1 Reply Latest reply on Aug 21, 2013 1:11 AM by Jared Conway

# Effect of curve in duct

Dear all,

I have a problem with a flow simulation when using Floworks. I want to do some simulations on quite some complicated duct geometries, but before I do that, I want to validate and learn more about floworks. Therefore I made several simple models.

I started with a very simple straight duct (length = 892,5, diameter=100) and simulated this with the following variables:

v gas = 10m/s

density 82kg/m3

roughness = 0

With manual calculations I found a Zeta of 0,0841 and a pressure loss (defined as Total pressure at inlet - Total pressure at outlet) over the duct of 345 Pa. I made a model within Floworks and found exactly as the manual calculations, so far so good.

Then a made an curve into the duct and kept all other variables equal(including length). I then found(outcome of simulations) that Zeta dropped to 0,0656 and that the pressure loss decreased to 269Pa. This is of course inpossible as curves in piping always lead to increased losses.

Can anyone explain what I did wrong and how I can solve this?

I have set-up my model as follows:

- Boundary condition at inlet lid = 10m/s inlet velocity (fully developed flow)

- Boundary condition at outlet lid = 5000000 Pa static pressure

- Use of Argon gas with intial condition set at 5000000 Pa

Thank you very much in advance,

With kind regards,

Peter

• ###### Re: Effect of curve in duct

What does the mesh look like?

What are you comparing it to? (Physical test? Hand calcs?)

What happens if you extend the pipe on both ends by like 10xD?

What do the rest of the parameters and plots look like? Are they pretty reasonable?