A recycling plant in Ohio bought the widest Bunting electrostatic separator on the quote sheet, chasing the throughput in the brochure. The feed going in was a wet, wide-screened automotive shredder residue, everything from fines to eight-millimeter chunks. By the end of the first shift the unit was moving closer to 200 kg/h, copper was reporting into the plastic tail, and two lots had to be re-run. Nothing was broken; the sizing was wrong, and the cost came back as lost product and a rejected lot.
The size band and the surface dryness of your feed set real tonnage, not roll width, and that is the whole game. Bunting quotes its throughput with a qualifier that decides the project: up to 1,000 kg/h (2,200 lb/h) per meter of roll width, depending on material type and particle size. Get the band and the dryness right and the machine's quoted number becomes reachable.
In the First Week, Pin the Feed Size Band
The first week on site is a dry run with no tonnage target yet, because you cannot size anything until you know the band you actually have. Pull a representative sample from the hopper discharge rather than off the top of the pile, split it down and sieve it, then write two numbers on the board: the particle-size band and the moisture content. Compare those against the quoted throughput at that band, because a wide, unscreened cut silently divides the headline number.
Dry electrostatic separation in industry is normally fed a narrow, closely sized cut, commonly a few hundred microns up to roughly two millimeters, and tighter is better because every oversized chunk and every fines cloud steals capacity from the cut that pays. The order for week one is fixed: sample, sieve, record band and moisture, and only then judge your tonnage against that figure. A plant that skips the sieve and orders a two-meter machine to compensate is buying an expensive fix for a screening problem.
By Day Three, Set the Conductivity Split
Bunting electrostatic separators use differences in electrical conductivity to separate and recover dry materials, so the machine does not sort by size or density. By day three the job is to force every particle to touch the roll as a single layer, let the field discharge the conductive fraction so it throws clear, and let the non-conductive fraction hold its charge and fall to tail. Each lever gets set and re-sampled one at a time, because two simultaneous changes make a bad split unattributable.
The feed has to arrive surface-dry; a film of moisture on a plastic particle conducts enough to blur the boundary, which is why a heater or dryer sits ahead of the roll on any humid feed. On the roll itself, confirm a level monolayer with no dust bridging the electrode gap, roll speed matched to particle mass so light flakes are not flung out with the conductors, electrode position giving a clean conductor throw, and a steady feed rate instead of surge feeding that buries the roll in waves. Change one, resample, log the result, and move to the next.
Wiring the 220 to 240 Volt Single Phase Unit
Input power is 220 to 240 volts, single phase. Read that as an installation requirement rather than a nameplate detail: the unit wants its own dedicated single-phase circuit sized for its draw, and it should not share a panel run with large three-phase motors or welders that drag the voltage around. Earth the electrode supply properly and check voltage at the terminals with the machine loaded.
A sagging single-phase supply shows up as a weak, inconsistent throw and erratic recovery long before it ever trips a breaker, so an operator blaming the feed is often looking at a supply problem. Confirm the circuit and the earth at the wall before the unit ships. Fixing it after install means a new circuit, a rescheduled electrical contractor and idle production days, a far bigger bill than the wire itself.
Three Months In, What Wears and What Corrodes
Three months in, the failures stop looking random and start following the service environment. Stainless steel components are specified to provide corrosion resistance, which handles bulk rust but not everything a recycling plant throws at a machine. Humid or salt-laden air condenses on cold electrodes overnight and starts pitting the fasteners and guards. Conductive dust from the feed films across the insulators, and the split degrades before any alarm sounds.
Abrasive grit polishes the roll surface so the throw angle drifts, and caked fines create dead spots that dump non-conductors straight into the conductor bin. The three-month inspection list is short and cheap: check stainless fasteners and guards for pitting, wipe the dust film off the insulators, look at the roll surface finish, and re-check electrode spacing with a gauge. Do these and you catch a slow drift; skip them and you find out through a rejected lot.
One Shift, Start to Finish, on Real Feed
Walk one shift on real feed from dry hopper to product bins. In the first hour the operator confirms the feed band and that the dryer is running at setpoint, then starts the vibratory feeder low and raises it until the roll carries a just-continuous monolayer, no pile and no gaps. By mid-shift the splitter positions are locked and the operator watches three things only: feed rate, roll speed and dust build-up, logging bin weights every hour so a drift is caught while it is small.
In the last hour the run stops, the roll is brushed, the electrode gap is wiped, feed and product bins are weighed, and one composite sample is pulled from each bin for the next morning's check. The order matters because it keeps every tunable variable observable one at a time. When recovery drops at two in the afternoon, the operator knows which step moved instead of guessing across six at once.
End of Run, Confirm the Split Held
End of run, prove the split held instead of assuming it ran. Weigh the feed and each bin, then re-sieve and assay the conductor and tail fractions to get recovery and purity, and compare the achieved kilograms per hour per meter of roll width against the same figure at the identical particle-size band, because comparing across different bands is meaningless. Three numbers belong on one sheet: feed band, moisture, and throughput at that band.
If purity is high but tonnage is low, the band or the feed rate is the lever. If tonnage is high but the tail still carries metal, the roll or electrode setting is. When those three numbers hold across three consecutive runs, the unit is doing what the spec quoted for your stream.
Before you order a model or sign the acceptance sheet, put that one figure from the first week back on the table: the band and moisture of your actual feed. If the band is wide or the feed is damp, the quoted throughput is fiction for your stream, and that single number decides the roll width you buy, the screen ahead of the machine, and whether a dryer is mandatory rather than optional.
