3 replies to this topic

[Art Montemayor]

As I have stated numerous times on our forums, one doesn’t design or “size” a jacket on a process tank. What you do is simply fit the largest jacket you can to the tank and then calculate what you can obtain from this “best effort” task. You are going to obtain whatever is the result of what you install on that tank – that is ALL you can do. You have a lot of other variables and unknowns working in this application that you haven’t mentioned - and that are very important:

• What type of jacket are you applying? There are various types. What type of baffling are you using – if any?
• Do you really expect to heat the tank contents to 100 ^oC with only the jacket?
• Is your process tank designed to withstand the 60 psig EXTERNAL PRESSURE?
• Is your jacket going to be built to withstand the 60 psig internal pressure? Is your jacket wall thickness going to be thicker than your process vessel?
• Is your process vessel stirred? Do you have any liquid mixer or apply forced convection inside the tank?

Engineers use jackets on process vessels and reactors to ASSIST and help in the overall heating and/or cooling process needs – never is this done to be the ONLY way to transfer heat into a process fluid. The most efficient (and cheapest) way to heat and/or cool a process fluid inside a tank is to:

• Use an external circulation pump with an external heat exchanger;
• Use an internal tank coil(s) with an efficient mixer (stirrer).

The jacket is used to assist and help the above devices do the bulk of the heat transfer.

The jacket, when used by itself, is an unsteady state heat transfer device and – as such – a complicated heat transfer problem. You have a convection heat transfer operation that has to be defined (is it natural or forced?) and a falling film condensation taking place on the external steam side. Besides that, the steam pressure in the jacket is NOT THE SAME PRESSURE AS THAT SUPPLIED. The dynamic steam pressure in the jacket is varying in accordance with the heat transfer rate – which is unsteady state and varying with the LMTD and the overall U. this is NOT AS EASY as you have painted it.

Can you tell us exactly what you have and what you propose to do – filling in the blank information that I describe above? We may be able to help you, but we need ALL the basic information and background.

[Art Montemayor]

Workingbugs:

Thank you for the additional basic data. Now I have a better grasp of what you are proposing to do.

You have a vessel that is nominally 8 feet in diameter and 10 feet in height, tangent-to-tangent. These dimensions give you a total capacity of 4,200 gallons with simple dished heads. For this application, I recommend an internal heating steam coil(s) with an efficient vertical agitator. This is much more efficient, economical, and much better to control and vary than a steam jacket. You can install a steam jacket if you like, to assist in the initial heat up of a batch if you think it is necessary. My engineering reasons for the recommendation are:

• It is not practical to heat a 4,000 gallon batch to 100 ^oC using steam in a jacket. It will take a very, very long time to accomplish it – if you are willing to wait for it to happen.
• A vessel that big will require a very thick shell to withstand the proposed steam pressure – and the jacket will probably also require a thicker shell. This is not only costlier, but it will add an increased weight on the foundation. For the relatively little heat transfer obtained, it isn’t worth it.
• Internal steam coils can easily take 150 psig steam – or higher – at little or no extra cost. The jacket can’t take that kind of pressure – and it is very costly.
• You can added considerable heat transfer area with the internal coils – and consequently heat up very rapidly due to the added area and the agitator’s more efficient forced convection effect; you are limited with the jacket.
• Since you are building the reactor from scratch, it is probably cheaper to install the coils now than the jacket. There is less welding involved with the coils than with the jacket.
• There may be an advantage with the jacket in the clean-outs required between batches, but you have an agitator installed that requires clean-out as well already, so the clean-out required by the coils is not that much added work or effort.
• If you do the detailed calculations and compare the two methods, you will find that the agitated internal coils heat up the batch so much faster that it is not a horse race. If you add the money value on the reactor time saved by heating up faster, you will quickly see the incentive for the internal coils. The only thing that can beat the internal coils is an external pump recirculation system with an external heat exchanger – but that costs more money.

I hope this experience helps you out.

P.S.: Dimple reactors suck. Nobody uses these anymore.