We are working on Unidirectional Flushing (UDF) model analysis (early stage of creating the model at this moment).
I have a question after reading your post and the Section 10.9 in Advanced Water Distribution Modeling and Management.
Q. 1 K is more accurate than k, in the table and we should be using K value for the model analysis, yes?
I guess we could do extensive calculation using the equation 10.15 to find K.
However, the k value we calculated using the old hydrant flushing test seemed to use 2.5" hose nozzles instead of 4.5" pumper nozzles.
Q. 2 When one conduct UDF, shouldn't they use pumper nozzles for better flushing? Or, does it depends on the pipe size? Is this something we can find out as we do the model analysis?
Q. 3 What are the common/recommended approach to find the accurate emitter coefficient? If we don't have good data, can we pick a number from the range of K value given for a nozzle size for the first flushing day and adjust it based on the first flushing results?
Since my question is about emitter coefficient, I hope it is OK to post it here.
Thank you for your advice in advance.
1. K is the overall emitter coefficient for the model element. k is an intermediate value used along the way. See page 451 in Advanced Water Distribution Modeling and Management for an explanation. Neither is "more accurate". K is what you enter into your model.
2. Your choice of nozzle depends on what velocity you want to achieve awhile minimizing the amount of damage that can be done by the flowing stream.
3. Yes. the best you can do is estimate K from the table and see if the measured flow equals what is in the model and make adjustments accordingly. But don't expect great precision. A value that works well for one hydrant may not do as well for another. And remember, you are not trying to determine if the velocity in the pipe is 2.357 m/s or 2.358 m/s but rather is the velocity 2 m/s or 3 m/s.
Answer Verified By: Sushma Choure
Thank you so much for your reply.
Your AWWA webinar has been helping!