift of hopper cars.
The recirculating soak sump holds about 2 m3 and turns over through the booms during the soak pass. Evaporation and drift leave through the top of the tunnel. Drift loss for this class of bay is tens of litres per car, not hundreds, precisely because rotary heads run at modest flow.
Makeup water returns the sump to level, and caustic doses on concentration. The working band is typically 2 to 5 percent for heavy film soils. It is dosed against strength rather than a fixed volume per car.
Fresh water stays once-through where hygiene demands it. The pre-rinse and the final rinse draw from the metered main. The water ledger then reads as three lines. Pre-rinse takes 0.20 m3, and the final pass adds 0.132 m3.
The HP rinse puts four running heads at 45 L/min for 60 s, giving 0.36 m3. The total is about 0.70 m3 of fresh water per car, against 1.05 m3 on the old once-through layout.
| Loop element | Working figure | Watch item |
|---|---|---|
| Soak sump | 2 m3 recirculated | Concentration, temperature, oil load |
| Filtration | around 100 micron band | Differential pressure across the bag unit |
| Pre-rinse | 200 L/min for 60 s | Keep dust out of the soak |
| HP rinse | 360 L/min per head bank, 60 s | Impact at the shell, not at the gauge |
| Final pass | 44 L/min for 180 s | Stripe-free overlap from the top boom |
Grit is the enemy the ledger hides. Hopper cars bring mineral dust into the sump. Grit in the bearing races of a rotary head slows it, then stalls it. The head then lays a partial pattern while the controller logs a normal-looking cycle time.
Filter changes, not head replacements, are the cheap insurance. The industrial water spray nozzle guide covers material and wear selection for these loops.
Maintenance and Spares
Nozzles in this service do not fail open and dramatic. They wear, pass more water at the same pressure, and quietly rob the neighbouring jets of impact. Published wear data puts the flow rise at end of life around 5 to 10 percent, which is the trigger used here.
| Item | Count per bay | Replacement basis |
|---|---|---|
| Rotary jet heads | 8 | Rebuild kit at 2 years, replace on pattern loss |
| Static orifices (booms + top) | 48 | Flow check at +5 percent, about 30 percent per year |
| Fan nozzles on the soak curtain | 16 | Wear on salt-laden film duty |
| Wand tips and hoses | 4 sets | Annual, abrasion duty |
The monthly routine is short. Section the gate valves and check flow at each header. Walk the patterns from the gantry with a visual. Log the differential pressure across the filtration unit.
The before bay burned roughly 90 orifices a year on rodding and replacement arithmetic. The after bay holds about 40 static orifices in service, and consumes spare parts worth a dozen orifices plus one head rebuild kit per year. The old habit of chasing plugged bars with a wire also ended, and that recovered a maintenance queue nobody booked.
Results
No plant audit figures appear below. These are working outcomes for this class of retrofit, built from the sizing arithmetic above and kept inside published bands for rotary head duty.
| Metric | Before (static bars) | After (rotary retrofit) | Basis |
|---|---|---|---|
| Fresh water per car | 1.05 m3 | 0.70 m3, a 33 percent cut | Phase volumes in the sump ledger |
| Caustic per car | 6.3 L once-through | 3.5 L on recirculation | 0.6 percent before, 0.5 percent dosed on volume |
| Rewash rate | 14 percent | 5 percent | Shadow zones swept by patterned impact |
| Cycle time per car | 15 min bay occupancy | 11 min | Less wand time, same tunnel speed |
| Nozzle replacements | about 90 per year | about 13 parts per year | Orifice count times wear bands |
| Boom pressure at head | 4.2 bar | 5 bar | Impact held by geometry, not motor size |
The rewash cut deserves the last word, because it multiplies everything else. Metered water per car falls 33 percent on the first pass. The rewash drop removes a further slice of second passes on top of that. At 75 cars a day, moving from 14 percent to 5 percent rewashes saves about seven car-washes of water, chemical, and bay time every day. Not one extra litre is touched on the first pass to buy it.
Why This Matters
The sizing method, not the hardware list, is the transferable asset. Intermodal tank cleaning shops, gondola bays, and locomotive washes all run the same four numbers. Those are impact at the target surface, overlap at the working pitch, dwell at the line speed, and power on the P^1.5 curve. Any rail car washing equipment study that skips one of the four will surprise somebody during commissioning.
The layout logic transfers too. Fixed bars fail on shadows because they flood; rotating heads succeed because they concentrate and repeat. A retained hand wand is not a concession, it is the right device for a crowded top deck. The decision checklist that came out of this bay is short enough to laminate.
- Map the shadows on one representative car before quoting a single nozzle.
- Set boom stand-off first, pressure second; 1/(throw) is a tax you want to avoid, not outrun.
- Size overlap from width = 2 * d * tan(theta / 2) at the worst-case distance, not the nameplate.
- Check the pump against flow at pressure at the head, with the filter dirty, in February.
- Keep wands where fixtures beat geometry, and meter water per car from day one.
A tank cleaning equipment selection built on that order behaves predictably across car classes. A bay that behaves predictably is a bay you can staff, permit, and audit.
FAQ
How does a rotary jet head work? Liquid under pressure enters the body and drives an internal gear set that spins one or more nozzles through a fixed pattern. There is no motor and no wiring in the wash zone. The pattern repeats every one to two seconds. Every point inside the swept band therefore takes many impacts per dwell window, not one continuous flood.
What pressure does a railcar wash boom need? For this class of exterior duty, 4 to 7 bar at the head is the practical band, and this bay runs 5 bar. Higher pressure buys little once throw is long, because impact at the surface is governed by the distance the jet travels before it lands.
What fails first on car wash nozzles in this service? Orifice wear from abrasive fines, then bearing wear inside rotary heads when filtration is skipped. Both show up as rising flow at constant pressure, before any visible pattern defect. That is why the monthly flow check is on the list.
Can a bay retire the high pressure spray wand completely? Rarely, and usually not on tank cars. The top deck has too many fittings at irregular stand-offs. Retain the wand for the deck and the true shadows. Let the automated booms carry the shell, so the wand time left is short and documentable.
The Final Call
BoreJet supplies nozzles and does the sizing arithmetic behind them: flow at pressure, pattern geometry, filtration, and drain detail for this class of bay.
Send the shadow map, the metered water figure, and the pump curve through the contact page. The response comes back with head count, orifice sizes, and the pattern that fits your dwell window.
Your Duty May Differ
Send the Vessel Drawing and the Pump Curve.
We size the head class, the count and the operating point for your tank, not a catalogue guess. A coverage test on the actual vessel beats any estimate.