3 replies to this topic

[ISMAR]

Dear colleagues,

In our new Amine Gas Treating and LPG Recovery Unit, now in detailed engineering phase, we chose TEMA rear head type of some HE as "T" and "S" floating head type instead of "U-tube" type for Condenser/Chiller and Reboiler which deal with two phase/changing phase in tube side. Tube side fluida is : LP or MP Steam, Feed Gas, and Deethanizer O/H. It seems that it's not a right choice due to there is two phase flow on the tube side which might cause redistribution problem. The liquids tend to settle to the bottom tubes and the vapor ends up in the upper tubes. Any references for this theory ?
My questions is :
- Is it still possible to use "T" and "S" type without any redistribution problem ? If it is, is there a big difference of design size between "T" or "S" type and the size of HE if we choose the "U-tube" type ?
- Any experience with this floating heads type in existing refinery which operate erratically with very low heat transfer due to this two phase redistribution problem ? We have some HE in our existing refinery with "T" adn "S" type for reboiler and condenser service which deal with two phase fluida in tube.

[Art Montemayor]

ISMAR:

I would never allow the use of "S" or "T" rear head design on any heat exchanger (and certainly not a reboiler!) unless my client ordered me to. And if that were the case, the client would have to take full liability on their performance. There may be some very good reasons for using such an expensive, inefficient, and leak-prone type of design -- but I have never, in my 44 years in engineering found any.

I am addressing this issue with over 10 years of direct, field experience in designing and operating Amine plants as well as also directing the design and fabrication of heat exchangers. I have simply never found any reasonable justification to even consider such a design in an Amine plant. What is more baffling is that an engineer would use the design in a reboiler. You don't state what kind of reboiler, so I have to assume a kettle type - which is the reboiler of choice in an Amine unit.

I would also strongly criticize the placement of the condensing fluid in the tube side; this simply doesn't make sense. You must be retaining a lot of basic data that justifies these type of abnormal decisions, because what you describe simply is not the normal, logical way to design heat transfer equipment. However, if you are dealing with a conventional Amine unit, then there is no reason to use these type of heads for the following reasons:

1) The required gaskets will eventually leak and cause unit shutdowns or contaminated product/ process fluids;

2) The design requires a considerable amount of machining and fitting; this drives the cost up.

3) The inherent nature of demanding a pull-through design forces the mechanical designer to increase the critical clearances between the outer tube limit and the inside diameter of the exchange shell; this immediately causes a lot of shell-side fluid by-passing and the efficiency goes down - in many units by as much as 50%! Mechanical designers and fabricators try to reduce this wasteful by-passing through the use of strip seals. But these inevitably get ruined, or corrode out as the unit is used and re-installed. The end result is that there is no positive, sure way to eliminate the by-passing completely.

4) The above clearances and requirements cause the exchanger shell to become bigger; the final design is much bigger and weighs more, requiring more space and structural support.

5) The inherent internal bolts that have to be employed are subject to corrosion, seizure, and ultimate failure. The option is to use expensive, special alloy bolting and higher equipment costs.

6) No matter what the mechanical design takes into consideration, the actual pulling out of such a tube bundle is a heavy and cumbersome chore! Anyone who has worked or supervised such a maintenance job will concur that if pull-out tube bundles can be avoided, they should! In these type of bundle design, because of the sealing strips, the pulling out is very hard and sometimes almost impossible when the strips jam inside the shell. This is a very important part of process design engineering that is often overlooked -- to the hatred and irritation of the maintenance department in the plant. As a junior engineer I still recall my first exchanger maintenance assignment that took us several days struggling to extract a "removable" tube bundle. We worked continuously through the first night, without eating or taking a break. The ultimate answer was to ship out the exchanger complete and have the bundle hydraulically extracted in a fabricating shop - with the end result that the bundle (and the shell) were seriously damaged and practically had to be reconstructed. The maintenance department never forgave the engineering department for this gross error in judgement and expertise and they proceeded to dis-inflate all the tires of the engineers' cars in the parking lots as a symbol of their feelings! My car was spared.

I could continue on and on giving you reasons not to employ this type of design - but you (or your company) will do what you decide to do, and obviously you already have by taking this into the detailed engineering phase. What is not logical is to use this pull-through type of bundle design in a reboiler. You should explain why and how you intend to apply this where it surely is not needed and only represents trouble and considerable additional costs. I am not writing this to be merely critical of your design, but to point out that unless you have some very, very extra-ordinary reasons (which you haven't stated) you shouldn't be using this type of heat exchanger design on an Amine unit or on reboiler service - especially with the condensing fluid in the tubeside.

Art Montemayor

[Art Montemayor]

ISMAR:

Your additonal basic data now explains your initial questions better. I now understand you basically want to know:

1) Can you place a condensing fluid in the tube side of a heat exchanger (with S & T internal heads) serving as a condenser?

2) Can you use a U-tube design for such an application?

You still give us a lot of other related information, but I can't decipher out any other concerns. My response are:

Yes, you can use an S & T head if you like; there is no difference in heat transfer operation between heat exchangers using "N", "S", "T", or "U" configurations. You will, on a practical and logical basis be limited to a 2-pass tube design on either of the configurations.

The problem with putting the condensable fluid in the tube side is that you will have the trade-off of having to vent off any included non-codensables at key, critical points to be identified with your physical configuration and construction. All involved tubes should be critically leveled or, preferentially, sloped to favor positive liquid drainage leading to the single liquid outlet by gravity. Anything that obstructs this from happening is going to affect your heat transfer, process conditions, and possibly worse, the stability of the operation. I have used the U-tube Kettle design to condense such fluids as CO2, NH3, Propane, Butane, CO, and othe clean, pure fluids in the tube side with no operating problems. However, I stress that the fluids were clean and pure with a minimal of non-condensables. I used the U-tube design because I could design a tube bundle that instead of being a true "U" shape, was really of a "Vee" shape - in other words, I bend the tubes such that I get a positive slope for liquid drainage when I introduce the condensable gas at the top of the tube bundle and allow it to flow downward through the tubes. This facilitates what I call vapor-liquid separation (& I believe you call it "redistribution") as well as the separation of non-condensables in the vapor space above the bundles liquid outlet. This method complies with the common and well-known rules of separating and purging non-condensables in equipment and heat exchangers.

You mentioned Amine exchangers and I don't see any problem that you present in that respect. I hope I answered your specific questions.

Art Montemayor
Spring, TX