Dear Miguel Ángel Garcia you can't assume a constant value of UA coefficient in an ORC under off design conditions due to the fouling factor, so if you have a heating stream that you know it properties and conditions, you must know the energy it transfer and if you varies the temperature the UA is a temperature function.

Dear, Miguel Ángel Garcia .....you can't assume a constant value of UA coefficient.

Hi, Miguel,

the short answer is no! but it can be assumed as a first approximation, especially if other parameters vary widely so its own variation is insignificant. To get an idea of the accuracy of saying UA = constant, use simple expression(s) for the heat transfer evaluations and insert values/properties for the fluids under consideration. Pick the extreme values and see what results are obtained. For example when using fluid maximum temperature and minimum temperature how does UA vary.

Before progressing to a shell and tube heat exchanger(s), consider heat transfer to a single tube.

To a reasonable approximation, the inside heat transfer coefficient is given by:

hd/k = 0.023*(Cp*Mu/k)**0.33 * (Ro*V*d/Mu)**0.67

Check how h varies with temperature.

If the viscosity varies widely, you can add a viscosity correction factor of

((Mu/Ro) bulk)/(Mu/Ro) surface)**0.14.

In the above expression, hi needs to be adjusted to hio by the ratio of internal to external diameters.

The heat transfer coefficient on the outside of a tube in cross flow is given by

ho = 0.26*(Cp*Mu/k)**0.3 * (Ro*V*d/Mu)**0.6

To get the overall value of U, we say, 1/U = 1/hio + x/k + 1/ho where x=wall thickness and k = thermal conductivity of wall. U will vary with flow due to the presence of the Reynolds Number.

The above expressions should be adjusted for the inside and outside fouling factors, but that can follow. You may find that the fouling factors in the literature make any variation in the transport parameters irrelevant. Thanks to the laptop, you can put the above expressions into it and adjust the components with the aid of Perry or Spiers to your "heart's" content. You will find, most probably, that you get too much data but it will give you extreme values and an indication of accuracy for when you decide to use the shell and tube heat exchanger software. If you have access to them, and the time, you may care to check your results when using HTRI correlations with HTFS correlations.

I think that I have answered your question - hope that it helps!

Regards Stephen.

Dear Miguel, you cannot assume a constant value of UA. The heat exchanger area A is fixed but the global heat transfer coefficient U depends of the properties of both flows: density, viscosity, mass flow rate, temperatures and, maybe the most important, the phase of the fluid. In off-design coditions, U must be calculated for each different condition.

I agree with your answer Mr. Marcio S.