He Asked Us for a Lift Station Five Metres Deep. We Set a Level Instead.

The client asked for a lift station four to five metres deep. It was a reasonable request and he had a reason for it. His collection main runs about two hundred metres from the buildings to the plant and falls somewhere between a metre and a half and two metres over that whole run, so the sewage arrives at the plant shallow, a little under three metres below grade, while the plant itself needs a floor at five. His conclusion was the obvious one: build a wet well at the inlet, put pumps in it, and lift the flow into the plant.

We did not build it. Not because the request was wrong, but because the plant already had a deep chamber with a pump in it, and that chamber could be made to do both jobs if we set one number correctly. The number was the top working water level in the equalisation tank. I have written before about what that tank is for and why we size it off the shape of the day rather than a percentage of the flow, and separately about a job that ended up with no equalisation tank at all because the levels we did not set decided the order of the chambers for us. This note is neither of those. This one is about deleting a structure from a job by choosing a level, and it is the note I said I would not spend inside the piece on uplift I published yesterday.

The arriving invert is the only number that decides this

A gradient of two metres over two hundred is somewhere between one in a hundred and one in a hundred and thirty. That is flat, but it is not unreasonable for the pipe size, and in any case it was already built. What mattered was not the slope. It was the level at which the pipe reaches the plant, because every hydraulic decision inside the plant hangs off that one figure and nothing inside the plant can change it.

Sewage leaves a building at an invert that was fixed when the building was built. That is the same fact that drives most of the above ground against below ground argument: put the tanks in the ground and the wastewater arrives by falling, so the only pumping in the plant is the pumping you chose for process reasons. Put them on a slab at grade and you lift every drop that enters, for the life of the plant. In that note I also wrote that where a lift station already exists, the case for going underground mostly disappears. This job is the mirror of that sentence. There was no lift station yet, and whether one got built depended entirely on where we put the water in the first chamber.

Our first chamber was going to have its floor at about five metres and its roof at grade. The sewer arrives at about two and seven tenths of a metre down. So the chamber is deeper than the pipe that feeds it. Gravity was never the problem. The only way to create a problem was to let the water in that chamber stand higher than the pipe.

What a separate structure actually costs, beyond the concrete

A lift station is not one line item on a quotation read scope by scope. It is a second hole in the ground with its own excavation, its own base slab, its own walls and cover, its own waterproofing and its own pour sequence, sitting a few metres away from a structure that is being poured anyway, and it lengthens the part of the programme that was already the long part. On a job where the civil works are already most of the cost and the footprint is laid out to share walls rather than duplicate them, a detached box is expensive in a way that is hard to see on a bill of quantities built out of rates, because every item on it looks small.

Then it carries equipment forever. A duty pump and a standby pump, a level control, a panel, a power supply run out to it, and a wear item that somebody has to replace, which is an operation and maintenance line rather than a capital one. We treat equipment that fails predictably as a design problem rather than a maintenance problem, and the quiet arithmetic of a lift station is that it adds a pump set whose only function is to undo a level difference. It does no treatment, and on a small plant an extra pump set is a far larger share of everything the operator has to look after. It appears in the running cost per cubic metre every month and it never appears on a lab sheet.

And it adds a failure mode with nothing behind it. We have written up a plant that lost its biomass because a pump stopped, and the general shape of that story is that a plant depending on a pump at its inlet has a single point at the front of the process. A chamber that fills by gravity, the way we also return sludge by gravity where the levels allow it, does not stop filling when the power goes off. It just fills, which is what a buffer is for.

The pump was already in the drawing

This is the part that makes the deletion honest rather than clever. Equalisation only works if what leaves the tank is genuinely constant, which means a transfer pump on level control and not a gravity outlet. That is a rule, not a preference, and I have argued it at length in the note on how we size the tank. So the plant was always going to have a submersible pump sitting in a deep chamber at its inlet end, with floats and a panel and a power supply.

The lift station the client wanted would have had the same equipment, in a separate box, forty or fifty centimetres of wall away from the chamber that already had it. Two pump sets doing one job in sequence: one lifting the flow in, one metering the flow forward. Combining them is not a saving trick. It is noticing that the second structure exists only because nobody wrote down what level the water in the first one would stand at.

So the equalisation tank was sunk to a floor of five metres, its full internal width given to it across the box, and its working range set between roughly two and nine tenths of a metre down and four and eight tenths. The top of that range sits below the arriving invert. The main never surcharges. Nothing is pumped into the plant. The pump that was always going to be there does the only lifting in the front end, and it lifts forward into the process, which is work the plant needed done anyway.

The level stops being an operational preference

Here is the obligation that comes with the decision, and it is the reason this is a design note and not a trick.

On an ordinary plant the high level float in an equalisation tank is a soft setting. Raise it and you buy a little more buffer volume, which on a site whose peak turned out to be concentrated into a few hours is a tempting thing to do at two in the afternoon with a tanker on the way. On this plant that setting is structural in effect. If the water in that chamber is allowed to stand above the arriving invert, the two hundred metre main becomes part of the storage, and a foul main holding sewage at a flat gradient is a different and worse object than a tank holding it.

So the high level is on the drawing with its reason written next to it, as a level referenced to the arriving invert rather than as a float position. That matters more than it sounds, because plants change hands between operators and the third person to hold the job has no memory of why a number was chosen. A good operator can carry a plant, but he cannot carry a decision nobody recorded. We also say it in the commissioning handover rather than leaving it to be discovered, because commissioning is where settings get adjusted by whoever is standing there.

