The Drawing Said BOD 300. Nobody Could Tell Me Where That Number Came From.
The drawing we were given said BOD 300 mg/L. It was written in the influent table like a measurement, in the same typeface as the flow rate and the tank dimensions, and nobody in the meeting questioned it. I asked where it came from. Nobody knew. Somebody had put it there years earlier, and every drawing since had copied it forward.
That is not unusual, and I am not telling the story to embarrass anyone. We ask for an influent lab result on every job and almost nobody has one — on our own project records it is closer to nine out of ten with nothing on file. So the number on the drawing is usually an assumption wearing the clothes of a measurement. The honest response is not to demand data that does not exist. It is to stop pretending we know the number.
We stopped designing to one number
What we do now on every job is design the same plant at several wastewater strengths and put them side by side. The client's own drawing basis is one column. A middle case we think is more likely is another. The strongest specification the client's own operation could credibly produce is the third. Same flow, same site, same discharge target — three influent strengths, and then the chamber volumes, the equipment and the running cost written out under each one.
It takes an afternoon. We do it on every wastewater treatment job we design, domestic or industrial. The arithmetic half of design is not where the difficulty sits any more, and there is no excuse for showing a client one column when three cost almost nothing extra to produce. What it buys is a conversation about risk that can actually be had, instead of a single number nobody is willing to defend.
It also changes what a sample is worth. Once the three columns are in front of someone, the gap between the cheapest and the dearest column is the price of not knowing. When that gap is large, a few thousand pesos of laboratory work is obviously the best money on the job. When the gap is small, we say so and nobody spends anything. The design is where the money on an STP is decided, and this is the clearest example of it I can give.
What moves when the strength moves, and what does not
The thing that surprised the client most is how little of the plant is actually sensitive to the influent strength.
The screening, the grit chamber and the equalisation tank are sized off flow and off the shape of the day, not off BOD. A building that discharges most of its load in a few hours needs the same buffer whether that flow is weak or strong — we have designed around a three-hour wash-down window on a production site and the equalisation volume was set by the clock, not by the laboratory sheet. The clarifier is sized on surface loading, which is again a flow number. Disinfection contact time is flow. The treated-water storage, if there is reuse in the scope, is sized off the use pattern.
So across the three columns, a good half of the concrete barely changes. That matters commercially, because the civil works is the long pole on the programme and at roughly ₱14,000 per cubic metre of capacity at a 3,000 psi mix it is also the largest single line. On one recent job the whole three-column exercise moved the tank block by less than a fifth. Your site will differ — the rate moves with access, water table, ground conditions and who builds it, and a client's own contractor building to our drawings has come in nearer ₱10,500 per cubic metre with no margin of ours in it.
Aeration is where the strength is actually paid for
What does move, and moves hard, is the air. Oxygen demand follows the organic load almost directly, so doubling the assumed BOD roughly doubles the machine. That lands on the capital cost of the aeration equipment and then on the electricity bill for the next fifteen years, which is why we put operating cost per cubic metre in front of the client rather than a purchase price on its own.
Being wrong in the two directions is not symmetrical, and this is the part worth saying plainly. Under-sized aeration is a compliance failure you cannot operate your way out of. Over-sized aeration is a machine that is too big for the tank, and the failure mode there is subtler: too much air shears the floc and you get carry-over at the clarifier, and you strip the anoxic conditions you need for nitrate. It is also usually louder, and a loud machine on a small plant gets switched off at night by somebody who is not the operator. So we do not simply take the strongest column and buy for it.
The way out of that trade is to make the plant turn down. Sizing one duty machine and one standby rather than a single large unit gives a step of capacity that can be brought in when the load justifies it. We have left a plant running at 1 mg/L of dissolved oxygen because it was working, and a plant designed for the strong column but running on the weak one should be allowed to do that rather than be forced to consume its design air every hour of the day.
Ammonia does not follow BOD, and that is the column that catches people
If you only vary one number you will vary BOD, because BOD is the number everybody has an opinion about. That is a mistake. On the sites where we get into trouble, it is ammonia that decides the design, and ammonia does not scale with organic load. A kitchen-heavy building can be strong in BOD and unremarkable in nitrogen. A building that has put a process line into the domestic line on one drawing decision can be ordinary in BOD and carry an ammonia load it never needed to have.
So our columns vary nitrogen independently, and the discharge target is read off the receiving water rather than assumed. Under the General Effluent Standards as amended by DAO 2021-19, ammonia as NH3-N is 4 mg/L for Class C and 3 mg/L for Class B; phosphate as P is 4 for Class C and 1.5 for Class B; nitrate stayed at 14 mg/L for both, which is why nitrate is rarely the first thing to fail. The full table of limits is set out on our compliance page, and establishing which class your outfall actually discharges to is our work, not the client's. What DAO 2021-19 changed and what it left alone is worth reading before anyone assumes a limit has moved.
