SAMPLING

I have written some articles on sampling, most notably Sampling, Analysis and Quantum Mechanics.   I will boil down what I have written before into something easier to read.

First:  very few of you take enough samples or run enough analyses, at least in Ohio.  In Michigan, the State Dept of Environmental Quality makes the plants take enough data to do proper analyses.  Remember, IF YOU DON'T MEASURE IT YOU CAN'T CONTROL IT!  Here are a couple of oddball examples:  An SBR plant doing biological Phosphorus removal had aerobic digesters.  When they supernated the digesters, they blew their Phosphorus limit for the day.  Why?  Because the spernatant had 70 ppm of Phosphorus, and the flow was about 40% of the average daily flow.  Luckily,  they were in Michigan and had to report that on their daily monitoring report.  A second example:  A superintendent supernates the anaerobic sludge storage tanks.  The supernatnant ammonia is about 600 ppm but she does not know that.  I was there that day and asked her to take some samples.  The primary effluent ammonia was 60 ppm, and the activated sludge effluent was 30 ppm.  We were able to divert effluent to some empty sotrage lagoons and recycle it through the plant, avoiding an exceedance.  For my help, I got taken off the project.  Never expect gratitude when you are a consultant.

I have also worked at Class IV plants that did not have enough data to properly calculate volatile solids reductions.

Next, you should all be taking composite samples for your MLSS, RAS and WAS sampling.  Why? Because samples are variable in time and in space.  If you take 10 MLSS samples, one adfter another, you will get a variatioin of about 5% from the average. As to variation over time, as the flows change during the day, some of the MLSS gets transferred from the aeration tank to the clarifier and then later, back.  If you have the operator run out first thing in the morning and take a MLSS sample, then calculate your wasting, and start your WAS pump after noon, you will not waste the right mass; I guarantee it.  It is better to get a 24 hour composite sample and run it the next day to determine  what your MCRT was that day, and base your wasting on history and trends, than to fall into the trap of todays data.

Plotting and trending - No I don't mean plotting your boss's downfall and trending your pay.  I mean your data.  You all have Microsoft Excel spreadsheets, so you can store, plot, graph and trend all you data.  So get to it.

I'll give you another lecture soon