Two questions about heat exchanger

@samko95-fWBI0p • Oct 21, 2024

Oct 21, 2024

1.3K

Hi every one
I simulate turbulent cross flow finned tube heat exchanger shown below by using fluent, and study the same case experimentally.

[IMG]

I've tow problems

There are two way to calculate Re number (Re based on duct hydraulic diameter or pipe hydraulic diameter). I think the second one is common but it require up to 200000 to gain turbulent, and i can't get this value experimentally, so what is your suggestion?
Nusselt number (Nu= h* hydraulic diameter/K)

I want to know how to calculate hydraulic diameter of the pipes with its fins.

I'm looking for answer or any site can answered me please
thank you

Like 0 Replies 21

Replies

Welcome, guest

Join CrazyEngineers to reply, ask questions, and participate in conversations.

CrazyEngineers powered by Jatra Community Platform

Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

Well, for flow in a conduit of circular cross section, the limits are like this:

Re < 2000 : Laminar flow

2000 < Re < 3000 : Transition flow

Re > 3000 : Turbulent flow

The limits of Re > 20,000 is for flow over flat plates. Hope this helps. And as for getting the flow turbulent, you would be surprised as to how a small velocity of the flow can get it to go turbulent.

Enjoy! Fluid Mechanics is awesome!
samko95

@samko95-fWBI0p • Oct 22, 2014

Thank you for your reply #-Link-Snipped-#
-I know what you talking about exactly, I'll clear more :
some researchers depend on tube hydraulic diameter and an others depend on duct hydraulic diameter,which way is the right or the best (useful) to measure Reynold number ( based on duct hydraulic diameter or pipe hydraulic diameter).
-the second problem, the hydraulic diameter of pipe is simple to calculate, but how to calculate the hydraulic diameter taking in consideration the fins.
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

- About your doubt: Whichever way you use to calculate the hydraulic diameter (either duct or pipe), make sure you do not use Re = 20,000 as your Turbulent flow limit. For flow in a circular conduit, the limits I mentioned hold for flow transitions from one regime to another.

Also, do explain I detail, preferably with a diagram or a photo, what do you mean by "duct" diameter.

- As for the hydraulic diameter of the pipe taking into consideration the fins, do attach a top view of the assembly (so that we can see how the fins are spaced) and a front view (so that we can see what is the thickness of the fins over the pipe). Then we can take an educated decision.

Disclaimer: I love FM/CFD. But I am in no way the most definitive source of knowledge. I am helping only because I love this subject and have done quite a fair amount of CFD in my Masters.

Also, this is not a Homework problem now, is it?
samko95

@samko95-fWBI0p • Oct 22, 2014

OK my friend, thank you so much for your attention. I'll be back with more details after one hour or less, because I live in IRAQ planet where the electric power always turned off (It is our great enemy ).
-Yes it is not, I prepare to write my Master Thesis
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

Ok. At your ease man 😀
samko95

@samko95-fWBI0p • Oct 22, 2014

-Firstly I was forgetting to explain that my questions are just about air side (external cross flow over the tubes), and in order to gain turbulent flow, Re will be equal at least 200,000(depending on tube diameter), is this right?

-Circular tube,Square duct,and Rectangular duct hydraulic diameters are shown in figure below

Front and Top views (fin diameter 44mm, tube diameter 19mm, 5mm pitch (distance between each two fins))
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

samko95
Firstly I was forgetting to explain that my questions are just about air side (external cross flow over the tubes),
Oh. I thought it was for the flow inside the pipe. This changes everything 😀 Yes, Re > 20,000 is right in that case.

samko95

Again, exactly describe what you want to discuss. Give your problem statement once again.
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

In his case, the fins have a greater lateral area for air cooling than that for the non-finned area. I was going to suggest this:

1) Calculate the thermal entry length. Since this is a turbulent flow inside the conduit (I assume this), the thermal entry length should be about 15D to 40D. The rest of the tube length will have a constant convection coefficient. Divide the rest of the tube (tube length - thermal entry length) into two lengths, one circular non-finned and the other with the fins (has an increased thickness because of the fins).

But that brought me to the lateral surface issue. How can we account for that in the finned part?
Ramani Aswath

@ramani-VR4O43 • Oct 22, 2014

The Re in such cases has to do with the economics of pumping large volumes of the gas across the bank. While increased Re does improve the gas side coefficient, the pressure drop increases costs. Also the gas side coefficient cannot be increased too much by just increasing the Re because the slope of the line is quite shallow at high Re. This may be more of an optimization problem rather than a purely technical one on the effect of increasing Re on the overall HTC.
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

Yes, probably. Still, how do we account for the different lateral areas and thicknesses on the same tube if we had to do a numerical simulation of this problem?

