**** Ulrich Bretscher's Wastewater Page ****



The Measurement of Flow in Clarifiers

Usually a clarifier is the weakest link in the chain of sewage treatment. A malfunctioning clarifier can spoil the very best biological treatment result. Unfortunately, nobody can look into the water body of a clarifier. Hence the distribution of the activated sludge flocs, the flow and possible turbulences, the settling and thickening of the sludge and how the sludge is conveyed to the hopper usually remains a secret.

What you may see are clouds of flocs near the effluent of the basin. What first and easily can be checked then by a probe is the sludge deposit at the bottom of the tank. The rest remains guesswork. As a chemical engineer, working among civil engineers, you are considered a wizard who may drop some magic elixir into a malfunctioning clarfier and then it's working again. No problem, since the clarifier was designed according to the rules specified by the authorities, what gave the civil engineers their alibi.

Fig, 1:
Typical clouds of activated sludge flocs at the effluent side of a final clarifier.
STP Einsiedeln, 1991

Of course, I had to come up with a more suitable and stalwart answer. The flow and the source of the floc clouds had to be investigated. In a next step then there were to study countermeasures. Sometimes that meant bench-scale or pilot plant tests for finding a cheap and predictable solution.

In the early 1970s two kinds of flow meters for the determination of flows were on the market:

  1. The Ott-propeller. (German: "Ott-Flügel") A low friction propeller with an electronic revolution counter.
    At a flow of at least 10 cm/s it worked reliable. But it was impossible to measure flows at less than 5 cm/s. And that's what I was looking for.
  2. A hot-wire flowmeter. This instrument allowed to measure flows within the millimeter per second range. In the presence of activated sludge flocs the measurement was found unreliable. After a short time the heated wire was coated by activated sludge flocs what hampered the heat flow.

So there were to find other, more reliable means for measuring the flow. The answer was a cylindrical drift body hanging on a thread guided by a fishing rod. The cylinders height is 15 cm and the diameter is 12 cm. It's made of a plastic screen with a mesh diameter of one millimeter, fixed to ring shaped polyvinylchlorid (PVC) reinforcements at both ends. For controlling the depth of the drift body beneath the water surface, the thread is graduated in 25 cm increments. A telescopic fishing rod allows easy positioning of the drift body at any spot inside the water body.

An absolute vertical position of the thread is essential and acomplished by means of following the drift body by covering the thread with its mirror image, reflected by the water surface. That's facilitated by the fact, that the angle between thread and mirror image is magnified by two by any given deviation. This effect allows a very accurate, vertical positioning of the device. Then, the location of the thread relative to some reference reveals the momentary position of the drifting body.

For many years I didn't alter the design of the drift body. I admit, I had never done tests for finding out the optimum design before. This was done later on by Judith Ueberl as a preparatory work on her thesis "Modelling of Rectangular Settling Tanks". Her findings are published in the Journal of Environmental Engineering, Vol. 123, No. 3, March 1997, pp. 259-268
She investigated the suitability of different shapes of drifting bodies, - amongst others also a ball shaped body. But finally she abided with my good old cylinder type drift body.

Fig. 2:
Method of velocity measurement.
Drifting body, here shown at some angle away from vertical (schematic)

An assistant, equipped with a telescope, follows the position of the thread relative to a meter rule, located at the opposite wall of the basin. Depending on the local velocity of the drift body, the time may be measured to pass typically 20 cm. Thus, the average local, horizontal velocity component of the flow is obtained.

Fig. 3:
The drift body. Hight 15 cm, diameter 12
Fig. 4:
Sitting on a pulpit constucted especially for this measurement by the operators of the STP. It allows a straightforward observation of the thread and its mirror image.
Fig. 5:
The string and its mirror image are offline. Don't mind the mirror image of the clouded sky.
Fig. 6:
The string and it's mirror image are coinciding now. That warrants the drifting body and its string are adjusted exactly in vertical position. Note the little red mark just touching the water surface. That tells you the body is floating at the planned depth.

The estimated error in velocity amounts to 5 - 15% and, thus, is relatively high. However, due to velocities of only 10-20 mm/s, the absolute error is smalI. Further, repeated measurements of selected profiles indicated consistent results. The exact determination of velocity is herein not a problem of significance, as the purpose is rather to compare flow fields.
The determination of suspended solids was straightforward as samples were taken by a small pump at the same positions as selected for velocity readings and then further analyzed by filtering through a 0.45 µm membrane. The results relating to concentration measurements of the suspended solids were the basis for the final contour plot of equal solid matter concentration, as shown in the next section. Likewise, velocity profiles were plotted, and the two of these parameters will be subsequently discussed.

Fig. 7:
Taking the profile of the suspended solids. The samples are pumped by a drilling machine fitted with an inexpensive little impeller pump, available in any hardware store. It conveys about 0.2 l/s.

Later the sample will be analysed by filtering and drying at 105 °C.

Guiding the suction pipe on the catwalk is Josef Bisig, an outstanding plant supervisor.