[Bug] Fitting Loss for Elbows/Bends

[richardbelcher]

Fitting loss for 90deg Elbow appears to be (from looking at the KFit method) L/D = 30. This is for a screwed joint (i.e small bore sizes), larger bore sizes are typically long radius r/d = 1.5 and have a L/D = 14.

I would suggest adding LR Elbows as the default and creating Threaded/Screwed 90deg Elbow as second choice in fitting list.

Picked this up in QA with other software modelling water lines that the frictional drop in DWSIM very close but that the fitting losses were near double.

Reference.

[richardbelcher]

For adding custom fittings will editing and adding to fittings.dat file work? or does modification of KFit method required?

[DanWBR]

@richardbelcher we need to modify the Kfit method

[DanWBR]

@richardbelcher any valid reference other than another software?

[richardbelcher]

Hi @DanWBR,

Crane Technical Paper 410 is the reference where screwed elbow K = 30fT and 90deg elbow r/d 1.5, K = 14fT.

There is important difference between fT and f (friction factor calculated from colebrook white). The link below explains to work out equivalent length for Crane TP410 fittings we must always use fT, which Crane developed assuming fully turbulent flow and commercial steel pipe roughness.

Once the equivalent length is calculated then this can be utilized with actual friction factor and pressure drop as the DWSIM modules are written.

https://www.eng-tips.com/threads/crane-410-fittings.173164/