12 replies to this topic

[uetchem]

Actually I am working on assignment of designing Feed Water Heater, with Super Heated steam entering at Shell side and Sub cooled liquid leaving Shell Side.
So I am to handle Two phase flow/ Condensation occuring on Tube banks .
I am in search of some reliable formulas which can give me solution of two phase heat transfer coefficient and Pressure drop Calculations.
Keeping in mind change of phase along the bank of tubes and changing vapor fractions as steam moves on.
I will really appreciate an urgent response in this regard.
Affan Sadek

[djack77494]

Affan,
I'm sorry to say that you will not have an easy time of things here. The problem's complexity is greatly increased if you have desuperheating or subcooling sections in addition to condensation. You have both. I can only refer you to the vast literature of excellent heat transfer texts that deal with this subject. The most practical solution, of course, is to purchase the software that makes this type of problem quite easy.
Good luck,
Doug

[pleckner]

Affan,

In the event you are hoping to get a better answer than given by Doug, I have to second Doug's response. Sorry.

[uetchem]

Affan,

In the event you are hoping to get a better answer than given by Doug, I have to second Doug's response. Sorry.

Dear Sir

Actually Mr. Doug referred me to some literature, but plz specify some use ful literature.
I have gone through Culson volume 6, Choppy, Ludwig, online Engineering Hand book from wolvin and ofcourse Kern. though I have found som useful information particularly from culson and had already completed the initial Design, but I am not satisfied yet bcz both choppy and Culson are cattering phase change or Gradual Condensation with the help of term Nr i.e. No of tubes in vertical row and thus not dealing it with changing vapor Fractions.
SO if you people can guide me to some more specific literature in light of your experience, it might be of great help

Regards
Affan

[djack77494]

Affan,
You've just metioned the "classics" in heat transfer. Though there are other heat transfer texts, I don't think they'll take you beyond what you've already found. Try google searches, reading ChE Resource article/spreadsheets, and, if they aren't sufficient, plan on running a commercial program (HTRI would be good) to get your answers. If you're not that familiar with heat exchanger design, I'd recommend you hire a consultant. Yours is not an easy problem.
Doug

[latexman]

Affan,

If the steam is put on the tube-side of a vertical, down-flow condenser, I think it will more closely approach the classical models of desuperheating (if applicable), condensing, and subcooling in a smooth, sequential, continuous process than on the shell-side, where the fluid flow characteristics are broken up into sections at each baffle and tube support.

[Radionise]

The effect of flow pattern must not be neglected as well. I think the annular flow model will be good enough to predict the pressure drop. Evaluation of the heat transfer coefficient in filmwise condensation has been outlined by Nusselt too. You can then use it to evaluate the local as well as average heat transfer coefficient. I'm sure you can find the info from some of the books below:

1. Boiling, condensation, and gas-liquid flow / P.B. Whalley
2. Measurement of two phase flow parameters / G.F. Hewitt.
3. Two-phase flow and heat transfer in the power and process industries / A.E. Bergles ... [et al.].
4. Two-phase pressure drop and void fraction in narrow channels / Adrian John Holt.


In terms of software, perhaps you would like to look at HTFS.

http://www.hyprotech...htfs/resnet.asp

They have an extensive inventory of information from their elobated research in the areas of heat transfer and two-phase flows. The information is only available to subscribed members.

Good luck.

[amna]

i m designing vacuum distillation column. the feed contains 0.2% dissolved/ entrained gas.
so i just need to know the method to find heat duty of two phase condenser. can any one help me out.or please provide any reference!

[David Southall]

I think we might be able to help you more if you constrain the problem a little more:

- What is the tube-side fluid in your proposed heat exchanger? Are you just desubcooling water?
- Is there a specific reason you want to place the condensing stream on the shell-side? Tube-side may be a better choice, depending on what the other fluid is.
- Have you thought about flow arrangements? Is the shell baffled?
- Is the exchanger horizontal or vertical?

What I think you will find when designing a 2-phase heat exchanger is that finding the heat transfer equations is relatively easy; carrying out the design is the harder, more time-consuming part. Here is what I suggest you do:

- Consider what your economical pressure drop is going to be. The hydraulics of the problem are linked to the thermal design.
- Draw-up your heating/cooling curve, distributing the allowable pressure drop along the curve in terms of the physical properties you use. Process simulators like Hysys are great for checking these and producing physical properties.
- Break the heat curve into heat transfer zones. Zones should be selected such that the hot and cold side heat curves can be assumed linear over each zone. If you are condensing against a single phase heating stream, then you may get away with 3 zones for your problem (desuperheating, condensing, subcooling).
- Specify a geometry (number of tubes/shell side passes, tube size and pitch etc). Based on the geometry, select the appropriate heat transfer correlations. Also select a material of construction.
- For each zone, note the end point conditions (temperatures, pressures, heat transferred etc). Calculate end point overall heat transfer coefficients, then take an average. Also evaluate log-mean temperature difference (or actual temperature difference if the end point temperature differences are equal). Also calculate any correction factors applicable to the geometry (crossflow, bypass etc).
- Use a re-arrangment of (duty) = (correction factors) * (mean overall htc) * (area) * (LMTD) to estimate the area required for each zone. Sum these areas.
- Compare the required area with the available area in your specified geometry. If insufficient area is available, go back and re-specify the geometry. This is not just about increasing/reducing the surface area; you should also be looking for opportunities to improve/reduce heat transfer film coefficients, and to make adjustments to correction factors.
- if you have a geometry that satisfies thermal requirements, calculate pressure drops. If excessive, respecify geometry. If you are way under pressure drop, your design probably isn't optimal, and you may want to re-specify to use more of the pressure drop. You will probably also want to check flooding and other flow phenomena.

