We Softened the Water to Save the Grease Trap. The Softener Drained Into the Grease Trap.
We put a softener on a commercial kitchen in Cebu about seven months ago. The reason was not scale in the boiler and it was not the dishwasher. It was the grease trap. The crust in that trap was calcium soap rather than grease, and once hard water has saponified kitchen fat there is nothing you can dose, skim or flush that will undo it. The only move left is upstream: take the hardness out of the water before it ever meets the fat.
It worked. For about four months the trap stayed workable and the skimming went back to being a weekly job instead of a fight.
Then the crust came back. Not everywhere — in one compartment, hard, and sitting lower in the water than the old crust had.
The softener was not the problem
The first thing we did was test the water at the kitchen taps. Hardness was where it should be. The resin was exchanging, the regeneration cycle was running, the unit was doing its job. We checked the service flow rate too, because an undersized softener can pass hard water during a breakfast peak while still testing soft an hour later. It was fine. The kitchen was getting soft water.
So the calcium was arriving some other way.
We followed the softener's drain line. It ran about two metres across the floor and discharged into the kitchen floor drain. The kitchen floor drain feeds the grease trap.
Every night, that softener was emptying its regeneration water into the one compartment we had installed it to protect.
What one regeneration actually carries
This is worth doing in numbers, because the shape of the answer surprised us.
The unit holds roughly 50 litres of resin. Regenerated at about 100 grams of salt per litre of resin, that is 5 kilograms of salt per cycle, and it recovers something on the order of 2.5 kilograms of hardness as calcium carbonate off the resin bed. All of that leaves in the backwash, brine draw and rinse — call it 250 litres of water in total.
Two hundred and fifty litres carrying 2.5 kilograms of hardness is about 10,000 milligrams per litre as calcium carbonate. The raw water feeding the building is around 280. So the drain line was delivering water roughly thirty-five times harder than the supply, straight into a chamber full of fatty acids, once a night.
The calcium in a single regeneration is, on its own, enough to bind something on the order of ten kilograms of fatty acid into soap. That is arithmetic and not a measurement — it assumes every calcium ion finds a partner, which it will not. But it does not need to be anywhere near complete to build the crust we were looking at.
We had not taken the calcium out of the building. We had concentrated it, put it in a bucket, and poured the bucket into the grease trap.
The number that looked reassuring was the wrong number
Before we found the drain line, we had already done the salt arithmetic on the treatment plant, and the plant arithmetic says there is nothing to worry about. Five kilograms of salt spread through a building discharging 30 cubic metres a day is under 200 milligrams per litre. Nothing in a conventional biological plant is troubled by that, and nothing in the plant removes it either — biology does not touch chloride, settling does not, and chlorination adds to it. It arrives, it dilutes, it leaves.
That daily average is true and it is useless. What matters is not how much salt arrives in a day but what volume it meets in the forty minutes it takes to arrive. On this building the regeneration drained at two in the morning, when the equalisation tank was at its lowest standing level of the day. Same mass, a tenth of the dilution.
On a plant with a properly sized equalisation tank that is still a non-event. On a plant with no equalisation at all, the slug goes into the aeration tank undiluted, and on a 5 cubic metre a day plant with an aeration tank of two or three cubic metres you are no longer in non-event territory. You are in the range where it is worth thinking about, every night, forever.
And in this case none of that mattered, because the brine never reached the plant as a slug anyway. It stopped in the grease trap and left its calcium behind.
What we did
We re-piped the softener drain. It now runs past the grease trap entirely and discharges into the equalisation tank, which is where a stream carrying nothing but dissolved salts belongs — downstream of the chamber whose job is fat, upstream of the tank whose job is smoothing out everything the building throws at it.
We changed the regeneration from a timeclock to a meter. The unit had been set to fire every night at a fixed hour whether the resin needed it or not, which on this kitchen's actual water use meant regenerating at about two-thirds exhaustion. Metered regeneration cut the number of cycles, and every cycle it removed is salt not bought and calcium not discharged. Roughly a third less of both, on their consumption pattern.
We moved the time. The installer had set it for two in the morning for a sensible reason — soft water available all day, no interruption during service — and we kept that reasoning but shifted it to the quiet hour of the late afternoon instead. It now arrives when the plant has flow and the equalisation tank has liquid in it.
And then we did nothing else. No neutralisation step, no dilution drum, no chemical. There was a version of this job where we built a small holding tank to bleed the brine out slowly, and we costed it, and it was a tank that treats nothing — the same argument we make when a client asks us to add something to a dosing product that does not need it. The pipe route and the clock setting did the whole job for the cost of two metres of pipe.
The part routing does not change
Worth being clear about this, because it is easy to oversell what we did.
The salt you buy is the salt that reaches the plant. Nothing consumes it anywhere in the building. Roughly half of each cycle's sodium leaves on the treated water over the following day and the rest leaves in the rinse, and the hardness comes off the resin either way. Re-piping and re-timing changed which hour it arrives in and what it was mixed with when it got there. The mass going to the receiving water was the same on Tuesday as it was on Monday.
So if a building's discharge limits or its receiving water make dissolved salts a genuine constraint, none of the above helps and the conversation is a different one — about whether softening is the right answer at all, and about which parameters that particular industry is actually held to. That is not a pipe decision.
Where this does need holding back
We have had one job where the arithmetic came out the other way. An industrial softener on a process line, several hundred litres of resin, 30-odd kilograms of salt per regeneration, feeding a plant treating about 20 cubic metres a day. There the slug was large relative to everything downstream of it and we did build the holding drum, bleeding it into the equalisation tank over several hours. Same reasoning, opposite answer, because the ratio of slug to receiving volume was ten times worse.
The water bill is usually enough to work out which case you are in, before anyone buys anything. Regeneration volume against daily flow, and the standing volume of whatever the drain line reaches first.
What does and does not travel
The crust diagnosis travels. If a trap keeps building hard crust in a building on hard water, softening the kitchen supply is a real fix and we would do it again — it is the whole reason we built out the softener work on the commercial kitchen side in the first place. So does the reminder that a trap only works if it holds the water long enough to separate anything, which is a retention time question before it is a chemistry question.
The drain-line mistake travels further than we would like. We did not specify where that drain went, and the plumber ran it to the nearest floor drain, which is what any reasonable plumber does. Nobody did anything wrong. A detail that nobody owned decided the outcome, which is the same way the odour complaint on another job came up a floor drain rather than through the slab.
The numbers travel least. Our hardness is Cebu limestone groundwater, and 280 milligrams per litre is a local figure, not a national one. A resin volume, a salt dose and a regeneration frequency are specific to one unit on one building's consumption. Our fixed-film media holds biomass in place, so in our experience a plant built that way shrugs off a salt slug that would visibly upset a purely suspended-growth tank — but we have also seen a plant lose its nitrifiers to something nobody thought was a shock at all, which turned out to be a pump, not chemistry. Every site has its own version of the detail that nobody specified.
If you want to see the kind of work this comes out of, or what a plant of a given size tends to cost before anyone walks the site, the estimator is a reasonable starting point, and we keep operations and maintenance support for plants we did not build as well as ones we did. Most of what we do is in and around Cebu, though the question of where everything you take out of a plant actually goes is the same question everywhere.
If you have a grease trap that keeps crusting over, or a softener and a treatment plant on the same building and no clear picture of how they talk to each other, book a free 15-minute consultation and we will go through the arithmetic with you.