Accuracy of transient calculations

[Saugrohr]

To those who make transient calculations:

According to your experience of transient computations with several codes in the market, which accuracy do you think you get in your results of transients, specially on pressure results?.

I use to consider +/-5 % of accuracy for instance for the maximal pressure calculated. But I have never checked with experimental results.

Thanks to share your experience

 

[BigInch]

5% is probably a pretty good number. I've gotten better (2%)... and worse.

"If everything seems under control, you're just not moving fast enough."
- Mario Andretti- When asked about transient hydraulics

http://virtualpipeline.spaces.msn.com

 

[rshayesteh]

I guessed that you are talking about CFD transient calculations. If so, this is very hard to answer since CFD calculation results and their accuracy depends on a lot of parameters such as the complexity of your model geometry, the flow charachteristics (i.e. laminar or turbulent, one or more phases, 2D or 3D flow etc.)the quality of mesh you are generating, the accuracy of solver and turbulence modeling you are using, the time steps for calculations and so on. So it is not possible to mention a standard range of accuracy. For a simple 2D, laminar, one phase model, you can get 99% accurate results, but if you are talking about modeling the flow whithin a 3D pump, you can expect up to 20%-30% error depending on above mentioned parameters.