15 replies to this topic

[bag]

dear all

i have a doubt on one of our current shell and tube heat exchanger.

shell side inlet: 45degC
shell side outlet: 55degC
tube side inlet: 250degC
tube side outlet: 180degC
countercurrent

is it practical to have such a big temp difference in tube side? what are the potential problems? and how to resolve or minimize the problems? can expansion bellow absorb the thermal expansion and how does mechanical installation cope with this thermal expansion?

i personally have never seen such a big temp difference before (70degC). normally i only allow 15-20degC max. i'm not sure if i'm too conservative..

[pavanayi]

bag,
you personally have not seen so many things . Even I personally have not seen so many things!!!

Anyway, temperature difference and its practicality etc always comes down to things like the area required for the heat exchange or ease of manufacturing.
Have you estimated the area required? What is the shell side fluid? tube side fluid? Flow rates? What is the overall process? Without any such information, its tough to give any suggestions.

Edited by pavanayi, 25 February 2011 - 09:26 AM.

[alokjaiswal]

Dear Bag,

The main function of Heat Exchanger is to exchange the heat between the two fluids separated by solid wall. One fluid will gain heat and the other fluid will release heat of the equal quantity. So, its the transfer of energy which is taking place. Hence, the temperature drop or increase will depend upon how much heat is getting transferred between the fluids.

It is not unusual to have temperature difference of 70 deg C or more. The basic equations of heat transfer is:

m.Cp.(T2-T1)
where m = mass flow rate
Cp = Specific heat capacity
(T2-T1) = Temperature difference

Hence, other factors i.e., mass flow rate (m) and specific heat capacity will also determine the third factor i.e., temperature difference (T2-T1).

Alok Jaiswal

[Art Montemayor]

Bag:

It is difficult to answer your specific question(s) because you don’t furnish all the basic data related to the heat exchanger in question. Normally, it is not practical to have such a big temperature difference. But that depends on what you are trying to accomplish with the specific heat exchanger and the constraints of the process. To answer all your specific questions, we need the Data Sheet for the heat exchanger.

We also don’t know if this is an existing, operating unit or one that you are designing or thinking about. If it is one you are designing, then the problem is more difficult to approach without knowing what the scope of work is and what the process involved has as its needs and all the detailed basic data – like fluid identification, conditions, etc.

If it is an existing heat exchanger, the big temperature difference between shell and tube can be resolved with a TEMA BEU type of unit. We need more specific information.

[bag]

Art:

sorry that i have not provided enough information.

the task is to cool down a certain amount of fatty stuff from 250degC down to 90degC after a single pass through a cooler ( or a series of cooler). one of the engineer specified 4 coolers in series to achieve final temperature 90degC. so the fabricator has come out with the thermal design: 250degC to 180degC after first cooler, 180degC to 135degC after the second one, 135degC to 110degC after the third one, and 110degC to 90degC after the fourth one.

this is a new heat exchanger. however, it's already being fabricated. i got to see the specs only yesterday and i am really worried about it.

'Normally, it is not practical to have such a big temperature difference. '
---honestly speaking i have the same feeling when i first saw the specs, though i can't explain exactly why i would have this feeling. theoretically speaking the thermal design looks perfect. but my concerns are thermal expansions, mechanical installations and maintenance issues during operation. if i have to design a suitable cooler for this task i would either find a plate type heat exchanger or design a series of coolers with each max 15-20degC temperature difference. i'm not sure if i'm too conservative.

'the big temperature difference between shell and tube can be resolved with a TEMA BEU type of unit. '
---the process fluid is running at tube side and they are fatty stuff, with high melting points and sticky. we are definitely looking for self-drainable type of heat exchanger. U-tube is therefore not considered.

[bag]

hi pavanayi,

thanks for your reply. the thermal design is definitely logical. i'm just wondering the mechanical part. my concern is whether this heat exchanger is practical or not.

[bag]

alokjaiswal:

seems like you have seen such a heat exchanger before? do you mind to share the operating performance of this heat exchanger with me?

[Art Montemayor]

Bag:

Why didn't you just furnish ALL the basic data at one time and save everyone a lot of time? Uploading the Data Sheet would have been so simple.

What you should be using is a spiral type of heat exchanger - which allows cleaning of either fluid with ease.

[pavanayi]

Bag,
In transfer line exchangers in ethylene plants, you can see tubeside temperatures drop in the first exchanger from 850°C to 600°C in the first unit (usually a set of several parallel doube pipe exchangers), then to 400°C in the second unit (shell and tube exchanger), then to around 200°C in the third one (shell and tube exchanger).

Startup from cold was at the rate of 50°C per hour, so it used to take 16 hours just to get to the process standby temperature.

Edited by pavanayi, 28 February 2011 - 07:11 AM.

[bag]

here comes the spec sheet.

Attached Files

•  spec sheet.pdf   36.19KB   51 downloads

[bag]

Bag,
In transfer line exchangers in ethylene plants, you can see tubeside temperatures drop in the first exchanger from 850°C to 600°C in the first unit (usually a set of several parallel doube pipe exchangers), then to 400°C in the second unit (shell and tube exchanger), then to around 200°C in the third one (shell and tube exchanger).

Startup from cold was at the rate of 50°C per hour, so it used to take 16 hours just to get to the process standby temperature.

i'm more interested in the step from 600degC to 400degC using shell and tube exchanger. is it a single exchanger or multiple exchangers in series?

also the double pipe exchangers. i have seen that before but i thought it's just piece of jacketed pipe. now i know it actually has a name double pipe exchanger! thank you

[pavanayi]

Bag,
Yes the 600°C --> 400°C and 400°C-->200°C are achieved by single exchangers in series. The first one had BFW converted to steam in shell side as was the previous double pipe exchangers, whereas the second one is used to preheat BFW.

