I'm working to check the design of a fixed heat exchanger, and I'm not sure about how this project was designed some year ago.

I fell that an expansion joint could be necessary, but I don't have any data for mean metal temperatures along shell and along tubes.

Would you rely on operating conditions to determine a MMT for tubes and shell?

In attachment I'm posting the design data available.

Hello fedeghi

There is a way that I use to solve this problem in the absence of any other data. This is what to do to get the mean metal temperatures:

For the shell, I take the average of the inlet and outlet temperatures. So, for the shell, the mean temperature would be: (32 + 35) / 2 = 33.5F. As the tubes experience the shell side and tube side fluids, the mean temperature can be taken as the mean of the four terminal temperatures: (32 + 35 + 120 + 40) / 4 = 56.75F.

This method can be used is you have no other way of accurately getting the mean metal temperatures.

Thanks Ray.
This approach is very similar to what I wanted to do, and I starting from this assumptions I would not use an expansion joint with such a small thermal difference.
Also because on shell side we have 6 bar, so an EJ might cause bigger problems, instead of helping.

Anyway, talking about this approach to define MMT, I think I will only rely on it when the fluids on both sides have similar thermal coefficients.
And also in that case, convective efficiency will always be unknown to me.
I think I will have to ask to our process dept more and more..

Thank you, bye!

Hi fedeghi

There is a formula in TEMA for determining the mean metal temperature, but it requires so much information, that I don't think anyone has ever used it. The mean metal temperature is complicated by fouling factors, film thicknesses and the degree of turbulence.