[Thakar]

I have performed hand calculations of HTC and Outlet Temperature in a Helical coil using equations from Heat Atlas(Gnielinski) and have obtained HTC:4739 W/m2k and T_outlet:52C.I am now trying to validate these results with ANSYS CFX.I have read all the posts related to this query and yet left with some questions.

My V_inlet:0.5m/s,T_inlet:10 C,T_wall:100 C,Tbulk for HTC:304K with Laminar flow as the Reynold number is less than the critical in the hand calculations.

My results are varying hugely with Inflation layers and its Growth rate(Eg: 2794 W/m2K and 37 C for 6 layers inflation to 3498 W/m2K and 45 C for 10 layers and 1,4 growth rate for both) and unable to get close to the results abtained from hand calculations.I find these results surprising becase it a laminar flow with Reynolds number:6800.Am I missing anything else?Thanks in advance

Sorry for so many questions...I am a beginner with CFD and CFX.

[evcelica]

Edit: I looked at your reference:
Oh, I see the critical Reynolds number for coils in that reference is larger.
The heat transfer coefficient in that reference does seem a bit large. (1.57x that of a straight pipe !?!)

[Opaque]

Careful here confusing regions with recirculating flow with turbulent flow.

Turbulent flow is defined based on flow instabilities, and current literature defines the transition between laminar and turbulent based on a Reynolds number not level of mixing or recirculation. There are many variations in the definition of the Reynolds number, some are geometry dependent and others are boundary layer based (like momentum thickness or displacement thickness).

From wikipedia

"In fluid dynamics, turbulence or turbulent flow is a flow regime characterized by chaotic property changes. This includes low momentum diffusion, high momentum convection, and rapid variation of pressure and flow velocity in space and time."

Recirculating flow regions do not have to be chaotic.

My 2 cents.

[ghorrocks]

Also: Hand calculations of things like heat exchangers can expect an error of 50%. It is expected that you will not match hand calculations with any precision.

Have a look at the general accuracy FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[evcelica]

What is the geometry? ID, coil diameter, and length?

[Thakar]

Thank you for taking time and answering this query.

evcelica,I have the CS dia (d) 0,01 m , coil dia (D) 0,1 m , length of the helical coil (L) 0,62837 m , Area of Wall surface (A) 0,019741 m2 , Number of Turns 2.So the Re_Critical is about 9319 and the Re from the formular w*d/v is about 6364.

So I assumed the flow to be Laminar but as opaque has suggested that Turbulent-Laminar condition cannot just be decided by Geometry.I have tried some simulations with Medium intensity K-Epsilon Turbulence model and ran a parameric simulation with inflation and growth rate as input and have huge variations in HTC and T_Outlet again.

Ghorrocks I am not trying to get the exact value as its impossible but just trying to understand how CFX is working with this problem.

[pimpa]

what's the wall thickness ?
According to Wikipedia, estimate HTC through a thin or thick pipe wall as follows,
might provide some upper bound on flow-to-wall HTC:

HTC_thin= k/x
HTC_thick= 2*K/(di*ln(do/di)

k= thermal conductivity of pipe mat
x= pipe wall thickness
di, do = pipe inner/outer dia

(that is, if not used already by the reference)

And use a turbulence model.

[ghorrocks]

Pimpa: The heat transfer coefficient you quote is the HTC for heat conduction through a pipe wall. The HTC coefficient this thread is talking about is for heat transfer from the fluid to the solid at the fluid/solid interface. We are talking about different things.

[pimpa]

Glenn: you are right, edited the post to reflect your comment.