How to avoid a Temperature cross over 2

[bond0718]

I would like to know how to avoid a temp cross over in a simple heat exchanger. I read adding more shells like 2-4 instead of 1-2 can be a way. But i didnt understand how. can someone explain.

thanks in advance

 

[dcasto]

Are you describing an internal cross because of condensing or vaporizing?

 

[djack77494]

It's not clear what you are attempting here. Why do you want to avoid a temperature cross? Did someone tell you the exchanger wouldn't work with one?

Your pursuit is a worthy one, but is not capable of a quick answer. Generally temperature crosses are best avoided with a single shell S&T exchanger. That is because the exchanger's effectiveness drops rapidly as the temperatures cross. However, there are configurations and strategies, including use of multiple shells, that can overcome this problem. I suggest you spend some quality time with a good book on heat transfer to really understand the problem.
Doug

 

[reena1957]

Dear Moscos,
When the cold fluid outlet temperature goes higher than the hot fluid outlet temperature, a CROSS is supposed to take place in a counter current exchanger. Beyond this point, the temperaure of cold fluid will be higher than the hot one and heat transfer will take place in the opposite direction to that desired. For example, if you are cooling a hot fluid to 40 Deg with cooling water and the cooling water outlet temperature rises to 45 Deg C, there is cross of about 5 DeG c. This cross leads to poor utilization of heat transfer area and is normally taken care of in design by the LMTD correction factor F. When there is no cross, the F is above 0.8. If there is a cross, it rapidly falls below 0.8. thus losing the effect of good Delta T and increasing the area required almost exponentially. So when a cross cannot be avoided, you split the exchanger into 2 or more shells and avoid the cross in each shell because the hotter cold fluid never meets the colder shell fluid. You get good LMTDs and good Fs. But the cost is higher due to more shells. I suggest you draw some temperature vs Length profiles with 1-2 and 2-4 exchangers to get a feel of what I have written. Best wishes.

 

[sshep]

A temperature cross is determined by the inlet and outlet temperatures for the two streams- if the cold stream outlet is hotter than the hot stream outlet then there is a cross. The choice of exchanger design does not determine whether there is a cross, rather the presence of a cross determines what designs will be feasible. When there is a cross, a counter current flow design is needed because heat is exchanged only from higher to lower temp.

By installing multiple shells with the flow in opposite directions, the countercurrent design is achieved by discrete units rather than continuously in one shell- flow through exchangers is countercurrent although the flow in the exchangers is multipass. This achieves the effect of dealing with the cross while avoiding the tubeside problems of designing a single shell in counter current flow.

The problem with designing counter current flow exchangers is the tubeside, specifically: reducing length while getting high tubeside velocity, and rear head design that allows pulling the bundle. Baffling can get shellside velocity high, but good tube velocities usually require multiple passes, otherwise you are stuck with an impractical long thin design.

There is further economic reason for stacking exchangers. The cheapest exchanger to build is the "E" shell with "U" tube- a "U" tube is obviously multipass tubeside and such a unit can only be made purely countercurrent via the troublesome long baffle "F" shell.

best wishes,
sshep