Inlet and outlet of the shell-side of a S and T heat exchanger

[reactorshell]

When considering the operation of a shell and tube heat exchanger, where a heating medium (hot oil) flows in the shell side of the exchanger, is it necessary that the pipings to the inlet and outlet to the HE be such that the outlet piping rises to a level higher than that of the inlet. I believe this actually helps in ensuring that the shell side of the exchanger is fully filled with the heating medium, thus eliminating the possiblity of any vapour buildup within the exchanger which could significantly affect the heat transfer coefficient near the top of the reactor.

Any comments, advice and suggestions are greatly appreciated.

 

[22082002]

One more thing:
To conserve heat that is lost from the shell to atmosphere(of course, one will have insulation, but still),nowadays, in refineries(and espacially in all catalytic units in refineries), the hotter fluid is on tubeside.
It is also known prior to design that catalyst dust, etc. will be carried with thi sfluid, but across three licensors/designers i.e. UOP,TOYO, Haldor topsoe, I have observed this way of putting hot fluid in shell side.

 

[reactorshell]

22082002,

I totally agree with you on the part of which side the hotter fluid should flow in. Unfortunately, the HE is already in place and therefore process modifications to it could be difficult.

Anyway, what are the possiblities of developing vapour pockets on the shell side of the exchanger? What would be the effect on the heat transfer? Have u ever experienced this?

 

[22082002]

Can you tell me the baffle cut in the exchanger, whether it's horizontal cut or vertical cut?
If it's horizontal, there shouldn't be any problem. There will be small hole or triangular notch at the middle i.e at bottom and top point of exchanger (sorry, I am not able to get the correct technical word for it), which will push the vapors to lower pressure side i.e. outlet of exchanger.
Meanwhile I will talk to some of my collegues and find out additional information.

 

[25362]

Reactorshell: What gas or vapour would you expect to be released from the hot oil upon being cooled ? Dissolved inert gas or lights from thermal cracking ?
Isn't the back pressure CV sufficient to keep this from happening ?

 

[MJCronin]

reactorshell,

Does the "hot oil" system have an expansion tank mounted high ? (This tank is typically installed on the suction side of the pumps)

If you do have an expansion tank, then air/gases should collect there. That is one of the purposes of this tank.

Alternately, you could consider installing a start-up vent on the shell of the heat exchanger. This may require review and approval as an "Alteration" under ASME Section VIII

Do you suspect that you have problems with collected gases or are you just guessing ? Unless the HX is not fuctioning properly or exhibiting problems, I see no reason to modify it.

MJC

 

[reactorshell]

sorry guys, have been out for quite some time...here goes...

22082002,

the baffles are ~10% horizontal cuts. I really don't recall any notch in the middle...i've look through the mechanical drawings and found nothing...could u describe it in details?

25362,

Yes, i have actually suspected some sort of dissolved inert gas or light-ends from the thermal cracking of the hot oil. We did have some discussions on whether the back pressure is sufficient to prevent the formation of the vapour pockets and basically my team is pretty divided on whether this is so. How can i find out anyway?

MJCronin,

As far as i know, there is no such expansion tank installed. But i will check this out. But we do have a start-up vent, although we have never used it, at least not during my time here in the plant.

Yes, there seems to be a problem with the HX, such that we are not heating the tube-side sufficiently.

Does anyone has any idea on checking for trapped vapour pockets?

 

[25362]

Looking to the problem from the other side. If you are heating a gas/liquid mix as feed to a reactor, you may have phase segregation on the process side. I have had that experience on a wax hydrotreating exchanger feeding a reactor. Besides, the process fluid may have thermally degraded so as to form a film of coke-like deposit on the tube walls. This fouling may impede obtaining the desired results.

 

[reactorshell]

25362,

I guess u can look at the problem from teh other side as well and reason that the heat transfer problem, if there is one, could occur on the tube side. In fact, i do have a working reactor and we do suspect that fouling could occur as a result of condensation of the vapour phase, leading to severe plugging of the tubes. The question is, where would the problem have first started? THe tube or shell side?

 

[25362]

Can you define the fluids on each side ? It might help to pop up some ideas.

 

[reactorshell]

The heating medium on teh shellside is a commercial heating oil, Sandotherm while the tubeside is basically a vapour phase mixture of primarily methyl phenyl carbinol and water.

 

[22082002]

1)In response to Feb, 21, message:I really don't recall any notch in the middle...i've look through the mechanical drawings and found nothing...could u describe it in details?

It is called drain notch, to ensure all the liquid can flow to the exchanger drain nozzle. Notch is only in bottom.

Sorry for replying late.

2)As regards to basic question, how will be the piping be: Hotter side coming from top and cold going from bottom of exhanger so outlet will be at lower elavation than inlet for shell side fluid. This is the first layout that comes to my mind owing to thermal & density gradient.

3)Is this vapor generation recent phenomena?
I had little experience of similar heating media, dowtherm, but then it had expansion bottle at highest elevation.(at least one floor above the user exchanger). I cannot recall having thermal carcking or loss of oil.

Have you given any thought/consideration to leak in exchanger(since shell side and tube side Pressure is not known)

4) Don't you have any oil filter for the circulationg heating oil? If yes, the frequency of choking can indicate you.

Regards,
SAA

 

[25362]

22082002 is right concerning heat losses to the surroundings. In hydrotreaters heating up a two-phase flow mix of hydrogen-rich gas and waxy stuff feeding a reactor, having the colder fluid on the shell side and the outlet nozzle pointing upwards, liquids may be left behind on a stratified flow regime and coke up inside the shell. I've seen that happening in one lube refinery having an upwards pointing outlet nozzle. The problem was solved by installing instead an electrically heated coil with one heating element per tube so the flow regime became turbulent and no stratifying of liquids reappeared ever since.