The Ammonia, Nitrate and Phosphate Field Guide
Ammonia, Nitrate and Phosphate
The three parameters that actually fail — what they are, what to measure, and what we have done about them on real plants in the Philippines.
If your laboratory result came back failed, we can usually guess the parameter before opening the file. It is almost never COD. It is rarely BOD on its own. With reasonable filtration it is not solids either.
It is ammonia, and sometimes phosphate.
This guide is written for the person who has to fix it — a Pollution Control Officer who may not be an engineer, or an owner deciding who to trust with the job. We have kept the chemistry to what you need and left out the rest. If you are an engineer and want the technical version with the full design calculations, ask us and we will send it.
1. Why these three are the ones that fail
Your treatment plant contains two completely different workforces living in the same tank. They look identical in the water. They behave nothing alike.
The first workforce eats the organic matter — what the laboratory reports as BOD and COD. These bacteria are fast, hungry and very hard to kill. They double in a matter of hours. Starve them for a weekend, shock them with a bad batch, wash half of them away, and they are back at work the next day.
The second workforce converts ammonia. These are a different animal entirely. They double roughly once every day or two, not every few hours. They need far more oxygen for the same amount of work. They stop if the pH drifts down. They are killed outright by chlorine, and when you kill them they do not come back tomorrow — they come back in two or three weeks, if you fixed the reason.
This is the whole explanation for the result that confuses everybody: clear water, passing BOD, failed ammonia. Your plant is doing the easy half of its job well. The half that your discharge permit is actually testing you on has quietly stopped.
2. The five things you can measure yourself
You cannot run a plant on laboratory results that take a week to come back. By the time the certificate arrives, whatever caused the problem happened ten days ago and you are guessing.
Everything below can be measured on site, by your own operator, in minutes. This is the same short list we use ourselves.
| What you measure | What it tells you | Where to take it |
|---|---|---|
| Dissolved oxygen (DO meter) | Whether the ammonia converters can breathe at all. Below roughly 2 mg/L the fast bacteria take everything and the ammonia converters get nothing. | In the aeration tank, at the busiest hour of the day — not at the blower, and not first thing in the morning when the building is empty. |
| pH | Whether the process has turned acidic. Ammonia conversion produces acid. Below about 6.8 you are losing speed you already paid for in tank volume. | Aeration tank. |
| Alkalinity (test kit) | Your buffer — the thing that stops the pH falling. This is the single most-missed reading in the country. | Aeration tank, and once on the raw incoming water so you know what you started with. |
| Settled sludge (Imhoff cone) | How much biomass you are actually holding, today, for the price of a plastic cone. It tells you more about tomorrow than any meter on this list. | Aeration tank. Fill the cone, wait 30 minutes, read the millilitres. |
| Ammonia, nitrate, phosphate (aquarium freshwater kit) | Whether the biology is working, before you pay a laboratory to tell you it was not. | Take one from the incoming water and one from the final water. Both. See below. |
| ORP meter | Whether a tank is genuinely aerobic, anoxic or gone septic. Useful when you are setting up denitrification. | Whichever tank you are trying to keep anoxic. |
Everyone tests the water leaving the plant. Almost nobody tests the water coming in. Without the incoming ammonia figure, nobody — including us — can tell whether your plant is removing 90% of the ammonia and still missing the limit, or removing none of it at all. Those are two completely different problems with two completely different price tags, and they produce the same failed certificate.
How we read the cone
We do not track MLSS or sludge age day to day on most plants. Sending samples away takes a week or more, and no operator will keep that up. What we do instead is read the settled sludge in an Imhoff cone and act on it.
Fill the cone with mixed liquor from the aeration tank, leave it 30 minutes, read the sludge level.
We are comfortable running at the low end of that range on one condition: that the plant has fixed-film media in it. The reason is in the next section.
We put the whole set together — Imhoff cone, DO meter, ORP meter, pH meter, alkalinity test kit and the freshwater ammonia, nitrate and phosphate kit — as one package, indicatively around ₱20,000. Ask us for the current price. You do not have to buy it from us; you do have to own it from somebody. A plant run on guesswork fails eventually.