As far as the question regarding the hydraulic diameter goes, I guess a weighted average of the finned length (square cross section, as #-Link-Snipped-# Sir mentioned) and the non-finned length (circular cross section) should serve your needs well.
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

#-Link-Snipped-# This discussion might interest you. We can surely use your insight here.
samko95

@samko95-fWBI0p • Oct 22, 2014

A.V.Ramani
I feel that 200,000 is pretty high. In any case friction factor for 10,000 to 100,000 does not vary much. The drawing shows individual fins and not a spiral wound one. You can try this exercise.
I assume that the flow is perpendicular to the axis of the tubes parallel to the fin surface. Estimate the velocity of the flow between the fins. Treat one pair of fins as a rectangular duct (projected image) with a = spacing between fins and b = height of the fins and estimate Re. That should be a good approximation.
Hi Ramani
Right, in this case I'll get max velocity. what about hydraulic diameter needed.
samko95

@samko95-fWBI0p • Oct 22, 2014

A.V.Ramani
The Re in such cases has to do with the economics of pumping large volumes of the gas across the bank. While increased Re does improve the gas side coefficient, the pressure drop increases costs. Also the gas side coefficient cannot be increased too much by just increasing the Re because the slope of the line is quite shallow at high Re. This may be more of an optimization problem rather than a purely technical one on the effect of increasing Re on the overall HTC.
I took all that in consideration Sir, this not every thing about my research. I made different perforation in the fins (differ in number and shape for each model), this will increase pressure drop but as a result increase heat transfer coeff. due to ..... etc.
I think we are going out of our project.
As mention the hydraulic diameter of tube is the diam. itself. the important question now, how can calculate Hd with fin ?
samko95

@samko95-fWBI0p • Oct 22, 2014

Shashank Moghe
In his case, the fins have a greater lateral area for air cooling than that for the non-finned area. I was going to suggest this:

1) Calculate the thermal entry length. Since this is a turbulent flow inside the conduit (I assume this), the thermal entry length should be about 15D to 40D. The rest of the tube length will have a constant convection coefficient. Divide the rest of the tube (tube length - thermal entry length) into two lengths, one circular non-finned and the other with the fins (has an increased thickness because of the fins).

But that brought me to the lateral surface issue. How can we account for that in the finned part?
It's is not clear (at least for me), please clear it more .
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

I am talking about the internal flow. The flow will have to be thermally developed for you to have a constant convection coefficient. That's why I asked you to divide the length in two parts: Thermal entry length and beyond. The "beyond" part of the tube will have a constant "h". The "h" in the thermal entry length will have a huge varying slope.

In reality the velocity and the temperature profile inside the tube will take some length to "stabilize" and assume a constant parabolic shape. This might be automatically taken care of if you are using ANSYS or any other CFD software though I am not sure.
samko95

@samko95-fWBI0p • Oct 22, 2014

#-Link-Snipped-# and A.V.Ramani
Thank you so much for your notes, and your attention to reply. I think I've got the answer ....
Dh=4a/b
where:
a:Ratio of free flow area (up, bottom, and sides of heat exchanger) to the stagnation area.
b:Total surface area for one side of heat exchanger to the total volume of heat exchanger.

Best regards
Shashank Moghe

@shashank-94ap1q • Oct 22, 2014

What is stagnation area? Do share your final abstract once you are done publishing! Best of luck!
samko95

@samko95-fWBI0p • Oct 22, 2014

Shashank Moghe
What is stagnation area? Do share your final abstract once you are done publishing! Best of luck!
stagnation area: where air collide the heat exchanger.
thank you so much and I am sorry for my weak language, I did not take any course to learn English language.
Ramani Aswath

@ramani-VR4O43 • Oct 22, 2014

samko95
I did not take any course to learn English language.
English is not our native language either.
Shashank Moghe

@shashank-94ap1q • Oct 23, 2014

Oh not at all! This was a technical term I did not understand. Your English is perfect. Don't worry about that.
Ramani Aswath

@ramani-VR4O43 • Mar 21, 2023

I feel that 200,000 is pretty high. In any case friction factor for 10,000 to 100,000 does not vary much. The drawing shows individual fins and not a spiral wound one. You can try this exercise.
I assume that the flow is perpendicular to the axis of the tubes parallel to the fin surface. Estimate the velocity of the flow between the fins. Treat one pair of fins as a rectangular duct (projected image) with a = spacing between fins and b = height of the fins and estimate Re. That should be a good approximation.