My reply may not be that of help but it clarifies simple facts.
1) You need to start going over your simulation setup to see what did you wrong. What assumptions/simplifications made you too car from reality. How do you judge if these are correct numerical results...etc.
2) People are here to help, so it's better to continue on your earlier post to let others also keep track of the case.
3) Share your model and sim setup to get more accurate and quicker response.
I read the other post before commenting here. Just measure the exit temperature and figure out the over all heat transfer coefficient between the two liquids. I imagine that the energy lost by the one fluid is gained by the other. That would be the average heat transfer coefficient. What you get in the plots are the distributed heat transfer coefficients which vary all over the place depending on the local conditions. You can also set a goal to get the average heat transfer coefficient over some selected set of surfaces, you can the get the min and the max as well. Seems like you are burning a lot of cycles on something that is pretty straight forward. What are the average heat transfer coefficients for the experimental set up? How are you calculating those? Just do the same thing from the simulation data. I am pretty sure you don't have a heat transfer coefficient direct sensor.
Mathematically, h=Q/( | Tfluid-Twall | ) at a stagnation point is infinite (Tfluid-Twall = 0).
I'm not sure how to remedy that, but, as Bill mentioned, you can estimate U from Q (the amount of heat transferred = difference in mass flow x specific enthalpy) and the 4 inlet and outlet temperatures. Calculate the LMTD from the temperatures and the U=Q/(A x LMTD).
why would Tfluid-Twall=0 anywhere when the fluid is conductive? There is still heat transfer when the fluid velocity is zero, is there not?.