3. Ammonia: what we check, in order
When ammonia fails, we do not assume the plant is too small. Nine times out of ten it is one of four things, and none of them are visible in the water.
Oxygen
Converting ammonia takes roughly 4.6 mg of oxygen for every 1 mg of ammonia nitrogen — on top of everything the BOD bacteria are already using. If your dissolved oxygen sits at 1 mg/L, the fast bacteria take what they need first and the ammonia converters get the leftovers, which is nothing.
We usually aim for around 4 mg/L. On some plants 2 to 3 has been enough. On others, anything below 4 and the water starts going septic and smelling. A dissolved oxygen figure that works on one plant does not transfer to another, and we do not carry a number from one site to the next. Measure yours.
Alkalinity — the one everybody misses
Every 1 mg of ammonia nitrogen converted eats about 7.1 mg of alkalinity. That is not a small number.
Worked through: a plant treating 50 m³/day with 30 mg/L of ammonia is converting 1.5 kg of nitrogen a day, which consumes about 10.7 kg of alkalinity a day. If the incoming water only carries 120 mg/L of alkalinity, it is bringing 6 kg. You are 4.7 kg short before you start. The process eats its own buffer, pH slides down through 6.5, and the ammonia converters slow and then stop.
The operator sees failed ammonia and assumes he needs more bacteria. What he needs is buffer. We have fixed ammonia on more than one plant with nothing but alkalinity dosing and a pH controller.
Locally we have used lime, soda ash and sodium bicarbonate. All three work. How you introduce it matters as much as which one you choose — some systems tolerate dosing it straight in, others need it fed slowly or the pH swings. That part is trial and error on your own plant.
Losing the bacteria faster than they grow
Because the ammonia converters grow so slowly, anything that removes them faster than they reproduce will eventually stop your nitrification. That includes wasting sludge too aggressively — and it also includes losing them by accident, which is far more common than people think. See the second case below.
Shock
Chlorine from a laundry line. Disinfectant down a floor drain during general cleaning. A strong slug from production at 2am. The BOD population shrugs it off. The ammonia converters die, and you find out fourteen days later on a laboratory sheet.
If the plant was passing and suddenly is not, count back two weeks from the sample date. Whatever happened on the plant that week is your suspect. These bacteria do not react on the day. They react late.
4. Three plants, and what actually fixed them
These are our own projects. Names are withheld, the engineering is not.
A meat processing plant — ammonia up to 200 mg/L
Designed for 20 m³/day, actually running about 10. Ham and cured meats, a large staff, and two problems arriving together.
First, they had combined their process water with the domestic sewage from the toilets. That single decision meant the whole combined stream now had to meet the stricter set of parameters — including ammonia, which the process side alone would never have been tested on. Second, they had heavy emulsified oil from cooking plus soap from washdown, and only a grease interceptor to catch it. That job needed dissolved air flotation and did not have it, so everything landed on the biological plant.
Ammonia failed, repeatedly. What fixed it: the plant had two aeration tanks, which is what saved us. We kept the first as moving media — it tumbles and self-cleans, which suits a tank still carrying fat. The second we built out as fixed media: bundles of polypropylene rope, made up like a jellyfish, about fifteen ropes to a bundle, hung down the full five-metre depth of the tank. Then we set the recycle so we were not throwing away alkalinity, and held the settled sludge around 200 mL.
Since then that plant has never failed ammonia.
A 16-room boarding house with shops below — 5 m³/day
A small system built inside an existing three-chamber septic tank. First chamber anaerobic and doubling as equalisation, second and third aeration, and an IBC tank above ground as the clarifier returning to the front.
We failed ammonia over and over, and for a long time we could not say why. The conditions looked right. We could grow the bacteria. We just kept losing them.
The cause turned out to be mechanical. The submersible pump feeding the clarifier kept burning out whenever the water ran low. Every time it failed, the third chamber overflowed — and out went the ammonia converters and the alkalinity with them. The client also dosed chlorine occasionally, which finished off whatever survived.
