CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   laminar and turbulent flow simulation in elbow fitting to find K values wtr Re (

prashanth1234 January 17, 2019 11:07

laminar and turbulent flow simulation in elbow fitting to find K values wtr Re
First of all i am new to fluent. I welcome myself to the club. I would be glad if someone replies for this Thread. and I am struck with this topic:confused:.

As we all know that 'K' value (minor loss coffefficient) changes for laminar flow wtr reynolds number and will be constant for turbulent flow in fittings.

Question is how to simulate both flows in the fluent for example 5mm diameter 90 degree Elbow fitting (both inlet and outlet has same diameter). The main thing i want to do is, i would like to find the relation between K and Re in laminar flow, approximate value of critical reynolds number and K value in turbulent flow. So that I want to use this data to simulate the flow in the elbow fitting in simscape software. Fitting was designed in solidworks and no data is available except dimensions.

without knowing the critical reynolds number, how can i chose the options laminar, k-epsilon, or k-omega? Is it also possible to simulate the flow separation at the bend?

I tried to check with all the methods like laminar and in K-epsilon i used standard, scalable and enhanced wall treatment... i got the same pressure values for laminar, standard and enhanced wall treatment but for scalable i got drastic changes in values(which idk why yet).

Any suggestions on where to start and also possibilities of this method and changes in the methods would be appreciated.

LuckyTran January 17, 2019 12:44

You won't be able to figure out critical Reynolds number because that's not a property of your elbow but of the flow state that runs up to the elbow from whatever strange plumbing you might have.

Supposing that your have long straight pipe upstream of your elbow, you can assume the critical Reynolds number is the same 2300 for pipe flows. K values are normally reported anyway with this assumption. Otherwise, you should get the K value of the elbow + whatever is upstream.

Flow separation will naturally be captured.

I would use enhanced wall treatment. Enhanced is a little better than standard wall functions. Scalable wall functions are weird and it makes sense that you get weird results with them. It makes sense that you get the same results with standard and enhanced wall treatment, but shouldn't the laminar one be different?

prashanth1234 January 18, 2019 05:30

Hello LuckyTran, Thanks for your reply.

I got the same values at low reynolds number for enhanced wall, standard and laminar. which i thought wierd too.

would you be able to suggest any book to understand these concepts in more detail before working on the fluent completely?

LuckyTran January 18, 2019 08:56

The Fluent manual is pretty much a textbook.

We'd have to deep dive the results to figure out why they look wonky. Regardless of what you read though, you probably won't be changing models.

One thing I forgot to mention is you should give an inlet velocity profile at the inlet of the elbow because it makes no sense to report K values for a uniform flow.

Other than that, I think you should just look at some tutorials and get a better feel for Fluent. Or try to simulate flow in a straight pipe and try to match the friction factor.

prashanth1234 January 18, 2019 11:06

yeah i was giving uniform flow to check for the k values... i will change the velocity profile.
Thanks for your suggestion i will work on it.

All times are GMT -4. The time now is 11:58.