The most common sceptical question we get about reuse is about water. Washing a container uses water; doesn't that undo the point? It is a fair question and the honest answer has always been a ratio — about 44 gallons to return a container to service against about 190 gallons of process water to manufacture its replacement.
That ratio was fine. We wanted it better, so in September we fitted a settling tank and a bag filtration stage to recover first-stage rinse water. Six months of numbers follow.
How it works
The first rinse on any container is the dirtiest and also the largest single water draw. Instead of sending it to the rinsate handling system, it now goes to a 500-gallon settling tank, sits for a minimum of four hours, and is then drawn off the middle through 50-micron and 10-micron bag filters into a holding tank.
That reclaimed water becomes the cold pre-rinse on the next container. It never touches a final rinse, never touches a sanitising step, and never goes near a food-grade cycle beyond the pre-rinse stage.
What it recovered
| Cycle | Containers | Fresh water before | Fresh water after | Recovery |
|---|---|---|---|---|
| Cold triple rinse | 1,844 | 22 gal | 13 gal | 41% |
| Hot caustic (sugar) | 961 | 45 gal | 26 gal | 42% |
| Hot caustic (protein) | 203 | 48 gal | 29 gal | 40% |
| Degreasing | 412 | 52 gal | 52 gal | 0% — excluded |
| Surfactant / defoam | 288 | 40 gal | 40 gal | 0% — excluded |
| Acid neutralise | 117 | 35 gal | 22 gal | 37% |
| Weighted average | 3,825 | 44 gal | 27 gal | 39% |
Where it does not work
Two cycles are excluded, for different reasons.
Degreasing rinse water carries emulsified oil that does not settle out in four hours and blinds a bag filter in about forty containers. We tried coalescing plates; the recovery did not justify the maintenance. That water goes to rinsate handling.
Surfactant rinse water is worse in a subtler way: it recovers beautifully and then defeats the next wash, because carried-over surfactant foams in the pre-rinse and reduces contact between the cleaning solution and the wall. We measured a measurable drop in first-pass cleaning success and stopped.
Cost
| Item | Cost |
|---|---|
| 500-gallon settling tank (used IBC-derived, naturally) | $680 |
| Transfer pump and controls | $1,240 |
| Filter housings, 50 and 10 micron | $820 |
| Plumbing and installation labour | $1,960 |
| Total capital | $4,700 |
| Filter cartridges, six months | $412 |
| Pump energy, six months | $188 |
| Water saved, six months | 65,000 gal |
| Sewer and supply cost avoided | $1,090 |
On pure water cost it pays back in about two years, which is unexciting. It was not built for that. It was built because we publish a water figure with every reuse claim and we would rather that figure kept getting smaller.
Where the number sits now
Net 27 gallons of fresh water to return a container to service, against roughly 190 gallons of process water to manufacture the replacement it displaces. Seven to one, from four to one.
We have said elsewhere on this site that reuse is not free and we will not pretend otherwise. It still is not. But the honest ratio got meaningfully better for $4,700 and a settling tank made from one of our own containers, which is a pleasing sort of circularity even if it is a small one.