Two fixes, in this order. We fitted an overload relay so the pump stopped burning out. Then we installed two or three rope bundles as fixed media, properly spaced so oxygen could still move between them.
It has passed ammonia regularly since. We had tried moving media here first and it did not survive — it kept getting stuck or escaping past the strainers.
A poultry dressing plant — 650 m³/day, and the opposite problem
Incoming BOD around 2,000 mg/L and ammonia 250 to 300 mg/L. Four aeration tanks, then another four or five behind them.
The first set ran at a dissolved oxygen of about 1 — which by the book is far too low. But the liquor was chocolate brown, the sludge settled properly and there was no septic smell. We left it alone. There is no point fighting the reality on the ground when the result is good.
Further along, oxygen sat around 6 and there was no ammonia left at all. Nitrification was working perfectly. Which created the next problem: all that ammonia had become nitrate, and now nitrate was the failing number.
The fix used what the plant already had. We converted several tanks ahead of the main aeration into anoxic tanks — no air — and let the plant's own incoming strength destroy its own nitrate. No methanol. No molasses. No purchased carbon of any kind.
5. Nitrate, and the ratio that decides your chemical bill
Nitrate is where ammonia ends up. Convert ammonia with plenty of oxygen and you get nitrate — so a plant that has just fixed its ammonia problem can find itself with a nitrate problem instead.
Destroying nitrate is the opposite operation. You take the oxygen away, and bacteria strip the oxygen out of the nitrate molecule instead, releasing harmless nitrogen gas to the air. That needs a tank with no aeration — an anoxic tank — and it needs food, in the form of the organic strength already in your wastewater.
Destroying nitrate takes roughly 4 parts BOD to 1 part nitrogen. So compare your incoming BOD to your incoming total nitrogen. If the ratio is above about 4:1, your wastewater already carries the food to destroy its own nitrogen and you should not be buying carbon at all. On the dressing plant above the ratio was 14 to 1 — more than three times what was needed. Buying methanol or molasses there would have cost several hundred thousand pesos a year to solve a problem the wastewater was already able to solve itself.
Two more things worth knowing about the anoxic tank, because they are the reason we install one even when nitrate is not failing:
- It gives you back alkalinity. Destroying nitrate returns roughly half the alkalinity that making it consumed. That is often the difference between dosing chemicals every day and not dosing at all.
- It costs almost nothing to run. A tank with no air in it uses no blower power. You are recycling water and letting biology do the work.
How hard we recycle depends on the plant. Where nitrate genuinely has to come down we typically run the return at three to four times the incoming flow, sometimes two. Where ammonia is low to begin with — say 10 mg/L against a nitrate limit of 14 — heavy recycling is unnecessary. We still build the pipework and the pump, and still run it at about one times flow, purely to bring alkalinity back to the front of the plant.
6. Phosphate: the ratio first, the chemical second
Phosphate is the simplest of the three to understand and the easiest to get badly wrong.
Bacteria need carbon, nitrogen and phosphorus in a rough proportion of about 100 : 5 : 1 to build themselves. If your wastewater arrives near that ratio, the biology takes the phosphate up as it grows and you generally do not have a phosphate problem at all. Phosphate becomes a problem when there is a surplus beyond what the bacteria can use — which is common where detergents are heavy.
When there is a genuine surplus, we bind it. A small rapid mixer goes in between the aeration tank and the clarifier, dosing ferric chloride or PAC, which locks the phosphate into a solid that settles out with the sludge. Dosage accuracy matters here — too little does nothing, too much is money into the drain and more sludge to haul.
Once the phosphate is bound into the sludge, that sludge must stay in the clarifier. It must never reach the chlorine contact chamber. If it does, the bacteria are killed there and release the phosphate — and the ammonia — straight back into the water on its way out of your plant. You will have paid for chemicals, produced extra sludge, and failed the test anyway.
On batch plants we go further and put a filter on the decant line rather than let the water go straight to disinfection, sending the filter backwash back to the front of the plant.
7. Two methods we will not tell you to use
Both of these work. Both are sold in the Philippines. We are including them because you will meet them, and because the arithmetic is the honest reason we do not design compliance around them.