Varying nitrogen across the columns changes three things at once. The aerobic volume needed for nitrification, because nitrifiers grow slowly and need residence time the carbon-eaters do not. The alkalinity, because nitrification consumes it and a plant that runs out stalls — this is the line where an apparent ammonia problem is really a chemistry problem. And the carbon available for denitrification, because the nitrate you just made has to go somewhere and it needs organic matter to reduce it. Two numbers divided tell you whether you have that carbon or whether somebody is going to try to sell you methanol. On the strong-BOD, low-nitrogen column you have carbon to spare. On the weak-BOD, high-nitrogen column you do not, and the answer has to be found in the layout rather than in a drum of chemical. We never let compliance depend on a chemical that may not be on the shelf in Cebu that week.
Phosphate gets its own row for the same reason. If you want the whole nutrient picture in one place, our field guide sets out the readings and the arithmetic we actually use.
Every column still has a floor
There is a limit to how far the weak column is allowed to shrink the plant, and the limit is not a calculation. A tank has to be enterable, cleanable and buildable. Below a certain size a chamber costs the same to form as the next size up, because the formwork and the walls are the cost and not the cubic metres. Small plants are harder to run than large ones, and a tank drawn tight to the weakest credible influent is a plant with no tolerance at all for the week the occupancy doubles.
So each chamber carries a minimum that the columns cannot go below, and we say in the quotation which chambers are sitting on their minimum rather than on the calculation. On a three-chamber tank of around 70,000 litres we have charged about ₱1.2 million including a proper top slab; the increments between the columns are rarely the interesting part of that figure. Again, site conditions move it.
The same floor logic is what lets us keep the design honest in the other direction. Where the columns diverge enough that the plant would really be a different plant, we say so, and then it is a decision about money and risk rather than a technical argument. We have priced the same building three ways before and found the technology label was not what decided it. A fixed-film stage — FBBR-style media in the aeration tank is the cheapest insurance against the strong column that I know of, because it holds biomass the clarifier never has to return and it makes the plant much less sensitive to being wrong about the influent in the first place.
What we actually put in front of the client
Three columns on one page. Under each: the influent assumed, the chamber volumes with the ones sitting on their minimum marked, the aeration duty, the chemical lines shown as estimates and marked "if required", the footprint, and the running cost per cubic metre. Then a short written statement of what would have to be true for each column to be the right one, and what it would cost to discover halfway through construction that it was the wrong one.
The assumptions and the risks go in writing before anything is signed, including the unflattering ones. A client has the right to know what we have guessed at. That statement travels with the quotation whether the enquiry starts as a new plant or as support on a plant somebody else built. If a supplier hands you one influent column and no statement of where that number came from, that is the question to ask — and it is a fair question to ask us too.
Two things we do not do. We do not choose the column for the client where the choice is really a business decision about how the building will be used. And we do not quietly design to the strongest case and present it as the only option, which is the comfortable thing to do and is also how a client ends up paying in concrete for our uncertainty. The flow you declare on the permit is a ceiling you set yourself, and the same discipline applies to strength: the number that goes in the design is a commitment, not a guess you can revise later.
If you want to see roughly where your own site lands before talking to anyone, the STP price estimator and the design generator will both take a flow and a strength and give you an indication, and the parameters-by-industry page will tell you which parameters your sector is actually held to. Run them at two or three strengths rather than one. That is the whole argument of this post in a form you can do yourself in ten minutes.
Where this does not transfer
This practice earns its keep when the influent is genuinely unknown, which in our experience is most jobs. It earns much less on a site with two years of its own laboratory records, where the sensible thing is to design to the data and spend the afternoon somewhere more useful. It earns less again on a plant that already exists and is failing, because there the influent can simply be measured and a retrofit is a different job from a new build.
The strength bands we consider credible come out of Cebu buildings and Cebu operations, drawn from the plants we have actually built and commissioned, in a warm climate where biology is generally on our side. A cooler site, a different industry, or a building whose use changes after handover will produce numbers we have not seen. And a plant designed across three columns is still only as good as its commissioning — a correctly designed plant can fail its first test for reasons that have nothing to do with the drawing. Three columns reduce the chance of being badly wrong. They do not make anybody right.
If you are sizing a plant now and the influent number on your drawing has no laboratory sheet behind it, that is worth an hour of conversation before it becomes concrete. You can book a free 15-minute consultation and we will look at what you have.