Back Calculate Heat Exchanger from Test Data

[pcarrollg]

Hi,

I'm looking to backcalculate the properties of a heat exchanger we've purchased based on test data to characterize the heat exchanger. This is an air to air heat exchanger and we're assuming it's dry air. What I'm looking to calculate is:

1. Surface Area
2. Heat Duty
3. Heat Transfer Coefficient.

The data I have is

-Tube side inlet and exit temperatures
-Shell side inlet and exit temperatures
- Tube diameter and size
- Shell I.D. and O.D.
- Tube side and shell side flow rates
- Nominal tube length

I think i'm most of the way there, i'm just stuck at the end with an NTU value and a UA, and I'm not sure how to go from there to my three items above. My steps were...

- Find all the properties of the shell side and tube side fluids
- Find shell side fluid capacity using: Capacity shell side = (mass flow rate shell side) *specific heat
- Find tube side fluid capacity using: Capacity tube side = (mass flow rate tube side) *specific heat
- find c_min/c_max using the previous two heat capacities
- find actual heat transfer rate, max possible heat transfer rate, and divide them to get effectiveness.
- Use effectiveness to find the NTU using the table on page 11 for a shell and tube one-shell pass heat exchanger
- find UA from UA = NTU*minimum heat capacity

I'm stuck after this since I don't know what the inner tube layout is. Any help is greatly appreciated.

 

[Caloooomi]

Could you not just look at the nameplate and check with the manufacturer for the tube layour?

 

[EdStainless]

You need to know what standard (HEI?) this was designed under, and then use those design rules in order to try and get a fit. They will have various correction factors and allowances built in.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[pcarrollg]

It is an NEN type fixed tubesheet.

Where would i find the design rules for this? It adheres to TEMA guidelines.

 

[MintJulep]

-Tube side inlet and exit temperatures
-Shell side inlet and exit temperatures
- Tube diameter and size
- Shell I.D. and O.D.
- Tube side and shell side flow rates
- Nominal tube length

With the green items above plus the number of tubes you can calculate directly the surface area.

From tube size and tube flow and fluid properties you can calculate coefficient of convection.

 

[pcarrollg]

Oh I think I understand. So I was looking to solve for this equation for the shell side area:

As = DsCB/Pt

where Ds = inner shell diameter
C = clearence
B = baffle spacing
Pt = Tube pitch

But I dont think there are any baffles. So instead I should be able to just use

As = pi*Ds^2/4 - pi*(Tube outer diameter)^2/4*number of tubes

And then the surface area would be

SA = pi*(Tube outer diameter)^2/4*number of tubes*nominal length


But I'm looking to check the surface based off of our test data. So i'm also looking for a way to estimate surface area based off of the change in temperature and flow rate in the heat exchanger.... if that is possible.

 

[marty007]

The three values you are looking to calculate:
1) Surface area: Calculating surface area from number of tubes, tube diameter, and tube length will give you by far the most accurate result. It will also be the result that you will trust the most because it is based on as-built geometry.

2) Heat duty: It's just the change in temperature multiplied by the flowrate and the heat capacity. You can calculate this twice, once for the tubeside and once for the shellside, and you will get slightly different answers due to the accuracy of your inputs. All three of your values will have uncertainties associated with them; instruments have a certain accuracy and specific heat will be based on correlations/exact chemical composition. Your calculated values should still be relatively close though.

3) Heat transfer coefficient: Most heat transfer textbooks will point you towards heat transfer coefficients for tubeside/shellside flow. Whatever you come up with though will undoubtedly be the least accurate compared to the surface area/heat duty values. You will need to calculate two coefficients (shellside/tubeside), each of which is based on correlations that come with pretty significant uncertainties. Whatever you come up with should be taken with a huge grain of salt.

Good luck,

 

[srfish]

Unless this is a very short shell-and-tube heat exchanger, there will be baffles.