You have to be careful with the 2-phase heat transfer zone. End point conditions only may be OK, as the 2-phase htc for a single component fluid can often be approximately constant. It would be a good check to evaluate a htc's at a couple of intermediate vapour qualities to check if this is the case. If not (e.g. high pressure drop, or multi-component), you will need to break the zone into smaller zones of vapour quality, evaluate htc's at each of these, then take an overall average. If you had a multi-component stream you would need to think about using a Silver correction. You will also need to break into smaller zones if a zonal LMTD analysis would give a different value to the endpoint-based LMTD (this is the case if the curves cannot be assumed to be linear).

This also only applies if the area available to both sides is equal. If not, then you would need to evaluate the UA required, and calculate the UA available (by summation of reciprocals for both sides).

There are a number of correlations out there, depending on what your geometry is, and whether you are condensing shell-side or tube-side (HTFS, Butterworth, Nusselt, Kutateledze, Boyko-Kruzhilin, to name a few). One book I like on the subject is "Heat Exchangers - Selection, Rating, and Thermal Design" by Sadik Kakac and Hongtan Liu. If you can get access to the HTFS Handbook on resnet, then the CE series ("Equipment"), CP series ("Process") and CM series ("Method") are the sheets you need on the subject of condensation. Bear in mind that gravitational effects and vapour shear will affect condensing coefficients - you may find yourself needing to calculate a gravitationally based htc, a vapour-shear based htc, then taking an RMS average of the two. Another (free!) resource is Leinhardt's "Another Heat Transfer Textbook" (use google to find the free download).

I suspect 3 zones will be enough for your problem (unless the tube-side is boiling and the end points don't coincide with the condensing side, in which case 5 zones or more are probably required). I would also suggest one or two htc checks along the condensing zone.

As the solution to this kind of problem is iterative, I strongly suggest that you build your calculation around a spreadsheet, so that you don't need to keep repeating calcs. I would also suggest that you do some background reading so that you settle on the 'right' configuration type in the first instance, so that you do not need to reconfigure your spreadsheet for different geometry types (this can be really time consuming). I would also suggest that you think about having the spreadsheet do the pressure drop calculation in parallel to the thermal calculation; it really is frustrating to arrive at a working thermal solution after several iterations and find that the pressure drop is way off. If you do the calc in parallel, you can optimise on both.

I know this is long winded, but this is the design approach I would be thinking of to design a 2-phase heat exchanger like this. As you mentioned that this is an assignment, I have consciously avoided anything more than the basic heat transfer equation - part of the fun is about digging through references, working out what the heat exchanger is going to look like, and then justifying it by argument before the calcs and by the results after the calcs!

Ask yourself what the requirements of the assignment are - are they to find htc methods, or simply to design a heat exchanger? If it is an academic design exercise, you could greatly simplify the problem by breaking into 3 zones, assuming a constant condensing coefficient, then finding a table of typical (or 'rule of thumb') values for htcs and use those for the design, possibly with a de-rate factor for conservatism.

I don't know what the full scope of your assignment is; if you also have to consider mechanical design, then you may need to think about things like tube vibration (there is an HTFS paper on this), a suitable material of construction for your pressure and temperature (take a look at something like an ASME materials table for allowable stresses in ASME VIII, if designing to ASME), and corrosion.

One final note - I always start a quest like this with a google search and a wikipedia search. You do have to be a bit careful with them sometimes, but it's worth it just to see what references other people are using. It took a couple of minutes to start finding articles on google and wikipedia, for example I found a wikipedia article relating to surface condensors where shell-side steam is condensed against tube-side water. No equations, but like I said, the equations are the easy part.

[Steve McGahey]

I've got no offering on the heat transfer side of this - as everyone says, it's complex, and going to be solved numerically.

As for the pressure drop with two phase flow, try to find the "Masonellian valve sizing handbook". This is available online (I think it's free).

There's a pile of equations inside this which apply to two phase flow and pressure drops in valves... perhaps there's something similar in there for flow through pipes! (Or, in a fix, perhaps you can find a Cv-equivalent conductivity (inverse of resistance) for your pipework)

[Zauberberg]

Masoneilan "Control valve sizing handbook" is available (free download) at:

http://www.dresser.c...ish/OZ1000E.pdf

[AA Mishra]

Sub cooled liquid is not leaving.

0.2 dissolved gas is not condensible.

Regards

[Chris Haslego]

In most cases, an application like this is going to be done in two (2) separate heat exchangers:

One to handle the desuperheating and condensing
One to handle the subcooling

There are exceptions of course, but it's extremely difficult to come up with a shell and tube design that can desuperheat, condense, and then form turbulent flow for good sensible heat transfer as well. Again, it's possible, but very difficult to design and rarely attempted.