Breakpoint chlorination
You can destroy ammonia with chlorine. It takes roughly 7.6 kg of chlorine for every 1 kg of ammonia nitrogen, and 8 to 10 in practice. A plant treating 50 m³/day at 30 mg/L of ammonia would need in the region of 12 kg of chlorine every single day, forever.
That is before you count what it does: chloramines, a heavy chloride load in your effluent, disinfection by-products, and a chlorine residual that kills the same biology doing your BOD removal. It converts a biological problem into a chemical bill that never ends, and it treats the certificate rather than the plant.
Zeolite filtration
Zeolite genuinely removes ammonia at low concentrations, and for polishing a nearly-compliant effluent it has a place.
The question the seller rarely answers is where the regeneration goes. The media fills up and has to be washed out, and every kilo of ammonia you captured comes back off it in a concentrated backwash. If that backwash goes down the same drain you were trying to protect, you have moved the problem thirty metres. You have not treated anything.
If you use zeolite, the backwash has to go back to the front of your plant to be treated properly — never to the discharge line, and never to the ground. If nobody has told you where the backwash goes, you are being sold a filter, not a solution.
8. Why we build plants that forgive an operator
Most operators cannot culture activated sludge well, and it is unfair to design a plant that assumes they can. Staff change. The person who understood the plant leaves.
This is the main reason we put fixed or moving media into most of our plants. Media gives the slow-growing ammonia bacteria something to hold onto, so they are not washed out every time something goes wrong at the clarifier. The plant becomes forgiving. In our own experience — and we say experience deliberately, because we cannot give you a clean scientific explanation for it — the fixed rope media holds nitrifiers noticeably better than moving media does.
That is not a licence to run the plant badly. You still need your oxygen, your pH and your alkalinity. It buys you margin on a bad week, which is a different thing from buying you permission to stop looking.
On the biological side we also use our own probiotic cultures, Active Cells, for start-up and for recovery after a shock — with one rule we apply to ourselves: we will not build your plant so that it depends on a culture, a media or a locked control panel only we can supply. If you replace us, the plant should keep working.
9. Your first thirty days
- Days 1–3. Measure. Dissolved oxygen at the busiest hour, pH, alkalinity, settled sludge in the cone, and ammonia in both the incoming and the final water. Write them down daily. You cannot fix what you have not measured.
- Days 3–7. Fix the oxygen. If DO is under 2, nothing else you do will work. Check the blower duty, check for blinded diffusers, check whether a blower has quietly been out of service.
- Days 5–10. Fix the buffer. If alkalinity is low and pH is drifting down, start dosing and introduce it gradually. Watch the pH respond before you increase.
- Days 7–14. Stop losing biomass. Look for the mechanical fault — a failing transfer pump, an overflow, a clarifier carrying solids over. This is the step people skip, and it is what was actually wrong on our boarding-house job.
- Days 14–30. Let it recover. The ammonia bacteria need two to three weeks to rebuild. Keep measuring. Do not change three things at once, or you will never know which one worked.
10. When it is not an operating problem
Sometimes the honest answer is that no amount of good operation will save the plant.
If your dissolved oxygen is healthy, your alkalinity is adequate, your sludge is in range, nothing mechanical is losing your biomass — and ammonia still fails — then the tank is too small for the load it is being asked to treat, or the process was never designed to nitrify in the first place. At that point more chemicals and more bacteria are money spent to postpone a decision.
Volume is the constraint that is expensive to change and cheap to check. It is also where we start when someone asks us to look at a failing plant, because it decides whether you are looking at an adjustment or a rebuild.
Send us your lab result
Bring your last certificate and, if you have it, the incoming ammonia figure. We will tell you which half of your plant is failing — including when the answer is that your system is fine and the sampling was not. There is no charge for that conversation.
Book a free 15-minute consultationEvery site is different. The figures here are the ones we work to and the results are from our own projects, but your wastewater, your tanks and your discharge classification are not the same as anyone else's — treat this as a way to ask better questions, not as a design. Written by Innovative Water Solutions Inc., Cordova, Cebu.