10 replies to this topic

[daraj]

When we export model results from aspen plus to EDR and try to design a heatex, a TEMA sheet is generated by aspen EDR with clean, service an ddirty coefficients. Sometimes I end up with U service > U dirty which I know is not correct while designing. I wanted to understand why this happens an dhow can i rectify this? it would help if I know how does EDR exactly go up finding the service coefficient. I understand dirty coeff includes fouling factors and clean is without fouling. But how cna service end up being smaller than dirty? what does this imply?

[Bobby Strain]

Read your post. Sentences are inconsistent.

 

Bobby

[Pilesar]

If the overall heat transfer coefficient needed to achieve the specified outlet conditions (U service) is greater than the overall heat transfer coefficient calculated for the fully fouled condition you specified (U dirty) then the heat exchanger will not be able to achieve the results you specified when it is fully fouled and therefore your design is inadequate. Maybe you need more surface area or higher velocities or change other of the many parameters involved in heat transfer equipment. Your exchanger should have some additional heat transfer capacity as a safety factor by design. This results in the difference between U service and U dirty.

[daraj]

Pilesar, thanks for your inputs. But here is my confusion. In order to calculate the "overall heat transfer coefficient needed to achieve the specified outlet conditions (U service) " how does EDR do this? does it calculate this from first principles? I am wondering how the calculation prcedure for this is different from that of the procedure for Udirty? My understanding is Uclean is calculated from first principles based on flow properties, inlet/outlet conditions, chosen geometry etc. To this, fouling factors is applied, and then you arrive at Udirty which is a lesser number than Uclean.

But I dont get how Uservice is arrived at by the software?

[daraj]

In other words, how are Uservice and Udirty calculations independent of each other is what I am trying to understand..is U service calculated based on below?

Q= Uservice.A.LMTD, where Q is heat that need to be transferred. But where does the value for A come from?  unless you know A, you cannot calculate Uservice 

[daraj]

I am running EDR in design mode, so I am trying to find A, I dont know A beforehand. But you need A to find the U..

[Pilesar]

The heat exchanger area is determined from the physical configuration (tube size and number) of the exchanger. In design mode, the software is analyzing specific, discrete configurations. It proposes designs that have a tube length, bundle diameter, tube spacing, baffle spacing, and everything else that would be in an actual exchanger. These designs are guesses based on heuristics. Some of these proposed designs will be too big or too small or not an efficient use of pressure drop. The software analyzes each of the proposed designs to try to find what it thinks are the best configurations. So you are correct that there must be an area to find the overall heat transfer coefficient. There must also be a specific tube material, fluid velocity, fouling layer, etc. Because these are discrete exchanger design configurations, neither Uclean nor Udirty will provide the exact exchanger performance the user specifies. In order to reach that exact performance, the U is allowed to float in the calculations. The software asks itself 'Given the physical exchanger design proposed, what heat transfer coefficient would be needed to achieve the exact result the user requests?' If the answer (Uservice) is larger than Udirty, then the exchanger performance will not be satisfactory when fouled. Note that the heat exchanger service requested may not always be possible given the constraints. EDR may propose multiple shells to help. The EDR design mode is a great method to find out the range of reasonable designs. For final design, there is usually additional optimization the user can make after transferring the configuration to rating mode. The final design is the product of the human designer using EDR as a tool. Evaluate the design in rating mode. What if the baffle spacing and cut were different? What if the exchanger tubes were a different material? What if the nozzles were different? What if the tube pitch were changed? The software in rating mode can help answer these questions. When I design an exchanger, I spend much more time in rating mode than in design mode!

Edited by Pilesar, 03 October 2023 - 05:01 PM.

[breizh]

Hi,

I encourage you to review the basic on design of heat exchanger.

Attached documents to support your work with steps.

 

Breizh

 

 

[IGC]

Datasheets produced from "design mode", are usually not worth the paper they are printed on in my experience...

[daraj]

IGC, what I am doing is just conceptual design/modeling for a rough budgetary estimate of process capital costs using aspen EDR/economic analyzer..so accuracy and precision need not be super high as needed in say a basic engineering package or something. But yes, your point taken regarding design mode and its limitations.

 

Pilesar, thanks.

"Note that the heat exchanger service requested may not always be possible given the constraints."

 

that may have happened in my case. Even though the run converged in EDR and TEMA sheet generated Uservice ended up being higher than U dirty. So looks like given te constraints I gave(which includes TEMA type, tube diameter, baffle cut percentages, allowable pressure drop etc) EDR was not able to find a design which cna provide effective surface area>required surface area. In that case I wonder why it doesnt give a warning or error, but simply reports a Uservice > Udirty which is not very useful. Aspen needs to fix this, right? or point to where the user can provide some relaxations in specifications to get to a lower Uservice.

[daraj]

breizh, thanks, will go through those docs