2 replies to this topic

[Mahesh@A&M]

Current reactor has a 2 1/4" wide baffle in the jacketed(spiral baffle) vessel. Annular space between jacket I.D and shell O.D is also 2 1/4".

We are purchasing a duplicate vessel but the drawings shows 2 5/16" annular space. This is due to increased stress values based on 2010 ASME code.

My question is can I keep same baffle width ( 2 1/4") allowing water slippage through 1/16" clearance? This vessel also has internal cooling coil where most of the heat transfer occurs. ( because shell thickness is 1.6"). jacket fluid is chilled water.

Thank You

[Art Montemayor]

Mahesh:

It is difficult to visualize what you mean without a sketch of your reactor and its jacket. Please refer to the attached Workbook to see what I understand.

Basically, I think you have a batch reactor (and a THICK one, at that) that you need to cool. The jacket – like almost all reactor jackets – is a “best effort” type of cooling design. It is called that because you have a set, fixed heat transfer surface area that is totally enclosed and not free for inspection and difficult to clean. The scope of work on all reactor jackets is basically: “get all the heat transfer you get and rely on other means to control the reactor temperature effectively”. The “other means” is related to the other usual options open to you for application:

• An external, recirculation, pumped loop with an external cooler;
• An efficient mixer / agitator combined with an internal cooling coil.

Of the above two options, the first one is the most reliable and dependable for close, efficient reactor temperature control. A jacket is usually not considered as a safe and reliable temperature control method. Therefore, the amount of detailed engineering that goes into a jacket design is mainly based on the above philosophy.

You want to get as much heat transfer as you can out of the jacket, so you employ baffles between the reactor shell and the jacket in order to stimulate turbulent flow that increases the convective heat transfer coefficient and also reduces the chances of developing static, dead spots and induced fouling in the jacket – which is the worse place you can have fouling due to the difficulty of cleaning it.

I would normally not allow any clearance between the spiral baffle and the reactor shell AND the cylindrical jacket. Any clearance you allow will invite coolant by-passing of the designed spiral path (which is based on achieving a decent coolant velocity (approximately 3-6 ft/sec) and defeat the purpose of installing the baffles. You need the baffles in order to achieve the desired coolant velocity; but you can’t achieve the velocity if you allow leakage and resultant by-passing of the baffles.

I don’t understand what you mean by “increased stress values based on 2010 ASME code”. Does this mean that the radial expansion of the 1.6” thickness of the cylindrical reactor will put excessive force on the baffles and the jacket? That means you are allowing 1/8” for expansion on the reactor’s OD. If that is the case, then you have to fabricate accordingly. But you could use plastic spacers between the baffle and jacket inner wall. However, these would probably have to be “specially” built and cost$.

Attached Files

•  Reactor Cooling Jacket.xlsx   11.03KB   285 downloads

[Mahesh@A&M]

Thank You Mr.Art

I understood from pressure vessel vendor that " metal stress values have increased in ASME 2010 code" . This will make our new vessel shell and heads thickness marginally less than current vessel made back in 1990. So now I have to increase the jacket baffle width to 2 5/16" to avoid water slippage between baffle tip and I.D of jacket(spiral baffles are welded to shell).

Your sketch is right. This vessel is an agitated vessel with draft tube, baffles and internal cooling coil (within the draft tube).



Thank You