Low Flow Accuracy of Palmer Bowlus Flumes

The Palmer Bowlus flume was designed for installation in existing manholes and in line with sewer piping / channels. These sorts of installations normally see nice average to upper range flows.

Under normal operating conditions – where the head in the flume is large in comparison to the length of the throat – Palmer Bowlus flumes are typically accurate to +/- 3-5%. However, for lower flows – where the head is low in comparison to the throat length – the accuracy of the flume decreases to +/- 5-6%. As the head gets lower, the importance of the boundary layer becomes greater, resulting in the decreased accuracy.

Palmer Bowlus flume accurately measuring industrial discharge

It is important to note that this behavior is different than other flume types. In general, other flume types exhibit the same level of accuracy over the entire range of their rated flows; accuracy does not vary by head.

As a result, Palmer Bowlus flumes should not be used on low flow applications or where the flow will exhibit wide swings. For those applications other flume types (HS / H or Trapezoidal) should be considered.

However, if a Palmer Bowlus flume must be used for a low flow application, consideration should be given to going down one flume size (i.e. use a 8” flume on a 10” line) or investigate the possibility of using a nested (dual range) flume configuration.

nested Palmer Bowlus flume configuration

Keep in mind that there are downsides to both approaches. For the smaller flume option: to avoid turbulence coming into the flume, when using a flume smaller than the line size, a smooth transition should be formed into the flume. If the nested flume option is used, the will be more standing water upstream of the flume that would normally be present and the upper range of the smaller, inner flume may be less depending upon how it is nested into the larger flume.