7 replies to this topic

[CS10]

Dear friends,

 

I would like to get some ideas to solve the issue on my HYSYS simulation. I am using heat exchanger type LNG exchanger in my simulation.

 

I have a problem with the exchanger. I have specified the temperature in and out, but the exchanger doesn't solve. I found the problem with temperature cross. can you help me to work this out? and what does temperature cross mean?

 

thank you,

 

Cozta

[Pilesar]

Leave the temperature unspecified for all the outlet streams of either the hot side or cold side. Specify instead a positive 'minimum internal temperature approach'. Review the temperature curves of the hot side and cold side of the exchanger. Temperature changes within an exchanger are generally not all represented by straight lines. The software divides up the exchanger mathematically into sections for calculation purposes. This will treat your one exchanger as a series of perhaps five or ten smaller exchangers in sequence. More sectional divisions approach a continual curve and better represent the real world situation, so make sure you have enough divisions. The temperature cross warning means that the requested heat transfer for at least one of these heat exchanger sections is mathematically impossible. Heat cannot transfer from a colder temperature stream to a warmer temperature stream. The problem may not be apparent by looking only at the temperatures of the feeds and products to the exchanger because of the non-linearity of the temperature curves.

[CS10]

Thank you Pilesar for your explanation on this issue. 

 

What do you mean by dividing the exchanger in sections? 

 

 

I have attached also a schematic drawing of the heat exchanger. For this type of the heat exchanger, I am using a simplified model.

 

I have not been able to simulate the model using the rigorous model (EDR model), because I am still confuse with the branch of the stream within the heat exchanger itself  (stream C and D).

 

Appreciate for your inputs.

 

Attached Files

•  HX.PNG   5.63KB   27 downloads

[Pilesar]

You will likely never need to use anything more than the simplified model. I suggest you pretend you do not know the EDR model exists.

As for dividing the exchanger into sections, this is something the computer program does to solve the numerical integration needed to determine the heat transfer. At each point of the exchanger surface, the heat transfer surface only 'sees' what is actually touching it and does not see the whole hot stream or cold stream. So each length of exchanger surface can be represented as its own little exchanger. I think the programs generally use Simpson's rule (look this up) for the integration. It is really an easy concept, but difficult for me to explain in writing a few lines. I hope someone else here can better explain.

[Pilesar]

If your D stream enters or leaves the exchanger in the middle, then you should reflect this in your simulation and model this by two LNG exchangers in series. 

[Bobby Strain]

Do as Pilesar suggests. From your description it sounds like you are taking a partial draw, D1, from stream C1. If so, it is a configuration I have never seen in more than 50 years of practice. But, I must admit, I have not seem them all. None the less, at the point D1 is split is where you should split the exchanger. Is your objective to provide design data for a new exchanger? Or something else. Regardless, as Pilesar points out, the task should be quite easy with simple exchangers.

 

Bobby

[CS10]

I am trying to build the model with HYSYS of the existing system that we have and to be used for reference for any simulation or optimization required.

 

Thank you for all you inputs and comments on the questions I have posted. It's really helpful for me to build the model. 

[davidjdcv]

If you share your simulation, I could make a comment on it.

 

Juan