4 replies to this topic

[niba87]

Hello everyone,

There has been quite a bit of information available on the topic of shell and tube heat exchangers in this forum. Extending beyond a single phase design, i wish to extend that single phase design to a two-phase design. The purpose of my design is to achieve the following:

Hot fluid: Ethylbenzene Vapour Feed to be condensed (20bar, 473 degress celsius)
Cool Fluid: Boiler Feed water at high pressure to form high pressure steam.

Most textbooks show detailed calculations for a single phase vapour-vapour or liquid-liquid system, so I was just wondering what other design considerations have to be considered?

Thanks in advance.

[breizh]

Consider this resource , it may support your query:

http://www.wlv.com/p...ok/databook.pdf


Breizh

Edited by breizh, 24 February 2012 - 01:44 AM.

[srfish]

Hello niba87

Is this a horizotal or vertical heat exchanger? It makes quite a difference.

Breizh gives a good literature source. Another source is Process Heat Transfer by Serth where he gives examples.

I don't have the thermodynamic property table for ethyl benzene but it appears that it is superheated at 20 bar and 473 C.

You want to avoid anything that would impede the flow of condensate.

For several more tips, refer to www.gulleyassociates.com.

[niba87]

Thank you Breizh and srfish for your replies. I am currently considering a horizontal heat exchanger, but with regards to cost estimation, would vertical exchangers have any advantages?

Breizh, the literature you have provided looks comprehensive. Thank you once again!

[niba87]

In addition, I understand that the velocity of fluid flowing through the pipe has to be limited.

Normal heuristics place liquid velocity between 1-2 m/s whilst that of vapour can be as high as 30-50m/s. In my case, the hot vapour will condense, whilst the boiler feed water will vapourise. Thus, the linear velocities of the fluids will thus change. Are there known heuristics available for this?

Also, Breizh, should the velocity change, the reynolds number would certainly be affected. As such, in the design, will I expect the convective heat transfer to change as well?

Thanks and regards
niba87