The exchangers had a throughput of 25 tonnes per hour hydrocarbon and 5 tonnes per hour steam (totalling 30 TPH) in the tubeside.

[alokjaiswal]

Dear Bag,
Find attached a data sheet of S&T Hex in Ammonia Plant which is used to cool down the Methanator outlet process gas from 361.4 deg C to 125 deg C by exchanging heat to Boiler Feed Water which gains the temperature from 110 deg C to 285 deg C. Also, its one single Exchanger with area of 613 M2 which is working till date.

As far as your Data sheet for the Exchanger, which you have attached, is concerned, the data are absolutely correct if you calculate the duty and then back calculate the area required for heat exchanger. You will definitely get the temperature gradient of 70 deg C.

Attached Files

•  114-C.pdf   1.47MB   36 downloads

[bag]

Dear Bag,
Find attached a data sheet of S&T Hex in Ammonia Plant which is used to cool down the Methanator outlet process gas from 361.4 deg C to 125 deg C by exchanging heat to Boiler Feed Water which gains the temperature from 110 deg C to 285 deg C. Also, its one single Exchanger with area of 613 M2 which is working till date.

As far as your Data sheet for the Exchanger, which you have attached, is concerned, the data are absolutely correct if you calculate the duty and then back calculate the area required for heat exchanger. You will definitely get the temperature gradient of 70 deg C.

i'm confident if looking at the datasheet alone. what i'm not confident is purely the mechanical part, which might leads to operation problems.

i noticed in your datasheet floating head and tubesheets are used. however in our case the head is fixed and it's the expansion bellow which is used to absorb whatever thermal expansion takes place. and if the expansion bellow is not enough (let's say 2-3 is required instead of 1), after certain cycles of operation, will there be any equipment rupture? and if more bellows required can they all put together at one side or they have to be distributed evenly along the shell? also in this case, is sliding plate required to be installed under the support, instead of fixed bolt holes?

[Art Montemayor]

Bag:

Since you have already committed to fabricating the heat exchanger indicated in the Data Sheet, we really are only caught up in an academic discussion regarding the high temperature difference, the quality of the fluid, and the mechanical characteristics that would have been more favorable. Therefore, all I can comment on will be for the sake only of drawing attention to what I perceive as decisions and steps that I would have done had I the opportunity to prepare the Data Sheet:

• As I stated, if the tube-side clean out feature is regarded as important and has a top priority, then I would not have selected a shell & tube unit. I would have opted for a spiral heat exchanger which would have been much smaller than what you are fabricating - and easier to operate and clean, and with inherent expansion capabilities.
• If you have to use a shell & tube (BEM), then you have done something undesirable in heat exchanger design: you have given the tube-side fluid a velocity that is 10 to 12 TIMES LESS than the recommended velocity would be. At 0.2 m/s you are stimulating bad heat transfer and fouling. You are winding up with an exagerated size due to the horrible heat transfer coefficient on the tube-side. Your low tube-side pressure drop reflects this.
• You have not mentioned it, but you should be checking for the need for shel side expansion - especially due to the 1" OD SS tubes. The rather long SS tubes will generate more expansion stress than normal; TEMA has guidelines to test for the need for shell expansion allowance. You should insist on your fabricator calculating all the built-up stresses between the tubes and the shell in order to make sure you have a safe unit. I would have designed for an outside packed floating head on this exchanger in orde to relieve all built-up tube stresses. I have done this before in several plants that I operated and they worked well.

For what my past experience is worth, that is what I would have done.

[bag]

Bag:

Since you have already committed to fabricating the heat exchanger indicated in the Data Sheet, we really are only caught up in an academic discussion regarding the high temperature difference, the quality of the fluid, and the mechanical characteristics that would have been more favorable. Therefore, all I can comment on will be for the sake only of drawing attention to what I perceive as decisions and steps that I would have done had I the opportunity to prepare the Data Sheet:

• As I stated, if the tube-side clean out feature is regarded as important and has a top priority, then I would not have selected a shell & tube unit. I would have opted for a spiral heat exchanger which would have been much smaller than what you are fabricating - and easier to operate and clean, and with inherent expansion capabilities.
• If you have to use a shell & tube (BEM), then you have done something undesirable in heat exchanger design: you have given the tube-side fluid a velocity that is 10 to 12 TIMES LESS than the recommended velocity would be. At 0.2 m/s you are stimulating bad heat transfer and fouling. You are winding up with an exagerated size due to the horrible heat transfer coefficient on the tube-side. Your low tube-side pressure drop reflects this.
• You have not mentioned it, but you should be checking for the need for shel side expansion - especially due to the 1" OD SS tubes. The rather long SS tubes will generate more expansion stress than normal; TEMA has guidelines to test for the need for shell expansion allowance. You should insist on your fabricator calculating all the built-up stresses between the tubes and the shell in order to make sure you have a safe unit. I would have designed for an outside packed floating head on this exchanger in orde to relieve all built-up tube stresses. I have done this before in several plants that I operated and they worked well.

For what my past experience is worth, that is what I would have done.

hi Art Montemayor

thank you very much for sharing your valuable industrial experience with me.

i have one question. you mentioned 'you have given the tube-side fluid a velocity that is 10 to 12 TIMES LESS than the recommended velocity would be. At 0.2 m/s you are stimulating bad heat transfer and fouling. '.

i'm just using my common sense to think about the low tube side velocity scenario. imagine the shell(cooling) side is the same, with a smaller tube side velocity doesn't it allow a longer residence time for tube side process flow to stay in the heat exchanger? if this is the case, more heat should have been exchanged from the hot side to cold side and therefore heat transfer is better. kindly correct me if i'm wrong.