Buffer volume then has to be won somewhere else

If the ceiling on the water level is fixed by a pipe, the tank cannot grow upward. Our working band came out at a little under two metres of depth, between the floor and the invert-referenced ceiling, and the volume had to be found inside that band, which is the same arithmetic as asking what a tank of a fixed size can actually treat. Two ways to do it: more plan area, or a deeper floor.

Plan area was the thing we had least of to begin with, which is the usual constraint on these jobs and the reason the depth available to a chamber keeps turning out to be the governing dimension rather than its footprint. So the floor went down, and the box ended up around eleven metres by six, eight cells, a little over a hundred and eighty cubic metres of tank, with the equalisation cell taking the full internal width. That is a deeper hole than the plant would otherwise have needed, and the depth was bought deliberately in exchange for the structure we did not build.

It also means the buffer volume cannot be checked by eye. You cannot look into the chamber and judge it, because the useful part of the tank is the part below a line you cannot see. It belongs on the drawing as a volume between two stated levels, which is also how it should appear in the design basis a client is asked to approve, and in anything a design tool hands him.

What the deletion cost us

Three things, and they are worth saying plainly because a design decision presented with no cost attached is usually a decision somebody has not finished thinking about.

The excavation got deeper, so the flotation case got worse. On this site the water table stands about two metres below grade, and dropping the floor increases the head of water under an empty tank. That is the case that sized the base slab, and it was handled with a toe extended past the wall line rather than with thicker concrete, which lands in whichever of the three civil routes the job is using. Deleting the lift station and solving the uplift are the same design conversation on this job, not two separate ones.

The transfer pump works against more static head than it would in a shallower tank, because it starts lower. That is a duty point, not a problem, but it has to be sized against the real low water level and not the drawn top water level, which is exactly the class of mistake that shows up as equipment chosen off a daily volume instead of an instantaneous rate.

And a deep chamber needs reachable access. A hatch over the pump, a lifting eye above it, and hardstanding close enough that a pump can come out without scaffolding. The same discipline we apply to getting solids out of a plant applies to getting a pump out of a five metre chamber: it is cheap to allow for at the pour and dear to retrofit, and on a sealed tank somebody forgot to measure the opening of it is not possible at all.

The screening question that comes with it

A pump at the inlet end of a plant is only as safe as what is upstream of it. The job mentioned above, the one that ended with no equalisation at all, went that way because the arriving sewer had no screen on it anywhere, and a pump working in unscreened sewage is a service call waiting to be booked. Here the screening and grit chambers come first, poured with everything else, so the equalisation cell receives screened flow and the pump in it is working in something sensible.

That ordering is not a detail. It is the thing that makes a combined chamber defensible, and if it had not been possible, the honest answer would have been a separate screened wet well after all.

What I would ask before agreeing to a second structure

If somebody has put a lift station in a drawing for your site, these are the five questions I would want answered before the concrete is priced. None of them is about the pump.

• At what level does the sewer actually arrive at the plant, who measured it, and when? A surveyed invert and an assumed invert are different objects, and the same caution applies to the flow figure the plant is being sized against.
• What is the top working water level in the first chamber, and is it above or below that invert? If nobody can answer in levels rather than in percentages, the question has not been asked yet.
• How many separate structures are in the civil scope, and what does each one do that the others cannot? Two boxes with the same equipment in them is the pattern worth questioning.
• What is the high level float set to, and what happens to the incoming main if somebody raises it? If the answer is that nothing happens, fine. If the answer is that the main backs up, that setting needs to be on a drawing.
• What head was the transfer pump sized against, measured from which level? The low water level is the honest starting point.

Those questions sit alongside the rest of what we walk a site for, and they are the kind of thing that gets settled at an ocular rather than in a review of a drawing somebody has already finished.

Where this does not transfer

This worked because there was depth available to buy. The ground allowed a five metre excavation and the client had the setback space around the box to work in. On a site with rock at three metres, or a boundary wall tight against the tank, or a water table close enough to the surface that going deeper makes the flotation case fail rather than merely get worse, the level cannot be set where it needs to be and the separate lift station is the right answer. It is the right answer on a plant standing above ground too, where the inlet lift exists by definition, for the reasons set out in that note.

It is also the right answer when the arithmetic runs the other way. If the only route to an invert-referenced ceiling is sinking the entire plant two metres further than it otherwise needed, you are paying for excavation, waterproofing, a worse uplift case and a deeper access problem across every cell of the plant, in order to delete one small box. A shallow screened wet well with a modest duty and standby set can be cheaper and simpler than that, and I would rather say so than pretend the combined chamber is always the better engineering. It was better here because the depth was going to be bought anyway, and the jobs behind us are where that judgement comes from rather than from a rule.

And where the main arrives deep already, none of this is a decision at all. The chamber fills, the pump meters it forward, and nobody ever has the conversation. Most of the plants where this goes wrong are the ones where the arriving level was never written down, which is a drawing problem rather than a construction problem, and those are the expensive kind.

One last caution on reading your own job against this one. Our numbers came from a main that was already built, a water table we had measured, and a volume the client had agreed, drawn as usual against more than one wastewater strength. Yours will differ, and the load your site produces, the standard you discharge to and the figure on your discharge permit all move the volumes before any of the levels get chosen. If you want a rough sense of what a plant of your size involves there is an estimator on the site, and the wastewater services page sets out how we work through a new plant from the survey onward, across Cebu and the rest of the country. If you would rather talk it through, book a free 15-minute consultation.