13 Replies Latest reply on Mar 9, 2018 6:17 AM by Michael Fidel

# Heat Transfer Coefficient

Im currently working on a project about a double pipe heat exchanger with helical baffles

with hot water on inner pipe and cold water on outer pipe. My end goal would be to compare

heat transfer coefficient on outertube when adjusting mass flow rate from 0.1 to 0.3 kg/s.
However, my problem is that how can i look for heat transfer coefficient and compare it to others?
and can i also graph it through the flow simulation?

• ###### Re: Heat Transfer Coefficient

As general guide, I'd say whenever you want to extract results in flow sim (cut plot, surface parameter, etc) it must be either on a plane (or offset distance from plane) or on a surface of a solid body. As you know, surfaces in SW flow sim are not accepted entities. Having said that if you are looking to extract results at a particular location, you can use a simple trick by bringing a dummy solid body to your flow sim, re-run, and extract results on its faces. But you must assure that the dummy component is excluded from flow sim project to prevent any affect from it on the flow.

• ###### Re: Heat Transfer Coefficient

Sir, thank ypu for replying, but hardly im  anewbie in this simulation thing. Would you mind sir if you could send pictures of hows? Thanks much in advance! Appeciated!

• ###### Re: Heat Transfer Coefficient

Unfortunately, I am not able to, but I hope one of my colleagues on Forum would do it for you or provide you with a better solution.

• ###### Re: Heat Transfer Coefficient

This image should answer your question.  It is a Surface Plot on all surfaces, and then you can select to show the contours of Heat Transfer Coefficient from the list of available parameters.

• ###### Re: Heat Transfer Coefficient

Sir, my goal is to graph the annulus side heat transfer coefficient. Should i compare the average value from 0.1 to 0.3? The value varies from 20,000 w/m^2/k which i think a big value for the length of heat exchanger is just 100mm. TIA

• ###### Re: Heat Transfer Coefficient

Just select the surface(s) in the dialog box that you want to plot the heat transfer coefficient (and not the option for "Use all faces").

• ###### Re: Heat Transfer Coefficient

You can adjust the minimum and maximum values on the legend to capture the values better.

And the reason why the values are so large is because the defined SI unit is per meter squared, which is quite large (if you think that this is the same unit that a house is sized in), so as an expert move you can create your own unit like this:

1. Right-click on the Input Data tree item, and choose Units.

2. Navigate to Heat, open with the + sign, and then go to Heat Transfer Coefficient.  Choose the box of the unit and from the drop down choose Custom Unit.  (Note that in the box next to this on the right is how you change the number of decimal places shown, which might be useful.)

3. Define a new unit system by putting in the conversion to and from the main SI unit system, which is easy here because it is (10^3mm)^2 = 10^6 mm^2.  You would multiply and then divide to the SI unit system appropriately. See below.

Now your value of 20,000 W/m^2/K will be shown as 0.020 W/mm^2/K.  So again make sure that you change the number of decimal places in the unit to at least 3 (which is the default for this unit anyway).

• ###### Re: Heat Transfer Coefficient

Here is the result of the new unit system:

• ###### Re: Heat Transfer Coefficient

Sir, when im trying to do a surface plot the heat exchanger, it only shows this:

But when i tried to do surface parameter on the external surface of the inner tube, it shows a large number 100,000 w/m2/K:

Im stuck here sir. Thanks for your responses sir, it helps me a lot.

• ###### Re: Heat Transfer Coefficient

A couple things that I noticed from your simplified model (but then I made changes to make it more like your original model).

1. Your "lids" were touching the geometry only at edges, and it should instead have a positive area of contact.  I usually like to use the OD edge instead of the ID like you had done.

2. The simplified model did not make physical sense because how would you realistically get water entering the pipe the way that you did it, so that's why I remodeled it back to more like what you originally had with inlet and outlets on the outer tubes opposite sense to one another.

3. I changed the meshing scheme to use a Global Level of 4 and setting a minimum gap thickness of 0.5mm.  Without any local mesh control included, these simple settings still creates a good mesh.

4. I used Fluid Subdomains to define different initial conditions for the cold side (13C) and hot side (38) fluids, and similarly used Initial Conditions for the solid parts in those two separate regions.  I set by Global Initial conditions for the baffle solid temperature at the average of the two sides ( avg= (38+13)/2 = 25.5).

5. (Not required but extra credit) I set a contact resistance between the Outer and Inner tubes, and resistance between baffle and the tubes.

6. It is always good to create Goals for a study.  I kept it simple with finding the Average Outlet Temperatures of the hot & cold side fluids separately, and another Equation Goal type could had been used to calculate the efficiency of the heat transfer, which is the (Actual Change in Temperature) divided by the (Maximum Possible Change in Temperature) or (38C - actual cold outlet) / (38C -13C).

7. I created a new Unit System as described earlier for Heat Transfer Coefficient of W/mm^2/K so that the area is on the same order as the dimensions of the part.  I created a Surface Plot and reset the minimum and maximum values to narrow down to values that show detail on the plot.

• ###### Re: Heat Transfer Coefficient

Thank you sir for your response!

So sir, the main thing to be adjusted is the geometry itself?

Another thing sir is that, is fluid subdomain setting necessary even when temperature from boundary conditions are set (38 - hot: 13 - cold)?

• ###### Re: Heat Transfer Coefficient

I was referring to the model that you had sent to me that you tried to simplify.  The lids on your original model appear to be OK.

It is not necessary to define fluid subdomains for this problem, you will still achieve the same result.  I defined it in this steady state calculation so that the two separate zones are initially set closer to what they would be when the calculation is done; no need to waste iterations on having the regions reach the temperature of the incoming fluid, whether it's hot or cold.

The main change that I recommend to you is the settings for the units and the legend values max & min.

• ###### Re: Heat Transfer Coefficient

Sir, even if i change the units and set the maximum and minimum values to the max min of plot, the surface plot only shows blue with 2 orange spots. I think i should get the surface plot results from above? What should be the problem sir?

Thanks for the response sir!