3 replies to this topic

[cusx]

Hi,

 

I'm designing a S&T HX, and came across a peculiar situation that I would like some input from. 

 

Hot in: 1098 C

Hot out: 530.2 C

Cold in: 503.3 C 

Cold out: 872 C

R: 1.54

S: 0.62

 

I tried to use the 1-2 Ft Correction formula, and it gave an error, which I presume, given the conditions as above, a 1-2 S&T HX is not suitable. I tried using the graph method, and as expected, the values were not able to intersect. 

 

 

Therefore I look up other graphs of Ft Correction and found out that the R and S values were only able to intersect at a 3 pass shell heat exchanger. 

Therefore, I made the conclusion that given the required inputs and outputs, I'm required to design a 3 shell pass HX.

 

 

But when I look at the TEMA Standards, there's no shell type for 3 pass shell. the maximum is only 2 ( Type F).

 

Can I just use a type F shell and assume the Ft factor is one ? i.e. Ft=1 ?

[PingPong]

With your temperatures you need 5 (five) shell passes, not 3.

 

So you would need to install 5 shells in series.

[cusx]

That is an interesting comment. May I know under what basis did you determine that I needed 5 besides the temperature?

 

5 Shells, is it a Type E or F Shell ? Can I just design a 1-1 HX to overcome the cross temperature ?

[srfish]

PingPong is correct. You need at least 5 shell passes. 4 shell passes gives a LMTD correction factor of only 0.6462. This is much lower than the recommended minimum correction factor of 0.80. This is for type E shells. You could use 3 type F shells. There are also ways of using fewer counter-flow heat exchangers in order to lessen the number of shells in series.