🧪 What are day-average samples for? The story of a certain order that we will remember for a long time
A wastewater treatment plant is no place to cut corners. It's a laboratory, an operations hall and a crisis management center all in one. But sometimes... well, that's just it. Sometimes you get an order that reads, "Take one sample in the morning and send the result." Simple. Too simple.
And then the story begins, which you then tell for years. Just such a story happened at our company. It started with a sample - and ended with emails with the client's lawyer, hourly analyses and jokes about fermentation vat lattes.
📦 Outsourced order - "one sample, thank you."
It was Thursday morning. Traffic fairly stable, inflow under control. An order from a customer from an external company came in: "Please take one sample of raw sewage at 7:00 a.m. It is to represent the entire 24 hours. Simple tests: BOD5, COD, suspended solids."
Even at first glance - something is not sticking. But the customer is our master. We do according to the guidelines. Spot sample taken at 7:00, prepared, sent to the laboratory. By 11 o'clock we have the results.
🔴 COD: 2960 mg/l
🟠 BOD5: 1540 mg/l
⚫ Total suspended solids: 1270 mg/l
Alert. That's 3x more than usual. The customer calls in a panic: "Is something wrong! Is the treatment plant working!"
🔍 Hourly analysis - or why a point sample is a bad idea
We log into the flow monitoring system. And what do we see? Around 6:30-6:50 a.m., an intensive industrial discharge - starchy, from food production - came into the system. To the eye - a load of the type "dense kisel". By 7:15 it was all over.
So the sample at 7:00? It hit exactly at peak of contamination. One moment out of a whole day. One impulse that absolutely does not represent the standard.
📘 What is a day-average sample?
It's bulk sample from the whole day - collected at regular intervals (e.g., every hour), often in proportion to the flow. It makes it possible to obtain average result, which realistically shows what goes into the treatment plant for 24 hours.
Without this, we can hit a "peak," a "hole," or a single incident that distorts everything. And then clients scare lawyers because the numbers are not what they wanted (read: not what they expected).
☕ Latte with COD - or what the samples looked like
The 7:00 sample literally looked like a cappuccino made of cellulose and fat. Frothy, slushy, milky layer and heavy smell. Against the background of other hours of the day - like sabotage.
And this is where the moral comes in...
📌 The moral: a sample is not a selfie - it's an average
- 📊 Take daily-average samples, not point samples, if the result is to be reliable,
- ⏱️ Monitor flows and correlate time with chemical results,
- 🧠 Educate customers - a sample at 7:00 means nothing without context,
- 📷 Because one photo in the morning can show something completely different from the whole day.
💬 And finally...
If you like this story - like it, share more and leave comment at Facebook or LinkedIn.
You can also buy us coffee if you like what we do: 👉 https://buycoffee.to/img/share-button-primary.png ☕
Wastewater Tales - A blog from the other end of the pipe. 💧🧤
English 
Polish 
Spanish