Parshall Flume

[sjohns4]

I'm working for a client with an existing 10' Parshall flume they use to measure the volume of their industrial effluent discharge. I pulled field measurements and had a survey of the bottom profile done.

The dimensions appear very close to what I find in my reference book, however, I was wondering how much tolerance is allowable for varius dimensions yet still allow the flume to provide accurate flow measurements?

For example, the throat width was exactly 10'-0" at the entrance, but was 9'-10.5" at the exit.

 

[RWF7437]

As far as I know, there is no way to answer your question based on theory or fundamental laws. You must calibrate any weir to get the most "accurate" flow measurements. How accurate do you need to be ?

 

[bimr]

Extreme accuracy is not claimed for a parshall flume. However, measurement is very dependable with minimum maintenance.

http://search.barnesandnoble.com/booksearch/isbninquiry.asp?r=1&ean=080198386X

Flumes that are correctly manufactured and installed, are intrinsically accurate. A particular throat geometry will consistently produce a given change in depth at a given flow rate. The real issue is not accuracy but resolution. Therefore if resolution is expressed as a percentage of flow rate, flumes will measure higher flows with greater resolution than lower flows. For a given flume and flow meter combination it is not at all difficult to resolve flows to within +/- 5 % of rate over the range of 10 - 100% of maximum capacity. If one is careful setting up and calibrating the depth sensor, it is possible to achieve resolutions better than +/- 2% of flow rate over the same range. Below ten percent of maximum flow resolution of flow rate drops off quickly due to the fact that flow through a flume is not a linear function of depth.

http://www.accuraflo.com/faq.htm#accuracy

http://www.usbr.gov/pmts/hydraulics_lab/pubs/wmm/chap08_10.html